American Diabetes Association 81st Scientific Sessions

June 25-29, 2021; Virtual; Full Report

Executive Highlights

  • Boy was there a strong sense of community in this year’s ADA Sessions. A whopping 11,611 people attended 191 educational sessions with over 900 faculty. Nearly 1,100 ePosters and 1,489 abstracts were accepted, 36 industry symposia shared updates on innovative drugs and technology, and 54 exhibitors were showcased in the Exhibit Hall. Given this massive amount of learning, it’s so beneficial that attendees can access all the content through September 28, 2021.

  • It was another data-rich year on the diabetes therapy front. A number of major ADA sessions focused on providing a closer look at cutting-edge, late-stage candidates like Lilly’s dual GLP-1/GIP receptor agonist tirzepatide, Lexicon’s SGLT-1/2 inhibitor sotagliflozin, and high dose semaglutide for obesity (2.4 mg) and type 2 diabetes (2.0 mg). While trials for all four of these candidates had already released topline results, full result presentations provided a chance for KOLs to debate the clinical implications and usages of each. Our major takeaway? Novel therapies are raising the bar for A1c and weight loss, and KOLs expect these short-term benefits to translate to impressive CV and renal risk reduction down the line. Ex-Sanofi GLP-1 efpeglenatide also made a splash at this year’s meeting, debuting positive phase 3 monotherapy trial results and, more importantly, CVOT results in AMPLITUDE-O. Lastly, we caught a first look at results from the NIH-funded GRADE study – the largest and longest study comparing the effectiveness of type 2 diabetes drugs, supporting GLP-1s and insulin over SUs and DPP-4 inhibitors. As these novel therapies gain approval, we look forward to seeing how their addition to the rapidly expanding diabetes therapeutic portfolio, as well as information from studies like GRADE, will continue to evolve care – upwards and onwards!

  • Interest in CGM remained as high as ever and (by far) our most read highlight from our day-of coverage was Dr. Amy Criego’s demonstration of International Diabetes Center’s successful integration of Abbott LibreView with Epic EHR. For many HCPs, EHR integration with CGM has been a sort of “holy grail,” and we’re hopeful that IDC’s model could be widely utilized across other diabetes and primary care clinics. Elsewhere, data continued to pour-in on the utility of CGM with both type 1s and type 2s. An interesting cost-effectiveness analysis showed adding FreeStyle Libre to Medicaid formularies would be essentially neutral from a cost perspective and help to avoid thousands of acute events per year. Given the significant evidence base for CGM at this point, we were also excited to look beyond CGM as a tool simply for insulin dosing and as a tool that enables virtual visits, an improved understanding of sleep and metabolism in people without diabetes, and extraction of new glycemic details in new drug studies.

  • The AID pipeline remains very active, headlined by soon-expected clearance of Insulet’s Omnipod 5 system. At ADA, we saw positive results from the hybrid closed loop in a preschool-aged pivotal trial, as well as confirmatory extension phase results from the children and adult pivotal studies. Real-world data from MiniMed 770G showed a mean Time in Range of 77%, not quite the outcomes achieved by MiniMed 780G in the real-world in Europe, but still meeting consensus targets. Out of the UK, CamAPS FX data showed strong results, bringing Time in Range from 48% to 63% in children and young adult type 1s. Out of the real-world, Control-IQ’s post-clearance study CLIO showed strong A1c improvements from Control-IQ across all ethnic groups. On the insulin delivery front, there was also a lot to be excited about with infusion sets. In two late-breaking posters (100-LB and 101-LB), Medtronic presented highly-anticipated results from the US pivotal trial (n=259) for its Extended-Wear Infusion Set (EWIS), showing that EWIS performed well with a survival rate at seven days of 75%. We also saw a feasibility study of Capillary Bio’s SteadiSet infusion set (701-P) showing 88% survival out to seven days (n=41 sets).

In this report, we provide our full coverage the ADA’s 81st Scientific Sessions. Titles highlighted in blue represent new coverage that wasn’t included in our daily highlights. Note that some talks may appear in multiple sections.

Our sections proceed as follows (you can navigate through these using our table of contents below):

  • Themes

  • SGLT Inhibitors

  • GLP-1 Agonists

  • Insulin Therapy

  • Novel Therapies

  • Type 1 “Cure” Therapies and Pathophysiology

  • Adjunct Treatments for Type 1 Diabetes

  • Treatment Algorithms, Strategies, and Guidelines

  • Glucose Monitoring – BGM and CGM

  • Automated Insulin Delivery, Pumps, and Pens

  • Digital Health, Telemedicine, and Decision Support

  • Beyond A1c and Hypoglycemia

  • Obesity

  • Prediabetes, Diabetes Remission, and Diabetes Prevention

  • Diabetes Complications

  • Policy, Epidemiology, and Education

  • NAFLD/NASH

  • Award Lectures and Additional Topics

  • Behavioral and Psychosocial Highlights

  • TCOYD and The diaTrive Foundation’s 14th Annual Diabetes Forum

  • Exhibit Hall

Table of Contents 

Themes

Diabetes Therapy

1. All in On Incretins: Tirzepatide and High Dose Semaglutide Raise the Bar for Diabetes and Obesity Therapeutics

While SGLT-2 inhibitors took center stage at ADA 2020, we were most struck at this year’s Scientific Sessions by the unprecedented A1c and weight loss efficacy seen with ‘next-gen’ incretin mimetics, particularly in combination and high dose formulations. Results from four different SURPASS studies of dual GLP-1/GIP receptor agonist tirzepatide, presented on Day #5 of the conference, demonstrated A1c reductions between 2.07 and 2.59% at the 15 mg dose. “Even more excitingly,” as Dr. Daniel Drucker (University of Toronto) said, tirzepatide delivered substantial weight reductions of -11% to -13.9% from baseline, also at the 15 mg dose. Of note, these trials were not specifically conducted for weight management, so we expect to see even greater reductions from the phase 3 SURMOUNT program in overweight/obesity. Speaking of weight management, Novo Nordisk’s high dose GLP-1 semaglutide also wowed in its 2.4 mg formulation. Semaglutide 2.4 mg delivered between 15 to 17% weight loss in the phase 3 STEP program for overweight/obesity, leading Day #2 of ADA 2021. Semaglutide 2.4 mg also conferred substantial cardiometabolic benefits in STEP 1, with a dramatic reduction in the percentage of patients with pre-diabetes over the course of the trial (from 45% to 16%), alongside a meaningful increase in normoglycemic patients (from 55% to 84%). During the session on STEP, Dr. Julio Rosenstock offered a powerful call-to-action to use semaglutide 2.4 mg as a first line therapy in people with obesity and new onset diabetes, in place of metformin. Following these stellar results, the SURPASS symposium featured a lively debate on preference between tirzepatide and semaglutide 2.4 mg, with KOLs calling both drugs “excellent molecules.” Notably, topline results for both these clinical programs had been released prior to the meeting, but we greatly appreciated the chance for KOLs to delve into the clinical implications of the data. Audience interest seemed equally apparent, given ample Q&A at both sessions.

  • A number of important incretin-related sessions also explored options that we imagine may play a role in specific populations of people with type 2 diabetes requiring treatment intensification. On Day #2, in SUSTAIN FORTE, semaglutide 2.0 mg strongly outperformed its predecessor semaglutide 1.0 mg in people with type 2 diabetes. Participants who took the higher dose as intended saw a significant 2.2% reduction in A1c from 8.9% at baseline vs. 1.9% with the lower dose. Soliqua, Sanofi’s GLP-1/basal insulin combination, emerged as a better alternative to premixed insulin for people with poorly controlled type 2 diabetes who are not able to meet their A1c goals on standard basal insulin/OAD therapy. In the SoliMix trial, Soliqua improved glucose control without weight gain or elevated hypoglycemia vs. traditional premixed insulin. Given that Soliqua circumvents two of the most common concerns about initiating or intensifying insulin therapy, we hope that all patients with poorly controlled type 2 diabetes will be able to access Soliqua to achieve their A1c goals. Be sure to catch our interview with Sanofi’s Vice President & US Medical Head of Diabetes Dr. Luigi Meneghini for more insight on Soliqua as well.

2. NIH-Funded GRADE Study Reflects Challenges and Triumphs of Comparative Effectiveness Studies

We’re always excited about data on the newest molecules in development; however, we’d be remiss not to take a moment to reflect on the significance of the NIH-funded GRADE study, which had its first readout at this year’s meeting. GRADE, led by renowned clinical trialist Dr. David Nathan (HMS), was a landmark study for a myriad of reasons; amongst them, the fact that it (i) was the largest (n=4,076) and longest study (five years average follow-up) comparing the effectiveness of type 2 diabetes drugs; (ii) sought to answer a highly clinically relevant question (“what is the best drug to take after metformin?”); (iii) enrolled a relatively diverse population (~20% of participants were Black, 4% Asian, and 3% were Native American/Alaskan Native; 18% were Hispanic/Latino); and (iv) was designed and conducted completely independently of drug manufacturers (though companies generously donated medications for the trial). Primary results from GRADE, which were presented on Day #4 of ADA 2021, showed that liraglutide and insulin glargine were most effective at keeping A1cs <7.0%, followed by glimepiride, then sitagliptin. On CV outcomes (which were deemed “preliminary,” as ~10% events had yet to be adjudicated), rates of “any CVD” – consisting of MACE, hospitalized HF, unstable angina, TIA, or revascularization – were lower with liraglutide (5.8% of participants developing “any CVD”), followed by insulin glargine (7.6%), glimepiride (8.0%), then sitagliptin (8.6%).

  • While these results provide some evidence as to which drugs are most effective, further analyses will be needed to best apply these findings to specific individuals. Big picture, GRADE study investigators hope to use these results to help “personalize” the use of glucose lowering drugs for type 2 diabetes. As Dr. Nathan said, “We want to know, for example, if drug X gives you more diarrhea and drug Y more hypoglycemia, and if you are African American maybe drug Z is better… We tend to treat type 2 diabetes as if it’s all the same, and it’s not. It’s about time we started finding out how these different drugs work in different people.” GRADE independent commentator Prof. David Matthews (University of Oxford) however mused that there is “very little data from GRADE” to “guide individualization of therapy, i.e., personalized medicine,” given so much variability within the allocation groups (i.e., a person who does better on the “worse” drug actually may have better results than a person who does worse on the “better” drug). This reflects a key challenge of comparative effectiveness studies – ambiguity of study results. In the future, however, further subgroup analyses by demographic, genetic, and physiological characteristics will help shed more light on this area. Another major drawback to GRADE’s study design was the exclusion of an SGLT-2 inhibitor (because SGLT-2 inhibitors had not yet been approved at the time of study design), rendering the data, in some KOLs’ opinions, obsolete. This represents yet another challenge of comparative effectiveness research – studies that collect much needed data on long-term CV or renal outcomes often will not be able to keep up with the latest in drug development. All in all, we find that GRADE does a fantastic job of representing both the triumphs and challenges of this kind of research, and we commend the GRADE investigators for pursuing what we believe to be a very worthwhile project.   

  • GRADE was the most prominent example of a comparative effectiveness trial at this year’s ADA; however, discussion of the importance of head-to-head studies emerged at a number of other sessions. In particular, presenters from both the SOLOIST/SCORED CVOTs for SGLT-1/2 inhibitor sotagliflozin and phase 3 SURPASS program for Lilly’s dual GLP-1/GIP receptor agonist tirzepatide highlighted that head-to-head trials would be helpful in determining the relative contributions of SGLT-1 inhibition and GIP receptor agonism (in the case of sotagliflozin, against an SGLT-2 inhibitor, and for tirzepatide, against a GLP-1). While we have yet to hear of any sotagliflozin vs. SGLT-2 inhibitor trials, we’ll get a chance to see data from the latter from the SURPASS-CVOT, an impressive head-to-head against dulaglutide 1.5 mg.

3. Efpeglenatide CVOT Confirms CV Efficacy of Exendin-Based GLP-1s

While “next-generation” incretins like tirzepatide (and even high dose semaglutide) stole the show, data from the AMPLITUDE-O CVOT helped clarify an important point of contention in the field – namely, that “traditional” exendin-based GLP-1s do indeed provide the same CV benefits as human-based GLP-1s like liraglutide (Victoza), dulaglutide (Trulicity), and semaglutide (Ozempic). In AMPLITUDE-O (n=4,076), exendin-based efpeglenatide (4 or 6 mg) met its primary endpoint of non-inferiority on 3P-MACE vs. placebo (p<0.0001), and furthermore, was found to deliver a 27% RRR (HR=0.73; 95% CI: 0.58-0.92, p=0.0069). As a reminder, all GLP-1s currently available have been engineered from one of two proteins – exenadin-4, found in the saliva of the Gila monster, or human GLP-1. Up until now, evidence on the CV benefits of exendin-based GLP-1s has been somewhat mixed: AZ’s exenatide narrowly missed superiority in the EXSCEL CVOT on 3P-MACE (HR=0.91; 95% CI: 0.83-1.00, p<0.001 for non-inferiority, p=0.06 for superiority), which thought leaders have broadly attributed to the trial’s pragmatic design (no run-in period, concomitant medication use, use of complicated and unwieldy reconstitution kits). In addition, results from the ELIXA CVOT for Sanofi’s Lyxumia (lixisenatide) were, as we called them at the time, resoundingly neutral, giving a hazard ratio for expanded 4P-MACE of 1.02 (95% CI: 0.89-1.17). A new sub-group interaction analysis of all GLP-1 CVOTs, presented by Dr. Naveed Sattar (University of Glasgow), however, largely put this controversy to rest – when including AMPLITUDE-O, the p-interaction between human GLP-1 homology vs. non-human GLP-1 homology on MACE jumped from 0.06 (suggesting a difference) to a very non-significant 0.39. As such, all GLP-1s give a relative risk reduction of ~15% on MACE, and this, as Dr. Sattar put it, “suggests strongly – almost completely definitively – that there is no difference in the benefits of GLP-1 based on human GLP-1 homology.”

  • Big picture, we believe broad insights on GLP-1 in terms of the cardio-protection as well as weight loss as well as glycemic-dependent nature could have big effects on getting more GLP-1 to those who would benefit from it. At present, while none of the GLP-1 therapies are generic, whether exendin-based or human GLP-1-based, in the future perhaps those that are given with greater frequency will become less expensive and perhaps major retailers like Target or others. With these results, we see ample rationale for bringing more “traditional” GLP-1s to a broader audience, given uniform CV benefits. While speculative, we wonder if a company like Walmart – which recently launched its own private label insulin analog (insulin aspart), has long had a highly discounted generic drug program, and serves millions of people with diabetes – may be interested in partnering with a company already manufacturing a GLP-1 to make it available to many more people (this may be possible given that unlike SGLT-2s, this class is not very homogenous) – we hope that perhaps a GLP-1 that is not selling as well and does not have as many advantages but is weight neutral and glycemic-dependent, may be brought to more people. The EXSCEL-CVOT trial is also particularly positive given simultaneous results from a phase 3 trial evaluating AZ’s exenatide in youth with type 2 diabetes, also presented at ADA 2021. We'll be curious to see if GLP-1 action so early on in diabetes development also impacts, or potentially staves off, long-term CV outcomes; of course, we’d also hope to see ample safety data if GLP-1 were given from such an early age. Up next, we’re keeping a close eye on Korea-based Hanmi to see if the drug developer decides to commercialize efpeglenatide (either on its own or through a licensing deal), discard development plans à la GSK’s Tanzeum (albiglutide), or wait to launch efpeglenatide in combination with another incretin hormone.

4. Heart Failure and Chronic Kidney Disease Fields March Forward with Nuanced Data on SGLT Inhibitors and Finerenone

Time and again, SGLT inhibitors continue to show impressive CV and renal outcomes. We were excited to see more data from the SCORED and SOLOIST-WHF trials, which investigated dual SGLT-1/2 inhibitor sotagliflozin in people with type 2 diabetes. As a reminder, SOLOIST-WHF showed a 33% risk reduction (p=0.0009) for the primary outcome of total CV death, HHF, and urgent HF visit in 1,222 patients admitted to the hospital for worsening heart failure. The SCORED trial (n=10,584) focused on patients with type 2 diabetes and an eGFR ≥25 and ≤60 and found a 26% risk reduction (p=0.0004) in total CV death, HHF, and urgent HF visits. The pooled data from the two trials found similarly positive results on the same primary outcomes (HR=0.72, p=0.000002). This trial shows CV and renal class effects of SGLT-2 inhibitors extend to more vulnerable patient populations, including patients with HFpEF, for whom there are no existing medications (though Lilly/BI’s SGLT-2 inhibitor Jardiance looks like it will soon be the first to gain approval). Of course, one of the biggest questions remaining is whether SGLT-1 inhibition truly has an additive benefit compared to the broad array of SGLT-2 inhibitors on the market. Moving to standalone SGLT-2 inhibitors, AZ’s Farxiga (dapagliflozin) also garnered significant interest at this year’s meeting for its potential in staving off diabetes in people with CKD. A new pre-specified analysis from DAPA-CKD found a higher frequency of new onset type 2 diabetes in the subgroup randomized to placebo (2.4/100 patient-years) vs. those randomized to dapagliflozin (1.5/100 patient-years). Although this result just missed the threshold of statistical significance (HR=0.62, 95% CI: 0.36-1.08), it corroborates a similar finding from DAPA-HF in which the SGLT-2 inhibitor gave a 32% relative risk reduction for new onset type 2 diabetes in people with HFrEF.

  • On top of SGLT inhibitors, our ears were also abuzz with discussion of Bayer’s novel, non-steroidal MRA finerenone; we expect that this important new compound will make its way into CKD treatment guidelines shortly and we’ll stay tuned on diabetes guidelines – certainly there has now been a lot published, including a standout NEJM piece in late 2020 “The Effects of Finerenone on CKD Outcomes in Type 2 Diabetes. Previously, results from the FIDELIO-DKD trial in diabetic kidney disease, presented at ASN 2020, showed a significant 18% relative risk reduction for adverse kidney outcomes (HR=0.82, p=0.0014). Moreover, finerenone reduced risk of a secondary CV endpoint by 14% (HR=0.86, p=0.0339). Notably, a late-breaking poster at the 81st Scientific Sessions showed efficacy of finerenone even in the absence of an SGLT-2 inhibitor, suggesting additive renal benefits when both drugs are used in combination. Additionally, another secondary subgroup analysis of finerenone showed the drug to have kidney and CV benefits regardless of A1c levels at baseline. In a Bayer-sponsored symposium dedicated to renoprotection, KOLs remarked that the development of finerenone marks a new era in which dialysis is no longer the primary method for treating severe CKD. We’ll be keeping an eye out to see how HCPs may decide to combine different drug classes to treat CKD and we’ll be watching out for results from the upcoming FINEARTS-HF trial, which Dr. Nasrien Ibrahim (Inova Medical Group) discussed as a potential treatment for HFpEF using finerenone.

5. New Therapies for T2D and Adjuncts for T1D Focus on Weight Loss and Early Signs of Renal Protection in Addition to Glycemic Control

We also caught various data readouts on novel therapy candidates for both type 1 and type 2 diabetes, ranging from promising pre-clinical poster presentations to more nuanced studies of newly-approved therapies on the market. Down the road, we’ll be excited to see whether these candidates make it to later stage trials – we remember, of course, when finerenone, as one example, was just a gleam in someone’s eye and when we got to work with diaTribe on letting people know about how to enroll. Interestingly, we noted that a substantial number of new studies used CGM metrics like Time in Range to provide more granularity on improved glycemic control (this trend was also apparent in type 1 “cure” studies). On top of “gold standard” A1c changes, studies explored candidates’ effects on weight loss, renal protection, glycemic variability, ease-of-use, and more. To name a few: in phase 2 studies, AZ’s dual GLP-1/glucagon receptor agonist cotadutide showed positive results on glycemic control, weight loss, and renal protection in type 2 diabetes and CKD or overweight/obesity. Hua Medicine’s phase 3 trial of glucokinase activator dorzagliatin showed its potential for improving beta cell function along with glycemic control in type 2 diabetes. Another China-based company, PegBio, with whom we’re not familiar, also debuted positive glycemic control data for its own glucokinase activator PB201, though only at the phase 1 stage. On the type 1 adjuncts front, REMD Biotherapeutics’ glucagon receptor agonist antibody volagidemab showed positive effects on glycemic control via A1c and CGM data. See below for a more detailed look at these candidates and more.

  • AZ’s GLP-1/Glucagon receptor agonist cotadutide. Results for phase 2 trials investigating cotadutide supported the candidate’s ability to improve glycemic control while promoting weight loss and renal protection. In a phase 2a trial in people with type 2 diabetes, stage 3 CKD, and a BMI of 25-45 kg/m2, receiving insulin and/or oral glucose-lowering drugs, treatment with cotadutide conferred a -27% change in plasma glucose AUC0-4h (p=0.001). CGM was used throughout the trial, and treatment with cotadutide led to +14.8% (+3.5 hours/day) Time in Range compared to -21% Time in Range for placebo as a secondary outcome (p<0.001). Though these results are impressive, the sample was fairly small with an enrollment of 41 participants. Treatment with cotadutide in this study was also associated with meaningful reductions in body weight (-3.69% from baseline vs. -0.21% with placebo, p<0.001) and in total daily insulin dose in subjects receiving ≥20 U baseline dose (35.2% reduction in dose for cotadutide; n = 14; p = 0.012). Early signs of renal protection were also shown with cotadutide treatment reducing UACR by 51% in patients with baseline micro- or macroalbuminuria versus placebo (p=0.0504) – though again sample size was limited at n=18.

    • These results were corroborated by another exploratory analysis in people with type 2 diabetes and overweight/obesity. Though cotadutide is primarily being developed for treatment of NASH (positive phase 2 data was presented at ADA 2020), the positive weight loss effects presented at EASD 2018 served as the rationale for this study that was performed to determine if cotadutide’s weight loss effects were the result of reduced energy intake, increased energy expenditure, or both. Individuals treated with cotadutide (n=18) lost significantly more weight than those on placebo (n=7) and consumed ~41% less during ad libitum lunch compared to placebo (p=0.011 for both analyses). However, treatment with cotadutide did not lead to significantly different total energy expenditure or resting energy expenditure compared to placebo, and though total energy expenditure decreased, cotadutide still led to weight loss. Preservation of resting energy expenditure may suggest glucagon receptor agonism. 

  • REMD Biotherapeutics’ glucagon receptor agonist antibody volagidemab. Positive phase 2 results, presented by Dr. Jeremy Pettus (UCSD), showed that adjunctive treatment with volagidemab improved glycemic control for people with type 1 diabetes. Glycemic control was measured by A1c and CGM data, and treatment with volagidemab led to significantly more participants achieving A1c <7% and reducing total daily insulin dose. Volagidemab also increased Time in Range by 6-7%, translating to ~1.5 hours/day, driven by reductions in hyperglycemia without increased rates of hypoglycemia. Without any reported serious adverse events in the trial, volagidemab represents a potential new adjunct therapy for type 1 diabetes that, through modulating glucagon signaling, can improve glycemic control moderately and without the unintended side effects (mainly severe hypoglycemia) of some very rapid acting insulins or other potential therapies like SGLT-2 inhibitors. REMD intends to advance volagidemab to phase 3 based on these results, with primary endpoints of change in baseline A1c and secondary endpoints of insulin use, CGM metrics, blood pressure, and comprehensive MRI liver fat measurements to assess NAFLD/NASH-related outcomes.

  • Zealand’s next-generation emergency glucagon dasiglucagon. Dasiglucagon was launched in the US under the trade name Zegalogue at the beginning of ADA, shortly following its FDA approval in March 2021. Results of a small study comparing the speed and success rate of dasiglucagon autoinjector administration to a traditional glucagon emergency kit in caregivers and adults with no diabetes experience showed higher success and ease of use for dasiglucagon. After the activity, participants also preferred using dasiglucagon, marking early positive reception for the brand, with 100% of the “bystander” group (untrained adults) and 94% of caregivers preferring the autoinjector and rating it as “easier and less stressful to use” compared to the traditional kit. This trial is reminiscent of high-fidelity manikin test for administering Lilly’s Baqsimi at AADE 2019, showing the progress in very easy to use rescue glucagon options since 2019 (Baqsimi, Zegalogue, and Xeris’ Gvoke). Next year, we’re looking forward to watching dasiglucagon transition from our “novel therapies” category into an established “diabetes complications” treatment.

  • Hua Medicine’s glucokinase activator dorzagliatin. Full results from dorzagliatin’s phase 3 SEED trial showcased the candidate’s promise as a new type 2 diabetes therapy treatment. Following positive topline results presented at ADA 2020, which showed a 1.1% reduction in A1c (from baseline of ~8.4%) at 24 weeks compared to an 0.5% reduction on placebo (p<0.0001), new data from this year’s ADA showed stable and sustained glycemic control over the trial period of 52 weeks. Reduced A1c and improved glucose tolerance were apparent as early as week four, but later treatment (halfway through the study) still led to A1c reductions below 7%, as seen in those who received dorzagliatin at the start of the study. Beyond glycemic control, dorzagliatin was able to confer improvement in beta cell function, as shown by significantly increased by HOMA 2-β versus placebo from week 24 to 52 (+3.28%, p<0.05).

  • PegBio’s glucokinase activator PB201. PegBio reported results from its phase 1 trial of PB201, which sought to assess safety, tolerability, and the PK/PD profile of PB201 as monotherapy in people with type 2 diabetes. Results showed that PB201 increased its clearance rate (AUC) and blood concentration (Cmax) in a dose-dependent manner, with dose-dependent effects on glycemic control. The highest dose (100/100 mg) conferred the highest TIR (63%), lowest A1c (7.10%), largest reduction in fasting plasma glucose (-1.47 mmol/L), and largest reduction in post-prandial glucose (-1.88 mmol/L). In this small sample (n=16), no hypoglycemia or adverse events were reported, supporting PB201’s linear PK properties, safety, and efficacy.

  • Also in their first in-human studies, Lilly debuted positive tolerability and dose finding results for its highly anticipated "triple G" GIP/GLP-1/glucagon receptor agonist candidate LY3437943 and PK/PD data for oxyntomodulin analog LY3305677, positioning the drugs for further study in type 2 diabetes, obesity, and NASH.

Diabetes Technology

1. CGM as a Platform: We Know About Glycemic Improvements and AID, But What Can We Learn In Virtual Visits, Sleep, and Even New Drug Therapies?

CGM has been the story at ADA (and in diabetes technology, more generally) for several years now. At ADA 2018, we wrote about Dexcom G6, Abbott FreeStyle Libre, Medtronic Guardian Connect, and Senseonics Eversense all making their ADA debuts as we called for CGM to be the standard of care. One year later, we highlighted studies demonstrating the glycemic benefits of CGM in the often-understudied populations of elderly adults, young children, and teenagers/adolescents. By ADA 2020, we were focused on why paying for CGM is not only economically feasible, but preferable to SMBG in many groups. Of course, diabetes technology at ADA 2019 and 2020 were also dominated by pivotal results from Tandem’s Control-IQ and Medtronic’s MiniMed 780G AID systems in respective years – both technologies are driven and enabled by CGM. In the last few years, it’s become clear that CGM is not just a device that can improve glycemic control in people on mealtime insulin. Instead, CGM is a platform that enables a better understanding of diabetes, patient behavior, and much more.

  • Throughout COVID-19, we continued to hear about the largely successful deployment of telemedicine and virtual care for people with diabetes. Perhaps underappreciated, however, is the role of CGM in helping to make the transition to telemedicine so successful. Indeed, it’s hard to imagine that diabetes providers and patients would have been able to move to a virtual care world ten years ago. Without the ability to use lab-measured A1c values, many providers reported relying on CGM-metrics such as GMI or Time in Range to assess glycemic control. For patients with reliable and fast internet, the ability to view AGP reports over video with providers was also extremely valuable. Of course, reliably downloading data from devices (both CGMs and connected BGMs) remained a common pain point, as referenced by Dr. Leslie Eiland on Sunday. Encouragingly, the next morning, Dr. Amy Criego demonstrated the successful Epic EHR integration between Abbott LibreView at International Diabetes Center, a model that will hopefully be used throughout diabetes care clinics. Altogether, most presentations we’ve seen, including one at ADA, seem to suggest glycemic control in people with diabetes at a population level did not worsen during the pandemic, and in many cases even improved slightly (e.g., ATTD 2021). This is a remarkable testament to power of CGM, telemedicine, and the resiliency and creativity shown by so many providers and patients throughout the pandemic.

  • In a compelling Saturday afternoon presentation, Dexcom data scientist Sarah Pickus presented results from a real-world observational study looking at the link between glucose and sleep. Participants without diabetes wore Dexcom G6 to measure glucose and Fitbit Charge 3 to measure sleep for ten days. Even though the participants did not have diabetes, each additional hour of sleep was associated with a 3.1 mg/dl decrease in mean daytime glucose. Of course, sleep is a critical component of health for people with diabetes (see Adam Brown’s article in diaTribe), but the study showed that CGM may indeed have value in people without diabetes in assessing overall health, a vision that many startups are trying to realize (e.g., Levels, January.ai, Supersapiens).

  • On Sunday, both our tech and therapy teams were stoked to see CGM data from Lilly’s phase 2 trial for its once-weekly basal insulin-FC. Over the past few years, we’ve heard a few calls to use CGM more in drug trials (see Keystone 2019, for example), but this is one of the few times we’ve actually seen CGM results from a trial. Of course, in a study on insulin, using CGM provides particular value in being able to measure hypoglycemia. In the study, Lilly’s basal insulin-FC significantly reduced the relative risk of total hypoglycemia by 30% and also conferred a 45% RRR on nocturnal hypoglycemia vs. degludec.

2. Outcomes for MDI vs. AID Stand in Stark Contrast: What Is Needed to Support Injection Users?

For a few years now, we’ve referred to smart pens and insulin dosing titration systems as a “nascent market.” Give all the activity in the area in the past year, it’s hard to continue calling the market “nascent.” In August 2020, Medtronic acquired the only FDA-cleared smart pen system in the US, Companion Medical’s InPen. In February, Lilly signed an agreement with Welldoc, announcing its intentions to submit its Lilly Tempo Button connected pen attachment and a mobile app to the FDA in 2021. A month later, Novo Nordisk officially launched its long-awaited connected NovoPen 6 and Echo Plus pens. Two months after that, Bigfoot received FDA clearance for its first-gen Bigfoot Unity Program, which was launched the day before ADA officially began. With all that activity around smart pens, we’d circled Dr. Bob Vigersky’s (Medtronic) presentation on real-world InPen data on our schedules before ADA started.

  • On Sunday afternoon, Dr. Vigersky presented results from a pre-post analysis of InPen users with poor glycemic control at baseline. In general, InPen delivered slight improvements in both GMI and Time in Range, with A1c typically improving by 0.1% and Time in Range by ~30 minutes/day for those with baseline GMI >8%. The analysis only included users with baseline GMI >8%, resulting in data from 423 real-world users. Following three months of InPen use, mean GMI fell slightly from 8.7% to 8.6%. Looking at Time in Range, mean Time in Range increased from 31% at baseline to 33% following three months of InPen, an improvement of 33 minutes/day. This improvement came from a reduction in time >180 mg/dl from 68% to 66%. Still, considering the massive amount of time this group of patients spent above range, it’s concerning to see that InPen initiation was only associated with a small improvement. To us, the study really highlighted the difficulty in MDI, and intensive insulin therapy more generally. In addition to the glycemic data, Dr. Vigersky also noted that with use of InPen, the average number of boluses delivered per day decreased from 3.7 to 3.6. This slight reduction in bolus frequency was offset by an increase in average dose amount. During the first month, the average bolus delivered was 7.1 units, compared to 7.7 units during the third month. We’re curious to see how longer-term use of InPen or other connected pen systems might change patient behaviors around insulin administration, particularly given the improvements demonstrated in this study were fairly modest. Despite the very concerning Time in Range and hyperglycemia results reported in this population, we relished seeing the real-world evidence (RWE) from these MDI users and hope to see more publications around MDI users in the future with more smart pens on the market.

  • The data on MDI users stood in fairly stark contrast with the continued roll of positive AID results in new populations, across age groups, in clinical studies, and in the real-world. Of course, it’s been a big spring for Insulet’s Omnipod 5 hybrid closed loop system leading up to ADA, and ADA continued that momentum. Saturday was the day of Insulet at ADA with positive results from Omnipod 5’s preschool-aged pivotal trial (+2.6 hours/day Time in Range), confirmatory extension phrase results from the primary children and adult pivotal studies, and encouraging patient-reported data from children, teens, and their caregivers in the pivotal study. Elsewhere, real-world data from Medtronic’s MiniMed 770G system showed a mean Time in Range of 71%, not quite the outcomes achieved by MiniMed 780G in the real-world in Europe, but still meeting consensus targets. Out of the UK, CamAPS FX data showed strong results, bringing Time in Range from 48% to 63% in children and young adult type 1s. Out of the real-world, Control-IQ’s post-clearance study CLIO showed strong A1c improvements from Control-IQ across all ethnic groups.

3. What’s Next in AID? Bi-Hormonal Systems, Infusion Set Technology, and More

This year’s ADA featured under-the-radar presentations on AID technology in the pipeline. This included presentations and posters on bi-hormonal systems, infusion sets, and user burden-reducing technologies that will drive the next generation of insulin delivery technology.

  • Several posters showcased the next generation of infusion sets, including those from Medtronic and Capillary Bio. In two late-breaking posters (100-LB and 101-LB), Medtronic presented highly-anticipated results from the US pivotal trial (n=259) for its Extended-Wear Infusion Set (EWIS), showing that EWIS performed well with a survival rate at seven days of 75%. The pivotal trial for EWIS completed back in November and as of May, the device has already been submitted to the FDA. In Europe, EWIS has been CE-Marked since February 2020 and a limited launch of the device began in Finland in November. Medtronic also read out preclinical porcine model data (692-P) on the safety and efficacy of its combined insulin fusion and glucose sensing device, which the poster referred to as “Duo.” The study found that Duo had a 67% seven-day survival rate (even better as the EWIS prototype in porcine models (64%)) and performed well as a glucose sensor. We also saw a feasibility study of Capillary Bio’s SteadiSet infusion set (701-P) showing 88% survival out to seven days (n=41 sets). We were excited to see the attention devoted to infusion sets because they are a major trouble spot for many AID system users and are sometimes overlooked as important spot for system improvement and reduction of patient burden.

  • Bi-hormonal AID systems in the pipeline also received much attention at this year’s sessions. Dr. Luz Castellanos (Massachusetts General Hospital) read out data from a short cross-over RCT (216-OR; n=20) showing that the bi-hormonal iLet AID system achieved +1.7 hours/day in range (77%) compared to the insulin-only system at 110 mg/dl while also reducing hypoglycemia. The study also compared the insulin-only and and bi-hormonal iLet system when configured to a 130 mg/dl setpoint target and showed impressive glycemic outcomes for both glucose targets. In the same session, we caught a read-out of a small, crossover inpatient RCT (215-OR; n=13) on DiaCon’s dual-hormone AID system, which uses a Dexcom G6 CGM, two Dana RS pump, and a locked Android that houses the algorithm as a controller. The small study found that the dual-hormone system achieved 69% Time in Range (+1.9 hours/day) in challenging, inpatient conditions. While still early-stage data, this is stirring data. As a reminder, in the US, the farthest-along bi-hormonal system is Beta Bionics’ bi-hormonal iLet system with Zealand’s dasiglucagon, which is slated for a pivotal trial in 2H21.

  • In addition to its impressive infusion set data, Medtronic shared news of additional AID-related projects in its pipeline. During Friday’s Medtronic Product Theater, VP of R&D Ali Dianaty referenced Medtronic’s work on bringing smartphone bolus capability to the MiniMed 700 series systems, something we heard for the first time at ATTD 2021. He also shared more details on Medtronic’s plan to bring smart bolus reminders to Medtronic’s AID users, which use the Klue Apple Watch and gesture-based tracking tool, which Medtronic acquired in December 2019, to automatically detect when a person is eating and remind the user to administer a bolus dose. During a session on the intersection of insulin delivery technology and nutrition, Dr. Bruce Buckingham (Stanford) shared data showing that by recognizing eating motions, Klue can detect meals 18-20 minute earlier than a blood glucose value-based algorithm. This holds important implications for Klue’s integration into Medtronic’s AID algorithms. As Dr. Buckingham argued, if Klue’s meal detection alerts were fed into an AID algorithm, that algorithm could then begin to administer insulin ahead of a rise in glucose, potentially limiting post-prandial hyperglycemia. Of course, this is mere conjecture at this point, but it would be an exciting advancement that could reduce the importance of meal announcements and pre-meal boluses to achieving glycemic control.

4. Digital Health Has a Blockbuster Year in Funding and Enrollment, But Where Are the Outcomes?

Throughout the COVID-19 pandemic, consumer, provider, and investor interest in digital health solutions skyrocketed. This momentum continued (at least from a funding perspective) well into 2021, as highlighted by the $14.7 billion in venture capital invested in American digital health companies in just the first six months of 2021. Major deals in 2021 so far include an eye-popping $540 million Series F raised by Noom, a $133 million Series E raised by Virta, $110 million Series D raised by Vida Health, $30 million Series D raised by Glooko, and exits from Steady Health (via acquisition by Carbon Health) and Better Therapeutics (via SPAC). on its 4Q20 update in February that it was launching its own corporate venture arm, Dexcom Ventures. Of course, Dexcom Ventures will be looking at sensing technology startups, but we also expect the fund to focus in on data analytics and remote monitoring companies with relevance to Dexcom’s core offerings.

Patient adoption of digital health tools also boomed over the past year. In May, Bloomberg reported that Noom’s revenue “reached $400 million last year [2020], up from $200 million in 2019.” If we assume Noom averages $35 per member per month, a conservative $400 million estimate for 2020 revenue would back out to a user base of right around one million. In 1Q21, Teladoc reported 658,000 total chronic care members, a figure which we believe represents Livongo members. That compares to ~328,000 Livongo for Diabetes members reported in 1Q20. Elsewhere, we picked up on bits and pieces of information from private digital health companies. Virta Health and Vida Health reported revenues growing at “nearly 200% YOY” and “more than [tripling]” in April and May, respectively.

  • Given all the interest around digital health in 2020 and 2021, we were somewhat surprised mobile apps weren’t a bigger topic at this year’s ADA. For some context, we counted 78 abstracts this year in our digital health category – as a note, that also includes telemedicine. That represents the same number of abstracts at ADA 2020 and a general decline from the digital health abstracts in the past few years (75 abstracts in 2019, 82 in 2018, and 95 in 2017). Given the audience of ADA (primarily diabetes HCPs), in our view, it’s possible that digital health companies are choosing to share data on their interventions in other settings.

  • Dr. Kasia Lipska (Yale University) made a similar point during a debate on the value of digital coaching apps, stating: “There’s a lot of money involved, huge revenues. These companies are worth billions of dollars. So what’s the evidence that what they’re doing works? Now, I’m going to disappoint you all, because there isn’t a lot.” Further, Dr. Lipska criticized the observational nature of most of the data that is available on digital coaching apps: “When you have no control group, you really don’t know the true treatment effect.” Dr. Anne Peters (USC) took the affirmative side of the argument and focused in on the immense potential for these digital solutions to personalize care and improve access. However, on the access point, Dr. Peters also noted that some apps can actually make it “harder for some of [her] under-resourced patients,” as they may not have the health and tech literacy required to engage with these programs. On that side, we were encouraged by results from a customized digital diabetes education program developed for older, Chinese American immigrants; however, that presentation also underscored the amount of personalization needed to achieve positive results in these groups. Dr. Peters also called for digital health programs to integrate better with the existing healthcare system (e.g., EHRs), so that providers could better monitor their effect.

  • For its part, we did see quite a bit of promising outcomes data from the Onduo virtual diabetes clinic. In one retrospective study, Onduo participants with type 2 diabetes offered Dexcom G6 saw a 0.7 percentage point improvement from 7.9% to 7.2% after one year (n=354). For Onduo participants who didn’t use Dexcom G6, A1c improvements were more modest, improving from 7.6% to 7.4% after one year (n=418). In another poster, results showed that short-term participation drove A1c improvements that were sustained even when participation was discontinued. This is an encouraging sign, as the durability of member engagement in digital programs remains a somewhat open question.

5. CGM Evidence Galore: Type 2s and Cost-Effectiveness

ADA 2021 built on increasing momentum to expand diabetes technology access and use to type 2s and to highlight the cost-effectiveness of CGM as a tool with a broad use case. Across the board, data on CGM in type 2s was a major highlight of ADA 2021 both in presentations, in posters, and in symposiums. It was exciting to see the breadth of data on CGM in type 2s spanning from efficacy in non-insulin-requiring type 2s to the cost-effectiveness of FreeStyle Libre and Dexcom G6 in type 2s. As we look to the second half of 2021, we expect that this momentum will only grow and move us closer to expanded use of CGM in type 2s.

  • Between sessions, posters, and oral presentations, a plethora of data was read out in support of CGM in type 2s. Highlighting this slew of evidence were data from a six-month prospective, single-arm study that assessed the effect of Dexcom G6 on clinical outcomes in type 2s on basal-only or non-insulin therapy (600-P; n=38). The study found that Dexcom G6 drove an A1c drop from 10.1% to 7.1% and that Time in Range increased +3.6 hours/day to 72% at six months. In a Dexcom Symposium, Dr. Bergenstal dug into evidence showing that while both intermittent and continuous use of Dexcom G6 resulted in A1c reductions and declines in hypoglycemia, type 2s see significant improvements with continuous use as compared to the intermittent use (-0.7 percentage points vs. -0.4 percentage points A1c reductions). To us, this reinforces what is clear to anyone with diabetes that has been able to use CGM – that there is significant benefit to persistent use over intermittent use in type 2s on MDI. That said, there certainly isn’t always, as large databases show – to learn more about this through dQ&A, write Richard Wood, our sister organization’s CEO.

  • The poster hall featured a trove of real-world data on CGM in type 2s. Real-world retrospective EHR analysis (67-LB) showed that type 2s see double the six-month A1c reduction that type 1s saw (-0.73 percentage points -0.79 percentage points vs. -0.35 percentage points -0.37 percentage points) after initiating Abbott’s FreeStyle Libre or Dexcom G5 and G6. Elsewhere in the poster hall, a large retrospective data claims analysis (n=82,983 type 2s; 83% on basal-only or no insulin; 77-LB) found significant A1c reductions for CGM (-0.18 percentage points) and rtCGM (-0.31 percentage points) users compared to BGM users. Another EMR analysis (64-LB) found that type 2s on basal-only therapy saw a -1.4% A1c reduction from 9.4% to 9.0% after three to six months of initiating FreeStyle Libre. Given Dr. Grace’s study, the MOBILE study read out at ATTD 2021 and that was contextualized during ADA 2021, the real-world Kaiser study published in JAMA, and these many posters and real-world studies, we are hopeful that the sudden rush of evidence in support of CGM in type 2s will drive insurance coverage changes and expanded use of CGM in type 2s. 

  • In another vein, several major posters provided data in favor of CGM’s real-world cost-effectiveness in type 1s and type 2s. In an exciting poster (136-LB), Abbott estimated that adding FreeStyle Libre coverage to all people with type 1 or type 2 diabetes on Medicaid would be effectively cost-neutral and that increased FreeStyle adoption would actually drive ~$19 million in annual cost savings. Another poster (136-LB) that included Medicaid and CHIP budget analysis showed that switching from BGM to FreeStyle Libre would save -$327/type 1 patient, -$101/insulin-requiring type 2 patient, and -$323/child or adolescent with diabetes. Those are huge cost-savings for people with diabetes and for the medical system as a whole! Looking at Dexcom G6 use, another claims data analysis poster (66-LB) found that in type 2s on Medicare (n=571), diabetes-related medical care costs decreased -$424/person/month after initiating Dexcom G6. Such savings are likely due to reductions in DKA and all-cause hospitalization, which Dr. Irl Hirsch (University of Washington) argued would save a lot of money in an Abbott symposium and in a poster (68-LB). Of course, the huge range of potential savings is striking, ranging from taking less insulin to needing less ER care or even HCP care or getting more exercise, which may be associated with less insulin use – there are so many areas of greater efficiency and productivity and better care.
  • Across the Atlantic, Swedish registry data (135-LB) found that FreeStyle Libre use was associated with an incremental cost-utility ratio of $24,917/QALY and $16,421/QALY in insulin-requiring type 2s with baseline 8%-9% and >9% A1c groups, respectively, both of which were significantly below the Swedish willingness-to-pay threshold. This data is notable for several reasons: (i) it reinforces the cost-effectiveness of CGM not only as a monitoring tool but as connected to therapeutic decision making and other interventions; (ii) it supports the cost-effectiveness of both FreeStyle Libre and Dexcom G6; and (iii) it highlights the cost-effectiveness of CGM in type 2s rather than solely type 1s.

Big Picture

1. Guidelines Galore! ADA/EASD Debut Draft Type 1 Diabetes in Adults Consensus Report; Key Sessions Spotlight Latest Recommendations on CKD and HF

Various professional organizations shared updates to their guidelines based on new scientific evidence from the studies performed in the past year, particularly for type 1 diabetes, chronic kidney disease, and heart failure.

  • As a new release, the ADA/EASD shared their 2021 draft consensus report on management of type 1 diabetes in adults, focusing on the report’s central goal to provide essential information on type 1 diabetes care to primary care physicians, given that the majority of people with type 1 are treated in that setting. Importantly, the session kicked off a ten-day comment period for patients, HCPs, and others to provide feedback, reflecting the organizations’ focus on broad stakeholder involvement. While previous guidelines were based on broad, non-specific guidelines, new recommendations were put into the updated report in a way that mirrors the 2018 ADA/EASD consensus report on type 2 diabetes management. The guidelines outline recommendations at the point of diagnoses, scheduling care, self-management education, psychosocial self-care, disease monitoring, selecting insulin or adjunctive therapies, addressing hypoglycemia, DKA, pancreatic transplantation, special populations, and novel therapies. Information on managing CV and microvascular risk factors were not included in the report, as committee members felt these topics were addressed comprehensively in other guidelines. These guidelines will hopefully help rapidly advance the standard of care for type 1 diabetes. For example, CGM was strongly recommended as the glycemic monitoring method of choice, further highlighting the importance of integrating technology into type 1 care for best patient outcomes. For insulin administration, those on the committee believe that ideal insulin replacement is best achieved with either multiple daily injections or pump therapy and that insulin analogues are the “treatment of choice” for type 1 diabetes. The report even goes so far as to recommend that HCPs should support patients who may be using off-label DIY loop systems, recognizing the benefit these systems can promote.

  • Kidney Disease: Improving Global Outcomes (KDIGO) task force also presented its 2020 Guidelines, which aligned greatly with the ADA’s recommendations for people with diabetes and CKD. Diet and exercise remain foundational, with metformin and SGLT-2 inhibitors recommended as first-line therapies. A1c remains the preferred measure for glycemic control, as eGFR greater than or equal to 30 mL/min/1.73m2 does not affect A1c measurements. However, CGM is recommended for people with advanced CKD and those who may be on dialysis; given CGM’s accuracy at all stages of CKD, we’ll be curious to see if later editions the guidelines elevate CGM to first-line blood glucose monitoring like in the ADA’s type 1 diabetes recommendations. If people do not meet glycemic targets or can’t take SGLT-2 inhibitors or metformin, KDIGO recommends starting a low-dose long-acting GLP-1 (preferred), DPP-4 inhibitor, SU, TZD, insulin, or alpha-glucosidase inhibitor. Looking ahead, we want to see if GLP-1s are further elevated in care, pending results from the FLOW trial for Novo Nordisk’s GLP-1 semaglutide in type 2 diabetes and CKD. The KDIGO guidelines also addressed that cost, injection burden, and dialysis may hinder patients from taking SGLT-2 inhibitors or GLP-1s, and while we appreciate the variety of suggested medications, we are curious if the organization will release a more specific hierarchy of recommendations for low-cost alternatives to give patients and HCPs more guidance. Beyond glycemic control, the KDIGO guidelines also stress aggressive kidney disease treatment through diet, exercise, and intensive use of ACEs/ARBs. Novel, non-steroidal MRA finerenone was notably missing from the 2020 guidelines (given that the drug’s trials had yet to readout), and we expect to see more information on how finerenone can be most effectively combined with other treatment options in the future.  More broadly, KDIGO also recommends use of the Chronic Care Model to create a team-based approach to kidney disease management.
  • Though not officially representing a professional organization at ADA, Dr. Biykem Bozkurt (Baylor) also weighed in on the newest insights for treatment of HF, heavily citing her work on the 2021 consensus report “Universal Definition and Classification of Heart Failure” (supported by the Heart Failure Society of America, Heart Failure Association of the European Society of Cardiology, Japanese Heart Failure Society, and Writing Committee of the Universal Definition of Heart Failure). Promisingly, heart failure treatment is shifting toward earlier treatment and prevention, especially in people with diabetes, hypertension, coronary artery disease, or obesity classified as “at risk for heart failure” – in large part fueled by recent data on SGLT-2 inhibitors. Beyond lifestyle changes, implementing an SGLT-2 inhibitor can lower risk for progression of HF – as shown in the hHF risk reductions reported in the EMPA-REGCANVASDECLARE, and VERTIS CV CVOTs. Other analyses have shown that biomarkers like NT-proBNP and hsTnT can be implemented into treatment algorithms to predict HF development and SGLT-2 inhibitor efficacy. Beyond use of SGLT-2 inhibitors alone (and not included in the Universal Definition report), Dr. Vanita Aroda (Brigham and Women’s) also posited in the same session that SGLT-2 inhibitors and GLP-1 receptor agonists may have complementary, non-overlapping clinical and physiologic benefits. Though there is not yet evidence from a head-to-head clinical trial to support this claim, existing CVOT data supports non-over-lapping clinical benefits of stroke for GLP-1s and hospitalization of heart failure and hard renal endpoints for SGLT-2 inhibitors (both drugs show benefits on MACE, MI, CV death, and broad renal endpoints). As Dr. Mikhail Kosiborod stated in the past, dual use of ACE inhibitors and beta blockers in HFrEF did not require a dedicated CVOT – which may also be the case for GLP-1s and SGLT-2s. As summarized by Dr. Aroda, GLP-1s seem to primarily act on the atherosclerotic and inflammatory pathways, while SGLT-2 inhibitors appear to modulate myocardial energetics and cardio-renal pathways.

2. Cell-Based Type 1 “Cures” Move Beyond Animal Models to First In-Human Studies; Future of Immunotherapies More Precarious Than Ever

While data readouts on the type 1 immunotherapy front were far sparser at this year’s ADA (compared to data from the TN-10 study of teplizumab, T1GER trial for golimumab, and phase 2 trial of Novo Nordisk’s anti-IL-6/liraglutide combination at ADA 2020), we caught two key updates on the beta cell replacement front at this year’s meeting. Based on early phase 2 evidence from one participant, ViaCyte’s stem cell-derived islet replacement therapy PEC-Direct (VC-02) produced endogenous insulin, as demonstrated by increases in glucose-responsive C-peptide levels, increased Time in Range, and reduced A1c. Importantly, this was the first example of implanted pancreatic progenitor cells producing endogenous insulin in a patient – a major step forward for the scientific field. We also heard from University of Chicago’s Dr. Piotr Witkowski who presented promising data on Sernova’s Cell Pouch in people with type 1 diabetes and hypoglycemia unawareness (using human donor-derived islets). Six out of seven participants showed successful Sernova Cell Pouch implantation and tolerance, and the most advanced patient who completed the trial protocol has been insulin independent for 14 months with a A1c of 4.8%. While both ViaCyte’s and Sernova’s results are encouraging, the findings are based on small sample sizes, so we are interested to see future readouts and potential variability between patients. Related to beta cell replacement therapy, a five-year follow-up from a trial of CXCR1/2 inhibitor reparixin suggested that it could be an effective immunosuppressant for patients undergoing islet transplantation (50% of patients in the reparixin and 25% in the placebo groups achieved insulin independence).

  • Forward momentum on type 1 immunotherapies has seemed to slow over the past year, however, especially given teplizumab’s recent CRL and discontinued programs from both J&J and Novo Nordisk on golimumab and anti-IL-21/liraglutide, respectively. Debate on immunotherapies and their trials emerged throughout various sessions. As a timely topic, there was a lively debate as to whether immunotherapies should even be tested in clinical trials while the COVID-19 pandemic ensues, with both sides offering solid points. During another symposium on immunotherapy, Dr. Stephen Gitelman (UCSF) emphasized that clinicians should not offer immunotherapies to patients outside of clinical trials due to a lack of demonstrated “robust, durable effect” in clinical trials. As per usual, calls for longer-term and combination immunotherapy trials were abundant – while there has not been visible traction gained in this area over the past year, we believe there is a great deal happening behind the scenes.

  • How to best stratify patients for type 1 diabetes immunotherapy remains a major area of investment and substantial developments could further ignite the field. Building on impressive follow-up data from last year and ADA 2019, preliminary data from a post-hoc analysis of the TN-10 Teplizumab trial suggested that participants’ six-month proinsulin:C-peptide (PI:C) ratios could predict the rate of progression of type 1 diabetes and responsivity to teplizumab treatment.

3. Pediatrics Update: Amid Rising Rates of Type 2 and High Therapeutic Failure, Newer Drug Classes Offer Promising Benefits

Epidemiological data from the SEARCH for Diabetes in Youth study highlighted a sobering trend of rising prevalence and incidence of pediatric type 2 diabetes. In especially concerning news, the incidence of type 2 in youth nearly tripled during the COVID-19 pandemic, with Black youth accounting for 77% of the new-onset type 2s. These findings reflect widening health disparities among racial groups, as over 90% of youth with type 2 come from minority backgrounds. The SEARCH study also found that poor glycemic control remained high among non-white youth diagnosed with type 1 between 2002 and 2016. Given that pediatric patients experience high rates of cardiovascular, retinal, and neurological complications compared to adults, there is dire need for earlier intervention and further research into therapy, drugs, and prevention.

  • RISE and TODAY studies point to high rates of therapeutic failure in pediatric population, suggesting that more effective treatments are needed to target diabetes’ especially aggressive presentation in youth. As a reminder, the RISE study revealed increased insulin resistance and hyper-responsive beta cells among youth with impaired glucose tolerance or newly diagnosed type 2, compared to adults of the same status. The TODAY study found that nearly half (46%) of youth type 2 participants experienced therapeutic failure early in their disease course. Indeed, the RISE study demonstrated that neither metformin alone nor metformin followed by insulin therapy was sufficient to combat the intense insulin resistance in pediatric patients. These alarming results led Dr. Caprio to boldly assert that these drugs are simply not strong enough to treat pediatric type 2s.

  • GLP-1s, SGLT-2 inhibitors, and inhaled insulin offer a glimmer of hope for pediatric patients, but much more research is needed. Results from a phase 3 trial of AZ’s GLP-1 exenatide in youth with type 2 diabetes showed efficacy in reducing A1c (p=0.012) and trended toward impacting secondary metabolic markers, but did not reach statistical significance, presumably due to small size and short trial duration. Likewise, during a session on SEARCH, Dr. Elvira Isganaitis (Joslin Diabetes Center) explained that studies of DPP-4 inhibitors and SGLT-2 inhibitors have found promising results in youth. Currently, injectable insulin, metformin, and liraglutide are the only FDA-approved medications for pediatric type 2, so future approvals for any of these drugs would mark an important expansion in the market. 

    We see far more use of inhaled insulin for people with T2D and believe that use will continue to expand as it is better understood.

    • On this note, in welcome news for children who struggle with injectable insulin, Mannkind’s phase 2 trial data demonstrated that Afrezza inhaled insulin produces pharmacokinetic results in pediatric patients with type 1 diabetes that mirror those found in adults; the inhaled insulin was also found to be generally safe and tolerable in the younger population.

SGLT Inhibitors

DARE-19 Symposium: Dapagliflozin Safe in Patients Hospitalized with COVID-19; Trend Toward Benefit on Organ Failure + Death Raises Question of a Future in Critical Illness

Highly anticipated full results from DARE-19, AstraZeneca’s trial of SGLT-2 inhibitor Farxiga (dapagliflozin) in patients hospitalized with COVID-19, were presented in an afternoon symposium on Sunday. Dapagliflozin was found to be very safe in patients hospitalized with COVID-19 and, while not superior to placebo, there was a numerical trend toward benefit with the SGLT-2. Given the initial questions surrounding safety of continuing SGLT-2s in patients hospitalized with COVID-19, we see these results as extremely positive. But perhaps most interestingly, DARE-19 raises questions about the potential of SGLT-2s to prevent organ damage in a wide range of critical illness, which the study investigators are eager to see explored in larger, broader, and better-powered trials. As a reminder, topline results for the trial were presented in April 2021 then detailed in full at ACC 2021.

  • Rationale, Design, and Baseline Characteristics: DARE-19 enrolled 1,250 patients across seven countries hospitalized with COVID-19 for ≤4 days, with O2 saturation ≥94% on ≤5 L/min, chest x-ray consistent with COVID-19, and ≥1 of hypertension, type 2 diabetes, ASCVD, HF, or CKD. Patients were excluded for critical illness, eGFR <25, type 1 diabetes, and prior DKA. Participants were randomized 1:1 to dapagliflozin 10 mg/day or placebo for a 30-day treatment period or until discharge. 99% of patients in each group completed the study.

    • Dr. Mikhail Kosiborod explained the polarized views on SGLT-2s and COVID-19. The virus causes a systemic illness that disturbs metabolism to cause widespread organ damage, and patients with cardiometabolic risk factors have compromised organ function at baseline that increases risk for further tissue damage with infection. SGLT-2s have been shown to favorably affect many of the deleterious processes of COVID-19, leading to the hypothesis behind DARE-19: dapagliflozin may confer protection from organ failure and death, and improve recovery, in the setting of COVID-19 hospitalization with cardiometabolic risk factors. However, he later noted, some professional and expert groups initially recommended the cessation of SGLT-2s upon admission for COVID-19 due to the possible risk of dehydration, DKA, and AKI. Indeed, the DARE-19 trial itself was the subject of some criticism when it began.

    • DARE-19 had two primary endpoints. The first was a “Prevention” endpoint of time to first major clinical event, including respiratory or cardiac decompensation, arrhythmia, doubling of creatinine, initiation of renal replacement therapy, or death. The second was a “Recovery” endpoint, which considered how patients changed from baseline to Day 30: improvement (discharge, hospitalized without oxygen), stable (hospitalized with oxygen), or deterioration (high-flow oxygen, organ dysfunction, or death). Of note, the second endpoint was elevated to a primary endpoint following the sharp decline in total admissions and mortality due to COVID-19 a couple months into the pandemic, as physicians became better at managing the illness.

    • At baseline, mean age was ~62 years old, 43% were female, and average baseline systolic blood pressure and oxygen saturation were normal. About 51% of participants had type 2 diabetes, 7% heart failure, 85% hypertension, 16% ASCVD, and 7% CKD. Additionally, about 30% were receiving systemic steroids, 18% remdesivir, 36% insulin, 35% an ACEi/ARB, and 22% statins.

  • Efficacy Outcomes: While there was a strong trend toward benefit with dapagliflozin on both primary endpoints, neither achieved statistical significance. Dr. Otávio Berwanger presented primary outcome data: On the first primary outcome of Prevention, there was no significant effect on time to organ failure or death with dapagliflozin (HR=0.80, 95% CI: 0.58-1.10, p=0.168). Event rates were 13.8% with placebo and 11.2% with dapagliflozin. There was no difference by diabetes status (p-interaction=0.668). On the second primary outcome of Recovery, a hierarchical composite endpoint, the “Win ratio” was 1.09 (95% CI: 0.97-1.22, p=0.14) (note: >1 favors dapagliflozin on this measure). The p-interaction for diabetes status was 0.222.

    • On the composite kidney endpoint of AKI, initiation of RRT, and all-cause death, there was also a non-significant trend toward benefit with dapagliflozin (HR=0.74, 95% CI: 0.50-1.07, p-interaction=0.739). The same pattern followed for all-cause mortality alone (HR=0.77, 95% CI: 0.52-1.16, p-interaction=0.817).

  • Safety Outcomes: There were numerically fewer serious adverse events with dapagliflozin compared to placebo, including less AKI (doubling of creatinine or AKI after discharge). Dr. Subodh Verma explained that only two cases of DKA occurred in patients with diabetes treated with dapagliflozin; both were identified through protocol lab monitoring and rapidly resolved. Over the course of the study, there were no significant changes in serum bicarb, eGFR, or hematocrit with dapagliflozin compared to placebo, regardless of diabetes status. DKA was monitored via active surveillance in all participants with type 2 diabetes via daily acid-base monitoring during hospitalization; treatment was interrupted for an increase in anion gap or decrease in bicarb.

  • Clinical Implications: Dr. Kosiborod emphasized that dapagliflozin was very well-tolerated in this extremely high-risk group, concluding that SGLT-2s should not be discontinued in the setting of COVID-19. Indeed, he argued that doing so would put patients at long-term risk for discontinuation, as evidence shows patients often do not restart medications after they are discontinued in the hospital. Further, given the rapid onset of clinical benefit for patients with HFrEF who start SGLT-2s, he argued there’s risk of losing that benefit just as rapidly. The risk of discontinuation, therefore, is technically unknown but potentially large. However, he explained during Q&A, there isn’t any evidence to support SGLT-2 initiation in a patient with COVID-19, especially without an indication for the drug (i.e., diabetes, heart failure, CKD). On a different front, the research implications of DARE-19 are significant, as the trial furthered the hypothesis that SGLT-2s may be organ-protective in a wide range of acute illness, including sepsis. He pointed to evidence from DAPA-CKD showing a significant benefit on non-cardiovascular death (HR=0.54, 95% CI: 0.36-0.81), comprised of infection, malignancy, and renal failure. Finally, he anticipated further data from the TACTIC-E study, and called for further study in larger trials outside of COVID-19.

  • Independent Commentary: Finally, Dr. Carolyn S.P. Lam provided the independent commentary on DARE-19, which was very positive. She praised the investigators for executing such a successful trial during a pandemic (when most trials were disrupted). In particular, Dr. Lam commended the flexibility of elevating the hierarchical composite endpoint to a dual primary endpoint when incidence of organ failure and death plummeted to 1/3 of that anticipated (from 34% to about 10%). Given this, however, there’s certainly value in conducting further study of SGLT-2s in acute illness. She also emphasized how daring (pun well-intended) the trial was in the first place, given concern about AKI and DKA in a vulnerable patient population, and was extremely positive about the study’s safety findings. If anything, she said, dapagliflozin was associated with lower risk of AKI, and the two DKA events were detected biochemically (not clinically), noting that the rate of DKA in DARE-19 was on par with prior SGLT-2 CVOTs.

Experts Sound Off on SGLT-2s/GLP-1s for Hospitalized Patients: Prescribe ASAP or Prescribe at Discharge?

Should SGLT-2 inhibitors and GLP-1 agonists be initiated in patients hospitalized for a CV condition? Drs. Sandeep Das (UT Southwestern, Dallas, TX) and Jennifer Green (Duke University, Durham, NC) debated this important question, though the experts agreed as much as they disagreed. Each acknowledged the impressive cardioprotective effects of SGLT-2s and GLP-1s. Dr. Das summarized data from the major CVOTs, pointing to significant risk reduction for three-point MACE with both therapy classes and to the pronounced heart failure benefit of SGLT-2s. He presented a meta-analysis of DAPA-HF (AZ’s Farxiga) and EMPEROR-Reduced (Lilly/BI’s Jardiance) showing a 13% drop in all-cause death and a 31% drop in heart failure hospitalization with an SGLT-2 vs. placebo – results that “changed the face of systolic heart failure care.” Dr. Das further highlighted the meaningful weight loss associated with GLP-1 agonists. In STEP 1, participants taking Novo Nordisk’s semaglutide (Ozempic) lost 12.4% more body weight compared to placebo-treated patients over 68 weeks – Dr. Das called this “absolutely stunning for a non-surgical option.” Dr. Green, who is on the ADA’s Standards of Care committee, underscored that SGLT-2s and GLP-1s are now recommended for type 2 diabetes patients facing high CV risk, which means they are indicated for approximately one-third of type 2s in primary care. Even though Dr. Green ultimately argued against initiating these treatments in an inpatient setting, she advocated for accelerated uptake overall and lamented the under-use to-date. According to NHANES, only 1-2% of type 2s with ASCVD or multiple risk factors for atherosclerosis are prescribed a GLP-1 agonist, and only 7% are prescribed an SGLT-2 inhibitor. Only 6% of type 2s with CKD stage 3 are prescribed an SGLT-2, while a mere 9% of type 2s with established heart failure are prescribed an SGLT-2. These numbers are surprising and disappointing, and while short-term cost can certainly be one factor, Dr. Green pointed out that only 10-15% of commercially-insured type 2s with ASCVD, heart failure, or CKD were prescribed an SGLT-2 in 2019 (indicating that cost and coverage are not the standalone factors to blame although we recognize those with very high deductibles and low incomes are in a difficult position). Both speakers acknowledged that starting these treatments in the hospital could address vast under-utilization, although their opinions diverged somewhat on safety, efficacy, and timing.

  • Safety. Dr. Das argued that hospitalists should seize the opportunity of inpatient care to initiate an SGLT-2 or GLP-1 where indicated. He emphasized low hypoglycemia risk with both drug classes and turned to DARE-19 results for additional evidence of safety. DARE-19 explored dapagliflozin as a treatment option for COVID-19 inpatients, and the SGLT-2 demonstrated a compelling safety/tolerability profile in this particularly sick, high-risk population. Dr. Green also mentioned DARE-19 in her remarks but focused on the two episodes of DKA in the dapagliflozin arm (vs. zero in the placebo arm). We should note that in both instances, the DKA was mild. That said, Dr. Green explained that 38% of all SGLT-2-related DKA occurs in an inpatient setting. There are reasons to be particularly cautious in treating already-hospitalized patients, and the ADA currently recommends stopping an SGLT-2 regimen during severe illness, including during and around the time of a surgical procedure. Dr. Green also noted that side-effects can be especially uncomfortable and harmful for patients recovering from a procedure. For example, side-effects of GLP-1s include nausea, vomiting, and poor appetite, which can slow down wound healing and rehabilitation.

  • Efficacy. Patients are highest-risk for a recurrent MACE event in the days following their previous event. Given this, Dr. Das suggested that it’s only logical we leverage the cardioprotective benefits of SGLT-2s and GLP-1s in the inpatient cardiovascular care setting. He pointed to a recent study in JAMA Cardiology which found that dapagliflozin was most beneficial in heart failure patients hospitalized within the past 12 months. Dr. Green noted that across the major diabetes CVOTs, almost all participants were started on an advanced therapy in an outpatient setting. She argued that there is insufficient evidence at this point in time to universally recommend initiation of SGLT-2s and GLP-1s in the hospital. However, as more data accumulates, guidelines could change (indeed, we have been impressed by the dynamic nature of the ADA Standards of Care over the past few years).

  • Opportunity cost. Dr. Das recognized that GLP-1s are tricky to prescribe given that they require titration to reach an effective dose as well as patient counseling on side-effects and the importance of adherence. He added that they are even less likely to be started in an outpatient cardiology clinic than SGLT-2s; we have certainly picked up on a reluctance to prescribe injectable therapy at cardiology conferences (it remains to be seen how oral GLP-1 agonists might change this, over time). As such, the inpatient setting is a can’t-miss opportunity to initiate a new cardioprotective drug, provide injection training, offer counseling on side-effects and risk mitigation, and set patients up for success with medication adherence. Dr. Das put it best in his conclusion: “Delaying therapy has a cost – preventable heart attacks, strokes, heart failure complications, and death.” Dr. Green raised the issue of increasing costs with a longer hospital stay. While unable to endorse a universal recommendation to initiate SGLT-2s and GLP-1s as part of inpatient care, she advocated for a sound alternative: “Prescribe these medications at discharge, and make a meaningful difference in a patient’s chronic disease care before they leave.”

SOLOIST/SCORED Symposium Closes Out ADA: SGLT-1/2 Inhibitor Sotagliflozin Delivers Significant Early CV Benefit in Pooled Analysis; KOLs Ask for Head-to-Head vs. SGLT-2 Inhibitor to Clarify Role of SGLT-1

New analyses of the SOLOIST and SCORED studies now show that early use of SGLT-1/2 dual agonist sotagliflozin leads to cardio-renal benefits in people with type 2 and chronic kidney disease or worsening heart failure.

  • Dr. Lawrence Leiter (University of Toronto) and Dr. Deepak Bhatt (Brigham and Women’s Hospital) began the session with a detailed description of sotagliflozin; while SGLT-2 inhibitors have already been shown to beneficially impact glucose, CV, and renal outcomes, there is a growing body of literature suggesting SGLT-1 inhibition may also provide additive benefits. Of note, SGLT-1 reabsorbs glucose in the GI tract, meaning it also lowers plasma glucose independent of kidney function. Importantly, Drs. Birtram Pitt and Deepak Bhatt recently published an article on this very subject in Circulation: “Does SGLT1 Inhibition Add Benefit to SGLT2 Inhibition in Type 2 Diabetes Mellitus?” Dr. Leiter briefly discussed data from Lexicon’s inTandem1 and inTandem2 trials, which showed sustained A1c benefit for patients with type 1 using sotagliflozin. Trials of sotagliflozin in patients with type 2 and moderate to severe renal impairment have also shown safety and efficacy. The SOLOIST-WHF trial contributes to the literature by focusing on patients with recently worsening heart failure, and the SCORED trial studies patients with Stage 3b and Stage 4 CKD with or without albuminuria who have been previously underrepresented in other SGLT-2 inhibitor trials.

  • SOLOIST-WHF enrolled 1222 patients with type 2 diabetes and worsening heart failure, stratified by ejection fraction of <50% or ≥50%. Patients who had been admitted to hospital with signs and symptoms of heart failure and treated with intravenous diuretics but had been stabilized and off oxygen with a type 2 diagnosis met the main inclusion criteria for the study. The randomized, double-blind, placebo-controlled trial began in hospital or within 3 days of discharge. Patients were randomized between the once-daily, orally administered sotagliflozin 200 mg arm or placebo arm. The primary endpoint, which had to be adjusted due to COVID-19 and funding shortfalls, was the total number of deaths from CV causes and HHF/urgent visits for heart failure. Secondary endpoints included total number of hospitalizations and urgent visits for HF, incidence of death from cardiovascular causes, incidence of death from any cause, total number of deaths from CV causes, HHF, nonfatal MI, and non-fatal stroke, among others. Because the trial ended early, patients had a median follow-up duration of only nine months.

  • The SCORED trial included 10,584 patients with type 2 and an eGFR ≥25 and ≤60. Subjects were randomly assigned to the sotagliflozin 200 mg arm or the placebo in this double-blind trial. The primary endpoint was the total number of deaths from CV cause, HHF, and urgent visits for HF. Several secondary endpoints, including total number of HHF and urgent visits for HF, deaths from CV causes, total number of deaths from CV causes, HHF, nonfatal MI, and nonfatal strokes, among others. The mean follow-up duration was 16 months, but this study was also stopped early due to financial issues related to COVID.

  • Dr. Subodh Verma (St. Michael’s Hospital) and Dr. Mikhail Kosiborod (Saint Luke’s Mid America Heart Institute) began the results portion of the symposium with a review of data from the SOLOIST and SCORED trials. For reference, results of both trials were published in NEJM in early 2021, and the results of a pooled CV analysis were presented earlier this year at ACC 2021. The SOLOIST-WFH trial, which investigated dual SGLT-1/2 inhibitor sotagliflozin in people with type 2 diabetes admitted for heart failure (n=1,222), showed a 33% risk reduction for the primary outcome of total CV death, hospitalization for HF, and urgent HF visit (95% CI: 0.52-0.85; p=0.0009; number needed to treat: 4). Impressively, this effect was significant by day 28 of the study. Treatment with sotagliflozin conferred similar benefits in total CV death and HHF (HR=0.68; 95% CI: 0.53-0.88; p=0.003) and first CV death or HHF (HR=0.71; 95% CI: 0.56-0.89; p=0.003). Sotagliflozin treatment was associated with higher rates of diarrhea (p=0.032), venous thrombotic effects (p=0.015), and severe hypoglycemia (p=0.037), but was well-tolerated with consistent benefit among otherwise. Dr. Verma did state that the trial had limitations, including an early stop and primary endpoint change mid-trial from composite MACE (while the investigators were still blinded to any results) , early use of sotagliflozin led to significant cardio-protective effects.

  • The SCORED trial showcased similarly positive results in its enrolled population of people with type 2 with CV risk factors, eGFR of 25-60 mL/min/1.73m2, and A1c ≥7% (n=10,584). Treatment with sotagliflozin led to a 26% risk reduction in total CV death, HHF, and urgent HF visit (HR=0.74; 95% CI: 0.63-0.88; p=0.0004; number needed to treat: 54). This effect was also significant early in the trial, by day 95. As in SOLOIST, sotagliflozin significantly reduced risk of first of CV death or HHF (HR=0.77; 95% CI: 0.66-0.91; p=0.001), and reduced risk in first of CV death, non-fatal MI, or non-fatal stroke (HR=0.84; 95% CI: 0.72-0.99; p=0.035) and total CV death, non-fatal MI, or non-fatal stroke (HR=0.77; 95% CI: 0.65-0.91; p=0.002). Sotagliflozin’s effect on total CV death, non-fatal MI, or non-fatal stroke was significant by 94 days into the trial. This trial also showed higher number of adverse event rates for diarrhea (p<0.0001), volume depletion (p=0.003), genital mycotic infections (p<0.0001), and DKA (p=0.022) with sotagliflozin versus placebo. Of note, the benefits outlined above and additional significant reductions in A1c, systolic blood pressure, and weight (p<0.001) were similar across subgroups.


  • Dr. Deepak Bhatt (Brigham and Women’s Hospital) presented the results of a meta-analysis of the SOLOIST and SCORED trials to showcase sotagliflozin’s benefit in reducing total CV deaths, HHF, and urgent hospital visits across ejection fractions. As the goal of the SOLOIST trial was to address patient health in the vulnerable period after hospital admission for worsening HF, new findings from a Poisson regression show that people treated with sotagliflozin were alive and out of hospital for 2.9 additional years per 100 patient years compared to placebo (p=0.027). Additionally, a pooled analysis of SOLOIST and SCORED shows that treatment with sotagliflozin conferred a significant 28% RRR on total CV death, HHF, and urgent HF visit (HR=0.72; 95% CI: 0.63-0.82; p=0.000002; number needed to treat: 31). This hospitalization data was also just published in the Annals of Internal Medicine. In participants with HFrEF (EF <40%), the relative risk reduction for this outcome was 22% (HR=0.78; 95% CI: 0.63-0.96; p=0.02; number needed to treat: 11). This grew to 39% RRR (HR=0.61; 95% CI: 0.40-0.94; p=0.02; number needed to treat: 7) in patients with HFmrEF (EF 40% to <50%) and 37% RRR (HR=0.63; 95% CI: 0.45-0.89; p=0.009; number needed to treat: 9) in people with HFpEF (EF ≥50%). Impressively, sotagliflozin also conferred a 27% RRR (HR= 0.73; 95% CI: 0.54-0.99; p=0.04; number needed to treat: 121) in this outcome in people with no history of HF (EF ≥ 50%).

  • Dr. Julie Lewis (Vanderbilt University) delved deep into the cellular mechanisms of SGLT-1/2 before turning her attention to renal outcomes of the SCORED trial. Perhaps the most important takeaway was that inhibiting these cotransporters reduces glomerular filtration rate and pressure – preserving renal structure and function by decreasing glomerular hypertension. The endpoint for renal outcomes was a composite score, which included the first sustained ≥50% decrease in eGFR, chronic dialysis, renal transplant or sustained eGFR <15ml/min/1.73m2 . A benefit with sotagliflozin was not shown to be statistically significant (p-value: 0.11) but seemed to be directionally beneficial – from our view, you couldn’t really assess this since the trial ended so quickly. Indeed, there were very few renal events that occurred in either treatment group prior to the study’s abrupt conclusion. Dr. Lewis pointed out that with only 16 months of follow-up, it would have been nearly mathematically impossible for subjects to decrease eGFR ≥50%. Dr. Lewis also noted that DAPA-CKD, which showed a 39% RRR for kidney outcomes, would not have shown statistical significance if it had been stopped at 16 months. Moreover, the sotagliflozin did show the initial drop in eGFR characteristic of other SGLT-2 inhibitors, making it “a good prognostic sign of long-term renal protection.”

  • Dr. Javed Butler (University of Mississippi) provided an independent commentary of the SOLOIST-WHF and SCORED trials. Primarily, Dr. Butler wondered whether SGLT-1 inhibitors offer additional cardiovascular benefit and proposed a head-to-head trial comparing the dual inhibitor to an SGLT-2 inhibitor. He also suggested that the cardiovascular benefits from SGLT-2 inhibitors may be a class effect, but considering the class variability seen within DPP4 inhibitors and GLP1 receptor agonist, this cannot be assumed. Therefore it is best to use drugs that have been tested. Turning to HFpEF, Dr. Butler suggested that despite the small sample size of patients with preserved ejection fraction, he was optimistic about the results. However, he is also looking forward to the release of data from upcoming trials with empagliflozin and dapagliflozin in this patient population to see if they confirm the results. It’s worth noting that the ethics of the early conclusion of the study were mentioned throughout the session and were brought up again by Dr. Butler in his independent review. Dr. Butler concluded by stating, “Many questions answered, many questions raised.” We look forward to future research investigating dual SGLT-1/SGLT-2 inhibitors to answer some of these new questions.

Question and Answer

Q: What is the mechanism for improved CV outcomes?

Dr. Subodh Verma: I can say that there are many different direct and indirect mechanisms that have been postulated. The direct mechanisms include potential effects on ion channels in the myocardium, in particular interactions with the sodium hydrogen and the late in-warding sodium current. There have been effects that may also regulate mytophagy/autophagy, which are survival signals at the level of the myocyte, and effects on cam-kinase-2-delta, which regulates sarcoplasmic reticulum calcium uptake and release.

From an indirect standpoint, there have been effects that are renal-cardiac. There’s a dominant effect in all the trials with respect to improving renal functions, and there are hemodynamic and non-hemodynamic effects through which improving renal function can improve cardiac function. One of these includes the release of erythropoietin, which I know is controversial, but nonetheless has been seen in a trial Drs. Bhatt, Leiter and I have been involved with which showed that early EPO release was seen with these therapies. There are also the effects on beta hydroxybutyrate, making the energy-starved diabetes heart making more ATP production yielded through an energy efficient pathway through beta hydroxybutyrate. Finally, the effects on reducing sympathetic activation. There are big connections between kidneys and central sympathetic discharge, which can drive reductions in sympathetic outflow and explains why blood pressure goes down and cardio function improves with no reflex rise in heart rate, plasma-renin activity, or norepinephrine turnover. Whenever there’s an answer that goes through 12 things, there are a lot of options.

I read Dr. Bhatt’s paper with Prof. Pitt in Circulation. The mechanisms I’ve just mentioned are dominantly SGLT-2 focused, but there is whole biology of SGLT-1 and its expression in the vasculature that may lead to the MACE reduction seen.

Q: There have been a lot of questions about kidney issues. Is there any benefit on glycemic control or the development of micro/macro albuminuria?

Dr. Julia Lewis: The SCORED study did demonstrate effective blood sugar lowering in patients with GFRs <30. Since there’s a decrease in filtration, that would suggest a benefit in SGLT-1 that wouldn’t be apparent if given only an SGLT-2 inhibitor. SGLT-1 receptors are found in the s3 segment of the proximal tubule, and although it is not a high-capacity receptor for absorbing sodium and glucose, it is an important part of the proximal tubule particularly in AKI.

Q: There are lot of questions about safety in general in hospitalized patients. Is there anything you want to say about the DARE-19 trial in context of if it is okay to start or continue SGLT-2 inhibitors in patients who are hospitalized for any condition, assuming they are not at risk of death?

Dr. Mikhail Kosiborod: Yes, this was studied in DARE-19 Trial – which evaluated dapagliflozin in patients hospitalized with COVID-19; and safety data show that dapagliflozin was well tolerated in this patient population. Overall, the number of serious adverse events (including acute kidney injury) were numerically lower with dapagliflozin versus placebo. DARE-19 data suggest that the risk of continuing SGLT2i in patients that have existing indications for this class and are acutely hospitalized appears to be very small if as long as patients are monitored when on the therapy.

Q: For final thoughts, is there something here with sotagliflozin and SGLT-1 inhibition that goes above other available SGLT-2 inhibitors?

Dr. Lawrence Leiter: We don’t have head-to-head comparisons versus SGLT-2s, but as Dr. Lewis said, both in SCORED and other trials, sotagliflozin has advantage of lowering glucose even in patients with significant renal impartment. We continue to believe glucose control is important, especially in regard to microvascular complications. We also saw decreases in blood pressure and weight. Sotagliflozin has gotten approval in Europe for treatment of type 1 diabetes but has not been marketed yet. This hasn’t happened yet in North America, but in people with type 1, there is good data with sotagliflozin that show it leads to improvements in A1c, less glycemic variation, more Time in Range, and some weight loss. It definitely seems like this is a useful drug, but everything I’m saying is off label in North America.

New Findings from DAPA-CKD: AZ’s SGLT-2 Farxiga Gives Consistent Renal & CV Benefits across Subgroups with Normoglycemia, Prediabetes, Type 2 Diabetes

Steno’s Dr. Frederik Persson presented a prespecified secondary analysis of DAPA-CKD showing consistent benefits with dapagliflozin regardless of baseline glycemic status (no diabetes, prediabetes, or type 2 diabetes). Recall that AZ stopped this renal outcomes trial early due to overwhelming efficacy with the SGLT-2 inhibitor dapagliflozin (Farxiga). Full results read out at ESC 2020, in which dapagliflozin gave a whopping 39% relative risk reduction on the primary composite endpoint of sustained ≥50% eGFR decline, end-stage kidney disease, or renal/CV death (HR=0.61, 95% CI: 0.51-0.72, p<0.001). When looking at the data by glycemic status, all three subgroups saw significant risk reduction for this primary outcome with dapagliflozin vs. placebo: Hazard ratio was 0.62 in the subgroup with normoglycemia, 0.37 in the subgroup with prediabetes, and 0.64 in the subgroup with type 2 (p=0.19 for the interaction). Dr. Persson displayed a graph charting primary outcome results across a continuous spectrum of baseline A1c, and here too there was stable renal protection with Farxiga (p=0.62 for the interaction). He also broke down secondary endpoint data, highlighting dapagliflozin’s ~29% relative risk reduction for CV death/heart failure hospitalization which was consistent across subgroups (p=0.43 for the interaction) and ~31% relative risk reduction for all-cause death, again similar across subgroups (p=0.25 for the interaction). Turning to safety, Dr. Persson acknowledged some early concerns that SGLT-2 inhibitors might cause hypoglycemia in people without type 2 diabetes, but there were zero hypo events among participants with normoglycemia or prediabetes. Now that multiple studies have been published investigating SGLT-2s in patients without T2D (e.g., DAPA-CKD, DAPA-HF, EMPEROR-Reduced), we imagine this worry over hypoglycemia has largely subsided. Additionally, Dr. Persson underscored that no DAPA-treated patients in the trial experienced DKA. For more, read the detailed results from this analysis just published in Diabetes Care. Of note, dapagliflozin was approved for CKD by the FDA in April 2021 and received a positive CHMP opinion earlier this week.

  • Of the 4,304 patients enrolled in DAPA-CKD, 68% had type 2 diabetes, 15% had prediabetes, and 17% had normoglycemia at study start. Normoglycemia was defined by A1c <5.7%, prediabetes by A1c between 5.7-6.5%, and type 2 diabetes by historical diagnosis and/or baseline A1c ≥6.5%. As expected, individuals with type 2 diabetes were older and had a higher baseline BMI relative to the other subgroups.

  • Dr. Persson noted that there was no change in mean A1c over the course of the trial for any of the glycemic subgroups, and that there was no significant A1c drop with dapagliflozin vs. placebo. The implication is that glucose-lowering is not the mechanism by which dapagliflozin achieves renal protection (or cardio protection, for that matter). Interestingly, a poster at this meeting reports lower incidence of new-onset type 2 diabetes with dapagliflozin vs. placebo in DAPA-CKD – we’re curious how this squares with the finding that dapagliflozin did not affect A1c relative to placebo and as always, we’d like to know how Time in Range affected people in the DAPA-CKD study, as well as all outcomes trials. While we know that “it’s expensive” to use CGM in these trials, the wealth of information that comes from the fantastic Time in Range tool is enormous. The knowledge, for example, that there is less “time above range” even if the A1c doesn’t change because “time below range” also declines would be incredibly more valuable than just hearing “A1c didn’t change.” It may well be that TIR, TAR, and TBR were also similar whether dapagliflozin was used or placebo, and this seems like relevant information to have.

New DAPA-CKD Analysis Shows Lower Incidence of New-Onset Diabetes with SGLT-2 Dapagliflozin vs. Placebo; Raises Questions around Potential Prediabetes Indication

According to a pre-specified analysis of DAPA-CKD, AZ’s SGLT-2 inhibitor dapagliflozin (Farxiga) reduced the incidence of new-onset type 2 diabetes relative to placebo. Results were displayed on a poster, authored by Dr. Peter Rossing (Steno Diabetes Center) and other DAPA-CKD investigators. Of the 4,304 participants enrolled in DAPA-CKD, 32% or 1,398 had no history of diabetes and a baseline A1c <6.5%. The present analysis counted cases of incident T2D over the duration of the trial (median 2.4 years follow-up), and found a higher frequency in the subgroup randomized to placebo (2.4/100 patient-years) vs. those randomized to dapagliflozin (1.5/100 patient-years). Although this result just missed the threshold of statistical significance (HR=0.62, 95% CI: 0.36-1.08), it corroborates a similar finding from DAPA-HF in which the SGLT-2 inhibitor gave a 32% relative risk reduction for new-onset T2D. Moreover, Dr. Rossing noted that there was a statistically significant effect observed for women in DAPA-CKD, who saw 79% risk reduction for new-onset diabetes with dapagliflozin vs. placebo (HR=0.21, 95% CI: 0.06-0.71, p=0.03 for the interaction). Taken together, these subsequent analyses of DAPA outcomes trials suggest a potential role for dapagliflozin to play in treating prediabetes and preventing the progression to type 2 for individuals at high risk – particularly those with established heart failure and/or chronic kidney disease.

The SGLT-2 market has shown consistent growth (+24% YOY in 2020) and AZ’s Farxiga is rising most rapidly, with revenue up 54% YOY to $625 million in 1Q21. Given the high prevalence of prediabetes in the US and globally, we suspect this could be another key market for the SGLT-2 class if the agents continue to show efficacy in diabetes prevention. Clearly, there are major regulatory barriers to getting a therapy to market for prediabetes; specifically, in the US, since there is at present no therapeutic pathway a present for prevention, FDA presumably would be very unlikely to approve a pill that treats it even for a small subsection of people with prediabetes who also had major risk factors and quickly escalating A1c. While some express lots of hope about the potential for earlier intervention during type 2 diabetes progression, others would say, hope is not a strategy. Still, since earlier interventions would keep patients from unnecessary complications, lower productivity, etc., and since diabetes is hitting those who are poor and those in underserved populations at disproportionately higher rates, we hope that NIH is receiving more requests to analyze these important questions – certainly they were asking for research requests and proposals by researchers that themselves represent these populations. See related talk by Dr. Sherita Hill at ENDO in March, 2021.

  • Recall that AZ stopped the DAPA-CKD trial early due to “overwhelming efficacy,” meaning the drug was so beneficial in protecting against adverse renal outcomes that it was unethical to withhold it from placebo patients any longer. Dapagliflozin gave a 39% relative risk reduction for the primary composite kidney endpoint (HR=0.61, 95% CI: 0.51-0.72, p<0.001). Data from the CREDENCE trial (which was also stopped early due to high efficacy) indicates a renal protective effect of J&J’s SGLT-2 inhibitor canagliflozin (Invokana) as well. Indeed, we’ve heard consensus among thought leaders that renal and heart failure benefits are class effects of SGLT-2s, opening up two new markets for these products in addition to type 2 diabetes. While it’s very early to say at this point how long it may take for prediabetes to eventually become a viable indication and another new market for SGLT-2s (or GLP-1s), we’d certainly think it is possible to design a trial around very high-risk people with pre-diabetes and multiple risk factors. Since so many with pre-diabetes never get T2D, this would need, of course, to be a very specific population and very carefully designed. We imagine there would be more and more multi-stakeholder funding for this, particularly in the wake of COVID-19 and the clear belief that if people were at healthier weights and had less T2D, outcomes would’ve been better.

Dr. Muthiah Vaduganathan Concludes SGLT-2 Inhibitors Show Promise in Treating HFpEF But Will Face Challenges Showing “Disease-modifying” Potential in Reducing Mortality

Dr. Muthiah Vaduganathan (Brigham and Women’s Hospital) argues that SGLT-2 inhibitors show promise in HFpEF, and is looking forward to results from EMPEROR-Preserved and DELIVER filling key evidence gaps. EMPEROR-Preserved, with an expected read out later this year, is investigating the use of 10 mg once-daily empagliflozin versus placebo in patients with heart failure with a preserved ejection fraction (HFpEF) with and without type 2 diabetes. DELIVER, with an expected 2022 completion, is evaluating the use of 10 mg once daily dapagliflozin versus placebo in patients with HFpEF with and without type 2 diabetes. To provide context for what results we may expect from these trials, Dr. Vaduganathan highlighted a series of studies showing that SGLT-2 inhibitors are mechanistically promising in targeting key aspects of HFpEF pathogenesis. However, he acknowledged that it will be challenging to show SGLT-2 inhibitors’ “disease-modifying” potential in HFpEF mortality due to low overall mortality and high competing risk of non-CV causes of death. Still, he is enthusiastic because reductions in hospitalization burden and improvement in quality of life will be clinically relevant to patients, clinicians, and regulators.     

  • Dr. Vaduganathan cited a series of studies to argue that SGLT-2 inhibitors are mechanistically promising to treat HFpEF. Dr. Vaduganathan cited the EMPA-TROPISM, SUGAR-DM-HF, Empire HF, EMBRACE-HF, and CREDENCE studies to show that SGLT-2 inhibitors attenuate adverse left ventricular remodeling, reduce left ventricular mass, lower intracardiac filling pressures, and lower blood pressure. In particular, Dr. Vaduganathan explained that SGLT-2 inhibitors’ ability to reduce left ventricular mass is promising for HFpEF treatment since left ventricular mass may be a critical issue for patients with HFpEF, especially for patients with a hypertensive phenotype with left ventricular hypertrophy. 

  • While promising, SGLT-2 inhibitors face considerable challenges in showing mortality reduction in HFpEF. While CV death is common in HFpEF, Dr. Vaduganathan listed numerous studies identifying high competing risks for non-CV death; across his cited studies, 30% to 60% of deaths were for non-CV causes. In general, comorbidity (obesity, diabetes, chronic kidney disease, or chronic obstructive pulmonary disease) is becoming increasingly more common in HFpEF compared to HFrEF, potentially making it harder to achieve treatment benefits for HFpEF. Furthermore, at low left ventricular ejection fractions, cardiac risk factors dominate for CV death/hHF/all-cause death, but at higher ejection fractions the contributions of demographics (particularly, age and sex) and non-CV risk factors become more dominant. On top of the difficulty in showing mortality reduction, Dr. Vaduganathan believes that the high use of background therapies in HFpEF may raise the bar for standard of care and make incremental treatment benefits from investigational therapies more challenging.  

VERTIS CV Post-Hoc Finds Greater Heart Failure Protection from SGLT-2 Ertugliflozin with Baseline Use of Diuretics

Dr. David Cherney authored a poster examining the CV effects of SGLT-2 inhibitor ertugliflozin (Merck/Pfizer’s Steglatro) based on concomitant cardiorenal therapies, including diuretics, RAAS inhibitors, and mineralocorticoid receptor (MR) antagonists. This was a post-hoc analysis of VERTIS CV, which read out on the final day of ADA 2020. In the CVOT, ertugliflozin showed neutral impact on three-point MACE (non-fatal MI, non-fatal stroke, and CV death) but a significant benefit on hospitalization for heart failure (HR=0.70, 95% CI: 0.54-0.90). The use of diuretics at baseline was associated with an enhanced benefit of ertugliflozin on heart failure/CV death (p=0.01 for the interaction) and on hospitalization for heart failure (p=0.02 for the interaction). The same trend was seen for loop diuretics, specifically, with a greater benefit of ertugliflozin on heart failure/CV death (p=0.006 for the interaction) and on hospitalization for heart failure (p=0.01 for the interaction). No significant interactions were observed with RAAS inhibitors or MR antagonists, nor did diuretics significantly mediate ertugliflozin’s effects on MACE or renal outcomes. Given the growing consensus that heart failure protection is a class effect of SGLT-2 inhibitors, we are keen to see any new data illuminating how this therapy class interacts with other medications to keep patients healthier for longer.

How DAPA-HF and EMPEROR-Reduced Trials Changed the Role of SGLT-2 Inhibitors in Treating HFrEF

Dr. Carolyn Lam (Duke NUS) suggests that “SGLT-2 inhibitors, not just dapagliflozin but perhaps as a class, may be considered foundational therapy for HFrEF.” Dr. Lam began her talk by referencing the four diabetes CVOTs that first hinted at SGLT-2 inhibitors’ effectiveness at reducing the risk of hospitalization for heart failure (hHF): DECLARE-TIMI 58, CANVAS, EMPA-REG OUTCOME, and CREDENCE. Dr. Lam summarized results from DAPA-HF, first read out at ESC 2019, which showed a 26% RRR on CV death/hHF/urgent HF for patients with heart failure with reduced ejection fraction (HFrEF) on dapagliflozin compared to placebo. Dr. Lam also reviewed results from EMPEROR-Reduced, presented at ESC 2020, which demonstrated a 25% RRR on CV death/hHF in patients with HFrEF on empagliflozin versus placebo. From both trials’ data, Dr. Lam recommended that SGLT-2 inhibitors become the fifth foundational therapy in HFrEF, along with ARNI, ACEi or ARB, MRA, and b-blocker classes. Specifically, Dr. Lam suggested using dapagliflozin and empagliflozin for HFrEF. While dapagliflozin has received approval for treatment of HFrEF irrespective of diabetes status in the US, Europe, Japan, and China, empagliflozin has only received European Commission approval for this indication and awaits regulatory approval in other geographies.  

  • From DAPA-HF sub-group analyses, Dr. Lam showed that dapagliflozin’s benefits were consistent despite varying diabetes status, background therapies, age group, and diuretic use. Dr. Lam emphasized that DAPA-HF was a “paradigm-shifting trial” because it included patients with and without type 2 diabetes and was the first to show that an SGLT-2 inhibitor may be a treatment for adults with HFrEF irrespective of diabetes status. Dr. Lam emphasized that the benefits of dapagliflozin were consistent irrespective of baseline type 2 diabetes status (27% RRR on primary endpoint for patients without T2D; 25% RRR on primary endpoint for patients with T2D) and that patients without diabetes had stable A1cs above hypoglycemic ranges. Interestingly, there was a reduction in new-onset diabetes in the dapagliflozin arm compared to the placebo arm. 

    • While adverse events are more likely in older patients, Dr. Lam asserted that the benefit/risk profile of dapagliflozin is just as favorable in older patients as in younger patients. This claim is supported by a greater absolute risk reduction in older patients but a similar relative risk reduction between all age groups.  In particular, the hazard ratios for the primary composite outcome (CV death, hHF/urgent visit, and all-cause death) remained below one across all age groups. 

    • The benefits of dapagliflozin are similar in patients regardless of diuretic use, dose, and background therapies. Alleviating concerns that SGLT-2 inhibitors may be harmful to patients on diuretics, Dr. Lam cited a study in Circulation that showed that the efficacy and safety of dapagliflozin was consistent regardless of diabetic use or dose. Noting that enrolled patients were very well treated with background therapies for HFrEF, Dr. Lam showed that dapagliflozin’s risk reduction remained consistent regardless of background therapy. Dr. Lam has pointed out in a previous talk at ESC 2020 that there is a lack of evidence on whether SGLT-2 inhibitors should be prescribed to a “de novo” HF patient with no prior CV therapies. This question remains to be answered.   

  • Dr. Lam argued for an SGLT-2 inhibitor class effect by comparing EMEPEROR-Reduced and DAPA-HF trials. A meta-analysis of DAPA-HF and EMPEROR-Reduced published in The Lancet confirmed consistent effects of empagliflozin and dapagliflozin on hHF in patients with HFrEF, which suggests that these therapies may also improve renal outcomes and reduce all-cause and CV mortality. Dr. Lam also noted the similar absolute risk reduction and number of patients needed to treat for dapagliflozin and empagliflozin in their respective CVOTs. 

  • Dr. Lam concluded by summarizing the many options healthcare practitioners have to treat HFrEF, suggesting a four-drug combination. Dr. Lam listed five pathways that have been proven to improve CV outcomes in HFrEF patients, which impact angiotensin 2, norepinephrine, aldosterone, neprilysin, and SGLT.  Recognizing that these five pathways can only be modulated by four drug classes (ARNI, b-blockers, MRAs, and SGLT-2 inhibitors), Dr. Lam concluded “the important concept is to try to get patients on this four-drug combination as quickly and safely as possible.”

Dr. Brendon Neuen and Dr. David Wheeler Discuss the Implications of the CREDENCE and DAPA-CKD Trials on Standard of Care for Chronic Kidney Disease

Following Dr. Lam’s and Dr. Vaduganathan’s presentations, the session’s focus shifted to discussing the implications of recent clinical trials examining the efficacy of SGLT-2 inhibitors to treat CKD. Dr. Brendon Neuen (University of New South Wales) analyzed the implications of the CREDENCE trial studying canagliflozin, and Dr. David Wheeler (University College London) surveyed results from the DAPA-CKD trial examining dapagliflozin. Due to the strong results from these trials, both speakers strongly advocated for the use of SGLT-2 inhibitors to treat CKD, agreeing that these therapies confer substantial CV and renal benefits. Dr. Neuen stated the “challenge for us now is to implement these findings for the benefit of our patients.”

  • Dr. Neuen asserted CREDENCE “emphatically confirmed the hypothesis that SGLT-2 inhibitors can safely reduce the risk of kidney failure.” He attributed this not only to canagliflozin’s “substantial risk reductions” for CV and renal outcomes and its reassuring safety performance, but he also highlighted CREDENCE’s strong design that “specifically [assessed] the effect of SGLT-2 inhibitors on kidney disease,” unlike previous trials such as CANVAS and EMPA-REG. Since participants in CREDENCE were at higher risk for kidney failure than the participants of CANVAS and EMPA-REG, Dr. Neuen argued the trial was more powered to determine SGLT-2 inhibitors’ effects on CV and renal outcomes, making it comparable in power to the groundbreaking RENAAL and IDNT RAS blockade trials nearly two decades ago – which enabled this clinical breakthrough on SGLT-2 inhibitors’ efficacy for CKD to finally occur after years of failure.

  • Dr. Neuen noted CREDENCE’s results “continue to provide important information that can guide the implementation of canagliflozin and other SGLT-2 inhibitors” in clinical practice to treat CKD. In particular, he called attention to a noticeable drop in eGFR observed in the canagliflozin-treated population followed by stabilization over time. While he mentioned clinicians should be acutely aware of this frequent eGFR dip since there is “substantial” individual variation in this reduction, he stated canagliflozin’s efficacy and safety profile is similar irrespective of the magnitude of the eGFR decrease, eliminating some concern from implementing it into treatment regimens. In addition, Dr. Neuen highlighted canagliflozin’s 30% mean reduction in albuminuria after six months, asserting the results “underscore the importance of monitoring albuminuria to assess long-term cardiovascular and renal risk,” since the associated reduction in cardiorenal risk was largely explained by the lowering in albuminuria. Canagliflozin’s convincing safety and efficacy profile encouraged Dr. Neuen to recommend SGLT-2 inhibitors be “routinely offered” in combination with RAS blockade to people with type 2 diabetes and CKD, and he emphasized those with more advanced CKD will “gain the greatest absolute benefit” from this treatment.

  • Dr. Wheeler, co-chief investigator of DAPA-CKD, reviewed results from the trial, emphasizing that they were extremely encouraging since the trial went further than CREDENCE by including individuals without diabetes and studying a population with a lower mean eGFR (more severely impaired kidney function). He was particularly impressed by dapagliflozin’s consistent benefits across several patient subgroups, noting it was similarly beneficial in reducing the risk of kidney failure, CV death, and HF hospitalization irrespective of diabetes status, CV event history, CKD cause, and CKD stage. Furthermore, Dr. Wheeler viewed dapagliflozin’s safety profile as “reassuring” for both diabetologists and non-diabetologists since no DKA was observed in those with diabetes, and those without diabetes did not suffer from hypoglycemia.

    • Like Dr. Neuen, Dr. Wheeler also highlighted an “acute drop” in eGFR in the dapagliflozin-treated population, suggesting more work is needed to understand the cause for this decrease and its implications. Nevertheless, he was encouraged that the eGFRs of the patients in the dapagliflozin arm quickly “leveled off” and ultimately declined at a slower rate than those in the placebo group. He believed it was clear dapagliflozin slowed the progression to end-stage kidney disease despite this initial decrease.

SGLT-2 Inhibitors Lauded for CV and Renal Benefits for Patients With or Without Diabetes, but are Under-Prescribed

Across the board, SGLT2 inhibitors have shown impressive effects on cardiovascular and renal outcomes, irrespective of a diabetes diagnosis. A series of presentations, sponsored by Boehringer Ingelheim and Lilly, highlighted results from numerous recent SGLT2 inhibitor trials. Notably, a recent meta-analysis combining data from EMPA-REG OUTCOME, CANVAS, DECLARE, CREDENCE, and VERTIS-CV, found a 15% relative risk reduction (HR=0.85, 95%: 0.78-0.93, p≤0.02) in overall CV death. Additionally, when a consistent threshold was used to determine kidney function, there was a 38% relative risk reduction (HR=0.62, 95%: 0.56-0.70, p≤0.09) in kidney disease progression. Furthermore, patients started seeing improvements quickly after treatment began. In the DAPA-HF trial, significant reductions in CV death and hospitalizations were evident at 28 days. Unfortunately, fewer than one in four patients with T2D and cardiovascular disease are on either an SGLT2 inhibitor or a GLP1 agonist. Dr. Lawrence Leiter (University of Toronto) expressed that given the abundance of remarkable data, the next step is to make sure patients are receiving these lifesaving therapies.

  • Dr. Javed Butler (University of Mississippi) reviewed evidence revealing SGLT2 inhibitors are effective in patients with or without diabetes. The DAPA-HF trial inclusion criterion of left ventricular ejection fraction (LVEF) ≤40% was independent of a diabetes diagnosis. Patients receiving dapagliflozin saw a 26% (p<0.00001) relative risk reduction in CV death or worsening HF events over two years. For patients with diabetes, the RRR for CV death or HF hospitalization was 25%, while for people without diabetes, the RRR was 27%. The nearly identical results suggest that diabetes is not crucial to the therapeutic mechanism of the SGLT2 inhibitor, and Dr. Butler concluded by saying that empagliflozin and dapagliflozin are the new standard of care for patients with HFrEF.

  • In patients with chronic kidney disease, Dr. David Cherney (University of Toronto) presented similar results – patients with and without diabetes had nearly identical outcomes. The DAPA-CKD trial, which enrolled 4000 patients with CKD, concluded earlier than expected in March 2020 due to its extremely promising results. The data showed subjects benefitted greatly from the study, with a nearly 44% relative risk reduction in renal-specific outcomes. Dr. Cherney noted that there is a fairly common eGFR “dip” that occurs when patients begin SGLT2 treatment but that this shouldn’t scare providers off because in the long run SGLT2 inhibitors demonstrate a protective effect.

  • Dr. Jennifer Green (Duke University) emphasized the update in ADA Standards of Care that adds SGLT2 to the treatment algorithm for high-risk patients, independent of A1c and metformin use. The recent change incorporated evidence from the VERTIS-CV, SCORED, SOLOIST-WHF, and EMPEROR-Reduced trials. The meta-analysis showed that the relative risk reduction on MACE outcomes for SGLT2-treated patients on metformin was about 7%, while patients not on metformin saw a reduction of about 18%. Thus, MACE reductions in both categories suggest providers should feel comfortable prescribing an SGLT2 inhibitor to their high-risk patients irrespective of their metformin use.

SGLT-2 Inhibitor Posters

Title

Authors

Details + Takeaways

Effects of Canagliflozin (CANA) on Major Adverse Cardiovascular Events (MACE) by Baseline EGFR: Pooled Hispanic Subgroup Analysis from the CANVAS Program and CREDENCE

Matthew R. Weir, Jagadish Gogate, CV Damaraju, Ricardo Correa-Rotter, Kenneth W. Mahaffey

  • Analysis of data from the double-blind RCT CANVAS Program (n=10,142) and CREDENCE trials (n=4,401) to determine impact of canagliflozin on MACE risk in Hispanic patients with different eGFR baselines (three subgroups: <45, 45-60, >60 mL/min/1.73m2)

  • Canagliflozin reduced the risk of MACE in the overall population by 17% (HR= 0.83, 95% CI: 0.75, 0.92) and in the Hispanic cohort by 29% (HR=0.71, 95% CI: 0.55, 0.92), regardless of baseline eGFR

  • Even Hispanic patients with a baseline eGFR <45 mL/min/1.73m2 saw a significant reduction in MACE (HR=0.51, 95% CI: 0.30, 0.85)

  • Safety outcomes in the Hispanic cohort were similar to that of the overall pooled data

Prescribing of SGLT2 Inhibitors in Primary Care: A Qualitative Study of Primary Care Physicians and Endocrinologists

Tamara Milder, Sophie Stocker, Melissa, Baysari, Richard O. Day, Jerry Greenfield

  • Qualitative study included interviews with 15 PCPs and 12 endocrinologists in New South Wales, Australia to evaluate the utilization of SGLT-2 inhibitors in primary care for type 2 diabetes

  • Identified two causes of provider hesitancy to prescribe SGLT2s: (i) limited knowledge of the benefits; and (ii) a shift in the focus of type 2 treatment from glycemia to cardio-renal risk protection

  • Though many PCPs recalled learning about SGLT2s from a pharmaceutical representative, many also stated that ideally they would be educated by endocrinologists, independent of pharmaceutical company sponsorship

Cardiorenal Outcomes with Ertugliflozin by Baseline (BL) Glucose-Lowering Agent (GLA): An Analysis from VERTIS CV

Bernard Charbonnel, Samuel Dagogo-Jack, Christopher P. Cannon, David Cherney, Francesco Cosentino, Darren K. Mcguire, Weichung J. Shih, Jie Liu, Annpey Pong, Ira Gantz, Robert Frederich, James P. Mancuso, Richard E. Pratley

  • Sub-analysis of RCT VERTIS CV (n=8,246) to assess the effect of ertugliflozin in patients with established ACSVD and type 2 diabetes by baseline glucose lowering agents (metformin, insulin, sulfonylureas, and DPP-4 inhibitors)

  • The effects of ertugliflozin on CV and kidney outcomes were NOT significantly modified by baseline glucose lowering agent use

  • Ertugliflozin was not associated with an increased risk of symptomatic or severe hypoglycemia

Does Sacubitril/Valsartan Combination Therapy Provide Additional Renal Benefits in Type 2 Diabetes Mellitus Receiving Canagliflozin?

Vishal Gupta, Vaishali Teli

  • Retrospective, one-year study compared matched patients on canagliflozin-only treatment to patients on canagliflozin and sacubitril/valsartan (SVC) combination therapy for T2D treatment (n=98)

  • No significant difference in eGFR (p=0.953) or UACR (p=0.92) between canagliflozin and canagliflozin + SVC groups

  • Sacubitril/Valsartan combination therapy does not provide any additional renal benefits in T2DM receiving canagliflozin

Ertugliflozin (ERTU) Delays Insulin Initiation and Reduces Insulin Dose Requirements in Patients with Type 2 Diabetes (T2D): An Analysis From VERTIS CV

Samuel Dagogo-Jack, Robert Frederich, Bernard Charbonnel, Jie Liu, Christopher P. Cannon, Harry Shi, Darren K. Mcguire, David Cherney, Francesco Cosentino, Weichung J. Shih, Urszula Masiukiewicz, Ira Gantz, Richard E. Pratley

  • Analysis of VERTIS CV (n=8,248) assessed the impact of ertugliflozin (5 mg or 15 mg) on insulin therapy dose requirements in type 2s with ASCVD and a baseline A1c of 7%-10.5%

  • At baseline, 47% were on insulin; the patients on insulin had longer duration of T2D (~16.2 yrs vs 10.1 yrs), higher A1c (8.4% vs 8.1%), and less frequent users of sulfonylurea (16.4% vs 63.1%)

  • In insulin-naïve patients, ertugliflozin significantly reduced the likelihood of insulin initiation by 30% and 36% for 5 mg and 15 mg groups of ertugliflozin, respectively (5 mg: HR= 0.70, 95% CI: 0.58-0.84, p<0.001 15 mg: HR=0.64, 95% CI 0.53-0.78, p<0.001) and delayed new insulin initiations by up to ~1.8 years vs. placebo

  • In patients on insulin at baseline, the change in insulin at month 60 was −2.47 U/day with 5 mg, −1.77 U/day with 15 mg, and +9.42 U/day with placebo

Glycemic Efficacy and Safety of Ertugliflozin (ERTU) in Patients with Type 2 Diabetes (T2D) and Chronic Kidney Disease Stage 3 (CKD 3): An Analysis from VERTIS CV

Samuel Dagogo-Jack, Richard E. Pratley, David Cherney, Bernard Charbonnel, Darren K. McGuire, Francesco Cosentino, Weichung J. Shih, Jie Liu, Robert Frederich, James P. Mancuso, Annaswamy Raji, Ira Gantz

  • Analysis of VERTIS CVOT (n=8,246) to assess glycemic efficacy and safety of ertugliflozin in patients with Stage 3 CVD(n=1,776), defined as eGFR of 30-<60 mL/min/1.73 m2

  • At week 18, patients with Stage 3 CVD saw significant reductions in A1c relative to placebo at -0.27% with 5 mg of ertugliflozin and -0.38% with 15 mg of ertugliflozin (p<0.001 for both groups)

  • Stage 3 CKD patients on ertugliflozin also saw significant adjusted reductions in bodyweight: -1.4 kg for the 5 mg- ertugliflozin cohort and -1.5 kg for the 15 mg – ertugliflozin cohort

  • Those on ertugliflozin also saw a significant reduction in  systolic BP relative to the placebo group (-2.9 mm Hg for 5 mg; -3.2 mm Hg for 15 mg)

  • No difference in overall adverse events, symptomatic hypoglycemia, hypovolemia, and kidney-related adverse events between ertugliflozin and placebo 

Empagliflozin Reduced the Total Burden of Events Leading to or Prolonging Hospitalization in EMPA-REG OUTCOME

Silvio E. Inzucchi, Christoph Wanner, David H. Fitchett, Bernard Zinman, Stefan Anker, Michaela Mattheus, Ola Vedin, Stefan Hantel, Søren S. Lund

  • Post hoc analysis to assess the effect of empagliflozin (10 mg or 25 mg) vs. placebo on all-cause hospitalization (ACH) and an ACH/all-cause mortality (ACM) composite in type 2s  (n = 7,020)

  • Empagliflozin reduced the risk of total events of ACH by 22% vs. placebo (HR=0.78, p<0.0001) and ACH/ACM by 24% vs. placebo (HR=0.76, p<0.0001)

  • The number of ACH/ACM events prevented by empagliflozin vs. placebo was 67.7 per 1000 patient-years (95% CI: 95.9, 39.5); the number needed to treat (NNT) over the 3-year trial to prevent one event was 4.9 (95% CI: 3.5,8.4)

Ertugliflozin in Older Patients with Type 2 Diabetes (T2D): An Analysis from VERTIS CV

Richard E. Pratley, Samuel Dagogo-Jack, Bernard Charbonnel, David Cherney, Francesco Cosentino, Darren K. McGuire, Margaret N. Essex, Philip L. S. Jones, Jie Liu, Ingrid A. Adamsons, David Lawrence, Weichung J. Shih, Christopher P. Cannon

  • Analysis of cardiorenal outcomes and safety of ertugliflozin (5 mg or 15 mg) in older patients in the VERTIS CVOT (n=8,238)

  • 50% of participants were ≥65 years (baseline A1c 8.1%), with 11% ≥75 years (baseline A1c 8.0%)

  • Cardiorenal benefits of ertugliflozin in >65 years sub-population were consistent with the overall cohort

  • Ertugliflozin was generally well-tolerated in patients ≥65 years and ≥75 years with no new safety concerns in these older patients

Comparative Safety of Empagliflozin in Routine Care Patients: Interim Results from the Empagliflozin Comparative Effectiveness and Safety (EMPRISE) Study

Elisabetta Patorno, Helen Tesfaye, Deborah J. Wexler, Robert Glynn, Mehdi Najafzadeh, Lily G. Bessette, Heidi Zakoul, Lisette Koeneman, Anouk Deruaz-Luyet, Sebastian Schneeweiss

  • EMPRISE assessed the effectiveness, safety, and healthcare utilization of empagliflozin vs. DPP-4 inhibitors, based on matched patient data from Medicare and two U.S. commercial claims datasets (2014-2019)

  • Over a mean follow-up of 6.3 months, patients who initiated empagliflozin had an elevated risk of DKA (HR: 2.0, RD: 1.4 per 1000 patient years) and a decreased risk of acute kidney injury hospitalization (HR: 0.66, RD: -8.6 per 1000 PY) compared with those starting DPP-4 inhibitors

  • Empagliflozin initiators had a similar risk of lower-limb amputations and fractures compared with DPP-4 users

Luseogliflozin Suppresses Fine Glycemic Fluctuation Late at Night: New Aspect of SGLT2 Inhibitor

Keishi Yamauchi

 

  • Study assessing impact of luseogliflozin (2.5 mg) on glycemic variation in six type 2s; glucose levels measured every 15 minutes from 11pm-3am by Freestyle Libre Pro

  • ~50% reduction of coefficient of variation (CV) of glucose levels: 11.6±4.8 pre-treatment, 4.1±2.5 post-treatment (p<0.01)

  • Suppressive effect not observed with sulfonylurea or DPP-4 inhibitors

  • Indicates a new effect of SGLT2 inhibitors, suggesting potential to reduce blood vessel damage via regulation of fine glycemic fluctuation, but the clinical significance of the findings is unclear

 

Phase 3 Results of Fixed-Dose Combination of Remogliflozin Etabonate and Vildagliptin in Indian T2DM Patients

Brij Mohan, Manoj S. Chawla, Rahul. R. Kodgule, Monika Tandon, Sagar Katare, Sachin Surawanshi, Hanmant V. Barkate

  • Phase 3 double-blind RCT comparing a fixed-dose combination of twice daily 100 mg remogliflozin etabonate and 50 mg vidagliptin to once daily fixed dose 25mg empagliflozin and 5 mg linagliptin in 182 Indian type 2s on stable metformin therapy and with an A1c of 8%-11% at baseline

  • No significant difference in 16-week Ac1 reduction between remogliflozin/ vidagliptin arm and empagliflozin/linagliptin arm (1.41% vs. 1.36%, p=0.82)

  • No significant difference in 16-week Ac1 reduction between remogliflozin/ vidagliptin arm and empagliflozin/linagliptin arm (28% vs. 31%, p=0.66)

Risk of Severe Complications of Urinary Tract Infection in Real-World Use of Dapagliflozin

Catherine Johannes, J. Bradley Layton, Daniel C. Bleachler, Ryan M. Ziemiecki, Ling Li, Heather Danysh, Jade Dinh. Phillip Hunt, Cecilia Karlsson, Hungta Chen, Alicia Gilsenan

  • Retrospective study found no increased risk of severe urinary tract infections (sUTI) for female type 2s who used dapagliflozin (n=17,265) compared to other glucose-lowering drugs (n=161,176) (HR=0.93; 95% CI: 0.73-1.18)

  • Serious UTI events included hospitalizations or ED visits from urosepsis and pyelonephritis

Risk of Acute Liver Injury in Real-World Use of Dapagliflozin

Catherine Johannes, J. Bradley Layton, Daniel C. Beachler, Ryan M. Ziemiecki, Ling Li, Heather E. Danysh, Jade Dinh, Phillip Hunt, Cecilia Karlsson, Hungta Chen, Alicia Gilsenan

  • Non-interventional, post-authorization safety study comparing the incidence of hospitalization for acute liver injury for new T2D users of dapagliflozin and new users of comparator glucose lowering drugs

  • Study included >28,000 person-years of dapaglifozin exposure and did not show any evidence of an increased risk of acute liver injury associated with dapagliflozin compared with other glucose lowering drugs (IRR=.85; 95% CI 0.59-1.24)

  • Results are consistent with DECLARE-TIMI trial, which found a null effect on acute liver injury from dapagliflozin relative to placebo

Economic Benefits of Initiating Empagliflozin vs. GLP-1 Receptor Agonists among Patients with Type 2 Diabetes

Pratik Pimple, Aditya Raju, Eileen Farrelly, Sharash Shetty

  • A retrospective matched cohort design assessed the all-cause healthcare costs of initiation of empagliflozin vs. GLP-1s in 43,588 type 2s

  • Empagliflozin all-cause costs were $273 lower than GLP-1 costs (p<0.001), driven primarily by reduced pharmacy costs, which were $195 lower (p<0.01), and reduced medical costs, which were $78 lower (p<0.01)

  • A sub-analysis containing 10,694 patients who were already on metformin and began GLP-1 or SGLT2 as a second-line treatment found that patients on empagliflozin paid $271 less for all-cause total costs compared to those who added GLP-1s

The Role of Industry and NIH/Academia in the Clinical Development of GLP-1 Agonists and SGLT2 Inhibitors

Brian R. Kocak, Zachary Kassir, Jing Luo

  • Investigated the role of NIH/academia in the development of GLP-1s and SGLT2s, which often have a high price because of high development costs, based on data from 200 GLP-1 and 104 SGLT2 phase 3 trials from clinicaltrials.gov

  • 90% of GLP-1 trials were funded by industry, but 52% were led by NIH-supported scientists

  • 89% of SGLT2 trials were funded by industry, but 42% were led by NIH-supported scientists

  • The first drugs approved in each class, exenatide and canagliflozin, received more funding than the subsequent drugs in each class

  • The data suggest that NIH funding – either directly or indirectly through intellectual leadership – played a significant role in the development of these drugs

Effectiveness and Safety of Empagliflozin in Routine Care in Europe and East Asia: Results from the Empagliflozin Comparative Effectiveness and Safety (EMPRISE) Study

Avraham Karasik, Stefanie Lanzinger, Elise Tan, Daisuke Yabe, Dae Jung Kim, Wayne Sheu, Cheli Melzer Cohen, Reinhard Holl, Kyoung Hwa Ha, Thomas Nystrom, Leo Niskanen, Majken Linnemann Jensen, Moe Kyaw, Julio Nunez, EMPRISE EU and East Asia Study Group

  • The EMPRISE study retrospectively compared the effectiveness and safety of empagliflozin vs. DPP-4 inhibitors in type 2s from Europe, Israel, and East Asia based on EHR data from 66,915 type 2s

  • Patients on empagliflozin had a 29% relative risk reduction (RRR) in HHF compared to those on DPP-4s (0.60-0.82), 26% RRR in stroke (0.62-0.88) and a 45% RRR in all-cause mortality.

  • Empagliflozin users had a higher rate of diabetic ketoacidosis than did DPP-4 users (HR=1.82, 1.12-2.95)

The Impact of Canagliflozin on the Risk of Neuropathy Events: Results from the CREDENCE Trial

Jinlan Liao, Amy Kang, Chao Xia, Clare Arnott, Gian Luca Di Tanna, Arun Krishnan, Carol Pollock, Rajiv Agarwal, George Bakris, Hiddo Heerspink, Adeera Levin, Dick de Zeeuw, Bruce Neal, Hong Zhang, David Wheeler, Bernad Zinman, Kenneth Mahaffey, Vlado Perkovic, Meg Jardine, Brendan Smyth

  • Analysis of the CREDENCE trial (n=4,401) comparing neuropathy-related adverse events in patients receiving canagliflozin vs. placebo

  • 48.8% of the participants had a diagnosis of neuropathy at baseline

  • Canagliflozin had no impact on overall neuropathy-related adverse events (HR=1.03, p=0.66), nor on specific neuropathy-related adverse events, including sensorimotor polyneuropathy (HR 0.93, 95% CI 0.69 to 1.25, p=0.63); diabetic neuropathy (HR 0.91, 95% CI 0.68 to 1.22, p=0.52), or non-autonomic neuropathy (HR 1.03, 95% CI 0.87 to 1.21, p=0.77)

Canagliflozin Improves Cardiovascular and Renal Outcomes across Broad Geographic Regions: Results from CREDENCE

Kathryn Cardoza, Amy Kang, Brendan Smyth, Tae Won Yi, Gian Luca Di Tanna, Carol A. Pollock, Dick De Zeeuw, David Wheeler, Christopher Cannon, Bernard Zinman, Meg Jardine, Clare Arnott, Adeera Levin, and Kenneth Mahaffey

 

  • Sub-analysis of the CREDENCE trial assessing the effects of canagliflozin on kidney and CV events in patients with T2D and CKD (n=4,401) across 34 countries

  • Canagliflozin reduced the risk of kidney and CV events across geographic regions

  • Canagliflozin use incurred a 30% relative risk reduction on primary outcome composite, which included end-stage kidney disease (dialysis, transplantation, or sustained eGFR <15ml/min/1.73m2), a doubling of serum creatine, or death from renal or CV causes (HR=0.70, 95% CI:0.59-0.82), with no evidence of heterogeneity across different geographic regions

  • Canagliflozin use was associated with a 31% RRR in CV death/HHF (HR=0.69, 95% CI: 0.57-0.83) and a 0.34% RRR in doubling of serum creatinine, ESKD, or death (HR=0.66, 95%CI: 0.53-0.81) with no evidence of heterogeneity across different geographic regions

GLP-1 Agonists

AMPLITUDE-O Full Results: GLP-1 Efpeglenatide Delivers 27% RRR on MACE, Becoming First Exendin-Based GLP-1 to Find CVOT Success in Patients with Type 2 and CV/Renal Disease

Headlining Day #4 of ADA, positive results of the AMPLITUDE-O CVOT for ex-Sanofi GLP-1 agonist efpeglenatide were published today in NEJM and outlined in detail during an afternoon symposium. As highlighted by Dr. Stefano del Prato (University of Pisa), over the past 30 years, while there have been some improvements to the prevalence of CKD in people with diabetes, there have been no major changes to the prevalence of CV complications. (We note we’re not sure about the last couple of years, since virtually all data we’ve seen is a couple of years old.) As represented by a 2019 meta-analysis of GLP-1 receptor agonist CVOTs, however, there has been a plethora of positive data on the effects of GLP-1s on CVD. That said, uncertainty about whether exendin-4 based GLP-1 agonists provide the same benefit has persisted, given that exenatide was not found to be superior to placebo on 3P-MACE in the EXSCEL CVOT (which we questioned at the time and have been doing so pretty much since 2018). Thus, the AMPLITUDE-O trial study of exendin-4 based efpeglenatide not only provides valuable information on efpeglenatide itself, but also, the exendin-4 GLP-1 class more broadly.

  • Dr. Carolyn Lam (Duke University) emphasized the global scope of this Sanofi-sponsored study, which included 4,076 participants from a whopping 28 countries. The randomized, double-blind, placebo-controlled trial examined whether a weekly injection of 4 mg or 6 mg of efpeglenatide was non-inferior to placebo on the primary outcome of MACE in people with type 2 diabetes who were at high risk for cardiovascular events. Following proof of noninferiority for the primary outcome, the statistical plan called for hierarchal analysis of superiority on MACE and several secondary outcomes: expanded MACE including coronary revascularization or hospitalization for unstable angina, renal composite of new or worsening nephropathy (new macroalbuminuria, sustained eGFR fall ≥40%, dialysis or renal replacement, or sustained eGFR <15 ml/min/1.73 m2), MACE or non-CV death, renal function outcome, and a composite outcome of MACE, non-CV death, heart failure, or the renal function outcome. Randomization was stratified according to region and current or future use of SGLT-2 inhibitors or neither. Participants were given efpeglenatide on top of pre-randomization therapy (which remained stable for the first 12 weeks of the study); the study continued until 330 MACE events occurred.

    • The baseline characteristics of the 4,076 patients enrolled in the AMPLITUDE-O Trial were as follows: mean age of 64.5 years, mean diabetes duration of 15.4 years, 87% white, 33% female, mean A1c of 8.9%, mean eGFR of 22%, 90% with prior CV disease, 63% of participants on insulin, and 15% of participants on an SGLT-2 inhibitor. Notably, this study has the longest duration of diabetes, lowest eGFR, highest A1c, highest insulin use, and highest use of SGLT-2 inhibitors at baseline compared to other GLP-1 agonist CVOTs. The retention and adherence of AMPLITUDE-O was quite high: 97% patients were retained throughout the trial with 99.9% vital status, and 81% of patients were on efpeglenatide at last visit. The subjects were followed for a fairly short time, just a median of 1.8 years.

  • Dr. Julio Rosenstock (University of Texas) continued with a discussion of efpeglenatide’s excellent metabolic results and positive safety profile. Overall, treatment with efpeglenatide reduced average A1c (-1.2%; p<0.001), weight (-2.6 kg; p<0.001), BMI (-0.92 kg/m2; p<0.001), systolic blood pressure (-1.5 mm Hg; p<0.001), and pulse pressure (-2.1 mm Hg; p<0.001) vs. placebo. Importantly, the substantial reduction in A1c occurred despite the use of additional glucose-lowering agents in the placebo group over the duration of the study. Consistent with other GLP-1 receptor agonists, treatment with efpeglenatide led to a moderate increase in heart rate (+3.9 bpm; p<0.001). Efpeglenatide was generally safe and well-tolerated, showing a similar safety profile to other GLP-1 receptor agonists. The discontinuation rate for adverse events was slightly higher among those receiving efpeglenatide vs. placebo (5.4% vs. 3.4%; p = 0.015), and the majority of discontinuations were for GI effects (3.3% vs. 1.8%; p=0.009). There was no statistically significant increased risk of pancreatic events, pancreatitis, pancreatic or other cancers, severe hypoglycemia, worsening of retinopathy, acute renal failure, injection reactions or any other spontaneously reported adverse events during the trial period.

  • Lead investigator Dr. Hertzel Gerstein (McMaster University) unveiled the positive cardiovascular outcomes reported in AMPLITUDE-O, with efpeglenatide showing both non-inferiority and superiority to placebo on primary and secondary outcomes and positive trends in exploratory outcomes. Efpeglenatide 4 mg or 6 mg conferred a significant 27% RRR (HR=0.73; 95% CI: 0.58-0.92; p<0.0001 for non-inferiority; p=0.0069 for superiority) on the primary outcome of composite MACE (first non-fatal MI, stroke, or death from CVD or undetermined cause). Dr. Gerstein emphasized that Kaplan-Meier curves for efpeglenatide and placebo diverged within the first three months after randomization, highlighting the GLP-1’s early efficacy. Per the trial’s design, additional outcomes were assessed for superiority hierarchically until results were unsignificant. The secondary expanded MACE, renal composite, and MACE/non-CV death outcomes all showed significant risk reductions from treatment with efpeglenatide (HR=0.79, p=0.20; HR=0.68, p<0.0001; HR=0.73, p=0.004; respectively).

  • When comparing 4 mg and 6 mg doses to placebo on the MACE outcome, efpeglenatide showed a dose-dependent effect, with the 6 mg dose conferring greater reductions in hazard ratio (HR=0.65 vs. HR=0.82). Exploratory analyses also trended in the positive direction for the effects of efpeglenatide. Though the trial was not powered to look at efpeglenatide’s effects on each MACE component outcome, efpeglenatide trended toward benefit in fatal or nonfatal MI (HR=0.95), nonfatal MI (HR=0.78), fatal or nonfatal stroke (HR=0.74), nonfatal stroke (HR=0.79), CV death (HR=0.72), all-cause death (HR=0.78), and heart failure hospitalization (HR=0.61).


  • When comparing MACE effects among subgroups, there was no evidence of differential effects by baseline SGLT-2 inhibitor use. According to Dr. Gerstein, this indicates that the overall 27% reduction in hazard ratio can be generalized to participants regardless of background SGLT-2 medication. Efpeglenatide’s effects were also independent of baseline eGFR and metformin use and consistent across a variety of demographic factors (see renal outcomes for a full rundown).

  • Up next, Dr. Richard Pratley (AdventHealth Translational Research Institute, Orlando) expertly broke down efpeglenatide’s renal results. Broadly speaking, efpeglenatide was shown to significantly decrease albuminuria and the development of macroalbuminuria (UACR 21% lower; 95% CI:14-28, p<0.001) and decrease the rate of decline in eGFR (0.87 ml/min/1.73 m2 higher; 95% CI: 0.27-1.51, p=0.005). As the third endpoint of its hierarchical assessment, efpeglenatide was assessed on the kidney composite outcome of (i) new macroalbuminuria (ACR >300 mg/g) and ≥30% rise from baseline; (ii) decrease in eGFR by ≥40% from baseline for ≥30 days; (iii) renal replacement therapy for ≥90 days; and (iv) eGFR <15 ml/min/1.73 m2 for ≥30 days. Impressively, efpeglenatide delivered a statistically significant 32% relative risk reduction on this renal composite (HR=0.68; 95% CI: 0.57-0.79, p<0.0001). Importantly, a pre-specific subgroup analysis, comparing the kidney composite based on sex, age (<65 years old or ≥65 years old), ancestry (White, Black, Asian, Other), region (USA/Canada, Latin America, Europe, Other), diabetes duration (<10 years or ≥10 years), baseline A1c (<8% or ≥8%), BMI (median <31.9 or ≥31.9), and eGFR (median <71.5 or ≥71.5) found similar effects across clinical subgroups (p-value of interaction non-significant). Of particular interest, a pre-specified subgroup analysis based on baseline SGLT-2 inhibitor usage (yes or no; 15.2% of participants using SGLT-2 inhibitors) found no difference in effect (p-value of interaction=0.38), signaling that efpeglenatide is just as effective on renal outcomes in people taking or not taking an SGLT-2 inhibitor.

  • Dr. Naveed Sattar (University of Glasgow) concluded the session with findings of a meta-analysis of eight major GLP-1 agonist CVOTs including the results of AMPLITUDE-O. While it is accepted that GLP-1s reduce CV events, all-cause death, and some renal outcomes in type 2, uncertainty remains on exendin 4-based GLP-1s and effects on HF and other renal outcomes. Overall, the meta-analysis revealed that all GLP-1s analyzed decreased 3-point MACE, all-cause mortality, heart failure hospitalization, and renal dysfunction without additional risk of severe hypoglycemia, retinopathy, or pancreatic side effects. Importantly an analysis on GLP-1 homology found no interaction based on human vs. non-human GLP-1 homology, and Dr. Sattar concluded that “almost definitively,” there is no difference in MACE outcomes based on GLP-1 homology. To close out this discussion, Dr. Sattar emphasized that the AMPLITUDE-O trial showed that efpeglenatide safely reduces major CV and renal outcomes in both low and high-risk populations with type 2 diabetes with the additional effects of reducing blood glucose levels, blood pressure, and weight.

  • The ever-eloquent Dr. Amanda Adler (University of Oxford) closed the session with valuable independent commentary. Overall, particularly given that Dr. Adler’s commentary is often characterized as steady and neutral, she definitely sounded rather positive. And who wouldn’t be? She highlighted efpeglenatide’s safety, efficacy on lowering risk of CV and renal disease, likelihood to satisfy regulators, use of a population outside of SGLT-2 inhibitors’ marketing authorization (i.e., low eGFRs), and clear labelling of results as being exploratory, to name a few. That said, Dr. Adler did point out a number of perceived shortcomings of the trial design and potential areas of controversy.

    • First, Dr. Adler emphasized that any publications on AMPLITUDE-O must explicitly state that the trial was primarily designed as a non-inferiority trial; during Q&A, Dr. Julio Rosenstock pushed back on this critique, however, and emphasized that while this is true, the data is just as supportive of superiority as data for other GLP-1s. We hadn’t remembered that point and thought the superiority was all the more impressive.
    • Second, Dr. Adler questioned whether a placebo comparator is truly clinically relevant. Dr. Adler was critical of the fact that the AMPLITUDE-O authors did not more explicitly discuss why they had chosen placebo as their comparator and proposed that putting efpeglenatide up against another GLP-1 with known CV benefit may have been a more meaningful decision. We can imagine cost may have had something to do with it. Dr. Adler, however, did find that placebo was likely the “correct comparator for renal disease,” given that efpeglenatide would be added on to existing renal therapies (rather than used to replace them).
    • Third, Dr. Adler questioned why the trial design had only controlled for region and SGLT-2 inhibitor use, as opposed to other cardioprotective drugs as well.
    • Last, Dr. Adler critiqued the inclusion of “potential future use” of SGLT-2 inhibitors as part of SGLT-2 inhibitor concurrent use, given that “not all doctors do what they say they’re going to do.”
    • In totality, we found Dr. Adler’s appraisal to be an interesting and fair assessment of the trial that acknowledged shortcomings, but overall, found the data quite compelling.  
  • As we wrote on Saturday, when results of efpeglenatide’s AMPLITUDE-M monotherapy trial read out, it was a bumpy road for the ex-Sanofi GLP-1. As background, efpeglenatide was first licensed from Hanmi by Sanofi in 2015. Its development program then stalled in 2019 when Sanofi discontinued its diabetes therapy R&D. As we understand it, Sanofi and Hanmi then entered into negotiations in May 2020, which resulted in the drug finally landing back at Hanmi in September 2020. Given these positive results, and Dr. Adler’s assessment that the data is “likely to satisfy regulators,” we’ll be curious to see if Hanmi moves to submit the drug for approval. As of late, Hanmi has been hard at work developing combination incretins, and plans to combine efpeglenatide with LAPS glucagon analog HM15136 have already been announced. Considering the immense amount of work that went into completing AMPLITUDE-O and its successful results, we’d love to see efpeglenatide reach the market and serve as another valuable option for people with type 2 diabetes. This is especially true given the likelihood they’ll have of receiving a label for renal protection as well as cardio protection.

Full SURPASS-2 Results for Lilly’s GLP-1/GIP Tirzepatide vs. Novo Nordisk’s Semaglutide Published in NEJM; 60% of Participants on 15 mg Achieve A1c ≤6.5% and 10% Weight Loss, Without Hypoglycemia

Kicking off with major news in the publication world, full results from the SURPASS-2 trial, assessing Lilly’s dual GLP-1/GIP receptor agonist tirzepatide vs. Novo Nordisk’s GLP-1 semaglutide (1.0 mg) in people with type 2 diabetes, were published in the New England Journal of Medicine today, along with an editorial by Dr. Katherine Tuttle, “Breaking New Ground with Incretin Therapy in Diabetes.” Publication was timed with the release of the SURPASS-2 poster at ADA (84-LB), of which we’ll hear further details on during a SURPASS symposium on Tuesday.

As a reminder, tirzepatide conferred statistically significant A1c and body weight reductions from baseline compared to semaglutide in all three treatment arms (5 mg, 10 mg, and 15 mg) by efficacy estimand (efficacy prior to discontinuation of study drug or initiating rescue therapy for persistent hyperglycemia). More specifically, tirzepatide delivered:  

  • A1c reduction: -2.09% (5 mg), -2.37% (10 mg), -2.46% (15 mg), -1.86% (semaglutide)
  • Weight reduction: -7.8 kg (-8.5%, 5 mg), -10.3 kg (-11.0%, 10 mg), -12.4 kg (-13.1%, 15 mg), -6.2 kg (-6.7%, semaglutide)
  • Percent of participants achieving A1c <7%: 85% (5 mg), 89% (10 mg), 92% (15 mg), 81% (semaglutide)
  • Percent of participants achieving A1c <5.7%: 29% (5 mg, not controlled for type 1 error), 45% (10 mg), 51% (15 mg), 20% (semaglutide)

Using the treatment-regimen estimand (efficacy irrespective of adherence to the investigational medicine or introduction of rescue therapy for persistent severe hyperglycemia), results were slightly less robust, though tirzepatide still delivered statistically significant A1c and body weight reductions across all three doses.

Along with these topline results, Lilly published data on two new pre-specified exploratory analyses. The first assessed the proportion of patients who achieved an A1c ≤6.5% and weight loss of 10% or greater, while not experiencing hypoglycemia <54 mg/dl or severe hypoglycemia. Very impressively, 32% of participants in the 5 mg arm, 41% in the 10 mg arm, and a whopping 60% in the 15 mg arm achieved this composite endpoint vs. 22% on semaglutide 1 mg. Of course, we’d expect higher rates from the semaglutide arm if given at higher doses, nevertheless, these are extremely positive rates for all four drug arms. Another exploratory analysis assessed tirzepatide’s effects on fasting lipids. At the highest 15 mg dose, tirzepatide conferred a 24.5% reduction in triglycerides, 23.7% reduction in VLDL cholesterol, and a 7.1% increase in HDL (“good”) cholesterol, supporting the drug’s possible efficacy on lipid endpoints as well. We’ll be back with more on tirzepatide on Tuesday.

Ex-Sanofi GLP-1’s Second Act? Efpeglenatide Meets Primary Endpoint in Phase 3 AMPLITUDE-M Monotherapy Trial; CVOT Data Coming Monday

It’s been a bumpy road for efpeglenatide, and we’re curious to see if the once-weekly GLP-1 will have a happy ending. As background, since  efpeglenatide was first licensed from Hanmi by Sanofi in 2015, its development program then stalled in 2019 when Sanofi discontinued its diabetes therapy R&D. As we understand it, Sanofi and Hanmi negotiations in May 2020 resulted in the drug finally landed back at Hanmi in September 2020; while we can’t say we particularly blame Sanofi for electing not to compete at every turn with Novo Nordisk and Lilly, we were very happy to see that the results for this trial would be announced. In sum, with this program, efpeglenatide’s AMPLITUDE-M monotherapy trial was the only one to be carried to completion, but – like Lexicon’s SGLT-1/2 inhibitor sotagliflozin, which wowed with CVOT data in SOLOIST/SCORED despite the trials closing early – efpeglenatide may see a second act, largely depending on truncated AMPLITUDE-O CVOT data.

  • Today, however, we were treated to a first glimpse of positive data on efpeglenatide from Dr. Juan Frias (National Research Institute, Los Angeles, CA), who presented results from AMPLITUDE-M. Efpeglenatide met its primary endpoint, delivering statistically significant A1c reductions at Week 30 (-1.6% at the highest 6 mg dose from 8.05% at baseline vs. -0.5% with placebo from 7.97% at baseline, p<0.0001). Furthermore, the rate of participants who achieved an A1c <7.0% or ≤6.5% was statistically higher than placebo across all three doses. These are excellent results and we are curious how low the cost of production could be. We wonder, if 95% of all PWD are type 2 and if this monotherapy (!) trial showed how well GLP-1 could do for those on metformin or (God forbid) a sulfonylurea only, wouldn’t there be opportunity for a company who cares about reducing CV risk to take this on at a slightly lower price in the name of increasing the volume of GLP-1 for patients? Or, perhaps this can be a way to move prices in the US lower so that those on self-pay or with huge deductibles can be able to afford GLP-1, a therapeutic class that results in major glycemic improvement, significant weight loss, less hypoglycemia, and greater insulin sensitivity? Oh, and satiety! Andiamo … look at these curves and buckle up.

  • On body weight, efpeglentide conferred significantly greater weight loss at the 4 mg and 6 mg doses vs. placebo. At the highest dose, ~28% of participants lost >5% of their baseline body weight and ~8% lost >10%. While 8% isn’t that much to write home about compared to recent high-dose GLP-1 results seen as well as tri-agonist results, presumably this is baseline and would keep patients from MDI for at least some time.

  • Safety and tolerability were consistent with the overall GLP-1 class. As expected, the majority of TEAEs were gastrointestinal, with slightly elevated rates of Level 1 hypoglycemia at the 4 mg and 6 mg doses (very low incidence of Level 2 hypoglycemia). Given these promising efficacy and safety effects, we will be curious to see if results on Monday that land efpeglenatide within the range of CV benefits seen across the class. Down the road, Hanmi also plans to combine efpeglenatide with other drugs like LAPS glucagon analog HM15136.

AZ’s Once-Weekly GLP-1 Exenatide Delivers Promising Results in Youth with Type 2

Results from a phase 3 trial of AZ’s GLP-1 exenatide in youth (ages 10-18 years old) with type 2 diabetes showed efficacy in reducing A1c and impacting secondary metabolic markers. After 24 weeks of treatment, youth treated with once-weekly exenatide 2.0 mg had a 0.36% drop in A1c, compared to a 0.5% increase in the placebo, as the trial met its primary endpoint (p=0.012). Exenatide treatment was also associated with favorable changes in fasting glucose, systolic blood pressure, and body weight, although these differences did not reach statistical significance vs. placebo given the small nature of the trial (n=83) and short duration (24 weeks). These results track well with the extensive data on exenatide in adults showing favorable effects on glucose, body weight, and more.

  • Regarding safety, adverse events generally tracked with the known safety profile of exenatide in adults. GI events were not elevated when compared to placebo, although rates of upper respiratory infections were elevated in the treatment group. Importantly, rates of hypoglycemia were higher in the treatment group (14% vs. 4%).

  • These results follow similarly promising results for Novo Nordisk’s Victoza (liraglutide) in a pediatric population. Results from the ELLIPSE trial (n=135) of liraglutide in children and adolescents (10-17 years old) with type 2 diabetes were just published in NEJM in 2019 and showed that treatment was associated with a 0.64% drop in A1c vs. a 0.42% increase in A1c in the placebo group (p<0.001).

  • We note that approval of these GLP-1 agents would represent a major breakthrough in the current landscape of pediatric type 2 treatment. Before the approval of liraglutide in youth-onset type 2 diabetes in 2019, only metformin and insulin were approved in this population, and it’s clear that more treatment options are desperately needed: In RISE Peds, neither metformin nor insulin was able to preserve beta cell function in children/adolescents with type 2 or prediabetes. Having multiple GLP-1 agonist approved for youth could be a game-changer (especially an easier-to-use once-weekly formulation as shown here), as a more advanced and potent molecule may support beta cell preservation, weight loss, and better glucose-lowering. Importantly, about 5,000 people <18 years old are diagnosed with type 2 diabetes in the US each year.

KOLs Debate Semaglutide 2.4 mg vs. 1.0 mg During Q&A; Dr. Julio Rosenstock Calls for Use as First Line Therapy in Newly Diagnosed Diabetes for Remission

Following the detail review of STEP data by Dr. Kushner and Dr. Wilding (see above), Dr. Lee Kaplan (Harvard Medical School) provided a nuanced perspective on the clinical implications of semaglutide 2.4 mg for obesity care, followed by a lively Q&A on semaglutide’s clinical applications between Dr. Kaplan, Dr. Melanie Davies (Leicester University), Dr. Wilding, and session chair Dr. Julio Rosenstock (Dallas Diabetes Research Center).

  • Dr. Kaplan highlighted that despite a mean 15-17% weight loss for people taking semaglutide 2.4 mg, there was significant variability between results. On the higher end of the STEP 1 trial, ~11% of patients lost more than 30% of their body weight (approaching the efficacy of bariatric surgery), and ~35% lost more than 20%; on the lower end, 7-13% of patients lost less than 5% body weight (31% with type 2 diabetes). During Q&A, Dr. Kaplan elaborated that this variability should be kept in mind when deciding which dose of semaglutide would best fit a patient – for some, the 1.0 mg dose may provide the most desirable balance of efficacy vs. risks. Further research on why people with type 2 diabetes experience less weight loss with semaglutide, and other weight loss management drugs more broadly, is also needed – as explained by Dr. Kaplan, one potential explanation is higher insulin levels in people with type 2 diabetes (and therefore insulin resistance); however, other studies suggest that this is likely not the only cause. Looking ahead, Dr. Kaplan has high hopes for the future of weight management drugs, particularly in regard to long-term obesity complications. While the majority of obesity complications are improved with just 5 to 15% weight loss, these complications (see graphic below) continue to improve with increasing weight loss, suggesting that higher efficacy combinations like Novo Nordisk’s semaglutide and cagrilintide (predicted to approach 25% weight loss) will even more greatly improve outcomes.

  • Following symposium presentations, we were particularly struck by a lively post-symposium Q&A that mirrored the talk we’ve been hearing for a while in the field, around which some clinically relevant questions arose, specifically, who should be prescribed what: (i) should people with type 2 diabetes and obesity first start at the 2.4 mg semaglutide dose indicated for obesity (presumably without reimbursement) or 1.0 mg dose for diabetes (with an unasked query, should they take a higher dose somewhere in between and try to get prescriptions filled in 25 days rather than 30); and (ii) whether people newly diagnosed with diabetes should immediately start semaglutide 2.4 mg (rather than traditional first line therapy). As to be expected, opinions on these questions varied greatly, with Dr. Rosenstock strongly in favor of the 2.4 mg dose in both situations. Clearly, while the reimbursement is a major question, we’re very interested in, um, what people with diabetes and obesity should do, particularly cohorts of newly diagnosed, very meaningful and unresolved weight issues, etc.

    • In the first clinical scenario, Dr. Rosenstock asked presenters to choose whether they would prescribe 2.4 mg or 1.0 mg semaglutide to a person with type 2 diabetes, a BMI of 32 to 33, and A1c of 7.5 to 8%. Citing data from STEP-2, Dr. Davies voiced her support for the 2.4 mg dose if the “clinical priority is to maximize weight loss;” that said, she noted that for some patients, the 1.0 mg dose may be “reasonable” to get “good glycemic control, CV protection, and achieve some weight loss.”

      • Dr. Wilding largely agreed with her stance but emphasized that in clinical practice, many individuals will likely want to first try the 1.0 mg dose and only uptitrate if weight and glycemic parameters are not met.

      • Comparatively, Dr. Rosenstock seemed very much in favor of starting the 2.4 mg dose in order to ‘hit the BMI hard’ and calling for a ‘paradigm change’ in how dose escalation is approached.

      • More in line with Dr. Wilding, Dr. Kaplan emphasized that learning to clinically nuance obesity drugs moving forward will be key. He stated, “I would also think that we’re going to see a broader changing paradigm. If we think about how we treat other diseases, we don’t always go to the maximum dose that’s available. We go to the dose necessary to achieve the clinical endpoint we want. You don’t always use the maximum dose of metformin or an anti-hypertensive. As these [obesity] medications become more powerful, what we’ll be able to do is customize therapies to individuals. There will be some people in whom you’ll have to go to 2.4 mg to get the repones you want, but even for the weight loss response, the variability from patient to patient suggests that there will be some patients where 1.0 mg or less or somewhere in between is going to be enough. One of the challenges is going to be to learn how to clinically nuance our therapy the way we do for other diseases.”

      •  To this, Dr. Rosenstock again rallied for a major paradigm shift stating that Dr. Kaplan’s methodology was the “usual thinking” and suggested that the fact that ~1/3 of patients would see a 20% reduction in body weight on the 2.4 mg dose but not the 1.0 mg dose should serve as sufficient rationale for starting higher.

    • Very notably, Dr. Rosenstock again supported the higher dose of semaglutide when discussing a person with a BMI of 31, newly diagnosed with type 2 diabetes, suggesting that the drug should be used as a first line therapy (in place of metformin). This was the first time we heard a KOL speak so strongly publicly in support of first tackling obesity with a higher dose GLP-1, to potentially reverse type 2 diabetes – typically we’ve heard concerns about reimbursement for what is not just a slightly higher dose but a far higher dose. We’ve also heard concerns about know-how related to titration and we look forward to hearing what dQ&A CEO Richard Wood can tell us about adherence, dosing, GLP-1 titration without help from trial PIs, etc. Dr. Wilding agreed that this would be a “fantastic approach” for patients looking to put their diabetes in remission, particularly in the UK where data from the DiRECT trial, supporting diabetes remission with a low energy diet, is well known. Dr. Davies added that the strategy might also be useful for helping people with newly diagnosed type 2 diabetes who have managed to lose weight via lifestyle medications maintain the weight loss, as this is often quite challenging. Dr. Kaplan once again brought a balanced conclusion to the table, stating, “We now have opportunity to try out these various approaches. If we’re serious about treating obesity like a disease, we have to figure out the best pathway forward. Clearly, it’s going to be different in different people. However, there are a lot of stakeholders in this business. It may turn out that the approach you’re referring to is best used in populations where the severity of comorbidities, likelihood of poor CV outcomes, or even just cost of care is greater, so you would – in those situations – jump to more intensive therapy like this. But of course, all of this will have to be worked out amongst the payers, governmental organizations, patients, and providers themselves.”

    • When asked if he expects a decrease in bariatric surgery with the launch of semaglutide 2.4 mg, Dr. Kaplan shared that he initially does not expect much of a change, given that so few people receive either bariatric surgery/obesity medication and that 90% of people in the US who receive bariatric surgery are self-referred. These self-referred patients will likely now choose themselves whether they prefer chronic medication vs. a single procedure, especially regarding relative risks and costs. Over the next couple of years, however, Dr. Kaplan, in fact, expects an increase in bariatric surgery, as more people will be “taking obesity seriously” because they are benefiting from drugs like semaglutide 2.4 mg.

SURPASS Symposium: KOLs Debate GLP-1 vs. GIP Contributions to Tirzepatide’s Standout Weight Loss and A1c Efficacy, Position Tirzepatide and Semaglutide 2.4 mg as Equally “Excellent Molecules”

To top off tirzepatide’s impressive showing at this year’s ADA, a Tuesday morning symposium chaired by the eminent Dr. Alice Cheng (Trillium Health Partners) took a deep dive into results from four phase 3 trials for Lilly’s dual GLP-1/GIP receptor agonist:

  • SURPASS-1 (tirzepatide vs. placebo in people inadequately controlled with diet and exercise alone), presented by Dr. Carol Wysham (Multicare Rockwood Clinic);

  • SURPASS-2 (tirzepatide vs. semaglutide 1.0 mg as an add-on to metformin), presented by Dr. Juan Frias (National Research Institute);

  • SURPASS-3 (tirzepatide vs. insulin degludec as an add-on to metformin and/or SGLT-2 inhibitors), presented by Prof. Francesco Giorgino (University of Bari Aldo Moro);

  • and SURPASS-5 (tirzepatide vs. placebo in people inadequately controlled with insulin glargine with or without metformin), presented by Dr. Michelle Welch (Diabetes and Metabolism Specialists).

No doubt, it’s been a busy few days for this important compound, particularly on the publications front – results from SURPASS-2 were published in NEJM with an accompanying editorial, “Breaking New Ground with Incretin Therapy in Diabetes,” by Dr. Katherine Tuttle, and results from SURPASS-1 were published in The Lancet, along with commentary, “Tirzepatide and the new era of twincretins for diabetes” by Dr. Tricia Tan. While a more detailed review of the data was certainly helpful (results below), we particularly appreciated the lively post-presentation Q&A session and introduction by Dr. Daniel Drucker (Mt. Sinai), who discussed the evolving understanding of GLP-1 and GIP. At present, there is still much to be learned about the true effects of GIP in humans. As explained by Drucker, GIP is currently thought to act on adipose tissue (increasing LPL activity and adipose tissue), bones (decreasing resorption), alpha cells (modulating glucagon secretion), beta cells (increasing insulin secretion), and potentially the brain by decreasing food intake. Dr. Drucker emphasized, however, that human studies of GIP alone have only delivered “modest” results on blood glucose.

As such, there is still significant interest in determining the GLP-1 vs. GIP contributions from tirzepatide. During Q&A, the question of whether a “magic ratio” of GLP-1/GIP exists emerged, given middling results from other GLP-1/GIP co-agonists and GIP alone, which speakers acknowledged is likely a key question of interest amongst drug developers. Dr. Drucker was the most steadfast in his answer, believing that tirzepatide is “predominantly a GLP-1 receptor agonist;” He stated, “Until someone proves to me otherwise, I think it’s the ability of tirzepatide to interact at the GLP-1 receptor in a special manner that’s predominant in its superlative activity. I can’t rule out a contribution from GIP, particularly on glucose control, but we really need more additional human data.” Dr. Drucker further noted that determining if tirzepatide has additional GIP action “will be hard to prove.” In the future, use of biomarkers of GIP receptor activation – for example, something related to GIP’s effects on bone – may be able to shed more light onto this important concept.

In another thought-provoking moment during Q&A, Dr. Cheng asked the speakers if they would prefer prescribing semaglutide 2.4 mg or tirzepatide (once approved) to their patients, given such strong data for both molecules. Dr. Carol Wysham pragmatically responded that the decision would likely be determined by formularies in the US, however, she also emphasized that factors like the pen’s design and how easily it could be used to titrate up slowly or in a personalized manner would impact her decision. Overall, she finds both molecules to be “outstanding,” and hopes a head-to-head trial will be conducted – both sentiments that the panel seemed to collectively agree on.

Multiple audience members in the chat asked speakers questions regarding the risk of medullary thyroid cancer, retinopathy, and pancreatitis with tirzepatide, which session speakers tried to quell. Dr. Frias summarized that there was no signal of risk for any of these conditions across the four trials (though patients with active retinopathy were not included). An impassioned Dr. Drucker also chimed in to say that he hoped to “bury” the two concepts of medullary thyroid cancer risk and pancreatitis risk, which have emerged due to FDA precaution. Dr. Drucker re-emphasized that while GLP-1 receptors are expressed on thyroid C cells in certain animal models, they are not present on C cells in primates or humans. Exenatide has also been in the clinic for over 16 years, and long acting GLP-1s for ten years, with no evidence of medullary thyroid cancer. Similarly, pancreatitis has not been identified as an issue across the plethora of existing GLP-1 CVOTs. While elevated amylase or lipase levels may be present, this is simply “GLP-1 talking to acinar cells” rather than a sign of pancreatitis. Instead, Dr. Drucker hopes more attention will be paid to the “true adverse events” of incretin agonists – the GI side effects.

Further research on whether patients who experienced GI side effects also experienced the most weight loss is also needed. As explained by Prof. Giorgino, in theory, adverse effects like nausea and vomiting could reduce food intake and thus lead to weight loss. Prof. Giorgino, however, does not think this is likely the reason for tirzepatide’s unparalleled weight loss, given that a much larger proportion of participants experience substantial weight loss than the population that experienced GI side effects. Read on for a detailed summary of trial results:

SURPASS-1

Dr. Wysham expertly reviewed results from the SURPASS-1 trial, which revealed superior A1c reductions and weight loss across all three tirezpatide doses vs. placebo. SURPASS-1 enrolled people with type 2 diabetes and an A1c between 7.0 to 9.5%, BMI ≥23, and who were naïve to type 2 injectable therapy, randomizing them to tirzepatide 5 mg (n=121), 10 mg (n=121), 15 mg (n=121), or placebo (n=115). Average duration of diabetes was 4.7 years, A1c 7.94%, and BMI 31.9. When taken as intended, tirzepatide conferred the following at 40 weeks:

  • A1c change: -1.87% (5 mg), -1.89% (10 mg), -2.07% (15 mg), +0.04% (placebo)
  • Percent of participants achieving A1c <7%: 87% (5 mg), 92% (10 mg), 88% (15 mg), 20% (placebo)
  • Percent of participants achieving A1c <5.7%: 34% (5 mg), 31% (10 mg), 52% (15 mg), 1% (placebo)
  • Weight reduction: -7.0 kg (-7.9%, 5 mg), -7.8 kg (-9.3%, 10 mg), -9.5 kg (-11.0%, 15 mg), -0.7 kg (-0.9%, placebo)
  • Percent of participants achieving ≥5% weight loss: 67% (5 mg), 78% (10 mg), 77% (15 mg), 14% (placebo)
  • Percent of participants achieving ≥10% weight loss: 31% (5 mg), 40% (10 mg), 47% (15 mg), 1% (placebo)
  • Percent of participants achieving ≥15% weight loss: 13% (5 mg), 17% (10 mg), 27% (15 mg), 0% (placebo)

Adverse events were primarily related to the GI system (nausea, diarrhea, dyspepsia, vomiting). Nausea was largely mild or moderate across all three tirzepatide arms and generally subsided as time went on. Importantly, no severe or clinically significant hypoglycemia (BG <54 mg/dl) occurred with tirzepatide.

SURPASS-2

Dr. Frias detailed the results of the highly anticipated SURPASS-2 trial, showing superior A1c reductions and weight loss for Lilly’s GLP-1/GIP dual agonist tirzepatide over Novo Nordisk’s semaglutide 1.0 mg. SURPASS-2 enrolled people with type 2 diabetes and an A1c between 7.0% and 10.5% on metformin monotherapy, randomizing them to 5 mg (n=471), 10 mg (n=469), or 15 mg (n=470) tirzepatide or semaglutide, the current best-in-class GLP-1 agonist (n=469) for 40 weeks. Average duration of diabetes was 8.6 years, A1c 8.28%, and BMI 34.2. Very impressively, when taken as intended, all three tirzepatide doses showed superiority to semaglutide for the following key endpoints:

  • A1c reduction: -2.09% (5 mg), -2.37% (10 mg), -2.46% (15 mg), -1.86% (semaglutide)
  • Weight reduction: -7.8 kg (-8.5%, 5 mg), -10.3 kg (-11.0%, 10 mg), -12.4 kg (-13.1%, 15 mg), -6.2 kg (-6.7%, semaglutide)
  • Percent of participants achieving A1c <7%: 85% (5 mg), 89% (10 mg), 92% (15 mg), 81% (semaglutide)
  • Percent of participants achieving A1c <5.7%: 29% (5 mg, not controlled for type 1 error), 45% (10 mg), 51% (15 mg), 20% (semaglutide)
  • Percent of participants achieving >5% weight loss: 69% (5 mg), 82% (10 mg), 86% (15 mg), 58% (semaglutide)
  • Percent of participants achieving >15% weight loss: 15% (5 mg), 28% (10 mg), 40% (15 mg), 9% (semaglutide)

The safety profile of tirzepatide was consistent with that of the GLP-1 class, featuring transient mild-to-moderate GI side effects and low incidence of hypoglycemia. SURPASS-2 also seemed to have an imbalance of deaths in the tirzepatide arm (four in the 5 mg arm, four in the 10 mg arm, and three in the 15 mg arm vs. one in the semaglutide arm), which Dr. Frias largely chalked up to COVID-19-related deaths and existing CVD (though some audience members still seemed to have questions about the imbalance). Overall, these results represent unparalleled A1c and weight declines. However, questions remain regarding the exact mechanisms by which the unimolecular dual agonist is acting, how to best manage tolerability concerns, and the real-world ramifications of achieving A1cs <5.7%. For highly informative perspectives on these quandaries, see our coverage of the SURPASS-2 topline results announcement, featuring commentary from Prof. Filip Knop (Director of Center for Clinical Metabolic Research, Gentofte Hospital, University of Copenhagen) and Prof. Philip Home (Emeritus Professor, Translational and Clinical Research Institute, Newcastle University), as well as our interview with Dr. Frias.

SURPASS-3

Prof. Giorgino reviewed results from SURPASS-3, supporting the use of tirzepatide as a “very compelling option for patients with type 2 diabetes who need to intensify their treatments with an injectable therapy.” Indeed, SURPASS-3 enrolled people with type 2 diabetes, A1c between 7.0 and 10.5%, naïve to insulin therapy, and on a stable dose of metformin, with or without an SGLT-2 inhibitor, randomizing participants to tirzepatide 5 mg (n=359), 10 mg (n=361), 15 mg (n=359), or insulin degludec (n=365). Average duration of diabetes was 8.4 years, A1c 8.17%, and BMI 33.5; 68% were on metformin alone, and 32% were on both metformin and an SGLT-2 inhibitor. When taken as intended, tirzepatide delivered the following benefits over 52 weeks:

  • A1c change: -1.93% (5 mg), -2.20% (10 mg), -2.37% (15 mg), -1.34% (degludec)
  • Percent of participants achieving A1c <7%: 82% (5 mg), 90% (10 mg), 93% (15 mg), 61% (degludec)
  • Percent of participants achieving A1c <5.7%: 26% (5 mg), 39% (10 mg), 48% (15 mg), 5% (degludec)
  • Weight reduction: -7.5 kg (-8.1%, 5 mg), -10.7 kg (-11.4%, 10 mg), -12.9 kg (-13.9%, 15 mg), +2.3 kg (+2.7%, degludec)
  • Percent of participants achieving ≥5% weight loss: 66% (5 mg), 84% (10 mg), 88% (15 mg), 6% (degludec)
  • Percent of participants achieving ≥10% weight loss: 37% (5 mg), 56% (10 mg), 69% (15 mg), 3% (degludec)
  • Percent of participants achieving ≥15% weight loss: 13% (5 mg), 28% (10 mg), 43% (15 mg), 0% (degludec)

Importantly, the trial protocol called for a relatively “intensive” insulin titration protocol, making the A1c superiority seen with tirzepatide even more impressive. As evidenced in the graphic below, participants in SURPASS-3 used significantly larger amounts of insulin vs. other recent GLP-1RA vs basal insulin trials.

Side effects were again primarily GI-related, with the highest rates in nausea: 11.5% at 5 mg, 22.5% at 10 mg, 23.7% at 15 mg, and 1.7% with degludec. Again, rates of nausea were largely transient, however, and risk of hypoglycemia was low. 

SURPASS-5

Closing out the SURPASS program, SURPASS-5 investigated the safety and efficacy of tirzepatide in combination with basal insulin therapy, demonstrating significant reductions in both A1c and body weight for the GIP/GLP-1 dual agonist vs. placebo when used as an add-on to insulin glargine. The trial enrolled people with type 2 diabetes and an A1c between 7.0% and 10.5% on an insulin glargine therapy regimen (with or without metformin), randomizing them to 5 mg (n=116), 10 mg (n=119), or 15 mg (n=120) tirzepatide or placebo (n=120) for 40 weeks. Average duration of diabetes was 13.3 years, A1c 8.31%, and BMI 33.4; mean insulin glargine dose was 0.40 U/kg/day. All three tirzepatide doses conferred significant A1c and body weight reductions compared to placebo:

  • A1c reduction: -2.23% (5 mg), -2.59% (10 mg), -2.59% (15 mg), -0.93% (placebo)

  • Percent of participants achieving A1c <7%: 93% (5 mg), 97% (10 mg), 94% (15 mg), 34% (placebo)

  • Percent of participants achieving A1c <5.7%: 26% (5 mg), 48% (10 mg), 62% (15 mg), 3% (placebo)

  • Weight reduction: -6.2 kg (-6.6%, 5 mg), -8.2 kg (-8.9%, 10 mg), -10.9 kg (-11.6%, 15 mg), +1.7 kg (+1.7%, placebo)

  • Percent of participants achieving ≥5% weight loss: 54% (5 mg), 65% (10 mg), 85% (15 mg), 6% (placebo)
  • Percent of participants achieving ≥10% weight loss: 23% (5 mg), 47% (10 mg), 51% (15 mg), 1% (placebo)
  • Percent of participants achieving ≥15% weight loss: 7% (5 mg), 27% (10 mg), 32% (15 mg), 0% (placebo)

While the 5 and 10 mg doses of tirzepatide did not confer a reduction on daily insulin glargine dose, participants in the 15 mg arm decreased their insulin needs by 11% vs. placebo.

Treatment-emergent adverse events were numerically higher in the tizepatide arms and included diarrhea, nausea, decreased appetite, and vomiting, as is typical of incretin therapies. The incidence of clinically significant hypoglycemia (blood glucose <54 mg/dl) was consistent between the tirzepatide and placebo groups. Overall, these results point to tirzepatide as a promising option as an add-on to basal insulin therapy, conferring robust improvements in glycemic control and body weight without increasing the risk of hypoglycemia.

SURPASS-1 Monotherapy Trial Results for Lilly’s GLP-1/GIP Tirzepatide in Type 2 Diabetes Published in The Lancet

Keeping the ball rolling on tirzepatide publications (check out SURPASS-2 in NEJM from yesterday), Lilly today announced that full results from the SURPASS-1 trial, assessing dual GLP-1/GIP receptor agonist tirzepatide in adults with type 2 diabetes, have just been published in The Lancet. Publication of SURPASS-1 was timed with an ADA President Select Abstract’s presentation of the full results by Dr. Julio Rosenstock (Dallas Diabetes Center). We note that SURPASS-1 was published in the broader Lancet journal (rather than The Lancet Diabetes & Endocrinology) – another clear signal of the significance of the data (in case any is needed). As reported in topline results released in December 2020, tirzepatide met all its primary A1c and weight endpoints in the trial, and audience members seemed particularly enthused by the unprecedented A1c declines seen in SURPASS-1.

  • Results. By efficacy estimand, defined as efficacy prior to discontinuation of study drug or initiating rescue therapy for persistent hyperglycemia, tirzepatide delivered:

    • A1C change: -1.87% (5 mg), -1.89% (10 mg), -2.07% (15 mg), +0.04% (placebo)

    • Weight reduction: -7.0 kg (-7.9%, 5 mg), -7.8 kg (-9.3%, 10 mg), -9.5 kg (-11.0%, 15 mg), -0.7 kg (-0.9%, placebo)

    • Percent of participants achieving A1c <7%: 87% (5 mg), 92% (10 mg), 88% (15 mg), 20% (placebo)

    • Percent of participants achieving A1c <5.7%: 34% (5 mg), 31% (10 mg), 52% (15 mg), 1% (placebo)

    • Change in FSG: -43.6 mg/dL (5 mg), -45.9 mg/dL (10 mg), -49.3 mg/dL (15 mg), +12.9 mg/dL (placebo)

  • Results. By treatment regimen estimand, defined as efficacy irrespective of adherence to the investigational medicine or introduction of rescue therapy for persistent severe hyperglycemia, tirzepatide conferred:

  • A1c reduction: -1.75% (5 mg), -1.71% (10 mg), -1.69% (15 mg), -0.09% (placebo)

  • Weight reduction: -6.3 kg (5 mg), -7.0 kg (10 mg), -7.8 kg (15 mg), -1.0 kg (placebo)

  • Percent of participants achieving A1c <7%: 82% (5 mg), 85% (10 mg), 78% (15 mg), 23% (placebo)

  • Percent of participants achieving A1c <5.7%: 31% (5 mg), 27% (10 mg), 38% (15 mg), 1% (placebo)

  • Change in FSG: -39.6 mg/dL (5 mg), -39.8 mg/dL (10 mg), -38.6 mg/dL (15 mg), +3.7 mg/dL (placebo)

  • A press release published by Lilly emphasized that the majority (54.2%) of trial participants were treatment-naïve and were relatively early in disease progression. At baseline, mean duration of diabetes as 4.7 years, baseline A1c was 7.9%, and baseline BMI was 31.9 kg/m2

  • Similar to a post-hoc analysis conducted for SURPASS-2, an analysis from SURPASS-1 revealed promising benefits on fasting lipids. At the 15 mg dose, tirzepatide reduced total cholesterol by 8.4%, triglycerides by 21%, LDL cholesterol by 12.4%, VLDL cholesterol by 19.8%, and increased HDL “good” cholesterol by 7.5% from baseline. We believe many experts will be wondering to what degree these lipid benefits serve as a signal of early evidence for tirzepatide’s SURPASS-CVOT.

  • Treatment discontinuation was fairly substantial at the 15 mg dose (21.5%) vs. 9.9% at 10 mg, 9.1% at 5 mg, and 14.8% at placebo – particularly since these patients are titrating with the experts. That said, Dr. Rosenstock emphasized that treatment discontinuation due to adverse events was low, peaking at 6.6% at the 15 mg dose. Overall safety results were in line with the overall GLP-1 class, with the most common side effects being gastrointestinal. Hypoglycemia rates were relatively low, though Level 1 hypoglycemia was elevated across all three arms of tirzepatide vs. placebo.

The tirzepatide train chugs on – in a late-breaking poster presentation (81-LB), Dr. Kristina Boye (Eli Lilly) unveiled promising but mixed patient reported outcomes from the SURPASS-1 trial, investigating dual GLP-1/GIP receptor agonist tirzepatide in people with type 2 diabetes. SURPASS-1 participants who received at least one dose of study drug (“modified intent-to-treat population”) were evaluated on the following measures at baseline and Week 40: (i) EQ-5D-5L, a five-dimension measure of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression; (ii) EQ-VAS, a vertical visual analog scale from “worst possible” to “best possible” health, (iii) Impact of Weight on Self-Perception Questionnaire (IW-SP); and (iv) Ability to Perform Physical Activities of Daily Living Questionnaire (APPADL).

  • No statistically significant difference in mean EQ-5D-5L change from baseline was identified between tirzepatide dose and placebo, despite all three tirzepatide doses demonstrating an improvement from baseline. Similarly, tirzepatide did not confer significantly improved APPADL scores vs. placebo at any dosage; To note, APPADL measures qualities like “getting up from the floor,” “household chores,” and “moderate” or “strenuous physical activity.”

  • On the more positive side, participants on all three tirzepatide doses experienced (5, 10, and 15 mg) indicated better overall health-related quality of life gains vs. placebo, as measured by EQ-VAS (p<0.05).

  • Tirzepatide 10 mg and 15 mg also significantly improved IW-SP score vs. placebo, indicating improved measures of body weight self-perception (“feel unhappy with appearance due to weight,” “feel self-conscious in public due to weight,” “feel unhappy due to comparing weight with others”). No statistically significant improvement was seen with the 5 mg tirzepatide dose. For reference, when taken as intended, tirzepatide conferred weight reductions of -7.0 kg (-7.9%, 5 mg), -7.8 kg (-9.3%, 10 mg), and -9.5 kg (-11.0%, 15 mg) vs. -0.7 kg (-0.9%) on placebo. Given that SURPASS-1 was not intended to be a weight loss trial, we find these early results to be quite promising.

  • Overall, we commend Lilly for publishing this important PRO data to supplement the A1c and weight benefits seen on paper. In terms of study limitations, the authors highlighted the fact that measures like EQ-5D-5L only apply to select health-related quality of life outcome. More patients in the placebo arm also had their postbaseline measurements taken on rescue therapy, and thus were excluded from analysis, which may have introduced some bias.

Novo Nordisk’s Semaglutide 2 mg Outperforms 1 mg Dose Across Baseline A1cs and Body Weights in Type 2 Diabetes; SUSTAIN FORTE Full Results in Press at The Lancet Diabetes & Endocrinology

Also on the incretin front, we caught the full results presentation for Novo Nordisk’s SUSTAIN FORTE trial, investigating semaglutide 2 mg vs. 1 mg in type 2 diabetes, presented by Dr. Juan Frias (National Research Institute). While the data has not yet been published in an academic journal, Novo Nordisk today announced that the study is currently in press at The Lancet Diabetes & Endocrinology. As a reminder, positive topline results for the trial were originally released in November 2020, and semaglutide 2 mg is now under FDA review following resubmission in May 2021. When taken as intended (i.e., participants fully adhering to treatment, not initiating other type 2 therapies), semaglutide 2 mg (n=480) conferred a significant 2.2% reduction in A1c from 8.9% at baseline vs. 1.9% with semaglutide 1 mg (n=481) (p=0.0003); this result was slightly moderated when using the treatment policy estimand (the effect of getting the drug prescribed), resulting in a 2.2% reduction in A1c for semaglutide 2 mg vs. 1.9% with semaglutide 1 mg (p=0.0098).

  • Today’s results presentation also included a variety of positive post-hoc analyses on A1c and body weight that reinforced semaglutide 2 mg’s efficacy as a higher dose option. When taken as intended, ~52% of participants achieved an A1c ≤6.5% vs. ~39% in the placebo arm. On body weight, ~59% of participants experienced a weight loss of ≥5% on semaglutide 2 mg (vs. 51% on semaglutide 1 mg), and ~28% experienced weight loss ≥10% (vs. 23% on semaglutide 1 mg). Weight loss and A1c benefits were also consistently superior at the 2 mg dose across baseline A1cs (<9% vs. ≥9%) and BMI (<35 vs. ≥35).

  • In terms of baseline characteristics, participants in the 2 mg arm had an average diabetes duration of 9.2 years, baseline A1c of 8.9%, and BMI of 34.8 kg/m2. Safety and tolerability were comparable between the two doses, with a small proportion (<5%) of treatment discontinuation due to GI AEs.

  • We’re happy to see the results from SUSTAIN FORTE, and we’re glad to see the database growing to support semaglutide as a relatively safe, well-tolerated, and more efficacious high dose GLP-1. As well, we’re very moved to have a trial of nearly 1,000 people with T2D, over half of whom are on SFUs, and to bring the focus back to why anyone should be on SFUs at all if they have A1cs over 7%. This said, we would have liked to have heard even more about the negativity surrounding SFUs and negativity around how there are people with diabetes who are clearly under-treated. This should bring some focus back, we hope, to, therapeutic inertia and how HCPs can get more support for coverage since we know how much time and administrative burden some HCPs have, even when their patients do not get the therapy for which the HCPs advocate. Looking ahead, we anticipate HCPs may escalate patients through the 1 mg, 2 mg, and 2.4 mg doses of semaglutide (with the latter being indicated for obesity under brand name Wegovy) if coverage is not an issue, though we’ll be curious to see if guidelines start to recommend higher doses at earlier stages of disease progression. With this kind of published data against SFUs (even though there was not a placebo arm), we hope the hammer can come down on medicines that simply aren’t doing the job they should. SFUs work for some, but with upwards of 50% of those with T2D not at their glycemic target, we can’t imagine what the overall glycemic control looks like for those on SFUs. NEJM did show a couple of weeks back that still fully 25% of those with T2D are on SFUs in the US (see Table 2) – they haven’t yet responded on questions about sub-group analysis. Those on GLP-1 or an SGLT-2 had reached the whopping adoption rate of 7% by 2018, truly unfortunate for a class that was introduced commercially in 2006.

 

Our Conversation with Dr. Michael Radin (Executive Director, Diabetes Cardio Renal Medical Director Team, Medical Affairs, Novo Nordisk)

Kelly Close: Hi Dr. Radin! We are so excited to speak with you!

Dr. Michael Radin: Hi Kelly! I’m very excited to get to talk to you as well. I appreciate all the work that you do. I know this is a very busy time for you as we’re all trying to listen in on the scientific sessions.

Kelly: Please, the privilege is really all mine. There's so much going on in the world of diabetes, and we do try to cover Novo Nordisk as closely as we can. It’s amazing to have the opportunity. And, again, thank you for the support of everything that Novo Nordisk does.

Dr. Radin: We definitely appreciate it. I think it's nice when we align, as both your organization and ours are trying to do best for patients. That's what it's all about at the end of the day.

On SUSTAIN FORTE

Kelly: Yes, absolutely! Well, to start, could you tell us how you are feeling about SUSTAIN? What are the important pieces that you would like to make sure are out there? Do you feel like there's anything that people aren't getting?

Dr. Radin: Yes, of course. Generally speaking, I think what was very exciting is that we were able to share data on diabetes, obesity and cardiovascular care overall in people with type 2 diabetes. And that’s really great because it also represented a shift in our focus going beyond just diabetes to all sorts of chronic cardiometabolic conditions. So, with that being said, I think what we are most excited about is our SUSTAIN Forte data.

Why do we think this data is so important? Well, according to the latest data from the National Health and Nutrition Examination survey, half the people in this country still aren’t reaching their glycemic goal. And if we look at our semaglutide molecule and the SUSTAIN program more broadly, we see that 70 to 80 percent of patients were able to reach that general A1c target of less than 7 percent but that means there were still 20 to 30 percent of people who needed additional glycemic control. Hence, our goal in conducting the SUSTAIN Forte trial was to see if going to a higher dose of semaglutide could help more of those patients get to their glycemic target. And so, we’re excited because the highlights of our data show us that the 2-milligram dose compared to the 1-milligram dose does in fact demonstrate additional glycemic lowering along with additional weight lowering.

I would also like to highlight that this increase in dosage did not come at the expense of additional adverse events, particularly GI adverse events, which we know is most common for the GLP1 receptor agonist class. When you look at the rates of nausea in both the 2-milligram and 1-milligram arms, they’re nearly identical at just under 15 percent. And when you look at the number of patients with the stop therapy because of a GI event, both arms are quite similar at just around only 3 %. So that’s the data we're really excited about.

Kelly: Wow! That is really a big deal. Especially on the nausea. We actually just really learned about that from Richard Wood, who is one of our partners here. He runs a sister organization called dQ&A. I don't know if you have heard of them before, have you?

Dr. Radin: I’ve heard a little of it. I don't know a whole lot about it to be perfectly honest.

Kelly: Okay, well I’ll just give you a quick background because it's really relevant to this. dQ&A has a panel of about 18,000 patients predominantly with type 2 diabetes spread all over the world. It's in about 10 countries now, mostly located in and around central and northern Europe. And that's where we learned that the rates of nausea were just sky high with Byetta-- I mean well above 15% of patients-- and that the real-world experience was actually not that similar to some of those really early trials. But now we see that with the power of this molecule, you've gotten that under control. It's kind of magical, isn't it?

Dr. Radin: It really is. I know we're really excited about it. Especially to be able to help so many people living with type 2 diabetes. It's really wonderful.

On Patient-Doctor Conversations

Kelly: Is there anything you think would be relevant for people with diabetes to be able share with their healthcare teams when they get a chance to read this?

Dr. Radin: More broadly on that, or just about this?

Kelly: Yes. We're really just looking to see how patients can join the conversation and potentially suggest things to their doctor. And so, I was curious if there was anything about that that you would feel comfortable speaking about.

Dr. Radin: Yes, I think when I look at that in terms of the big scheme, we talk about the issue of therapeutic inertia, which in and of itself, of course, is not a new issue and an ongoing problem, again, reflected in much data. This data is just one example of that. And then you have COVID 19 and the horrific effect of this pandemic, particularly in people with chronic conditions like obesity and type 2 diabetes and cardiovascular disease, exactly what you pointed out. And so, if there's any silver lining to maybe try to take from this, I think it is the opportunity that COVID has had to shine a spotlight on these issues, almost as to say: look, therapeutic inertia is a huge problem; people with type 2 diabetes aren't getting to their glycemic targets; people with type 2 diabetes and cardiovascular disease aren't getting glycemic medications that can help lower the risk of a future event; people with obesity aren't getting the treatment that they need. And look what happens, unfortunately. Look at the higher risk that these people are under. And so, I hope this really drives physicians and healthcare providers more broadly to step up and take action, sort of a burning platform, if you will. Moreover, I hope it empowers patients to not settle and to go to their healthcare providers and ask what more they can do to get better.

Kelly: Yes. That is so great. I love seeing how well this is doing. I think that you are really paving the way for so many stakeholders to really care about obesity and to really care about type 2 diabetes.

On PIONEER Trial

Dr. Radin: I really appreciate that and in regards to what we were talking about earlier in glycemic control, I would even say, some of our data that we've generated for the ADA reflect that. I mean, for example, I could talk about one of our post-hoc analyses of oral semaglutide. We took the PIONEER 1 trial, which was looking at oral semaglutide in people with diabetes who were not on treatment yet– so just on diet and exercise– and we did a post-hoc analysis. These people on average had diabetes for three-and-a-half years of type 2 diabetes. In our analysis, we asked what happens if you treat them within a year or less of onset of their diabetes versus you waiting until more than a year. And what a significant cutoff if you look at A1c targets of less than eight, less than seven, even less than six near normal glycemia, substantially more patients reach those targets if you get to them within that first year. Again, the intent there is to say: look, you’ve got to do a better job not just addressing therapeutic inertia but getting to people early because it makes a difference.

Kelly: Yes. I love that. I just want to just pause for a second to make sure I got all of that.

Dr. Radin: Yes. Absolutely, I’ll pause as well...

On the Usage of Telehealth

Kelly: There is therapeutic inertia going on and there are also other kinds of inertia that are going on that play into it. And then there are also things that are just about the constraints on the medical system and what we expect of doctors and their healthcare teams.

And I don't know if you might have any comment on that, but I'm just sort of curious about the pandemic. Now people are using telehealth. We had that one weird moment in the middle where doctors were like, nope, nope, nope, nope, nope. We're going back. We're going back. We need the facility fee. We're going back.

We're not going to do telehealth. That was like for one minute I felt like and now I feel like most of the places in the US have acknowledged, okay, at least we will be using that some of the time. But, just broadly I mean, I'm so grateful that all these people are doctors and that they give up a lot too.

Dr. Radin: I think two comments come to mind around that, one is absolutely with telehealth and especially for chronic conditions like type 2 diabetes and even obesity. What's most important is to talk to the patient. To be able to talk to the patient, to have that time to take a good history.

To understand what is happening in their lives, that is impacting their care so you can do good shared decision making together. And telehealth is a wonderful platform for that because, you don't have the barrier of the patient having to make the trip to the office if something came up at the last minute. All of those barriers go away with telehealth.

It's a wonderful tool to supplement. I think you still need some inpatient visits. Of course. You still need some physical exam there. But what a wonderful tool. And you're right, that's a system issue. And that I think many healthcare providers have embraced some certainly haven't, but I think the majority have.

And it comes down to reimbursement. Will they get reimbursed for it? Because if they do, then I think they embrace it as well. I know a lot of colleagues that I have that really cherish being able to -- I’m not going into the office. I can do a whole day of telehealth and I’m taking care of my patients and talking to them and it’s wonderful.

I think that does address a barrier for people. And the other one, I just think it's multidisciplinary care, team-based care. To expect one physician to go it alone with all the data and all the information and everything else that's out there, it's too much.

And when you formulate these multidisciplinary teams that can come together to look at the patient and address the patient holistically, that's also, I think, another way to get around some of the barriers and issues in our current healthcare system.

On Sulfonylureas

Kelly: Yes. I know we only have five minutes left. Can we spend that time talking about sulfonylureas? I'm going to try to have my blood pressure just stay at a normal level. I mean, honestly, why is anyone even prescribed sulfonylurea.

Dr. Radin: I don't know that I have the right answer for you because we're in the same camp as you and especially again on days like today when you have medications for people living with type 2 diabetes like semaglutide.

For example, where you can get better glycemic control with a benefit of weight loss as opposed to weight gain without having a significantly increased risk of hypoglycemia as opposed to having an increased risk of hypoglycemia. And then again, if it's -- oh, well, sulfonylurea, at least it's a pill.

Well, now there's an oral GLP1 receptor agonist available, too. That excuse there. There's so much wonderful evidence that points to better available therapies. I'm with you. I wish I had an answer for you. Those who are motivated to continue to do so.

Kelly: Well, I wonder about maybe if there could just be more publications. I’m trying to remember for all of SUSTAIN what has been published already.

Dr. Radin: Well, I mean, we have, in all of SUSTAIN programs and even within the PIONEER programs. Multiple studies have looked head-to-head versus sulfonylureas, but they address sulfonylurea in terms of many of the background therapies are just accepting the fact that so many people out there are on a sulfonylurea and knowing that healthcare providers could feel confident that they could do that and then maybe even withdraw the sulfonylurea because of the response that they’re seeing. Even SUSTAIN FORTE, for example, all patients who are on a background of metformin, they could get back of sulfonylurea and actually half of them work.

Kelly: Yes. And do you have specifically the subgroup analysis of the people on the sulfonylurea part of it?

Dr. Radin: I don't have that. That actually hasn’t been performed yet on SUSTAIN FORTE. I don't have that. And I’d have to look up if we've done the post talk analysis for some of the others. We've done it. I know, for example, across the PIONEER program, we looked at responses across just various background therapies.

And the bottom line there is that whether they’re on metformin, sulfonylurea or even SGLT2 inhibitors, we saw a consistent effect of the molecule.

Kelly: Well, thank you so much for your time. I don't want to overextend, but this is very, very helpful. I love this program. I mean, to really be able to show to us what is happening. And we salute you as always and thank you for what you're doing.

Dr. Radin: Well, same here and thank you for taking time to meet with us during your busy schedule, and it's really a pleasure to speak with you. I was really excited for this interview. I’m very happy to meet you virtually, I look forward to when we can do this in person.

Kelly: Okay. I appreciate it so much. Thank you so much. And take very good care, we’ll talk to guys soon. Goodbye.

 

AZ GLP-1/Glucagon Receptor Agonist Cotadutide Delivers +14.8% Time in Range and Early Evidence of Renal Protection in Phase 2 DKD Trial; Advanced to Phase 2b vs. Semaglutide 1.0 mg

We also caught promising phase 2a data on AZ’s dual GLP-1/glucagon receptor agonist cotadutide in diabetic kidney disease (n=41). Poster 674-P highlighted the randomized, double-blind, placebo-controlled trial, which recruited patients with type 2 diabetes (A1c 6.5-10.5%), Stage 3 CKD (eGFR 30-59 mL/min/1.73 m2), and a BMI of 25-45 kg/m2, receiving insulin and/or oral glucose-lowering drugs. On the trial’s primary endpoint, mixed meal tolerance tests (MMTTs) at Day 32 revealed a -27% change from baseline in plasma glucose AUC0-4h with cotadutide vs. +3.7% change with placebo (p=0.001). Notably, CGM was used throughout the trial, and results are being posed as secondary outcomes. Cotadutide also conferred impressive improvements on Time in Range: +14.8% (+3.5 hours/day) change from baseline; while the results indicate a 21% Time in Range reduction for placebo (p<0.001), given the small size, it’s hard to make much of his. While most would wonder about the pronounced reduction in the placebo group, we believe most experts would just look to see what happened with TIR with a larger group – this was also a relatively short trial. We’re more interested in what change the treated group saw, 15% better TIR, and to see if that outcome was sustained in a larger group and for a longer time. We are also curious if GMI was measured and tracked – that would be a corroborating metric. Notably, the cotadutide group also had meaningful reductions in body weight (-3.69% from baseline vs. -0.21% with placebo, p<0.001) and decrease in total daily insulin dose in subjects receiving ≥20 U baseline dose (35.2% reduction in dose for cotadutide; n = 14; p = 0.012). These body weight and insulin dose reduction differences in cotadutide and placebo appeared to rebound during the follow-up period.

  • Notably, the phase 2 trial provided early evidence that cotadutide may have renoprotective effects, as represented by reduced albuminuria. UACR was reduced by 51% vs. placebo at Day 32 in patients with baseline micro- or macroalbuminuria (n = 18; p = 0.0504). 

  • Safety and tolerability results were largely in line with the broader GLP-1 class. Treatment-related adverse events were higher in the cotadutide arm (71%) vs. 35% for placebo, largely driven by gastrointestinal side effects. We were also concerned to see one death in the cotadutide group due to diabetic ketoacidosis, and want to know more information on this. Though we assume many will wonder about the SAE, we assume it was deemed unrelated to cotadutide, given that the candidate has been advanced to phase 2b – the patients were fairly sick at baseline given their EGFR levels alone. Heart rate was also increased in the cotadutide group after 32 days of dosing (14.13 bpm vs 3.14 bpm for placebo; p<0.001).  

  • Overall, we’re looking forward to watching this compound and to see AZ advance an incretin-based therapy to mid-stage trials is exciting, considering the company’s current stronghold in SGLT-2 inhibitors with Farxiga (dapagliflozin) for those with and without diabetes. Noably, today’s poster confirmed that cotadutide is now being studied in a phase 2b study vs. Novo Nordisk’s semaglutide 1.0 mg in diabetic kidney disease, set to complete in March 2022. AZ is also studying cotadutide in NASH, with phase 2b data expected in 2H21 – we are interested to know what phase 3 would look like if it gets to that point since the assumption is that although liver disease is a very unmet need, NASH trials still need to be very comprehensive and show outcomes. While admittedly, this represents some conjecture, see December 19, 2020 Hepatology, “Nonalcoholic Steatohepatitis: Current Thinking From the Division of Hepatology and Nutrition at the Food and Drug Administration.”

AZ’s Cotadutide Phase 2 Results Show the GLP-1/Glucagon Dual Agonist Confers Weight Loss by Reducing Energy Intake

Dr. Rajna Golubic (Institute of Metabolic Science at the University of Cambridge) presented findings from an exploratory phase 2 trial of cotadutide in people with type 2 diabetes and with overweight and obesity (n=28). Though cotadutide is primarily being developed for treatment of NASH (with positive phase 2 data presented at ADA 2020), the positive weight loss effects presented at EASD 2018 and of GLP-1 agonists and glucagon individually served as the rationale for investigating this mechanism further. The randomized, double-blind, placebo-controlled, “short but intense” study aimed to determine if cotadutide’s weight loss effects were the result of reduced energy intake, increased energy expenditure, or a mix of both. Enrolled participants had A1c ≤ 8% and BMI > 28 ≤ 40 kg/m2. Individuals treated with cotagutide (n=18) lost significantly more weight than those on placebo (n=7) and consumed ~41% less during ad libitum lunch compared to placebo (p=0.011 for both analyses). According to least-squares means analyses, treatment with cotadutide did not lead to significantly different total energy expenditure or resting energy expenditure compared to placebo. Though total energy expenditure decreased, cotadutide still led to weight loss, and preservation of resting energy expenditure may suggest glucagon receptor agonism. Interestingly, treatment with cotadutide at doses up to 300 μg daily led to lower time spend in activity – which could explain lower total energy expenditures instead of weight loss being the sole driver for such. Dr. Golubic suggests that future studies could incorporate exercise programs to further parse out this relationship.

  • Dr. Golubic also reported safety data, with more participants taking cotadutide (94%) experiencing adverse events than those taking placebo (71%). Specifically, treatment with cotagutide up to 300 μg/day was associated with more nausea (72% vs 29%) and vomiting (22% vs 14%). While heartrate slightly increased in both treatment groups, those taking cotagutide showed a trend toward decreased systolic blood pressure compared to placebo. Overall, only two serious adverse events (coronary artery thrombosis, grade 3 hematoma muscle) occurred in patients taking cotadutide (n=1 for each event).

Exploratory PIONEER 2 Findings Demonstrate Long-Term A1c Control with Oral GLP-1 Rybelsus

A new analysis of PIONEER 2 data for Novo Nordisk’s Rybelsus (oral semaglutide) found that participants spent more time with an A1c <7.0% and <6.5% with the oral GLP-1 than with the study’s SGLT-2 comparator empagliflozin (Lilly/BI’s Jardiance). An A1c <7.0% was achieved for a mean of 27 out of 52 weeks with oral semaglutide, compared to 19 weeks with empagliflozin (for an A1c of <6.5%, the mean was 16 and 7 weeks, respectively). In total, 78% of the oral semaglutide group and 60% of the empagliflozin group had an A1c <7.0% at least once during the study, and 46% vs. 28% sustained that achievement for ≥38 weeks. The OR of having an A1c <7.0% at week 26 and 52 was 4.12 with the oral GLP-1 vs. empagliflozin (95% CI: 2.94-5.76, p<0.0001). Unsurprisingly, there was no difference in the rate of hypoglycemia between groups (≤2% had severe or confirmed hypoglycemic events). While patients in PIONEER 2 were relatively uncomplicated (participants were uncontrolled on metformin alone, and exclusion criteria included eGFR <60, for example), finding that roughly half of the Rybelsus arm achieved quality glycemic control for the majority of the treatment period with a once-daily tablet is highly encouraging. This exploratory analysis provided a novel look at the data from Novo Nordisk’s extensive phase 3 program for oral semaglutide, though assessment of “time in glycemic control” with serial A1c measurements may not tell the full story about daily glucose control with oral semaglutide. As ever, we’d love to see the relative impact of these therapies on Time in Range for people with uncomplicated type 2 diabetes. Still, there’s little doubt for us that an oral GLP-1 such as Rybelsus is or will become the most appealing and effective second oral medication for many people with type 2 diabetes.

  • As a reminder, PIONEER 2 (n=822, A1c 7.0-10.5%) demonstrated superior A1c lowering with oral semaglutide 14 mg compared to empagliflozin 25 mg, the higher dose of Jardiance marketed by Lilly/BI. See our coverage of full results from ADA 2019. Note that oral semaglutide was escalated to from 3 mg to 7 mg at 4 weeks, then to 14 mg after 8 weeks; empagliflozin was started at 10 mg then increased to 25 mg after 8 weeks. This novel analysis used on-treatment without rescue medication data, which may demonstrate a stronger effect than an intention-to-treat analysis would. Baseline A1c was 8.1% in both groups.

  • Novo Nordisk launched Rybelsus on the US market in the fall of 2019, following September 2019 approval. See our most recent commercial report on Rybelsus here.

Novo Nordisk Symposium Explains Why GLP-1s Have Risen to “Top of the Pecking Order” in Diabetes Care

Novo Nordisk hosted a symposium spotlighting GLP-1 agonists and featuring an all-star panel of Dr. James Gavin, Ms. Davida Kruger, and this year’s Banting Medal recipient Dr. Jens Holst. Each speaker offered compelling answers to the session’s overarching question: Why have GLP-1s reached the top of the pecking order for managing type 2 diabetes? Indeed, ESC now recommends GLP-1 agonists as first-line therapy for people with type 2 diabetes facing high CV risk, and ADA endorses their use irrespective of baseline A1c or background metformin. An audience poll found that >70% of HCPs in attendance recommend a GLP-1 product as a patient’s first injectable (ahead of basal insulin), which Ms. Kruger said she was “thrilled” to see. We should note that real-world utilization of GLP-1s remains low (far lower than we’d like) due to still-limited reimbursement and cost constraints. We, too, were pleasantly surprised by the poll result. It suggests effective provider-facing education about the multifaceted benefits of GLP-1s and hopefully signals a decline in clinical inertia to initiate GLP-1 treatment earlier in the course of diabetes progression – but access/affordability must also be addressed.

  • While positive CVOT results have generated the most buzz around GLP-1 agonists recently, Dr. Gavin emphasized that the class’ glucose-lowering efficacy is noteworthy on its own. If we rank all the diabetes medications available today by their mean A1c reduction in drug-naïve patients, four of the top five are GLP-1s: injectable semaglutide (Novo Nordisk’s Ozempic), liraglutide (Novo Nordisk’s Victoza), dulaglutide (Lilly’s Trulicity), basal insulin, and oral semaglutide (Novo Nordisk’s Rybelsus). In a similar ranking of second-line drugs added to metformin, three of the top five are GLP-1s: injectable semaglutide, oral semaglutide, premixed insulin, dulaglutide, and basal-bolus insulin. Of course, Dr. Gavin also touted the cardio protection, renal protection, and weight loss that come with GLP-1 treatment. He highlighted that Victoza, Ozempic, and Trulicity are all FDA-approved to reduce MACE, that Trulicity is the only product with an indication for primary prevention, and that Victoza is the only GLP-1 option for pediatric type 2s.

  • Ms. Kruger outlined the patient perspective on why GLP-1s are on top of the pecking order. Compared to insulin, a GLP-1 gives the same if not better glucose control without causing hypoglycemia or weight gain. As Ms. Kruger explained, “hypoglycemia and weight gain are what we worry about when we start basal insulin.” An abundance of data shows that GLP-1s do not increase risk for hypo and actually stimulate weight loss. She shared anecdotes of patients who feared hypoglycemia due to episodes they witnessed in family members, or who were unhappy when insulin caused 12 lbs of weight gain over six weeks. Adherence to GLP-1 therapy is improved by its no-hypo and weight loss features (provided the medication is titrated slowly to minimize GI side-effects), and Ms. Kruger underscored the importance of good adherence. She mentioned that an increasing number of patients are hearing about GLP-1 agonist options via friends or television commercials (some even come into clinic singing one of the marketing jingles), and that level of engagement boosts adherence as well. It’s interesting to note growing patient awareness of the GLP-1 class alongside provider education, and we so hope this translates to greater volume of GLP-1 prescriptions.

  • Dr. Holst commented on the science behind GLP-1 agonists and the mechanisms underlying their remarkable efficacy. While the precise mechanism of action that leads to CV benefits remains unknown (and we have even less understanding about renal protection, according to Dr. Holst), he drew a distinction between short- and long-acting GLP-1s. The short-acting agents in the class, including exenatide and lixisenatide, have a lower exposure rate, which is likely why they did not demonstrate statistically significant risk reduction for three-point MACE in their respective CVOTs. The long-acting agents all did (liraglutide in LEADER, semaglutide in SUSTAIN 6, dulaglutide in REWIND, albiglutide in HARMONY); Drs. Holst and Gavin both agreed that AZ’s long-acting formulation of exenatide (Bydureon) narrowly missed the threshold for superiority in EXSCEL due to a pragmatic trial design and “just bad luck.”

    • Dr. Holst also praised the impressive science that led to the development of oral semaglutide: “I once delivered a lecture on oral GLP-1 titled ‘mission impossible.’ Luckily they went ahead anyway, but it’s a really special development in science.” He explained that only ~1% of each oral semaglutide dose is effectively absorbed into the bloodstream (99% is lost), but that sufficient concentration builds up over time because the agent is so long-acting (in other words, a short-acting GLP-1 could not be taken as a pill). Given the molecular complexity of Rybelsus, there must be a relatively stringent protocol around administration (e.g., take it in a fasting state with 4 oz. of water, withhold food and other medications for at least 30 minutes). Rybelsus was launched in the US in late 2019 and commercial performance has been strong to-date ($118 million revenue in 1Q21), with the product serving as a growth driver for Novo Nordisk’s overall diabetes/obesity portfolio.

GLP-1s are Cost-Neutral Compared to Standard Care in Adult Type 2s Following First CVD Hospitalization; Nonsignificant Trend Toward Cost Reduction with GLP-1s

In an insightful poster (82-LB), Dr. Marc Evans (AstraZeneca) and colleagues demonstrated that GLP-1s are cost-neutral compared to the standard of care in GLP-1-naïve adult type 2s following their first CVD-related hospitalization. Using data from 1,712 adult type 2s who were initiated on GLP-1s and 122,334 adult type 2s who received standard care in the six months following their first CVD-related hospitalization, the researchers found that the added cost of treating T2D patients with CVD with GLP-1s is offset by significantly lower inpatient and outpatient care costs. Prior to hospitalization for CVD, total treatment costs were similar for both groups ($2,011 per-patient per-month [PPPM] for those on GLP-1s vs. $1,934 PPPM for those receiving standard care). While statistically equivalent at baseline (($2,011 per-patient per-month [PPPM] vs. $1934 PPPM for GLP-1s vs. standard care cohorts), total treatment costs were lower, although not significantly, for patients receiving GLP-1s compared with standard care  after initiation for CVD ($3,854 vs. $4,288 PPPM). These findings were driven by the significantly lower inpatient and outpatient costs and significantly higher drug costs of GLP-1s compared with standard care. Specifically, the total drug cost for GLP-1s was $1,046 compared to $675 PPPM for standard care while the GLP-1 cohort saw lower in-patient ($1,147 PPPM vs. $1,458 PPPM) and outpatient ($1,660 PPPM vs. $2,155 PPPM) costs compared to standard care.  Given that the 2019 ESC and ADA/EASD guidelines recommend GLP-1s as first- or second-line antidiabetic medication for T2D patients with CVD, this study is an important assessment with regards to treatment cost, which poses a substantial barrier for the approximately one-third of T2D patients who have CVD.

GLP-1 Agonist Posters

Title

Authors

Details + Takeaways

Efficacy and Safety of GLP-1RAs with or without Baseline SGLT2i: Post-hoc Analysis of the SUSTAIN 10 Trial

Matthew Capehorn, Andrei-Mircea Catarig, Ofir Frenkel, Charlotte Hindsberger, Michel Marre, Hermione Price, Richard E. Pratley

  • Post-hoc analysis of the SUSTAIN 10 trial comparing safety/efficacy of GLP-1 agonist therapy (semaglutide or liraglutide) in patients with and without SGLT-2 inhibitor therapy at baseline

  • Both GLP-1 agonists were associated with reductions in A1c, body weight, and systolic blood pressure from baseline to week 30, regardless of SGLT-2 inhibitor use

  • Reductions in A1c, body weight, and systolic blood pressure were greater with semaglutide than liraglutide

  • No unexpected safety concerns arose in the trial, and the rate of study discontinuation due to adverse events did not differ between participants with vs. without SGLT-2 inhibitor therapy 

  • Results support the addition of a GLP-1 agonist on top of existing SGLT-2 inhibitor therapy

Cost-Effectiveness of Subcutaneous Semaglutide vs. Empagliflozin as Add-On Therapy for Typse 2 Diabetes

Margaret Zupa, Ronald Codario, Kenneth J. Smith

  • Cost-effectiveness evaluation of semaglutide vs. empagliflozin in people with type 2 diabetes and cardiovascular disease

  • Compared to empagliflozin, semaglutide was associated with a 0.053 QALY gain at a cost of $1,006 (i.e., $18,897 per QALY gained)

  • Semaglutide was preferred over empagliflozin in 55% of model iterations

  • Results suggest that for most patients with type 2 diabetes and cardiovascular disease, semaglutide is a more cost-effective option than empagliflozin

Use of Glucose-Lowering Agents According to Cardiovascular Risk Factors in People with T2D and with or without CVD in the CAPTURE Study

Guillermo Dieuzeide, Abdullah M. Alguwaihes, Jose Luis A. Leon, Fahri Bayram, Patrice Darmon, Timothy Davis, Kirsten T. Eriksen, Tianpei Hong, Thomas J. Jensen, Csaba Lengyel, Ofri Mosenzon, Guiseppina T. Russo, Shinichiro Shirabe, Katerina, Urbancova, Sergio Vencio, Gourav Yadav

  • Post hoc analysis of the CAPTURE study to identify patterns diabetes therapy use among people with type 2 diabetes with and without CVD

  • Regardless of CVD status, therapy use was more common in men, people aged 55 to 75, and those with diabetes duration >10 years, A1c >7 and <9%, diagnosed hypertension, or an eGFR >60 ml/min/1.73m2

  • Newer drugs (GLP-1 agonists and SGLT-2 inhibitors) were less commonly prescribed than insulins and SUs

  • Low use of newer drugs suggests delayed introduction of these agents despite treatment guidelines encouraging the use of these agents for people with types 2 diabetes and high CV risk

Online Patient Education on GLP-1 Receptor Agonists in Patients with T2D and CVD Prompts Real-Life Changes

Amy Larkin, Anne Le

  • Measured the impact of online education activities on knowledge, confidence, and intent to act in people with type 2 diabetes and CVD (n=28,893)

  • Following the online education activity, 85% of patients planned to discuss heart disease risk with their HCP and 87% reported increased confidence asking how to reduce their risk

  • The online activity was associated with a 54% improvement in recognizing that GLP-1 agonists can treat type 2 diabetes and lower heart disease risk

Addition of a GLP-1 Agonist to Insulin Therapy in Type 2 Diabetes

Colleen Gavigan, Thomas W. Donner

  • Retrospective analysis assessing the efficacy of GLP-1 agonists (liraglutide, semaglutide, or dulaglutide) as an add-on to insulin therapy in people with type 2 diabetes (n=81)

  • 46% of patients on prandial insulin at baseline discontinued it within a year of beginning GLP-1 agonist therapy

  • There were no significant differences in change in A1c or percentage of patients who came off prandial insulin based on BMI (< or ≥30), insulin dose (< or ≥0.8 units per day) or diabetes duration (< or ≥10 years)

Novel Microsphere Formulation Developments of Long-Acting GLP-1 RA Including Exenatide, Liraglutide, and Semaglutide

Jin G. Jung

  • Investigated the pharmacokinetic profiles of novel sustained release (SR) versions of several GLP-1 agonists (exenatide, liraglutide, and semaglutide) in rats

  • All three SR-GLP agonists showed extended-release profiles ranging from 1 week (SR-liraglutide) to 4 weeks (SR-exenatide and SR-semaglutide)

Discovery of Nonpeptide, Orally Available Small Molecule GLP-1 Receptor Agonists

A-Rang Im, Jung-Eun Park, Dahae Hong, In-Gyu Je, Jung Ho Kim, Min Whan Chang, Chang-Hee Hong, Jae Eui Shin, Te-Ik Sohn, Yearin Jun, Eunhye Jang, Yeongran Yoo, Woojin Jeon, Kyungmi An

 

  • Investigated the safety and efficacy of novel oral small molecule GLP-1 agonists in cynomolgus monkey

  • All compounds demonstrated potent in vitro activity, excellent sensitivity against a class B G-protein coupled receptor related to GLP-1, and a strong safety profile

  • All compounds increased insulin secretion an intravenous glucose tolerance test

Mobile Health Application as a Real-World Data Resource: Self-Recorded Weight Reduction with Once-Weekly Semaglutide

Ulrik Bodholdt, Sophie Birot, Andrei-Mircea Catarig, Umut Erhan, Filip K. Knop

  • Assessed self-recorded weight loss among users of mobile health app designed to assist people with types 2 diabetes starting once-weekly semaglutide (n=1,506)

  • Mean self-recorded weight loss was 4.55% (4.65 kg/10.25 lbs) from baseline to week 18

  • Mobile health app data can add to the body of RWE on semaglutide

Impact of Early Initiation of Dulaglutide on A1c in Patients with Type 2 Diabetes

Meredith Hoog, Joseph L. Smith, Maria Yu, Jennifer Peleshok, Reema Mody, Michael Grabner

  • Retrospective, real-world observational study assessing glycemic control among people with type 2 diabetes who started dulaglutide on top of 1, 2, or 3+ oral agents

  • Dulaglutide was associated with A1c reductions of -1.39%, 1.30%, and 1.01% over 6 months among people on 1, 2, and 3+ oral agents, respectively  

  • A1c reductions were more pronounced among people taking fewer oral agents at baseline, supporting earlier dulaglutide initiation

Higher Doses of Dulaglutide Induce Weight Loss in Patients with Type 2 Diabetes (T2D) Regardless of Baseline BMI: Post-hoc Analysis of AWARD-11

Enzo Bonora, Juan P. Frias, Raleigh Malik, Anita Kwan, Sohini Raha, Angelyn Bethel, David Cox

  • AWARD-11 post hoc analysis assessing weight loss with different doses of dulaglutide

  • The 3 mg and 4.5 mg doses of dulaglutide were associated with greater A1c and body weight reductions compared to the 1.5 mg dose

  • In the 3 mg and 4.5 mg dulaglutide groups, average absolute reductions in bodyweight increased with higher BMI categories, but average percent weight loss was similar regardless of BMI

Real-World Effectiveness of Once-Weekly Semaglutide (sema OW) from a U.S. Commercially Insured and Medicare Advantage Population

Vincent Willey, Nitin Shivappa, Joshua Noone, Caroline Swift, Simo Du, Yurek M. Paprocki, Hiangkiat (Jason) Tan

  • Real-world study assessing the effectiveness of semaglutide among a population of commercially insured and Medicare patients (n=3,010)

  • Semaglutide was associated with significant A1c reductions in all age groups examined 

  • Across all age groups, those with baseline A1c ≥9% experienced the greatest A1c reductions (≥ 2.0%)

  • After semaglutide initiation, the percentage of patients who achieved an A1c <7% and <8% increased from 23% to 52% and from 50% to 79%, respectively

Evaluating Antidiabetic Medication Treatment Patterns in T2D Patients Initiating Once-Weekly Semaglutide (sema OW)

Vincent Willey, Nitin Shivappa, Caroline Swift, Joshua Noone, Simo Du, Yurek Paprocki and Hiangkiat (Jason) Tan

  • Retrospective, observational cohort study examining treatment patterns associated with one-weekly semaglutide (sema OW) initiation in adult type 2s (n=3,081) from US commercially insured + Medicare Advantage population

  • 28% of patients had a starting A1c>9%, 65% of patients were on metformin, 60% of patients were on ≥2 T2D medications, 30% of patients on ≥3 T2D medications

  • Sema OW more likely to be initiated as an add-on rather than a switch across antidiabetic classes; this trend reversed in patients who were taking many T2D medications (sema OW introduced as a switch in 70% of patients taking ≥4 T2D drugs)

  • About one-third of patients reduced the number of T2D medications they were taking after 90 days after their A1c reading

Greater Adherence and Persistence with Dulaglutide Compared with Semaglutide at 6- and 12-Months Follow-up in U.S. Real-World Data

Reema Mody, Janna Manjelievskaia, Maria Yu, Elizabeth Marchlewicz, Raleigh Malik, Nicole Zimmerman, Debra Irwin

  • Retrospective, observational, real-world study compared adherence and persistence among 190,246 matched type 2s dulaglutide and semaglutide users at 6- and 12-month follow-ups

  • Baseline: ~72% on metformin, ~23% on SGLT-2is, ~50% male, median age ~53 years

  • Patients on dulaglutide had greater medication adherence (54% at 12 months) than patients on semaglutide (43% at 12 months); consistent within subgroups based on age, dosage, baseline obesity, and insulin use

  • Relative to semaglutide, dulaglutide participants had higher persistence both at 6 months (71.9% dulaglutide vs. 62.2% semaglutide, p<0.0001) and 12 months (55.5% dulaglutide vs. 45.3% semaglutide, p<0.0001)

Greater Proportions of Subjects Achieved Composite Endpoints with Once-Weekly Semaglutide vs. Comparators in SUSTAIN Trials

Peter Gæde, Andrei-Mircea Catarig, Kathleen Dungan, Charlotte Hindsberger, Jens Øllgaard, Stephen Bain

  • Post-hoc analysis comparing A1c and body weight endpoints using semaglutide to active comparators (sitagliptin, exenatide extended release, insulin glargine, dulaglutide, canagliflozin, liraglutide)

  • Greater proportion of subjects achieved an A1c <7.0%, weight loss ≥5%, and systolic BP reduction ≥5 mmHg (Endpoint A) with semaglutide (once-weekly) than with the comparator therapies; same was true of the more stringent endpoint B (A1c <6.5%, weight loss ≥10%, SBP reduction ≥5 mmHg)

  • Proportion of patients across SUSTAIN trials meeting endpoint A: 12.4-19.1% [0.5 mg]; 18.9-36.7% [1.0 mg] vs. 0.8-15.7% [comparator] and meeting endpoint B: 2.5-8.0% [0.5 mg]; 5.0-14.8% [1.0 mg] vs. 0.2-3.0% [comparator]

Reduced Glycemic Variability with Once-Weekly Semaglutide vs. Active Comparators in Post-hoc Analysis of the SUSTAIN Program

Richard E Pratley, Vanita R. Aroda, Charlotte Hindsberger, Yasemin Kose, Margit Kaltoft, Esteban Jodar

 

  • Post hoc analysis of SUSTAIN 2-4, 7, 8, and 10 comparing within-day glycemic variability of once-weekly semaglutide  vs. active comparators in type 2s (n=5,680)

  • Once-weekly semaglutide led to significantly larger reductions in SD of SMBG than comparators

  • Changes in SD of SMBGs were closely aligned with changes in A1c from baseline

Achievement of Near-Normal HbA1c with Early Initiation of Oral Semaglutide: An Exploratory Subgroup Analysis of PIONEER 1

Vanita R. Aroda, Lars Bardtrum, Klaus Kallenbach, Anne Moeller Nielsen, Julio Rosenstock, Enrique C. Morales-Villegas, Melanie J. Davies

  • Post hoc analysis of PIONEER 1 trial (n=703) to examine impact of early initiation of oral semaglutide on glycemic efficacy, body weight, and achievement of A1c targets

  • Greater A1c and body weight reductions seen for oral semaglutide 14 mg vs. placebo for both T2D duration ≤1 year and >1 year

  • Higher proportion of patients starting semaglutide within ≤1 year of diagnosis reached glycemic targets than those starting semaglutide in >1 year: 45% of those on oral semaglutide 14 mg achieved A1c <6% vs. 31% in the >1 year group

Incorporating Treatment Pauses, Dosing Flexibility, and Education to Support GLP-1RA Therapy Persistence

Vanita R. Aroda, Robert Bauer, Alun L. Davies, Eskil B. Kreiner, Peter J. Lin, Richard E. Pratley, Stephen C. Bain

  • Evaluation of medication management strategies in PIONEER 6 trial to support continuation of oral semaglutide in context of intolerability due to GI-related adverse events

  • Type 2s (n=3,183) on oral semaglutide underwent dose escalation (3 mg to 7 mg after 4 weeks, then to 14 mg after 8 weeks) with education to address GI tolerability issues

  • Discontinuation of oral semaglutide (temporary and permanent) mostly occurred during the initial dose escalation period

  • 23% of patients on oral semaglutide had ≥1 treatment pause; of those individuals, majority (72%) had just one pause; median pause duration was 21 days

  • 75% of patients restarted oral semaglutide after the first AE-related treatment discontinuation

  • Results emphasize potential for education-based approaches to encourage treatment persistence on GLP-1s

Kidney and Metabolic Benefits by Adding GLP-1 Agonists and/or SGLT2 Inhibitors on Metformin in Obese Type 2 Diabetes (T2DM) Patients: 24-Month Real-World Data from Both Urban and Rural Clinics

Ioannis Zoupas, Athanasia Papazafiropoulou, Maria Xenou, Dimitrios Lygnos, Evangelos Fousteris

  • Prospective observational study of 253 patients with type 2 diabetes on metformin for 24 months. Patients were divided into three groups and administered either GLP-1 SGLT-2,  or both GLP-1 and SLGT-2 therapy in addition to metformin

  • Patients on GLP-1 therapy saw the largest A1c reductions from 8% at baseline to 5.9% at 24 months; patients on SGLT-2s saw A1c reductions of 1.1% from 7.7% to 6.3%; patients on combination therapy saw A1c reductions of 1.9% from 7.7% to 5.8%

  • eGFR improved most significantly among patients on combination GLP-1, SGLT-2 therapy from 98 to 106 (p=0.002)

A Real-World New User Cohort Study of Metformin vs. SGLT2i or GLP-1RA on the Risk of Cardiovascular and Renal Outcomes in Type 2 Diabetes Mellitus Patients with Cardiovascular Risk Factors

Caroline Foch, Arthur Allignol, Emmanuelle Boutmy, Kerstin Brand

  • Head-to-head cohort study comparing metformin vs. SGLT-2s or GLP-1s on the risk of cardiovascular and renal outcomes in patients with type 2, as measured in terms of heart failure hospitalization, myocardial infraction, stroke or CKD development

  • Patients on SGLT-2s (n=5,181) and patients on GLP-1s (n =6,164) were matched to 10,361 and 12,326 comparable patients on metformin

  • Treatment with SGLT-2s showed decreased risk of hospitalization for heart failure compared to metformin (HR=0.45); GLP-1 use associated with  decreased risk of hospitalization for stroke compared to metformin (HR=0.51)

  • GLP-1 and SGLT-2 use both associated with increased risk of CKD compared to metformin (HR=1.52, HR=1.36 respectively)

Insulin Therapy

Soliqua (Insulin Glargine/Lixisenatide) Shows Superior A1c Reductions and Weight Loss Over Premixed Insulin in SoliMix Readout

In a highly anticipated oral presentation, Dr. Julio Rosenstock unveiled the results of the SoliMix trial, showing improved glucose control without weight gain with Sanofi’s GLP-1/basal insulin combination therapy Soliqua over traditional premixed insulin. SoliMix enrolled people with type 2 diabetes and an A1c between 7.5% and 10% who were on a regimen of basal insulin and one or two OADs, randomizing them to once-daily injections of Soliqua, a fixed-dose combination of insulin glargine and GLP-1 agonist lixisenatide (n=443), or twice-daily injections of the premix insulin Novolog (n=444) for 26 weeks. Soliqua met both of the trial’s co-primary endpoints, including non-inferiority to Novolog in A1c reduction (-1.3% vs. -1.1%) and superiority to Novolog for body weight change (-0.7 kg vs. 1.2 kg). Soliqua also met all of the trial’s key secondary endpoints, including superiority to Novolog in A1c reduction, a greater proportion of participants achieving an A1c <7% without weight gain (27.5% vs. 12.4%), and a greater proportion of participants achieving an A1c <7% without weight gain and without hypoglycemia (19.4% vs. 7.0%). Very importantly, Soliqua was also associated with lower incidence of hypoglycemia (6.3% [n=28] vs. 12.9% [n=57]) and severe hypoglycemia (0.2% [n=1] vs. 0.5% [n=2]) compared to Novolog. No severe adverse events were reported in the trial, though participants randomized to Soliqua experienced more GI side effects (10.4% vs. 2.3%), as is expected with GLP-1 agonist therapy. Overall, the SoliMix results point to Soliqua as a better alternative to premixed insulin for people with type 2 diabetes who are not meeting their A1c goals on typical basal insulin/OAD therapy.

  • In Dr. Rosenstock’s view, the SoliMix results highlight Soliqua as a therapy that circumvents two of the two of the main concerns about initiating or intensifying insulin therapy among patients and clinicians alike: hypoglycemia and weight gain. To this end, a secondary analysis demonstrated that participants on Soliqua showed greater improvements in patient-reported outcomes over the 26-week trial compared to those on Novolog. Specifically, participants using Soliqua showed higher self-reported therapy adherence, diabetes management, and psychological health, and twice as many study participants (27.5% vs. 15.0%) and their physicians (29.9% vs. 15.3%) reported that treatment with Soliqua had been effective (defined here as complete control of diabetes).

  • The SoliMix results were also highlighted in an afternoon product theater for Soliqua. Ms. Katherine Whitmire (Southwest Medical Associates), a nurse practitioner and CDE, contextualized that the SoliMix results are particularly impactful for the growing patient population of older adults with type 2 diabetes. Over 25% of adults over the age of 65 have type 2 diabetes, and these individuals have unique clinical needs. Since older adults are at a higher risk of cognitive decline and institutionalization, the complexity of their diabetes treatment regimens is important to consider. Older adults with type 2 diabetes are also at higher risk of hypoglycemia compared to younger adults. With these considerations in mind, Soliqua is an excellent choice – offering not only improved glycemic control but also fewer injections and lower hypoglycemia risk compared to traditional premixed insulin. For more, see the press release from Sanofi and our interview with Sanofi’s Vice President & US Medical Head of Diabetes  Dr. Luigi Meneghini below.

Interview with Dr. Luigi Meneghini (Vice President & US Medical Head of Diabetes, Sanofi)

Kelly Close: Hi, Dr. Meneghini. Thank you so much for taking the time to speak with us today. If you could start off with some big picture thoughts on Sanofi’s presence at ADA and the SoliMix trial, that would be great. Then, we can go into our questions.

Dr. Luigi Meneghini: I think we're really excited for the ADA this year for a number of reasons. One, I think we continue to be committed to diabetes. I think you've seen the results of one of the trials, the SoliMix trial, which is going to feature Soliqua front and center. From an endocrinologist perspective, when I saw the result, it was pretty mind blowing.

That's the reaction that I've seen with many of my colleagues, and I would love to talk more about the SoliMix results and their implication in clinical practice. We also have quite a bit of information surrounding Toujeo and trying to find the place for second generation basal insulins in the treatment of diabetes. There have been some discussions around that and we're focused on showing the value and relevance surrounding these therapies.

Rhea: Great. Thank you so much. We've written about some of the SoliMix trials, could you start by going backwards a little and explain the rationale behind the trial design and why you want to start this trial?

Dr. Meneghini: Great question. The trial was designed to understand what alternatives there might have been for advancing basal insulin therapy. In individuals with type 2 diabetes not well controlled on basal insulin, it's often a challenge to get them to target.

You could, for example, switch to a premix insulin, which is commonly done outside of the US and in many areas in the US. You could go to basal plus sequential prandial doses of insulin. You could go to full basal bolus, or you could add a GLP-1.

The rationale was that, if we come in with a fixed ratio combination, we could give both the basal component and the GLP-1 component in one injection once a day. We compare this to a very common intensification method, which was switching patients from basal insulin to twice-daily premix. The rationale for this study was that we could compare these two approaches and find out what the results would be.

Ursula: Thank you. As a follow up, what do you think is the most important takeaway of the new data set?

Dr. Meneghini: Great. I would assume that you've seen the data, which showed that Soliqua once daily, compared to premix twice daily, titrated to the same targets with the same frequency but led to better glycemic control and better reduction in A1c. It did that with less hypoglycemia and less weight gain. When you ask patients and providers what approach they prefer, it was clear in all measures of the TRIM-D effectiveness score that Soliqua was preferred to premixed insulin.

In essence, showing that in those really difficult to treat patients with type 2 diabetes or who are not well controlled on basal insulin, starting once daily Soliqua versus twice daily premix led to better effectiveness, safety, and simplicity is spectacular. This gives us another approach and actually a better approach for intensification beyond basal insulin.

Claire: Our next question is whether you have any baseline data on age or diabetes duration of the participants in this study?

Dr. Meneghini: Yes, if you go to the baseline demographics, it shows that they're in their 60s. They've had diabetes duration of around 13 years. That's not uncommon for individuals who are on basal insulin therapy and whose A1c’s are not where they should be.

Kelly Close: We’re really happy that people will be getting access earlier and that they'll have the coverage. I hope that the commercial payers really plan on making a big deal about this as well.

Luigi Meneghini: That's a good point. I know that we’ve worked quite hard at making insulin-based therapies more accessible and affordable.

Kelly Close: I think that we were also interested to hear how early people can be on this therapy. people being able to be on a more optimized therapy at the right time is kind of the dream.

I really applaud and salute that you said specifically that you're going to try to work with people with very high A1cs. Is that still the case? And if there's anything that you might say about if you feel like in the field there is more professional CGM use or there's more intermittent use in people with type 2. They're not really “supposed” to be able to get it paid for until they're on MDI, but that kind of creates a circular thing sometimes where then they don't get the right medical therapeutic intervention.

Dr. Meneghini: The first question I think I heard was, is the focus for Soliqua in individuals with high A1cs? That's clearly one of the focuses for Soliqua. That is a patient population that is poorly controlled, has very high A1c, are at very high risk for developing vascular complications, and have an urgent need to get better glycemic control. There we see Soliqua as an opportunity to get these individuals under better control, but that is not the only focus. SoliMix, as you saw, was studied in individuals who had uncontrolled diabetes- about half of them probably with an A1c above 9%. Many of them had A1cs under that. In that group of individuals on basal insulin, we were able to get them close to the target. The A1c at the end of the study was around 7.3% safely, without any weight gain, with little hypoglycemia compared to premixed, and so on. That is another group of individuals that I think we can focus on. From an endocrinologist perspective, we put individuals with diabetes on premix insulin because we may not think that they can manage with basal plus or basal bolus type of therapies.

I can see Soliqua being somewhat of an alternate in thinking, “how do I advance therapy beyond basal insulin?” From an endocrinologist perspective, I think it's an exciting study. And it brings up some really wonderful opportunities.

Kelly Close: Yes, I just have one quick comment on that, and then I’ll send it back to the team. I think there are so many great studies about GLP-1s, and this and that. We were talking to Richard Wood at dQ&A the other day about how great it is that GLP-1s get so much attention. There are a lot of people who are on GLP-1s, but it's not enough and that's just not in the narrative about GLP-1s.

Dr. Meneghini: That's an excellent point, Kelly. There are no therapies that we know of right now that can arrest the progression of type 2 diabetes. Therefore, we know that at a certain point, we'll need to add a basal insulin.

The combination of a basal insulin with a GLP-1 agonist has clearly shown to be very effective in individuals at all stages of their diabetes journey. So that's a great point. Thank you.

April: I can take the next question. Have there been any subgroup analyses done exploring differences in outcomes between participants on Soliqua with an SLGT-2 inhibitor versus those not taking an SLGT-2 inhibitor?

Dr. Meneghini: Those are all very, very interesting post hoc sub-analyses that are going to be done right now. We have the presentation that you will see at ADA that focuses on the primary analysis of effectiveness, safety, and side effect profiles. You'll see at the ADA that there is an analysis of hypoglycemia during the night. There will be also an analysis that looks at patient reported outcomes in the controls, examining them by SGLT-2 use at baseline. I'm looking forward to that as well- excellent question.

Ashwin: What is your response to healthcare practitioners who may view GLP-1 agonists as pre-insulin therapies, making a combination GLP-1 agonist/basal insulin less preferable to them?

Dr. Meneghini: I think GLP-1s are excellent therapies. They’ve come to the forefront and have a lot of beneficial effects. With regards to the fixed ratio combinations, as we saw in SoliMix, they are highly effective in those individuals that have had diabetes for a period of time and are not well controlled.

Either they have very high A1cs -- above 9% or on basal insulin and their A1cs are still not at the target that they want them to be. For those individuals, a combination of basal insulin and a GLP-1 is an excellent combination. It simplifies things because it's not two injections that you have to start either simultaneously or sequentially. It’s just one injection, one time a day. It is simple. Because of the combination and the slow titration schedule, it mitigates some of the side effects that we might see with GLP-1s and mitigates some of the weight issues that we see with insulin therapy. It has its place. But GLP-1 therapies obviously are here to stay, and thankfully so.

Kelly Close: Yes. There are so many endocrinologists out there who are like, “that might not be exactly the right ratio for my patients.” It's kind of pointing out, “this is why I don't want to use this fixed ratio. I'll just leave them on MDI, or I'll just try to put them on something else.”

It’s kind of the perfect being the enemy of the good. I had a question about the ratio. It seems like there is pretty broad acceptance that there's CVD risk reduction with GLP-1 as a class effect. Do you think that that is getting to be well known?

Dr. Meneghini: Yes. I'm not going to tell endocrinologists how to practice, but I would suggest that we focus our therapy decisions by sharing those decisions with people with diabetes. There, I think that you will see the presentations on the patient reported outcomes. They are significantly better with Soliqua than they are with the premix insulin.

In response to your question, Soliqua is an effective therapy. We are identifying the types of individuals that would most benefit from it. I think that's a therapy that would be very effective in the hands of primary care physicians. If endocrinologists want to simplify or ask their patients what their preference might be, that could be also an option.

On your second point around cardiovascular risk, yes, lixisenatide showed that it was safe from a cardiovascular perspective. There is a poster that we have that uses a BRAVO risk estimation that has looked at modeling some of those risk factors. I don't want to say any more on that, but there can be some clues as to what we could expect based on the impact of Soliqua on patient metabolism.

Kelly Close: Yes, I'll just hand it to Rhea to finish up. I guess for us, as much as people can be urged to ask their doctors about getting at least professional CGM; that's also how we hope to lead them all to better options in terms of the therapy.

Dr. Meneghini: I absolutely love the fact that you and diaTribe are driving those conversations around time and range and the use of CGM, because it is one of the things that has made the most difference in the lives of people with diabetes. So, thank you.

Kelly Close: And many, many others out there with us. Thank you for the encouragement on that. That means a lot to us.

Rhea: I think for our final question, we’d just like to ask how you think Soliqua fits into Sanofi's broader diabetes portfolio strategy.

Dr. Meneghini: I think it fits very nicely. I think we are defining the types of individuals that would most benefit from Soliqua: the ones that have a high A1c, the ones that are not well controlled on a basal insulin, possibly the ones that are not controlled on a GLP-1 because there's the simplified addition of insulin therapy.

There are other populations that we are interested in exploring in the next year or so through real-world evidence and innovative ways of looking at data in clinical practice and trial populations, like elderly populations that might benefit from a Soliqua approach versus a premixed or other more complex insulin approach.

Thank you for your question. It's been really wonderful. Thank you for allowing us to come in front of you and just discuss this wonderful trial, SoliMix, that we're all excited about.

Kelly Close: Yes, as are we. Just to have focus on those two patient populations is immense. And yes, I wouldn't be worried about saying that people on GLP-1 only may not have the best strategy. Thank you so much again.

Phase 2 Trial Suggests Afrezza is Safe and Effective for Pediatric Patients with Type 1 Diabetes, Phase 3 to Launch in 4Q21

Exciting new data shared via poster (923-P) from MannKind revealed that Technosphere Insulin (TI), an inhaled rapid-acting dry-powder prandial insulin marketed as Afrezza, showed pharmacokinetic results in pediatric patients that mirrored results in the adult population. The central goal of the trial was to understand the pharmacokinetics and safety and tolerability of TI in children and adolescents (which has not previously been studied), as the drug is currently only approved in adult patients. The four-week phase 2 clinical trial was made up of 30 subjects, aged 8-17 years old with type 1 diabetes. The study measured serum insulin and blood glucose prior to and after TI was inhaled, and doses were titrated to reach a post-prandial glucose level of 110-180 mg/dl. Overall, the study found that serum insulin increased quickly after administration, reaching its peak about 10-15 minutes after inhalation, and returned to baseline after two hours. Additionally, a decrease in postprandial glucose excursion was observed 30-60 minutes post-inhalation – similar to the pharmacokinetic profile seen in adults. Although 76% of participants reported treatment emergent adverse events (TEAE), this distribution was consistent with data from the adult population, with a cough in the initial stages being the most commonly reported side effect. Perhaps more importantly, no severe hypoglycemic events were reported throughout the duration of the study.

As a result of the study, researchers feel the therapy is safe and well-tolerated enough to continue development. In a press release released in tandem with ADA, presenting author Dr. Kevin Kaiserman, Vice President, Medical Affairs and Safety of MannKind, said, “The pharmacokinetics and safety results provide us the rationale to move forward with conducting a phase 3 safety and efficacy study in pediatric patients with diabetes, which we expect to initiate in the fall.” Given the large burden that injecting insulin takes on a patient and their body, we are excited at the thought of expanding an easier mechanism of insulin delivery to a greater patient population.

  • To be eligible for the phase 2 study, subjects had to have the clinical T1D diagnosis, had to have been using insulin for more than a year and have a stable basal-bolus regimen for more than six weeks. The subjects were 60% female, 83.3% white, had a median age of 12.7 years, and had a mean duration of T1D of 4.3 years.

  • Financially, MannKind is in the “best position” it has been in the last four years, according to CEO Dr. Michael Castagna during the company’s 1Q21 update. In March 2021, MannKind restructured its debt to stabilize its financial position. Mr. Steven Binder, CFO, remarked in May 2021 that the additional cash and reduced debt cost will allow more resources to be funneled into pediatric trials. Following the strong results of this phase 2 trial, the company hopes to enroll its first patient in the phase 3 Afrezza trial in 4Q21.

MannKind Poster Shows Afrezza Inhaled Insulin Leads to Significant Increases in Daytime Time in Range for Uncontrolled Type 2s

An exciting poster from MannKind showed significant increases in daytime Time in Range for patients with uncontrolled type 2 diabetes using Technosphere Insulin (TI), an inhaled insulin with ultra-rapid onset. This retrospective analysis of the Levin study followed 20 patients with type 2 who had just begun TI over the course of 12 weeks in order to determine the effect of prandial inhaled insulin. During the daytime period, there were significant decreases in mean CGM values, significant decreases in time spent in hyperglycemia, and significant increases in Time in Range without a significant increase in hypoglycemia. Most importantly, daytime CGM Time in Range increased significantly from 38.7 percent to 62.1 percent– that’s an average increase of over four hours spent in daytime Time in Range per day! As a reminder, Mannkind’s Afrezza is the only commercially available inhaled insulin, providing an innovation way to alleviate needle-phobia and simplify insulin administration. These encouraging results suggest that this user-friendly insulin has potential to reduce long-term complications of diabetes, particularly among those with poorly controlled type 2 diabetes.


  • The data show impressive improvements regarding hyperglycemia, with time spent in type 1 hyperglycemia among those on Afrezza approximately half that of the baseline group (30.5 percent versus 60.5 percent, respectively), and time spent in type 2 hyperglycemia among those on Afrezza approximately one-fourth that of the control group (6.7 percent versus 25.7 percent, respectively). Similarly impressive decreases were seen in daytime CGM: the mean CGM values decreased significantly from 195 mg/dl to 153 mg/dl. For background, results from the Levin trial also showed a 1.6% reduction in A1c in 12 weeks. We’re excited by these promising results, particularly in light of Dexcom’s launch of the Global Movement for Time in Range yesterday, and all the ongoing Time in Range Coalition work at diaTribe and look forward to seeing growing awareness and use of Afrezza in the coming months. 

  • This is an especially exciting development given our reporting on MannKind’s decision to participate in the 2022 Medicare Part D Senior Savings Model in March and its Afrezza Assist programs, both of which help expand insulin affordability. As a reminder, the Medicare Part D Senior Savings Model caps the co-pay for a 30-day insulin supply at $35, ensuring better access to this life-saving drug for low-income people. AfrezzaAssist is MannKind’s newly launched telehealth program, which will help streamline the prescription and reimbursement process for so many patients.

Once-Weekly CGM Data Further Bolsters Safety and Efficacy of Lilly’s Phase 2 Once-Weekly Basal Insulin-FC in Type 2 Diabetes

Building on positive phase 2 results presented at ENDO 2021, Dr. Christof Kazda (Eli Lilly) today unveiled compelling CGM data (Dexcom G6) on Lilly’s once-weekly basal insulin-FC (BIF), providing further granularity on the drug’s comparable glycemic control vs. daily insulin degludec. As a reminder, BIF met the phase 2 trial’s primary endpoint by demonstrating non-inferiority to insulin degludec on the primary endpoint of change from baseline A1c after 32 weeks of treatment. The new data show that BIF was able to meet this endpoint with similar Time in Range, time ≤70 mg/dl, and time >180 mg/dl; this finding was particularly impressive given that both BIF arms were titrated to higher target glucose values vs. degludec. As a reminder, insulin degludec was titrated to a fasting glucose target of 100 mg/dl, and BIF-A1 and BIF-A2 were titrated to 140 mg/dl and 120 mg/dl, respectively, due to “safety reasons.”

  • In regard to hypoglycemia (time ≤70 mg/dl), a relative risk analysis revealed that BIF-A1 significantly reduced the relative risk of total hypoglycemia by 30% and also conferred a 45% RRR on nocturnal hypoglycemia vs. degludec. These results are not entirely surprising given the higher FPG target for BIF-A1, however, are still compelling given comparable Time in Range between BIF-A1 and degludec. While BIF-A2 did not significantly reduce either overall or nocturnal hypoglycemia, relative risk of both conditions were comparable to degludec.

  • Finally, an analysis of duration of time in hypoglycemia showed that hypoglycemia with BIF did not seem to vary depending on time post-injection at Week 32. 

  • In terms of study population, a total of 399 participants previously treated with basal insulin and oral antidiabetics were recruited to the study – 51% were women, mean BMI was 32.2 kg/m2, duration of diabetes ~15 years, baseline A1c 8.1%, and ~30% of participants were on a sulfonylurea. 

  • These CGM results provide further rationale for the clinical development of BIF, which is currently advancing through two larger phase 2 trials: one in insulin-naïve patients with type 2 diabetes (n=264) and another  in individuals with type 1 diabetes (n=254), both expected to complete in September 2021. Big picture, once-weekly basal insulins have been of high interest due to their potential to improve compliance and adherence to insulin therapy and prevent therapeutic inertia, and for their potential co-formulation with once-weekly incretins.

Length of Hypoglycemia Duration Consistent Between Novo Nordisk’s Once-Weekly Insulin Icodec and IGlar U100 in Type 2 Diabetes

An intriguing post-hoc analysis presented today by Dr. Robert Silver (Southern New Hampshire Health) compared the time of CGM-based hypoglycemic episodes (IG <70 mg/dl for at least 15 minutes) between insulin icodec and insulin glargine U100, revealing similar duration between the two, irrespective of insulin icodec titration method or use of an initial loading dose. Given its extended duration, icodec’s safety profile and hypoglycemia risk are of particular interest. In order to quantify and compare the duration of hypoglycemic episodes, Dr. Silver analyzed data from two phase 2 trials of insulin icodec (i) “A Randomized, Open-Label Comparison of Once-Weekly Insulin Icodec Titration Strategies Versus Once-Daily Insulin Glargine U100” by Dr. Ildiko Lingvay, comparing three icodec titration strategies in insulin naïve adults vs. insulin glargine U100 (n=205); and (ii) “Switching to Once-Weekly Insulin Icodec Versus Once-Daily Insulin Glargine U100 in Type 2 Diabetes Inadequately Controlled on Daily Basal Insulin: A Phase 2 Randomized Controlled Trial” by Dr. Harpreet Bajaj, investigating the value of two different switch approaches to insulin icodec vs. insulin glargine U100 in insulin-experienced adults (n=154). Hypoglycemia was measured using double-blinded Dexcom G6 data.

  • Starting with the latter ‘titration study,’ insulin icodec had similar lengths of hypoglycemic episodes across the three titration methods and U100. Median overall hypoglycemic episode duration (IQR) was 35.0 min (20.0, 70.0) for icodec titrations A (SMBG 80-130 mg/dL ±21 U/wk) and B (80-130 mg/dL ±28 U/wk) and insulin glargine U100 (80-130 mg/dL ±4 U/day), and 39.0 min (24.0, 70.0) for icodec titration C (70-108 mg/dL ±28 U/wk) – while similar in length, these results reflect a slight uptick in duration when using a more straight glycemic target. Similar lengths of duration were also found regarding time spent <54 mg/dl (Level 2 hypoglycemia) during a hypoglycemic episode and for nocturnal hypoglycemia. Furthermore, when looking at the full 16 weeks of the trial, there was no apparent clustering of hypoglycemic episodes at any time point.

  • These results were largely corroborated by the insulin ‘switch trial.’ Median overall hypoglycemic episode duration (IQR) was 40.0 min (20.0, 75.0) for icodec with a loading dose, 40.0 min (25.0, 80.0) for icodec without a loading dose, and 35.0 min (20.0, 60.0) for insulin glargine U100. Similar lengths of duration were also found regarding time spent <54 mg/dl (Level 2 hypoglycemia) and nocturnal hypoglycemia. Hypoglycemic episode clustering was also not apparent.

  • While these duration results are certainly promising, we note that certain insulin icodec arms demonstrated elevated rates of CGM-based hypoglycemia. Specifically, in the ‘titration trial’ (732-P), icodec titration C had a statistically significantly greater amount of time <70 mg/dl vs. insulin glargine U100, although this was somewhat expected giving the lower titration target. In the insulin ‘switch trial’ (731-P), time <70 mg/dl was significantly greater for icodec with a loading dose vs. insulin glargine U100 (p<0.001; no difference for icodec without a loading dose).

  • Big picture, these results contribute to the growing safety base behind Novo Nordisk’s once-weekly insulin icodec. Icodec is currently progressing through the phase 3 ONWARDS program, and we hope to see similar uses of CGM-based safety metrics applied to the larger trial population. 

Results from Phase 2 MemAID Trial of Intranasal Insulin on Cognition Led to Faster Walking Speed, Lower Insulin Resistance, and Higher Cerebral Blood Flow in Participants with T2D

Dr. Vera Novak (Harvard Medical School) presented results of the MemAID trial of intranasal insulin Novolin’s effects on memory in healthy older adults and older adults with type 2 diabetes (n=244). Before detailing the trial’s findings, she provided an overview of the study’s rationale: As diabetes accelerates brain aging by five years and insulin resistance has been linked to cognition and slow walking, intranasal insulin may lead to cognitive protection by improving memory, insulin signaling, and the brain’s glucose metabolism. To measure Novolin 40 IU’s effects on cognition, the phase 2 trial’s primary outcome was cognition, as measured by normal and dual task walking. Additionally, an MRI substudy that analyzed cerebral brain flow and vasoreactivity was performed, along with a safety/efficacy trial on A1c, glycemic control, insulin resistance, and weight.

  • Results found that treatment with Novolin led to faster walking versus placebo in those with type 2 diabetes, with significant differences in the normal and dual task walking. Treatment with Novolin did not lead to significant differences in executive function or verbal memory at baseline, during treatment, or post-treatment in the diabetes cohort. However, in the control cohort (treated with INI), Novolin led to significantly improved normal walking during treatment, executive function during and post-treatment, and verbal learning post-treatment.

  • Results of the MRI substudy showed that Novolin treatment increased relative cerebral blood flow in participants with type 2 diabetes (p=0.03). Specifically, this increase was seen during rest in the right medial-prefrontal cortex, an area of the brain that has been linked to decision making and long-term memory retrieval. This increased vasoreactivity was also correlated to the primary outcome of faster dual task walking. 

  • Safety data showed that Novolin led to decreased insulin resistance in those with type 2 diabetes, albeit without any difference in A1c, weight, and weight circumference compared to placebo. Blood glucose did decrease significantly in the diabetes arm (p=0.03), but only during the treatment period – a bounce-back to baseline levels post-treatment would explain why no change in overall A1c was observed and we wouldn’t expect that without continued treatment here would be continued improved glycemic numbers. Time in Range data would help clarify, through GMI, even if A1cs weren’t available during COVID-19, what the change was through nearly six months. Novolin use was also not associated with serious or moderate adverse events, and while “only 13” participants discontinued use from adverse effects, we would like to learn more about what prompted that, especially since the T2D part of this trial was under-enrolled.

  • This study has major limitations. Notably, the type 2 diabetes group was under enrolled, and a safety study in those with type 2 and type 1 diabetes was discontinued due to futility and a 65% drop out rate. We are unclear if another safety trial has been started. Additionally, Dr. Novak asserted that the trial was impacted by the COVID-19 pandemic severely – canceling 23 assessments and seven MRI scans and censoring ten participants. In addition, only 169 (79 T2DM patients and 90 non-DM controls) of the 244 randomized participants completed the study. That’s a high dropout rate for both arms, even considering COVID-19. We are not sure what was required in the clinic for the trial of participants, but a multitude of other trials during COVID-19 went well due to the CGM workaround where GMI was used as a proxy for A1c. Due to these issues, some main outcomes were under-powered statistically, limiting observed differences between the Novolin group and placebo and result generalizability. Ultimately, this seems like a trial beset with a range of issues, some certainly related to the pandemic and others not. We look forward to learning more about how cognition is being tested and assessed as this could have major positive implications for people with diabetes who may not previously have pursued such advanced treatment.

Treatment with Insulin Degludec Incurs 54% Lower Likelihood of Experiencing Nocturnal Hypoglycemia Compared to Glargine U100 Treatment (n=51)

ADA Presidents’ Select Abstract awardee Dr. Julie Brøsen (Nordsjællands Hospital, Denmark) presented results from the overnight sub-study (n=51) of the HypoDeg trial comparing insulin degludec and glargine U100 effects on occurrence of nocturnal hypoglycemia in patients with type 1 diabetes prone to severe nocturnal hypoglycemia. Participants in the two-year, ten-center cross-study were randomized to receive either insulin glargine or degludec over the course of 196 nights. Of the total cohort, 33 participants had 57 nights (29%) with instances of hypoglycemia. 67% of nights with hypoglycemia occurred during insulin glargine treatment while only 33% occurred during degludec treatment. People treated with insulin degludec were 54% less likely of experiencing a nocturnal hypoglycemic event £70 mg/dl, 64% less likely of experiencing a nocturnal severe hypoglycemic event (<54 mg/dl), and 62% less likely to experience symptomatic nocturnal hypoglycemia (confirmed £70 mg/dl) compared to those treated with insulin glargine (p=0.05 or p<0.05 for all).

Advancing from Basal Insulin (BI) to Premix Insulin Results in Better Glycemic Control, Lower Hypoglycemia (and Nocturnal Hypoglycemia) Risk, and Improved QOL Outcomes

We caught new data from the SoliMix trial (746-P), on BI’s iGlarLixi and its effects on Nocturnal Hypoglycemia risk. Primary results showed that switching to once-daily iGlarLixi (n=442), compared to standard care with BiAsp 70/30 (premix insulin) (n=441) resulted in better glucose control with less nocturnal (00:00-6:00) hypoglycemia vs. twice-daily BiAsp 70/30. Additionally, incidence and rates of ADA Level 2, Level 1, and any symptomatic nocturnal glycemia were lower with iGlarLixi vs BiAsp 70/30 and no severe (ADA Level 3) nocturnal hypoglycemia was detected in either treatment. The table below dives deeper into the results demonstrating 1.4X more hypoglycemia events and 1.9x more symptomatic hypoglycemia between bedtime and waking for patients on BiAsp 70/30 compared to those using iGarLixi. Nocturnal hypoglycemia is certainly a large concern for patients living with diabetes and we are excited to see the potential for newer therapies to reduce this risk.

  • Turning to quality-of-life outcomes, patients participating in the trial rated their insulin therapy in terms of treatment burden, daily life, diabetes management, compliance, and psychological health in a pre-post analysis. Notably, patients rated iGarLixi (n=443) more positively across all sub-categories compared to BiAsp 70/30 (n=443) with the most significant differences in diabetes management and treatment burden.

Insulin Posters

Title

Authors

Details + Takeaways

Higher Derived Time-in-Range with IDegLira vs. Insulin Glargine U100 in Patients with T2D

Athena Philis-Tsimikas, Vanita R. Aroda, Christophe de Block, Liana K. Billings, Elise Hachmann-Nielsen, Andreas Liebl, Ramsathish Sivarathinasami, John M. D'Cruz, Ildiko Lingvay

  • Compared dTIR (derived Time in Range) in patients with T2D taking IDegLira (insulin degludec/liraglutide) and IGlar U100 based on data from DUAL V trial (+metformin; n=557) and DUAL VIII (+oral antidiabetic drugs; n=1012)

  • IDegLira group had higher proportion of patients achieving ≥70% dTIR at end of treatment compared to IGlar U100 (DUAL V: 62% vs. 60%; DUAL VIII: 50% vs. 26%)

Effectiveness of Fixed Ratio Combination of IGlarLixi in People with Type 2 Diabetes Mellitus: Experience from Routine Clinical Practice

Janos T. Kis, Gabor Nagy, Peter Stella

  • 353 adults with uncontrolled T2D were given the fixed ratio combination of insulin glargine 100 U/mL and lixisenatide (iGlarLixi) over six months

  • 61% of patients (n=215) achieved primary endpoint of ≥1% A1c decrease by week 26

  • Glycemic variables improved significantly during follow-up (A1c from 8.9±1.3% to 7.4 ±1.0%; p<0.001) & mean body weight decreased from 90.5 kg to 88.2 kg

  • Only 37 subjects (10.5%) reported 1+ symptomatic hypoglycemic episode; no severe hypoglycemic episodes reported

Addition of a GLP-1 Agonist to Insulin Therapy in Type 2 Diabetes

Colleen Gavigan, Thomas W. Donner

  • Retrospective analysis assessing the efficacy of adding a GLP-1 (liraglutide, semaglutide, or dulaglutide) to the treatment regimen of 81 type 2s on insulin at baseline

  • Change in A1c from baseline was -0.94% at 3 months (p = 0.0007), -0.40% at 6 months (p = 0.06), and -0.58% at 12 months (p = 0.02)

  • Mean weight decrease over 12 months was 5.64 kg (p<0.0001)

  • Baseline BMI, insulin dose, and duration of diabetes did not significantly predict response to GLP-1 (i.e. A1c change)

Long-Term Safety of Intranasal Insulin (INI) in Insulin-Dependent Type 2 Diabetes (T2DM-IDDM): A Safety Substudy of Memory Advancement by Intranasal Insulin in Type 2 Diabetes (MemAID) Trial

Laura Aponte Becerra, Brahyan Galindo Mendez, Faizan Khan, Christos Mantzoros, Peter Novak, Vasileios Lioutas, Long H. Ngo, MemAID Investigators (J. Trevino), Vera Novak

  • RCT assessing the safety of long-term intranasal insulin (INI) use in insulin-dependent type 2s

  • 14 MemAID participants were randomized; 9 started treatment (5 INI, 4 placebo), 5 completed treatment & follow-up in total

  • No INI-related serious adverse events; 13 hypoglycemic events during study: 9 asymptomatic level-1 (2 INI, 7 placebo); 4 asymptomatic level-2 (2 INI, 2 placebo)

Mylan Insulin Aspart (MYL-1601D) Is Bioequivalent to Novolog (U.S.-Licensed Insulin Aspart) and NovoRapid (EU-Approved Insulin Aspart)

Yaron Raiter, Ulrike Hövelmann, Anoop Chullikana, Mark Liu, Charles M. Donnelly, Tracey Lawrence, Nilanjan Sengupta, Gopinath Ranganna, Abhijit Barve, Gopu C.L.

  • 66 subjects completed phase 1, randomized, double-blind, crossover 12-hour euglycemic glucose clamp study assessing the bioequivalence of MYL-1601D with NovoRapid and NovoLog

  • Geometric LS-mean ratio for 90% CI of PK endpoints (AUCIns_0-12h and CIns_max) & 95% CI of PD endpoints (AUCGIR_0-last, GIR_max) were within bioequivalence acceptance interval of 80% to 125%

  • Mean Insulin Aspart Concentration vs. Time by Treatment and Mean Glucose Infusion Rate Profile Curves were similar across all three treatments, as were safety profiles

Cellular Internalization and Localization of Once-Weekly Basal Insulin Fc (BIF)

Catherine Volk, Chen Zhang, Julie S. Moyers

  • Assessed in vitro cellular internalization of basal insulin Fc (BIF) compared to native insulin

  • Human Insulin Receptor internalization EC50 was significantly greater for BIF (4.67; n=3) than human insulin (0.0616; n = 3)

  • BIF and human insulin co-localized with Endosome Antigen 1 and had low levels of lysosomal localization

  • BIF stimulated IR internationalization to a similar maximum level, but had decreased potency compared to human insulin

Proposed Biosimilar Gan & Lee Insulin Lispro (GL-LIS) Shows Pharmacokinetic (PK) and Pharmacodynamic (PD) Bioequivalence (BE) vs. U.S.- and EU-Licensed Insulin Lispro (LIS)

Eric Zijlstra, Tim Heise, Marcel Ermer, Jia Lu, Michael G. Wilson, Leona Plum-Moerschel

  • 38 male volunteers completed phase 1 euglycemic glucose clamp study of three-period crossover design comparing GL-LIS to US- and EU-LIS (Humalog)

  • AUCLIS_0-12h and CLIS_max (PK parameters) had point estimates between 90% and 97%

  • AUCGIR_0-12 h and AUCGIR_max (PD parameters) point estimates for GL-LIS vs US-LIS: 101.5% and 100.8% respectively; for GL-LIS vs EU-LIS: 103% and 104%, respectively

  • The results demonstrate both PK- and PD-bioequivalence of GL-LIS to US- and EU-licensed LIS formulations

Efficacy and Safety of Insulin Glargine 300 U/mL (Gla-300) in People with Type 2 Diabetes Mellitus (T2DM) Uncontrolled on Basal Insulins (BI): ARTEMIS-DM Study

Bipin Sethi, Khalid Alrubeaan, Mustafa Unubol, Maria Aileen N. Mabunay, Mubarak Naqvi, Baptiste Berthou, Valerie Pilorget, Gustavo Frechtel

  • 372 type 2s with high A1c values on basal insulin from Asia, Middle East-Africa and Latin America switched to Gla-300 in this Phase IV study over 26 weeks

  • A1c fell 0.8% to 7.9% at week 26; 19% achieved <7% A1c target at Week 26

  • 30% of participants reported adverse events; 20% had a hypoglycemic event, only 1 (0.3%) was severe hypoglycemia

  • Results demonstrate that switching to Gla-300 with optimal titration was associated with improved glycemic control and low incidence of hypoglycemia in adults with uncontrolled T2D on basal insulin

Proposed Biosimilar Gan & Lee Insulin Aspart (GL-ASP) Shows Pharmacokinetic (PK) and Pharmacodynamic (PD) Bioequivalence (BE) to U.S.- and EU-Licensed Insulin Aspart (ASP)

Leona Plum-Moerschel, Ernestos H. Uhrmacher, Eric Zijlstra, Jia Lu, Michael G. Wilson, Matthew E. Barton, Tim Heise

  • 36 male volunteers completed phase 1 euglycemic glucose clamp study of three-period crossover design comparing GL-ASP to US-ASP (NovoLog) and EU-ASP (NovoRapid)

  • CASP_max and AUCASP_0-12h (PK parameters) point estimates between 96% and 98%

  • CGIR_max and AUCGIR_0-12 h (PD parameters) point estimates for GL-ASP vs US-ASP: 97% and 96%, respectively; for GL-ASP vs EU-ASP: 104% and 102%, respectively

  • Results demonstrate PK- and PD-bioequivalence of GL-ASP to US- and EU-licensed ASP formulations

Proposed Biosimilar Gan & Lee Insulin Glargine (GL-GLA) Shows Pharmacokinetic (PK) and Pharmacodynamic (PD) Bioequivalence (BE) to U.S.- and EU-Licensed Insulin Glargine (IG)

Tim Heise, Leona Plum-Moerschel, Grit Andersen, Jia Lu, Michael G. Wilson, Eric Zijlstra

  • 113 male patients with T1D completed phase 1 euglycemic glucose clamp study of three-period crossover design comparing GL-GLA to US-IG (Lantus) and EU-IG (Lantus)

  • CM1_max and AUCM1_0-24h (PK parameters) point estimates between 98% and 103%