And we’re back! After outstanding AADE pre-conference sessions, the first day of official conference programming featured sessions abuzz with conversation about CANVAS results (presented in full at ADA 2017). This CVOT found cardioprotective benefits to J&J’s SGLT-2 inhibitor Invokana (canagliflozin) alongside a nearly 2x risk for lower limb amputations. This presents a complicated risk/benefit profile, and educators are clearly interested in discussing real-world implications of this data – it’s hard to know, of course, the implications, given that comparisons between CVOTs are so challenging to make. See CANVAS commentary from Drs. Kittie Wyne, Katherine O’Neal, and Eden Miller below. We also include four diabetes technology highlights, covering insights from the great Drs. Bruce Bode, Jennifer Sherr, and more!
Diabetes Therapy Highlights
1. Ohio State’s Dr. Kittie Wyne returned to AADE to advocate for combination therapy, using the 2017 AACE treatment guidelines as a framework for her (highly compelling) argument. In addressing the amputation signal seen for J&J’s SGLT-2 inhibitor Invokana (canagliflozin) in CANVAS, Dr. Wyne called for a dedicated clinical study of patients with diabetes and peripheral vascular disease, which we think makes great sense. In the meantime, comparisons are hard to make, in our view.
2. Dr. Katherine O’Neal summarized the most important findings gleaned from diabetes CVOTs so far, highlighting a potential cardioprotective class effect for SGLT-2 inhibitors and more mixed data on CV benefit from GLP-1 agonist CVOTs.
3. In a Janssen-sponsored product theater, Dr. Eden Miller addressed the ~two-fold risk for lower limb amputations seen in the CANVAS trial for SGLT-2 inhibitor Invokana (canagliflozin). Notably, amputations didn’t come up in detail during her prepared remarks, but this safety signal arose during Q&A due to keen interest from educators in the audience. While Dr. Miller acknowledged that amputations are not to be minimized, she also emphasized that this risk is very low in the real world and can be well-managed with diligent monitoring.
4. Dr. Julie Gee presented original research to demonstrate the critical role that diabetes educators play in patient-centric care, where the healthcare professional functions to empower patients to engage in their own health and to effectively manage their diabetes. “Basically, CDEs rock – that’s what I found,” Dr. Gee concluded. Hear, hear!
Diabetes Technology Highlights
1. After Dr. Bruce Bode (“the hardest working man in diabetes”) and Ms. Lisa Kiplinger reviewed the need for and promise of insulin dose titration software, Dr. Bode shared some speculation on the landscape: he believes all three insulin companies (Novo, Lilly, and Sanofi) are committed to get the doses of their insulin devices to go to the cloud ASAP, “hopefully by the end of 2018.” He also added that Glytec is working with multiple companies to incorporate continuous glucose data into its titration system. We hope this connected pen and titration field starts to rapidly move to commercialization and scale from the big players, since it’s desperately needed and no real products are widely available yet.
2. Yale’s Dr. Jennifer Sherr shared her enthusiasm for use of the MiniMed 670G in adolescents, as well as the “lightbulb” moment she realized the power of automated basal insulin delivery.
3. A panel of experts that included representatives from BD, Onduo, WellDoc, Voluntis, and Healthy Interactions discussed the progress in diabetes management made possible by technology. The group expressed excitement for educators’ role in shaping the future of digital health, and explored why the field has not yet taken hold at his point.
4. Ms. Trish Comrie-Scheer provided valuable teaching tips for patients on the 670G system, focusing on setting realistic expectations and fostering trust in the system.
Diabetes Therapy Highlights
1. Getting to Goal with Combination Therapy, with Dr. Wyne
Ohio State’s Dr. Kittie Wyne returned to AADE to advocate for combination therapy, using the 2017 AACE treatment guidelines as a framework for her (highly compelling) argument. The ADA/EASD position statements perhaps have wider influence (“more often than not, these are the guidelines people hear about”), but Dr. Wyne described the merits of the AACE algorithm – namely, that it recommends earlier intervention with combination therapy if patients are diagnosed with A1c ≥7.5%, and recommends earlier initiation of basal insulin if patients are diagnosed with A1c >9%. That said, Dr. Wyne outlined hurdles to implementing combination treatment regimens in the real world, both on the provider-side and the patient-side. While a simultaneous approach to pharmacotherapy can be effective, clinical inertia often leads HCPs to wait 18-20 months before switching or adding agents, rather than the recommended three months – how depressing! Patients tend to equate the number of pills/injections they take to the severity of their diabetes, interpreting more medications as a sign of failure (type 2 patients are particularly averse to starting insulin injections for this reason, and Dr. Wyne explained how she sets expectations from day one that insulin will likely be needed, positioning it as a question of “when?” not “if?”). Fixed-dose and fixed-ratio combination products circumvent both of these obstacles, according to Dr. Wyne – providers can get their patients on a single tablet or injection that offers superior efficacy and a milder side-effect profile vs. component monotherapies. Moreover, Dr. Wyne emphasized how combo therapy allows you to target many aspects of the “ominous octet” (if not all eight) at once, echoing Dr. Susan Cornell’s perspective from Pharmacology Boot Camp the previous day. Notably, during this pre-conference session, Dr. Cornell defended GLP-1/SGLT-2 co-administration because this duo of drugs corrects all eight dysfunctions in the ominous octet. Our sense is that commercial enthusiasm for combination therapy lags significantly behind clinical enthusiasm, which is distressing from our view given the high number of patients not at their glycemic targets. Basal insulin/GLP-1 agonist fixed-ratio combos were perhaps the most highly-anticipated new therapy class in recent diabetes history – Dr. John Buse has gone so far as to say that Novo Nordisk’s Xultophy (insulin degludec/liraglutide) may be “the most effective anti-hyperglycemic agent on the planet” – and yet real-world providers in the US seem reluctant to prescribe them, now that Xultophy and Sanofi’s Soliqua (insulin glargine/lixisenatide) are finally available. To this end, we appreciated Dr. Wyne’s endorsement of earlier intervention with combo therapy, and we very much hope to see this take root in real clinical settings. We’re eager to see better reimbursement prospects for these products as they become more established on the market, particularly as “value-based therapy” takes hold.
- In addressing the amputation signal seen for J&J’s SGLT-2 inhibitor Invokana (canagliflozin) in CANVAS, Dr. Wyne called for a dedicated clinical study of patients with diabetes and peripheral vascular disease. When asked to share her two cents, she admitted “I really don’t know the answer,” but briefly discussed how many people with amputations in the CANVAS trial had baseline peripheral vascular disease, which heightened their risk for this adverse event. She mentioned that much more research is needed to better understand which therapeutic agents are safe and effective for this particular element of the diabetes patient population. Referring to the broader implications of CANVAS results, Dr. Wyne shared, “I can’t tell you we should take patients off this drug and put them on another,” because this outcomes trial has left us more questions than answers, at least for now, until post-hoc analyses further elucidate the safety signal. She added, “if someone’s doing well, and is stable on any therapy, I wouldn’t recommend changing it.”
2. Dr. O’Neal Reviews CVOTs Completed To-Date, Notes Benefit of SGLT-2 Inhibitor Class
Dr. Katherine O’Neal summarized the most important findings gleaned from diabetes CVOTs so far, highlighting a potential cardioprotective class effect for SGLT-2 inhibitors and more mixed data on CV benefit from GLP-1 agonist CVOTs. She set the stage by establishing diabetes as a CV disease risk equivalent. Regardless of age group, risk for CV events is magnified 2.5-4x in people with diabetes vs. the background population, and rates of CV death are 1.7x higher in people with diabetes vs. those without, according to Dr. O’Neal’s slides. We add that the recent CDC Diabetes State Burden Toolkit attributed 31% of diabetes-related deaths in 2015 to CV morbidity; without a doubt, CV disease is the leading cause of mortality among people with diabetes. All this makes CV risk reduction a critical component of diabetes management, which means providers have to be acutely aware of the differential CV effects of various drug classes, and of various agents within the same class.
- On SGLT-2 inhibitors, Dr. O’Neal reviewed the 14% relative risk reduction for the primary endpoint of three-point MACE (non-fatal MI, non-fatal stroke, CV death) seen for both Lilly/BI’s Jardiance (empagliflozin) in EMPA-REG OUTCOME and for J&J’s Invokana (canagliflozin) in CANVAS. The EMPA-REG OUTCOME study set off excitement for cardioprotective diabetes drugs – until then, these trials had demonstrated CV safety, at best – and CANVAS helped corroborate this as a very real benefit in improving CV outcomes. An FDA Advisory Committee recommended a new CV indication for Jardiance by a tight 12-11 margin in June 2016, and the agency followed suit with an expanded indication for the reduction of CV death in December. The DECLARE CVOT for dapagliflozin (AZ’s Farxiga) is expected to complete in the second half of 2018 according to AZ management. The VERTIS CV trial for Merck/Pfizer’s ertugliflozin (still pending FDA approval, with a decision expected by end of year) is expected to complete in October 2019. Both of these trials will give prescribers a better idea of any potential cardioprotective class effect for SGLT-2 inhibitors. Dr. O’Neal noted that researchers don’t yet understand the exact mechanism of CV benefit from empagliflozin or canagliflozin, suggesting that upcoming trials should help clarify this as well. Across her whole presentation, Dr. O’Neal seemed to be most impressed by the CV data for SGLT-2 agents. In our view, it’s enormously exciting that Jardiance is now the first diabetes drug with a CV indication – patients are accustomed to taking medicine to lower glucose, but now there’s a therapy available that can actually help prevent CV death. We heard from J&J management that the company plans to file with the FDA by end of September, requesting a new CV indication for Invokana.
- GLP-1 agonists have shown more inconsistent data in CVOTs, suggesting that cardioprotection may not be a class effect here. While Victoza (Novo Nordisk’s liraglutide) demonstrated an impressive 13% risk reduction for three-point MACE in the LEADER trial, Adlyxin (Sanofi’s lixisenatide) showed neutral CV effects in ELIXA, and topline results from the EXSCEL trial revealed neutral CV effects for Bydureon (AZ’s exenatide once-weekly) as well. An FDA Advisory Committee recently voted 17-2 in favor of granting a CV indication to Victoza based on LEADER data (the EMA recently approved this CV indication for Victoza’s European label). Ultimately, cardioprotection may not be a class effect for GLP-1 agonists although it may also be that some percentage of trials could be too short to show the impact. We look forward to the presentation of full EXSCEL results at EASD 2017 for more insight into these differential CV effects. One theory (pure speculation, at this point) is that human GLP-1-based drugs are showing a CV benefit (this applies to both liraglutide and Novo Nordisk’s not-yet-approved semaglutide), while exendin-4-based drugs are not (i.e. lixisenatide, exenatide once-weekly, and exenatide in a mini-pump as Intarcia’s not-yet-approved ITCA 650).
- On DPP-4 inhibitors, Dr. O’Neal described the signal for heart failure hospitalization associated with AZ’s Onglyza (saxagliptin) in SAVOR-TIMI. The EXAMINE trial for Takeda’s Nesina (alogliptin) found an imbalance in heart failure hospitalizations, with more occurring in the alogliptin group vs. placebo, but this did not meet criteria for statistical significance. Still, the FDA added warnings for heart failure to the Onglyza and Nesina labels in April 2016. Notably, the TECOS trial of Merck’s DPP-4 inhibitor Januvia (sitagliptin) reported a decidedly neutral hazard ratio of 1.00 for heart failure hospitalization, but the FDA issued a Complete Response Letter for inclusion of this data on the product label. Overall, Dr. O’Neal didn’t voice her personal clinical opinion one way or the other on DPP-4 inhibitors and any possible heart failure risk.
- After reviewing the evidence, Dr. O’Neal concluded that the current sequence of drugs in diabetes treatment algorithms should not be changed. She defended metformin as first-line therapy, pointing to positive CV effects associated with the drug in UKPDS. While we recognize metformin’s advantage of low cost and reasonable safety/efficacy, we also want to note the danger of putting too much faith in metformin: Too many patients remain on first-line therapy with no changes or additions to their treatment regimen for far too long, due to clinical inertia and other real-world factors. Now that we have cardioprotective therapies in our treatment arsenal, we’d love to see those used earlier on in the course of disease. In fact, Dr. Jay Skyler recently argued at Keystone 2017 that metformin is overemphasized in guidelines and that DPP-4 inhibitors should be phased out of diabetes management, replaced with GLP-1 agonists and SGLT-2 inhibitors, agents that might offer cardioprotection rather than neutral CV effects and possible heart failure risk. We do see DPP-4 inhibitors as having the cleanest side-effect profile of any drugs and believe they have a place with some newly diagnosed patients in particular.
3. Educators Ask: What Do We Make of the Amputation Signal in CANVAS?
In a Janssen-sponsored product theater, Dr. Eden Miller addressed the ~two-fold risk for lower limb amputations seen in the CANVAS trial for SGLT-2 inhibitor Invokana (canagliflozin). Notably, amputations didn’t come up in detail during her prepared remarks, but this safety signal came up during Q&A due to keen interest from educators in the audience. While Dr. Miller acknowledged that amputations are not to be minimized, she also emphasized that this risk is very low in the real world and can be well-managed with diligent monitoring. In the full integrated dataset from CANVAS and CANVAS-R, there were 187 lower-extremity amputations, occurring a rate of 6.3/1,000 patient-years in the canagliflozin arm vs. 3.4/1,000 patient-years in the placebo arm. Dr. Miller likened this to increasing the chances of a rare event from 1% to 2%, or to “doubling your chances of getting hit by an asteroid” (to be clear, she underscored that this statement was only a joke to illustrate a point about base rate bias, and that amputation-related concerns are of course valid). Moreover, Dr. Miller explained how the vast majority of amputation events in CANVAS were linked to a precipitating factor, often a diabetic foot ulcer. She encouraged vigilant monitoring – “every time my patients come in, shoes are off” – and stronger patient education around proper foot care in diabetes.
- This echoes a view we’ve heard from many thought leaders in the field: Janssen’s VP of Medical Affairs for Cardiovascular and Metabolism Dr. Robert Cuddihy told us that amputations in the CANVAS trial were usually preceded by an infection or some other warning sign, and Global Therapeutic Head of Cardiovascular and Metabolism Dr. James List suggested that this safety finding could kick-start a movement toward better patient education on foot care. We’d love for J&J to be a leader in this initiative. UCLA’s Dr. Anne Peters shared that Invokana sometimes results in greater A1c-lowering and weight loss for her patients vs. Lilly/BI’s Jardiance (empagliflozin). She supported Dr. Miller’s opinion that HCPs don’t need to switch all their patients off of Invokana, especially since there are ways to avoid lower limb amputations with regular monitoring in real-world practice.
- We’d also add that CV morbidity/mortality seems to have a greater burden within the diabetes patient population vs. lower-extremity amputations (CV disease was the catalyst for 31% of diabetes-related deaths in 2015, according to the CDC), so this safety signal shouldn’t completely overshadow canagliflozin’s significant 14% risk reduction for three-point MACE (non-fatal MI, non-fatal stroke, and CV death). That said, a risk/benefit analysis of CANVAS showed that for every 1,000 patients treated with canagliflozin for five years, 23 fewer MACE events and 17 fewer heart failure hospitalizations could be expected, but also 15 additional lower limb amputations (10 at the level of the toe/forefoot, five above the ankle). This is a murky risk/benefit profile that will need much more unraveling through post-hoc analyses of CANVAS and CANVAS-R; to date, very respected diabetes care providers continue to voice support for Invokana and for the SGLT-2 class, despite the FDA issuing a boxed warning for amputations on all canagliflozin-containing medicines, and despite the EMA investigating this risk for all SGLT-2 inhibitor products.
- Dr. Miller was limited in what she could say about Invokana’s CV benefits since these haven’t yet been incorporated on the product label in any form, but she urged educators to read the NEJM paper on integrated CANVAS results. We learned from Dr. List that J&J plans to file a Supplemental New Drug Application (sNDA) with the FDA requesting a CV indication for Invokana by end of September, and we’d be happy to see this product join the ranks of Jardiance, which is now indicated for the reduction of CV death.
4. Promotion of Diabetes Self-Management: Study Reveals “Diabetes Educators Do It Better”
Dr. Julie Gee presented original research to demonstrate the critical role that diabetes educators play in patient-centric care, where the healthcare professional functions to empower patients to engage in their own health and to effectively manage their diabetes. Via an online survey sent to a database of AADE-accredited CDEs (n=225, which corresponds to a 30% response rate, fairly high for a survey), Dr. Gee’s team collected data on beliefs related to the idea of patients as “self-managers” (as measured by the Clinician Support for Patient Activation Measure [CS-PAM] scale) and on how extensively diabetes educators employ strategies to foster self-management (as measured by the Clinician Self-Management Scale). The study population of CDEs scored remarkably highly on both scales, indicating strong belief in the importance of self-management paired with a strong tendency to support self-management in their practice. In the domain of activating and inspiring patients to take on the role of self-manager, the study population exhibited an average CS-PAM score of 77.7 out of 100 – the highest seen in the literature to-date among any healthcare specialty, according to Dr. Gee (we think this is very impressive). This CS-PAM score translates to endorsement of ideas like “Patients should want to be involved as a full partner with me in making decisions about their care” and “Patients should know what each of their prescribed medications is for.” In terms of clinical practice to support self-management, the study population scored an average of 4.3 out of 5 on the Clinician Self-Management Scale – also very high. This translates to high levels of agreement with statements like “Tell the patient you will be their coach, but that they are the one who has to carry out the plan” or “Ask the patient what change s/he wants to focus on.” There were no significant differences in CS-PAM or Clinician Self-Management Scale scores according to CDE’s age, years of experience, or discipline (registered nurse, dietician, pharmacist, etc.) Moreover, Dr. Gee’s research revealed a significant correlation between CS-PAM score and Clinician Self-Management Scale score, indicating that the more favorably CDEs view self-management practices, the more likely they are to adopt clinical strategies that promote this. These findings exemplify CDEs’ leadership in the ongoing movement toward less paternalistic care that instead invites the patient to be a shared decision-maker. “Basically, CDEs rock – that’s what I found,” Dr. Gee concluded, hence her creative talk title, “Proof That Diabetes Educators Do It Better.” Hear, hear! We sincerely hope that Dr. Gee’s findings help underscore the absolutely critical work of diabetes educators, some of the most patient-empowering members of the care system. The need for increased awareness of CDEs’ work was illustrated best by one audience member, who remarked during Q&A: “We’re changing lives, but nobody knows that and nobody values us. CDEs are being scrapped in many health centers. We need to promote this information and educate everyone out there about who we are.” This was met with a standing ovation – one hugely deserved!
Diabetes Technology Highlights
1. Dr. Bode: All Three Insulin Companies Committed to Doses -> Cloud, Hopefully by End of 2018; Glytec in talks with CGM players
After Atlanta Diabetes Associates’ Dr. Bruce Bode (“the hardest working man in diabetes”) and Ms. Lisa Kiplinger overviewed the need for and promise of insulin dose titration software, Dr. Bode shared some speculation on the landscape. Dr. Bode believes all three insulin companies (Novo, Lilly, and Sanofi) are committed to get the doses of their insulin devices to go to the cloud ASAP, “hopefully by the end of 2018.” Dr. Bode noted that Novo Nordisk is working with their own disposable pens, Sanofi is working with pen caps, and Lilly is working with pen cartridges with Companion Medical. We were interested to hear the news of Novo Nordisk’s work on disposable pens, since doing so would seem like an enormous disposable device change/addition in just 17 months! That Sanofi and Lilly are working on this was not surprising, given respective work with Common Sensing and Verily (Sanofi) and Companion Medical (Lilly). More details from our recent coverage are below. Dr. Bode also noted that Glytec is working with multiple companies to incorporate continuous glucose data into its titration software – great to hear, since current iterations only use BGM data. We’re glad to hear of all of these – given the dire need to use insulin more intelligently through data – though the field needs these to move to commercialization, scale, reimbursement, and outcomes. Many have been calling for dose capture for years, but no products are widely available yet (Companion Medical’s InPen might be the first, expected to launch this year following FDA clearance in 2016).
- We learned recently that Novo Nordisk is piloting a Novo Pen 5 Plus with NFC capabilities at 10 clinics in Sweden – Bluetooth is in the roadmap, which is clearly needed to realize the true vision of continuous, hassle-free data upload and continuous titration. For now, NFC on the Novo Pen 5 Plus enables data upload via a Glooko/Diasend NFC pad. We’re glad to see Novo Nordisk taking its first steps in digital health, including launching a data analysis and education app with Glooko two weeks ago (Cornerstones4Care Powered by Glooko). The company’s new digital health unit must be learning a lot and we hope to see big things from its collaborations with Glooko and eventually IBM Watson.
- Dr. Bode’s point did bring up a good question: Which will patients prefer – disposable pens with connectivity via attachments/caps (e.g., BD’s smart pen needles, Bigfoot/Timesulin dose capture device, Common Sensing) or durable/reusable pens with connectivity built right in (e.g., Companion Medical InPen). Where can a more sustainable business model be built? What will patients, providers, and payers be willing to pay for? What is lower hassle in the current healthcare systems around the world? How valuable will insulin dose data be and what device price premium might it command? Will connectivity + paired apps emerge as a competitive advantage for different insulins, or will insulins become a commodity and the true differentiator will actually be the apps/education built around them? We are huge proponents of this field and hope to see many, many products launch for capturing the dose data and using it effectively!
- Glytec’s outpatient eGlycemic Management System (eGMS) is currently partnered with Telcare, Livongo, and Agamatrix, Roche (this last one is news to us!), and the company is working with additional connected devices – this presumably includes CGM, since passive glucose data collection is essential for making this product simple and low burden. The growing body of literature on titration has shown that basing adjustments solely on SMBG data is very effective, but we can only imagine that the depth and breadth of glucose data offered by CGM would result in even better outcomes. As a reminder, Bigfoot acquired Timesulin in June and partnered with Abbott last month, giving it all of the pieces required to generate insulin dosing recommendations for MDIs based on continuous glucose data. Said Ms. Kiplinger, “You see the FDA clearance dates for these products are almost all 2017 … the field has really rapidly changed. Last time this happened was when the Internet came about.”
- It’s no surprise that Sanofi and Lilly are actively exploring the development of connected pens and caps. Sanofi has invested in Common Sensing (Gocap manufacturer – very exciting data at ADA), and Lilly has invested in Companion Medical (FDA-cleared InPen manufacturer; launch most recently expected this year). Further, both companies have embraced dose titration via apps (see Sanofi’s My Dose Coach and Voluntis partnership and Lilly’s Go Dose), and enhancing these systems with passive dose capture makes logical sense. It will be interesting to watch these three pharmaceutical giants enter the tech field – at what pace will products come out, how will the companies outsource vs. build internally, and how will they maintain products over time in an ever-changing app world? The insulin players know how to compete on drug products, but the tech world is a giant cultural shift. How will this play out? We assume once one major player launches a connected pen, the others will most certainly have no choice but to follow. Right now, it’s a question of game theory – who will be first? Connected pens feel inevitable at this stage – undoubtedly a “when” more than an “if” – meaning all three should be investing in connected delivery devices now; the R&D must be ready to translate to commercialization in the next couple years. Passive dose capture paired with glucose data and titration software strikes us as one of the richest areas for insulin innovation – and far less expensive and risky than bringing a brand-new insulin to market (though we’ll take that too).
- According to Dr. Bode, Glytec is on track to treat more than 100,000 patients this year, more than doubling since last year! The system is clearly very scalable, and based on retrospective data presented at that same ATTD presentation, is proving effective – patients (n=5,718) admitted with hyperglycemia to one of seven hospitals arrived at their prescribed target blood glucose in 0.8 days from a starting average of 262 mg/dl. Once at target, 68% of blood glucose readings remained between 70-180 mg/dl. Hypoglycemia was very unlikely once target had been reached and in the next 24 hours, with just 0.001% of time below 40 mg/dl and 0.01% of time below 70 mg/dl, respectively. Wow!
- Dr. Bode reminded the audience that “it’s a dosing problem, not an insulin problem.” Patients frequently discontinue their insulin regimens due to a number of barriers, and if they didn’t, the US health system could save an estimated ~$5,000 per patient per year – assuming 30% of people with diabetes take insulin, that adds up to a shocking $45 BILLION per year! He rattled off a number of alarming statistics, even for those following the field: 31% of patients never fill their script; only 50% modify their dose after an episode of hypoglycemia; 40%-60% experience hypoglycemia; 40% experience hypoglycemia in the first month; 50% of patients taking basal insulin are not at their A1c goal; and 77% discontinue insulin within 12 months if they experience hypoglycemia in the first six months. Not only could insulin dose titration make dosing of insulin safer and more effective, but it would also crucially make patients think that insulin is safer, resulting in better adherence, health, and cost savings.
- Using a titration software such as Glytec’s, Dr. Bode estimated that a CDE could feasibly monitor and treat up to 300 patients at any given time. Ms. Kiplinger explained that the duty of the educator (at least in her clinic) is to educate the patient (when and why to check blood glucose, how to use the system, what to expect), coordinate the care team, and monitor the patient (for safety). Since the software takes care of the rest, noting when a patient is in need of an adjustment and calculating the new dose, the educator simply has to fill in the cracks and support patients. We wonder how clinician-facing titration software will compare to those that are patient-facing – do the former offer a better framework to support the user, resulting in better engagement? Patients-facing software that mostly runs on its own is more scalable, but will drop-off be higher?
2. Dr. Jennifer Sherr Excited About 670G in Adolescents, Shares Her 670G “Lightbulb Moment”
Yale’s Dr. Jennifer Sherr shared her enthusiasm for use of the MiniMed 670G in adolescents, as well as the moment she realized the power of automated basal insulin delivery. She was most excited about the adolescent data from the 670G pivotal study, noting that “teenagers are going to be teenagers,” and the hybrid closed loop system should enable improved glycemic control, while still allowing young patients to be themselves. (In addition to overnight, the big benefit in teens in the pivotal came in stemming post-breakfast highs.) Indeed, the 670G pivotal found that adolescents were in Auto Mode a solid 76% of the time (vs. 88% in adults), which stayed roughly consistent in real-world use at the Barbara Davis Center in one-year data (ADA 2017). In the recent customer training phase, median time in Auto Mode in ALL users improved to 92%, though adolescent-only data was not broken out. Like other 670G prescribers we’ve heard from, Dr. Sherr was particularly taken with the tight control achieved overnight via dynamic insulin delivery. Her lightbulb moment came when she was looking at a patient’s data and realized that insulin delivery was halted for two whole hours. In her own words, “I agonize over basal rates, and I realized there’s no way I can get this degree of control [with manual insulin adjustments].”
- Dr. Sherr wrapped up her presentation with a brief discussion of the limitations and strengths of the pivotal trial, acknowledging that the lack of a control group and imbalance between the two-week run-in and three-month study phase data are problematic. Still, the data was collected over a broad age-range at 10 centers and strongly suggests the 670G to be a safe and effective device, at least in the engaged group in this study.
- In the coming years, we hope to see Medtronic’s closed-loop devices move to simplifying the user interface, optimizing training and prescribing, minimizing daytime user burden and improving control further (e.g., automatic correction boluses), reducing sensor calibrations, adding Bluetooth and smartphone connectivity, personalizing algorithms further, and dropping costs.
- Dr. Sherr was impressed with the new features of the 670G, highlighting the highly accurate Contour Next Link 2.4 BGM. The automatic wireless transmission of blood glucose data to the pump minimizes entry errors and shaves seconds off diabetes management time. It’s also a major plus to have this highly accurate meter driving calibration for the Guardian Sensor 3. While a few seconds might not seem significant, as Dr. Sherr noted, when managing a chronic disease over a lifetime, every few seconds really count.
- As a reminder, Medtronic has had some sensor shortages near-term, which are presumably gating a wider launch of the system to the full 20,000+ Priority Access Program participants. As of Keystone in mid-July, ~1,000 patients had used the 670G, implying it had reached a couple hundred additional people beyond the Customer Training Phase (~750 people) and pivotal study (124 participants).
- Dr. Sherr kicked off the 6 AM symposium with the infamous T1D Exchange curve indicating that only 25% of patients with type 1 diabetes are reaching their A1c goal. To our delight, she added that it may be time to move beyond A1c and talk about other metrics like time-in-range, referencing a conference “last month” (the Glycemic Outcomes Beyond A1c Workshop hosted by the diaTribe Foundation). Indeed, we’d love to see benchmark data on T1D Exchange registry members – what fraction spend 25%, 50%, 75%, and >75% of the day in 70-180 mg/dl?
3. Educators to Play an Important Role in Shaping Digital Health; Roadblocks Associated with Adherence, Burden, and Reimbursement Ahead
A panel of experts discussed the progress in diabetes management made possible by technology, expressing excitement for the role educators will play in shaping the future of digital health. Given their in-depth experience troubleshooting diabetes therapies, BD’s Ms. Rita Saltiel-Berzin noted the potential educators have to be instigators of change in the technology landscape, noting that some team members “have no idea what kinds of questions to ask” when designing digital health solutions. Since questions often arise in real-time, it will be critical to understand what patients ask and to help them tackle teachable moments – these almost never occur in the presence of providers. While there is certainly a lot to be excited about (Voluntis’ Ms. Ellie Strock pointed to the freshly-announced FDA pre-certification process for app clearance), there is still much to be accomplished. Mr. Paul Lasiuk, CEO and co-Founder of Healthy Interactions, outlined four key reasons why, despite his estimate that ~$8 billion has been invested in digital health this year, a large chunk of the capital most likely won’t prove fruitful: (i) There’s a massive disconnect between consumer and healthcare use – getting people to be active in their healthcare management has not translated very well, thanks in part to a systemic culture of viewing clinicians as responsible for fixing problems; (ii) The most common diabetes demographic, that of a slightly older population, views the smartphone strictly as a phone and not a device with wide-ranging uses; (iii) In a space occupied by 165,000 apps (of which more than 1,100 are diabetes-related), the digital health world is a crowded, confusing place with ambiguous regulation; and (iv) Reimbursement issues are a major barrier. For all of these reasons, Mr. Lasiuk cautioned that digital technology is not a silver bullet, and easing of patient burden must be emphasized. These were fair criticisms, though we must remember the field is at a nascent stage and still laying important groundwork now for future scale. As WellDoc’s Vice President for Clinical Advocacy Malinda Peeples pointed out, Welldoc has been able to effectively translate their products for real-world use and have seen excellent senior engagement in their technology; in fact, those over 60 years-old are some of their most active users! Dr. Deborah Greenwood echoed Mr. Lasiuk, pushing for simplicity. She recalled a poignant statistic indicating that if people with diabetes did everything providers asked of them, they would need to allot 2.5 hours/day to their self-care – insane, especially because today’s technology hopes to diminish burden. We agree that patients, CDEs, and others on the front line will absolutely need to be the first consulted as manufacturers approach digital health, as only they can comprehend the lived experience of diabetes and how an intervention would fit in and add value.
- One of the most common clinician/educator fears is whether technology will one day replace in-person healthcare – when prompted with this question, the panel responded with a resounding no. Onduo’s Paula Leclair noted that, as technology evolves, so will the job of the diabetes educator, especially as lifestyle coaches become more integrated (much like the industrial revolution created many new jobs and changed the jobs of farmers, but didn’t necessarily put them out of work). Still, as Ms. Saltiel-Berzin pointed out, the in-person aspect is critical for accountability – a patient may not feel as accountable to Alexa or Siri or an app as much as a real person. (Of course, incentives, joyful experiences that make life better, and smart remote monitoring might change this.) Furthermore, as per Ms. LeClair, Siri is not exactly whom she would want to talk to if in DKA. Instead, she views technology as an adjunct that is part of the patient-provider relationship. Mr. Lasiuk agreed, defining human interaction as the cornerstone of care, with technology as a facilitating toolkit.
Questions and Answers
How are we going to pay for this? How will our insurers and Medicare pay for educators to see patients in the digital scene?
Dr. Greenwood: I was really disappointed to see Medicare not cover the digital DPP. I thought this would be our breakthrough. Supposedly they will cover digital make-up classes but that was a huge blow and I was expecting they would cover it.
4. Tips on How to Manage Patient Expectations on the 670G
Ms. Trish Comrie-Scheer provided valuable teaching tips for patients on the 670G system, focusing on setting realistic expectations and fostering trust in the system. As we have heard time and time again (recently during a panel at Keystone), it’s critical that patient expectations of hybrid closed loop systems are managed appropriately. Ms. Comrie-Scheer suggested emphasizing that this is not a panacea designed to eliminate all user interaction; patients are still responsible for mealtime boluses and correction doses, as well as calibration, setting temporary targets for exercise (if desired), etc. In fact, like many new devices, 670G demands a higher level of engagement at first, and it will be important to remind patients of this before starting the system. She also recommended explaining to patients that glucose levels will not be perfect, and the system will initially be conservative, titrating insulin delivery in a more personalized manner over time. (Previous total daily doses inform the algorithm’s aggressiveness.) Echoing commonly held sentiments, Ms. Comrie-Scheer discussed the difficulties in encouraging patients “to let it ride” – trusting the system can often be challenging for patients accustomed to aggressively managing their diabetes. She suggested reminding patients that the algorithm is usually smarter than we are, and that it may act to normalize blood glucose more slowly than patients are used to (actually a benefit, considering the dangers of insulin stacking and rage boluses). According to Ms. Comrie-Scheer, patients may also initially struggle with new terminology; many are used to religiously monitoring A1c levels, thus making the interpretation of time-in-range a difficult transition requiring new skills and training. She further advised providers to educate patients on how to analyze percent time spent in the low and high glucose ranges – we hope to see robust training materials and guidance built out on this front.
- Ms. Comrie-Scheer provided a touching case study of a 14-year-old type 1 patient, who achieved massive improvements using the 670G – time-in-range increased by seven hours per day (from 44% to 73%)! See some of our favorite quotes from the patient and her family below.
- “This is the thing that allows mommies to sleep through the night.”
- “It’s really exciting … it keeps my sugar at a steady level.”
- “With type 1 diabetes, you’re making decisions about your blood sugar minute by minute … that’s a huge burden especially on a kid … this new pump frees up a lot of brain space to think about the kinds of things a kid should be thinking about.”
-- by Ann Carracher, Abigail Dove, Brian Levine, Payal Marathe, Maeve Serino, Adam Brown, and Kelly Close