April 1-4, 2017; Orlando, FL; Days #3-4 Highlights – Draft

Executive Highlights

Our team has waved goodbye to the “Theme Park Capital of the World” after the wild ride that was ENDO 2017! The last 1.5 days of the meeting was chock-full of intriguing data – from real-world studies and phase 3 meta-analyses, from early-stage therapies, from novel therapeutic targets, from UVA’s modified hybrid closed loop system, and much more. We also heard fascinating takes on the soda taxes, future directions for CVOTs, and how to digest the FOURIER outcomes results for PCSK9 inhibitor Repatha. Below, we include our top 13 highlights, followed by additional detailed discussion and commentary.

Each year, this meeting seems to get better and better and we can’t wait to see what next year brings – mark your calendars for the centennial (!!) meeting March 17-20 in Chicago (one of our very favorite conference cities!)

Top 13 Highlights

1. A poster presentation featured data from the Sanofi-sponsored real-world DELIVER 2 study, demonstrating significantly lower risk of hypoglycemia for type 2 diabetes patients who switched to Toujeo (insulin glargine U300) versus other basal insulins.

2. Dr. Adrian Vella presented results from the phase 2 AGATA study of vTv Therapeutics’ glucokinase activator (GKA) TTP399, demonstrating A1c reductions and weight loss in adults with type 2 diabetes.

3. UVA’s Dr. Mark DeBoer presented data on a modified hybrid closed loop system (Tandem pump + Dexcom G4 + DiAs control algorithm) in children 5-8 years old. In a 68-hour “resort study,” the kids (n=12) spent 73% of the time between 70-180 mg/dl, a marked improvement from 47% under usual home care. Mean glucose dropped remarkably from 190 to 152 mg/dl – whoa! We’re glad to see AID studies now moving into younger groups, an area with huge need.

4. Two posters on meta-analyses of the SUSTAIN phase 3 program for Novo Nordisk’s once-weekly GLP-1 agonist semaglutide highlighted the agent’s (i) weight loss and A1c-lowering efficacy regardless of background oral therapy and (ii) its consistent lowering of both fasting and postprandial glucose.

5. According to UNC’s Dr. Shu Wen Ng, the sugar-sweetened beverage industry has reported a ~40% reduction in sales in Philadelphia since the soda tax passed in 2016.

6. In a poster presentation, Dr. Ou Li shared fascinating preclinical data for XOMA 129, an insulin receptor negative modulator for the treatment of insulin-induced acute hypoglycemia. Also, Dr. Kirk Johnson presented promising preliminary phase 2 data for insulin receptor antibody XOMA 358 in 13 patients with post-bariatric surgery hypoglycemia

7. University of Pennsylvania’s Dr. Michael Rickels presented five year results from a study of metabolic and beta cell functional outcomes in people with type 1 diabetes following human islet transplantation. All subjects achieved near-normal glycemic control (A1c < 6.5%), with 10 of 11 maintaining A1c<6.5%, and 70% were entirely insulin-independent at five years post-transplantation, with stable beta cell secretory capacity over time.

8. Dr. Livio Luzi presented intriguing evidence that noninvasive electromagnetic brain stimulation can contribute to weight loss, potentially by changing the composition of the gut microbiome.

9. University of Colorado’s Dr. Neda Rasouli offered her perspective on where we stand – results from nine modern diabetes cardiovascular outcomes trials (CVOTs) in hand – and where we need to go in order to make these long, costly trials more useful for clinical decision-making.

10. Just a couple weeks following presentation of FOURIER results for Amgen’s PCSK9 inhibitor Repatha (evolocumab), Louisville’s Dr. Harold Bays offered commentary suggesting that a longer outcomes trial would reveal even greater benefit and cautioning that we must refrain from claiming a rdioprotective class effect at this time.

11. The great Dr. Jay Skyler, hailing from the University of Miami, gave an excellent overview of the overall disappointing collection of studies in type 1 diabetes primary prevention, secondary prevention, and intervention. A large majority of studies in each domain have had limited to no effect, though some immunomodulatory approaches in cases of new onset type 1 diabetes have shown a transient or potential effect.

12. In the 2017 Roy O. Greep Award for Outstanding Research Lecture, University of Pennsylvania’s Dr. Klaus Kaestner showed how beta cells can be epigenetically modified to reproduce in vitro and in vivo.

13. “We need a statin for weight loss – a drug that is well-tolerate, cheap, efficacious, and has hard endpoints and clinical data.” Speaking to a packed room (impressive for the very last session of the meeting!) Denver Health Medical Center’s renowned Dr. Daniel Bessesen lamented that too few patients are on obesity pharmacotherapy currently and was none-too-optimistic about the current options in the pipeline.

Top 13 Highlights

1. Real-World Study Shows Lower Hypo Risk for Patients Who Switch to Toujeo versus Other Basal Insulins

A poster presentation featured data from the Sanofi-sponsored real-world DELIVER 2 study, demonstrating significantly lower risk of hypoglycemia for type 2 diabetes patients who switched to Toujeo (insulin glargine U300) versus other basal insulins. According to observational data from electronic medical records, patients (n=1,894) on on Lantus or Levemir (75% and 25% of the cohort at baseline, respectively) who switched to Toujeo experienced 33% fewer hypoglycemic events over six months than their counterparts who switched to other basal insulins, including Sanofi’s Lantus (insulin glargine U100), Novo Nordisk’s Levemir (insulin detemir), and Tresiba (insulin degludec) (7.98 vs. 5.32 events per 100 patient-months; p<0.01). Furthermore, patients who switched to Toujeo additionally experienced a 48% reduction in hypoglycemia events requiring inpatient or emergency department care (3.82% of patients vs. 1.97%; p<0.01) versus those on other basal insulins. Both groups experienced a significant reduction in A1c over the six-month study period, falling from a baseline of 8.9% to 8.4% in the patients who switched to Toujeo and 8.5% in those who switched to a different basal insulin (p<0.01). There was no significant difference in the magnitude of A1c improvement between the groups, indicating that Toujeo’s reduction in hypoglycemia risk came without compromising blood glucose control – Toujeo was also non-inferior to other basal insulins in terms of A1c goal attainment. The observational nature of this data limits our understanding of how insulin dosing compared between the patients on Toujeo versus other basal insulins, and it is possible that hypoglycemia was underreported since no blood glucose or CGM data was collected. Furthermore, notably, this study only looked at patients who were already on a basal insulin who switched to a new basal insulin – presumably, patients who were well-controlled on Lantus or Levemir would not have switched in the first place (unless forced to by formularies) and thus would not have been included in the study. That said, DELIVER 2 provides valuable information that Toujeo can improve hypoglycemia (and lower A1c) for type 2 diabetes patients in a real-world setting. This data further supports findings from randomized, controlled clinical trials indicating that Toujeo – and next-generation basal insulins overall – are a major step above first-generation analogs, particularly in light of their hypoglycemia benefits and the greater stability associated with the insulins. We would have loved to see a separate analysis of outcomes following switches to Toujeo vs. switches to Tresiba only. Tresiba has already demonstrated significant hypoglycemia reductions in the dedicated SWITCH clinical trial program. Continue on to our detailed discussion and commentary section below for more on DELIVER 2, and see Sanofi’s press release summarizing the results here.

2. Phase 2 Data for vTv Therapeutics Glucokinase Activator TTP399

Dr. Adrian Vella presented results from the phase 2 AGATA study of vTv Therapeutics’ glucokinase activator (GKA) TTP399, demonstrating A1c reductions and weight loss in adults with type 2 diabetes. The six-month, double-blind trial randomized 190 patients with type 2 diabetes to one of four arms (i) TPP399 400 mg; (ii) TTP399 800 mg; (iii) DPP-4 inhibitor sitagliptin (Merck’s Januvia); or (iv) placebo. Patients treated with TPP399 800 mg experienced a mean placebo-adjusted A1c reduction of 0.9% at six months (baseline A1c 8%, p<0.001 vs. placebo), which was comparable to sitagliptin’s placebo-adjusted A1c reduction of 1%. Treatment with TPP399 400 mg produced a non-significant A1c reduction of 0.2%. There were no significant changes in fasting plasma glucose with either dose of TPP399 compared to placebo, suggesting that the candidate has an effect on postprandial glucose. The overall group treated with TTP399 800 mg experienced a mean placebo-adjusted weight loss of ~1.5 kg (~3.3 lbs), but the result was not statistically significant. That said, a pre-specified sub-analysis of weight in patients with baseline body weight >100 kg (220 lbs) was positive: TPP399 800 mg produced a statistically significant weight loss of 3.4 kg (7.5 lbs) in these patients (p<0.05). In addition, TPP399 800 mg was associated with a significant decrease in fasting plasma glucagon (19.6 mg/dl, p=0.012) and increase in HDL cholesterol (3.2 mg/dl, p<0.05). Notably, Dr. Vella reported that TTP399 did not cause hypoglycemia or increase blood pressure or triglycerides. There were also no detrimental changes in liver enzymes associated with TPP399 therapy. Overall, this is very reassuring as several candidates GKA candidates (AZ, Roche, Takeda, Merck, Amgen, Pfizer, Daiichi Sankyo, and Zydus Cadilla) have been discontinued in previous years due to hypoglycemia and/or liver toxicity concerns – notably, TPP399 is differentiated as a liver-selective GKA and the data thus far suggests that it may be able to avoid some of these safety concerns. We previously saw topline results from the AGATA study in August and we’re glad to see that the full results continue to paint a promising picture for this candidate. Furthermore, as Dr. Vella noted, GKA is a novel diabetes drug mechanism that may be complementary to existing drug classes, which suggests potential for added benefit with co-administration or even combination therapies.

3. Hybrid Closed Loop with DiAs, Tandem, Dexcom Safe and Effective in 5-8 Year Olds: 73% TiR, -38 mg/dl Drop in Mean Glucose

UVA’s Dr. Mark DeBoer presented data in a press conference indicating that a modified version of the DiAs hybrid closed-loop system (Tandem pump + Dexcom G4 Share + DiAs control algorithm) is safe and effective in children ages five to eight years old. In a 68-hour “resort study,” the kids (n=12; mean age 7 years) spent 73% of time between 70-180 mg/dl, a marked improvement from 47% during the 68-hour usual care home period (immediately preceding or following the resort stay). Time >180 mg/dl was halved (26% under closed loop vs. 52% at home) and hypoglycemia <70 mg/dl was comparable (<2% in both cases). Remarkably, mean glucose dropped very significantly from 190 to 152 mg/dl on the system – that would translate to a more than 1% estimated A1c reduction, assuming the results were consistent over time. Total daily insulin also dropped non-significantly from 19.9 to 18.6 units. The system was adapted with a lockout screen so that the kids couldn’t alter the insulin parameters (0/12 parents reported that their kids tampered). These results are very promising, especially because the children were extremely active, taking an average of 13,300 steps per day (rock climbing, hiking, and even riding a mechanical bull). This is yet another study demonstrating how automated insulin delivery can drive glycemic improvements across the board in groups young and old. All of the participants were pump-wearers prior to the experiment, though notably, 75% were not on CGM. We believe the pediatric group is where AID devices may have the most benefit, helping patients achieve better glycemic outcomes with less family work and stress (especially overnight).

• As a reminder, the system used in this trial is very similar to the TypeZero-powered system being tested in the International Diabetes Closed Loop trial, which will serve as Tandem’s hybrid closed loop pivotal (not yet open for recruitment, according to ClinicalTrials.gov). Tandem expects a launch of its hybrid closed loop by the end of 2018, though the IDCL study is only testing the system in 14+ year olds; presumably this trial presented today could serve as a pilot for a larger study. We’re not sure of Tandem’s plans for a pediatric indication down to age five.
• On a related note, we learned earlier at ENDO that the Medtronic 670G pediatric study (ages 7-13) is almost complete, will read out mid-summer, and be submitted to FDA in the fall. This is obviously the first commercial AID device that will be available for such a young age group.

4. Post-Hoc Analyses of SUSTAIN 1-5 Show Semaglutide’s Glucose-Lowering and Weight Loss Efficacy on Two Posters

Two posters on meta-analyses of the SUSTAIN phase 3 program for Novo Nordisk’s once-weekly GLP-1 agonist semaglutide highlighted the agent’s (i) weight loss and A1c-lowering efficacy regardless of background oral therapy and (ii) its consistent lowering of both fasting and postprandial glucose. The first post-hoc analysis (n=3,133 across SUSTAIN 2, 3, and 4) divided participants into three groups according to baseline medication regimen: metformin, metformin/sulfonylureas, and other (including TZD monotherapy, TZD/metformin, or TZD/sulfonylureas). SUSTAIN 2 compared semaglutide vs. sitagliptin (Merck’s DPP-4 inhibitor Januvia) over 56 weeks, while SUSTAIN 3 compared semaglutide vs. exenatide extended-release (AZ’s once-weekly GLP-1 agonist Bydureon) over 56 weeks and SUSTAIN 4 compared semaglutide vs. insulin glargine (Sanofi’s Lantus) over 30 weeks. In all three trials, semaglutide achieved superior weight loss and A1c reductions for patients regardless of background oral therapies (p<0.05 for all comparisons). For example, participants taking metformin monotherapy at baseline randomized to semaglutide lost a mean 3.6 kg -4.4 kg (7.9 lbs-9.7 lbs) body weight on the 0.5 mg dose and a mean 5.5 kg-6.2 kg (12 lbs-14 lbs) on the 1.0 mg dose – in contrast, participants taking metformin at baseline randomized to sitagliptin lost only 1 kg-2 kg (2.2 lbs-4.4 lbs), those randomized to exenatide lost 2.5 kg (5.5 lbs), and those randomized to insulin glargine gained 1 kg (2.2 lbs) on average. Similarly, A1c dropped by an average 1.2%-1.3% with semaglutide 0.5 mg and by 1.3%-1.6% with semaglutide 1.0 mg among participants on background metformin monotherapy, vs. 0.5% with sitagliptin, 1% with exenatide, and 0.7% with insulin glargine. This data provides further confirmatory evidence that semaglutide offers superior A1c and weight loss efficacy regardless of what other drugs a patient may be taking and further supports a role for semaglutide as an oral intensification therapy.

• The second poster, a meta-analysis of SUSTAIN 1-5, reported that a 1.0 mg dose of semaglutide led to significantly greater reductions in fasting plasma glucose vs. placebo (in SUSTAIN 1), sitagliptin (in SUSTAIN 2), exenatide (in SUSTAIN 3), insulin glargine (in SUSTAIN 4), and as an add-on to basal insulin vs. placebo (in SUSTAIN 5). Fasting blood sugar also declined by a significantly greater margin with the 0.5 mg dose in SUSTAIN 2 vs. sitagliptin and in SUSTAIN 5 when added to an insulin regimen (p<0.0002 for all comparisons). Postprandial glucose fell by a significantly greater margin with 1.0 mg semaglutide vs. all other agents as well (p<0.02 for all comparisons). A 0.5 mg dose of semaglutide also achieved significantly greater postprandial reductions in SUSTAIN 4 vs. insulin glargine and in SUSTAIN 5 as an add-on to insulin therapy (p<0.003 for both comparisons). Novo Nordisk filed a New Drug Application for the drug in early December, and given these promising results for potency and superiority on several key metrics central to diabetes management, we’re hoping for a swift FDA approval (a regulatory decision is expected in 4Q17).

5. Stats on Philadelphia and Mexico Soda Taxes from an Economist – ~40% Sales reduction in Philly!

According to UNC’s Dr. Shu Wen Ng, the sugar-sweetened beverage industry has reported a ~40% reduction in sales in Philadelphia since the soda tax passed in 2016. She briefly mentioned this statistic in Q&A, while noting that, overall, “the jury’s still out” on the success of the soda tax in Philadelphia. The tax only passed this past November and it’s unclear how the 40% drop compares to ongoing trends in falling soda consumption in the US overall – has the Philadelphia soda tax accelerated the decline in consumption and how does this decline compare to other parts of the country? The 40% figure is of course preliminary and based on companies’ self-reports, but it seems high in considering that Mexico’s one peso/liter excise tax on sugar-sweetened beverages, which only reduced purchases by 7.6% in its first two years (2014-2015) – which was a very impressive feat in and of itself! Regarding the Mexican soda tax, a subject of Dr. Ng’s research, she attributed much of its success to the fact that a significant portion of the tax has been passed on to consumers on average. Dr. Ng noted some heterogeneity, however, on the basis of the type of beverage (there was more pass-through for carbonated drinks than for fruit juices because soda companies face less competition in the carbonated beverages market), package size (pass-through was lower for larger packages, which is concerning because it was already cheaper per unit to buy big bottles), and by region of the country (there was only 7% pass-through in one southern domain). Despite this heterogeneity, Dr. Ng reported that average ml/capita/day purchased declined 6% (-12 ml/cap/d) since 2014. Even more promising, the decline was more pronounced (9%; -19 ml/cap/d) among lower socioeconomic households, the population most vulnerable to type 2 diabetes and obesity. The effect could be accelerating as well – Dr. Ng noted that these changes were more pronounced in year two than in year one. Impressively, a national sales drop of 9.6% in 2016 has correlated with a 7% increase in bottled water sales. Clearly, the tax is minimizing consumption in the way that was intended, but – as was pointed out in Q&A, the jury remains out on the impact of substitutes (like artificially-sweetened diet sodas) on long-term health outcomes. For more on the soda tax, including Dr. Ng’s take on key components of a successful soda tax policy, see our detailed discussion and commentary below.

6. New Data for XOMA Insulin Reception Antibodies XOMA 358 and XOMA 129

In a poster presentation, Dr. Ou Li shared fascinating preclinical data for XOMA 129, an insulin receptor negative modulator for the treatment of insulin-induced acute hypoglycemia. XOMA 129 is a human Fab antibody fragment that binds to the insulin receptor to inhibit signaling. In this study, XOMA 129 was evaluated for the treatment of insulin-induced hypoglycemia in minipigs. The trial evaluated XOMA 129 via three administration methods: IV, intramuscular injection, and subcutaneous injection. The data indicated that both IV and intramuscular administration of XOMA 129 45 minutes after insulin injection was able to produce a rapid and precipitous rise in glucose levels – 2 mg/kg of IV XOMA 129 brought glucose levels back to baseline in about two hours while 10 mg/kg administered via IV and 5 mg/kg administered intramuscularly brought glucose levels to ~250% of baseline within 2-3 hours. In contrast, in minipigs that were not given the XOMA 129 rescue therapy, the glucose levels continued to fall below hypoglycemic thresholds for at least the full first hour, before stabilizing at around 50% of the baseline glucose level. Even more excitingly, the poster presented data indicating a potential application for XOMA 129 as a preventive therapy for nocturnal hypoglycemia. Investigators administered 9 mg/kg XOMA 129 subcutaneously 90 minutes prior to the insulin injection in minipigs that were fasted for 2.5 hours (to simulate nocturnal hypoglycemia conditions). Those pre-treated with XOMA 129 did not experience a drop in blood glucose following insulin administration, while those without XOMA 129 on board experience a rapid drop in glucose down to around 50% the baseline glucose level. Principal investigator Dr. Ou Li shared that XOMA hopes to eventually position XOMA 129 as a subcutaneous preventive therapy for hypoglycemia – as an example of potential clinical use, Dr. Li suggested that this may be particularly helpful for children who may be trending low in the evening on a day when they’ve been particularly physically active, etc. It may also have utility in nightly therapy for those at high risk of chronic hypoglycemia. Canadian biotech company Zucara Therapeutics’ somatostatin inhibitor, slated as a nocturnal therapy, acts by restoring the glucagon response of alpha cells, essentially restoring the “counter-regulatory response” to low blood glucose. In contrast, XOMA 129 appears to have a direct action on the insulin receptor, so we’re not sure it would help restore the counter-regulatory response or hypoglycemia awareness. Nonetheless, XOMA 129 could prove a valuable alternative option to current unwieldly glucagon rescue kits and we’re so pleased to see its continued progress. We previously saw preclinical data for this candidate at ENDO 2016 and we’re hoping to see clinical data in humans in the future. Dr. Li hinted that XOMA may seek a clinical development partner for this candidate – XOMA is a relatively small company and XOMA 129 has the broadest potential indication within its pipeline by far, so we definitely see the logic of partnering with a larger company with more experience in diabetes. On the other hand, several soluble or nasal, easy-to-administer glucagons are in the competitive landscape and we’ll be curious to see how XOMA 129 rescue therapy compares to these next-generation glucagon products.

• Dr. Kirk Johnson presented promising preliminary phase 2 data for insulin receptor antibody XOMA 358 in 13 patients with post-bariatric surgery hypoglycemia, a complication in which patients routinely experience severe postprandial hypoglycemia due to hyperinsulinemia. Patients in this study were randomized to single administration of one of three doses (3, 6, or 9 mg/kg) of XOMA 358, administered via IV infusion over 30 minutes. Patients were monitored via CGM and bedside glucose for five days pre-dose and 11 days post-infusion. At both times, across all three doses, the mean time to postprandial hypoglycemia following treatment was longer than at baseline. The response appeared to be dose-dependent and, as expected, the single antibody infusion appeared to have greater efficacy five days after treatment compared to 11 days after treatment. Furthermore, the study results suggested that XOMA 358 may have an effect on nocturnal hypoglycemia: Four of the six patients treated with the highest 9 mg/kg dose, had hypoglycemic average nighttime glucose levels, and all increased by ~20% (p<0.05); the duration of nighttime hypoglycemia was also reduced by ~50% in all six patients. Overall, Dr. Johnson characterized XOMA 358 as safe and well-tolerated with no clear anti-drug antibody formation. Dr. Johnson also emphasized the observed duration of action for XOMA 358 is 1-2 weeks and the candidate has a three-week half-life – as a result, he shared that the company expects administration will be at most every two weeks in order to produce a sustained hypoglycemia prevention effect.

7. Human Islet Transplantation Study Five Year Follow-Up

University of Pennsylvania’s Dr. Michael Rickels presented five year results from a study of metabolic and beta cell functional outcomes in people with type 1 diabetes following human islet transplantation. Eleven people with longstanding type 1 diabetes (average disease duration of 28 years at baseline) with undetectable C-peptide levels and severe impaired awareness of hypoglycemia received intrahepatic islet transplants with approximately 9,500 islet equivalents per kilogram of body weight. All patients achieved near-normal glycemic control (A1c < 6.5%), and at the five year follow-up, A1c < 6.5% was maintained in 10 of 11 subjects, and 70% were entirely insulin-independent. Dr. Rickels’ team attributes these clinical outcomes to long-term functional changes in the transplanted islets, most notably establishment of a sufficient beta cell secretory capacity (ranging from ~40-90 μU/ml insulin on average over the five year study period). The single patient who experienced islet graft failure was found to have recurrent autoimmune diabetes, showing a sharp spike in GAD65 antibody levels approximately 3.5 years after the transplant. Dr. Rickels previously reported a significant A1c reduction (7% to 5.6%, p<0.01) and significant increases in total, hepatic, and peripheral insulin sensitivity for all 11 patients 6-7 months following transplantation, suggesting that islet transplants can produce immediate benefits, but are liable to biological challenges in the longer term. Dr. Rickels’ studies together suggest that the initially grafted islet beta cell mass must provide a reserve capacity for insulin secretion to help the human islet graft to resist metabolic exhaustion over time (at least for five years). This work further contributes to the evidence that islet transplantation could be an effective, long-term beta cell replacement therapy for certain people with type 1 diabetes. However, the scalability of this method is limited due to the very small number of available donor islets compared to the total population of people with type 1 diabetes – the field is working toward stem cell-derived unlimited islet cell sources, but the promising candidates thus far remain preclinical or in very early clinical stages of safety evaluation. Furthermore, it is and will be important to identify factors that predict responders and non-responders to the therapy, and to consider the balance between the benefit of islet transplantation on long-term glycemic control versus the risk and expense of the procedure. Even with all that said, until effective encapsulation or immunomodulation strategies are available, islet transplantation requires chronic use of immunosuppressive therapy which could lead to additional unintended adverse effects. We will be keeping a close eye on this study, and are eager to learn how well the transplanted islets function beyond five years.

8. Transcranial Magnetic Stimulation Alters Microbiome Composition and Induces Weight Loss

Dr. Livio Luzi presented intriguing evidence that noninvasive electromagnetic brain stimulation can contribute to weight loss, potentially by changing the composition of the gut microbiome. People with obesity (n=14, average baseline BMI=38 kg/m²) who underwent 15 sessions of deep transcranial magnetic stimulation (dTMS) to the insula and prefrontal cortex lost significantly more body weight at the end of the five week study period than those who received sham stimulation (-3% vs. -2% from baseline, p<0.01) and also experienced a significant reduction in self-reported food craving (p<0.001). According to a comparison of fecal microbiota before and after treatment, participants subjected to dTMS additionally showed an increase in abundance of two “good” bacterial species associated with anti-inflammatory properties, Faecalibacterium and Phascolarctobacterium. Interestingly, the magnitude of these microbiome compositional shifts correlated with improvements in blood glucose and changes in circulating levels of various hormones and neurotransmitters, including insulin (p<0.05), ACTH (p<0.01), norepinephrine (p<0.05), and prolactin (p<0.01). Based on these very early findings, Dr. Luzi hypothesizes that dTMS modulates hormone and neurotransmitter secretion in the brain, leading to shifts in gut microbiota composition to improve metabolism. Of course, it is important to bear in mind that this is a preliminary study with a small number of participants and a small effect size – clearly more work is required to validate these intriguing findings. That said, Dr. Livio’s results hint at the potential of dTMS as a non-invasive, non-pharmacological, and virtually side effect-less obesity therapy (or component of obesity therapy) – that is, if stimulation techniques can be targeted more precisely to produce a larger weight loss effect. Perhaps more immediately, this work is suggestive of a “gut/brain axis” beyond the familiar “gut/metabolism axis” that canonical microbiome studies have revealed, adding yet another a layer of complexity to the biology of obesity. The immense interest in the microbiome and its connection to diabetes and obesity has been growing for some time – there is still much to learn on this frontier, but we’re glad to see some opportunities for translational therapy based on the microbiome.

9. Missed Opportunities for CVOTs According to Dr. Neda Rasouli

University of Colorado’s Dr. Neda Rasouli offered her perspective on where we stand – results from nine modern diabetes cardiovascular outcomes trials (CVOTs) in hand – and where we need to go in order to make these long, costly trials more useful for clinical decision-making. She emphasized that the FDA and EMA guidelines for CVOTs emphasize the demonstration of cardiovascular safety and CVOTs designed to meet these regulatory requirements thus far have enrolled very high CV risk populations and included composite CV endpoints in order to accumulate enough events for analysis in a relatively short time period. She acknowledged that these studies have provided the field with very useful information already – we have reassurance that DPP-4 inhibitors, GLP-1 agonists, and SGLT-2 inhibitors very likely do not increase CV risk and, rather unexpectedly, we’ve learned that SGLT-2 inhibitors empagliflozin and GLP-1 agonists liraglutide very likely have cardioprotective benefits in high-risk patients with a history of cardiovascular disease. That said, many questions remain in terms of incorporating CVOT findings into clinical practice. She questioned whether current CVOTs are optimally designed to provide information on how to treat the general population of people with type 2 diabetes – she suggested that perhaps the “usual” primary outcome of composite MACE, the patient enrollment criteria, the power and sample size (powered for superiority or not), and the use of a glycemic equipoise design should be revisited. After all, trials thus far have often demonstrated heterogeneous effects on the components of the primary composite MACE outcome and, she pointed out, only a relatively small proportion of patients in general diabetes clinical practice have atherosclerotic cardiovascular disease, which is usually a key enrollment criterion for these trials. Further, she argued that for drugs in which another drug in the class has demonstrated cardioprotection, the CVOT should be powered to demonstrate superiority. She applauded Merck/Pfizer for modifying the CVOT for SGLT-2 inhibitor ertugliflozin to power for superiority following the reveal of the EMPA-REG OUTCOME results, for instance. While this can help shed a little more light on whether the cardioprotective benefit of EMPA-REG OUTCOME is a class effect or drug specific, Dr. Rasouli pointed that it’s still difficult to draw these conclusions when each diabetes drug conducts its own separate CVOT, completing at different times with differing enrollment criteria. Looking to future CVOTs that have not yet reported or initiated as well, Dr. Rasouli noted that the field will need to grapple with the use of diabetes drugs with demonstrated cardioprotection (empaglifozin and liraglutide) in the control arm – how might this affect future results?

• Dr. Rasouli suggested that cost is the main barrier to uptake of cardioprotective drugs, even in people with type 2 diabetes and a history of atherosclerotic disease. Based on the number needed to treat and cost of liraglutide, she estimated that it would cost around a million dollars to prevent one CV event. She called upon the healthcare providers in the room to engage in advocacy on this issue – “It’s our responsibility to change this.”

10. Commentary from Thought Leaders on the Groundbreaking FOURIER Results

Just a couple weeks following presentation of FOURIER results for Amgen’s PCSK9 inhibitor Repatha (evolocumab), Louisville’s Dr. Harold Bays offered commentary suggesting that a longer outcomes trial would reveal even greater benefit and cautioning that we must refrain from claiming a cardioprotective class effect at this time. Notably, the 20% risk reduction for the secondary endpoint in FOURIER (three-point MACE consisting of non-fatal MI, non-fatal stroke, and CV death) was driven by statistically significant risk reduction for MI (HR=0.73; p<0.001) and stroke (HR=0.79; p=0.01). The hazard ratio for CV death was 1.05 in favor of placebo, but this association did not reach statistical significance. If Dr. Bays’ predictions for a longer study are correct, evolocumab might show risk reduction for CV death with longer-term data collection – the median follow-up in FOURIER was only 26 months (~2.2 years), and an accumulated anti-atherosclerotic effect would presumably mean less CV mortality. In reference to Sanofi/Regeneron’s Praluent (alirocumab), the other PCSK9 inhibitor currently on the market, Dr. Bays underscored that not all LDL-lowering drugs improve CV outcomes. According to Dr. Bays, the CV benefit of the PCSK9 inhibitor class as a whole has not been established, and we must wait patiently for Sanofi/Regeneron’s CVOT ODYSSEY Outcomes to report (expected early next year). Indeed, the ODYSSEY Outcomes trial is continuing as planned following a scheduled interim analysis last November – this move doesn’t indicate positive, neutral, or negative CV results per se, but it was somewhat disappointing given the tremendous LDL-lowering efficacy of Praluent in phase 3 trials and given Sanofi management’s early comment that the CVOT could be halted early due to overwhelming CV efficacy. That said, while we understand that clinical trial purists will argue that we must wait for the data before drawing any conclusions, we do believe that most cardiologists view Repatha and Praluent as essentially interchangeable in clinical practice and would assume that the cardioprotection demonstrated for Repatha would apply to Praluent as well. There’s certainly less questions about a potential class effect or mechanism of action for the PCSK9 inhibitor class compared to SGLT-2 inhibitors and GLP-1 agonists (which, after all, weren’t developed explicitly to reduce cardiovascular events). We’re very optimistic that ODYSSEY Outcomes will add evidence for a PCSK9 cardioprotective class effect come early 2018, especially since, reassuringly, the Spire-2 CVOT for Pfizer’s now-discontinued bococizumab also demonstrated CV benefit for another PCSK9 inhibitor agent. For more from this Amgen-sponsored breakfast symposium, including perspective on statins and PCSK9 inhibitors from a patient advocate with familial hypercholesterolemia, see our detailed discussion and commentary below.

11. Dr. Jay Skyler on T1D Prevention and Intervention Landscape

The great Dr. Jay Skyler, hailing from the University of Miami, gave an excellent overview of the overall disappointing collection of studies in type 1 diabetes primary prevention, secondary prevention, and intervention. A large majority of studies in each domain have had limited to no effect, though some immunomodulatory approaches in cases of new onset type 1 diabetes have shown a transient or potential effect. Dr. Skyler specifically was pessimistic about the TRIGR study, evaluating casein hydrolysate formula as a type 1 diabetes preventive measure  (on the basis that cow’s milk-based formula might be a potential environmental trigger for type 1), despite a modest effect observed in a previous study from Finland. Dr. Skyler was more optimistic about combination immunotherapy – an approach under investigation in his group’s DIPIT trial, which is expected to begin enrolling later this year – may have some promise. Dr. Skyler has previously asserted that other antigen-based or immunomodulatory therapies “bring in the infantry, but leave the artillery,” whereas DIPIT and related studies engage multiple immune mechanisms and could therefore prove more robust. Preventing and reversing type 1 diabetes have proven difficult, but we are encouraged by the breadth of ongoing trials.  See below for tables containing Dr. Skyler’s near-unabridged list of type 1 prevention and intervention trials – our type 1 diabetes cure and prevention competitive landscape also contains of an overview of several ongoing efforts.

12. Epigenetic Modification Induced By Neat Molecular Toolkit Promotes Replication of Mature Beta Cells

In the 2017 Roy O. Greep Award for Outstanding Research Lecture, University of Pennsylvania’s Dr. Klaus Kaestner showed how beta cells can be epigenetically modified to reproduce in vitro and in vivo. As beta cells age, methylation of functional gene promoters decreases (correlating with increased expression), while methylation of proliferative gene promoters increases, reducing expression and therefore replication capacity. Dr. Kaestner’s lab aims to reverse the latter methylation to restore a dividing phenotype. Specifically, monoallelic expression of cell cycle inhibitor p57 is caused by allele-specific methylation, preventing transcription of a lncRNA – when demethylated, this lncRNA is transcribed and inhibits expression of p57, allowing the cell cycle to proceed. Thus, the researchers targeted a TAL effector linked to TET1 – TAL effectors bind to DNA and can be directed to any sequence; TET1 is a passive demethylator – to a region of DNA adjacent to the methylated site. With this novel system, when the fusion protein is expressed in beta cells, one would expect demethylation and subsequent lncRNA expression, followed by a loss of p57 and replication. In an initial in vitro study, sequencing data showed selective demethylation at the desired spot, along with reduced p57 expression and ~doubled expression of cell cycle markers PCNA1 and Ki67. How does the method translate to the in vivo setting? Lab members induced diabetes with STZ in NSG mice. Next they infected re-suspended islets with a virus (to make them overexpress the TAL effector-TET1 fusion protein) and injected them subcutaneously into the hyperglycemic mice. Adding BrdU, a synthetic thymidine analog that incorporates into the DNA of dividing cells, to the drinking water allowed detection of proliferating cells. Initial analyses suggest that the protocol worked – they observed a three-fold increase in proliferative beta cells (marked by BrdU and C-peptide). It is still early days for this group, but the impressive system at the very least offers a model to further explore beta cell replication. At most, this epigenetic pathway could hold the key to coaxing mature, beta-like cells to replicate, one of the holy grails of stem cell research. We hope to see this advance to clinical trials ASAP, be it with autologous or allogenic cells!

13. Dr. Daniel Bessesen: “We need a statin for weight loss; Overview of Exisitng Obesity Pharmacotherapy

“We need a statin for weight loss – a drug that is well-tolerated, cheap, efficacious, and has hard endpoints and clinical data.” Speaking to a packed room (impressive for the very last session of the meeting!) Denver Health Medical Center’s renowned Dr. Daniel Bessesen lamented that too few patients are on obesity pharmacotherapy currently and was none-too-optimistic about the current options in the pipeline. He believes that there is a perception issue driving the low frequency of physician prescriptions for obesity drugs. Providers believe that medications are dangerous, that they don’t work, and even that obesity patients should be treated differently from, say, diabetes patients. Dr. Bessesen was stunned (“oh my god!”) as he recited a quote from the great Dr. Gordon Schiff: “promoting and prescribing drugs to treat obesity does a disservice to our patients, society, and ourselves...” To fan the flames, these drugs are frequently not covered by insurance. Dr. Bessesen doesn’t have much confidence in the pipeline, but sees a bright spot in certain combination therapies, such as canagliflozin/phenetermine, dapagliflozin/exenatide, and pramlintide/phentermine (along with dual/triple agonists). And before signing off, he made it clear that while he advocates for the prescription of these drugs, supplementing them with behavior modification will always be more effective than the pharmacological intervention alone. See below for his comments on existing weight loss drugs.

Detailed Discussion and Commentary

Posters

DELIVER 2 Study: Lower Risk of Hypoglycemia after Switch to Insulin Glargine U300 vs. Other Basal Insulins in Patients with Type 2 Diabetes on Basal Insulin in Real-World Clinical Settings

Fang Liz Zhou, MD (Sanofi, Bridgewater, NJ), Fen Ye, MD (Sanofi, Bridgewater, NJ), Vineet Gupta, MD (Accenture, Florham Park, NJ), Rishab Gupta, MD (Accenture, Florham Park, NJ), Rishi Agarwal, MD (Accenture, Florham Park, NJ), Jukka Westerbacka, MD (Sanofi, Paris, France), Paulos Berhanu, MD (Sanofi, Bridgewater, NJ), and Lawrence Blonde, MD (Ochsner Medical Center, New Orleans, LA)

A poster presentation featured data from the Sanofi-sponsored real-world DELIVER 2 study, demonstrating significantly lower risk of hypoglycemia for type 2 diabetes patients who switched to Toujeo (insulin glargine U300) versus other basal insulins. According to observational data from electronic medical records, patients (n=1,894) on on Lantus or Levemir (75% and 25% of the cohort at baseline, respectively) who switched to Toujeo experienced 33% fewer hypoglycemic events over six months than their counterparts who switched to other basal insulins, including Sanofi’s Lantus (insulin glargine U100), Novo Nordisk’s Levemir (insulin detemir), and Tresiba (insulin degludec) (7.98 vs. 5.32 events per 100 patient-months; p<0.01). Furthermore, patients who switched to Toujeo additionally experienced a 48% reduction in hypoglycemia events requiring inpatient or emergency department care (3.82% of patients vs. 1.97%; p<0.01) versus those on other basal insulins. Both groups experienced a significant reduction in A1c over the six-month study period, falling from a baseline of 8.9% to 8.4% in the patients who switched to Toujeo and 8.5% in those who switched to a different basal insulin (p<0.01). There was no significant difference in the magnitude of A1c improvement between the groups, indicating that Toujeo’s reduction in hypoglycemia risk came without compromising blood glucose control – Toujeo was also non-inferior to other basal insulins in terms of A1c goal attainment. The observational nature of this data limits our understanding of how insulin dosing compared between the patients on Toujeo versus other basal insulins, and it is possible that hypoglycemia was underreported since no blood glucose or CGM data was collected. Furthermore, notably, this study only looked at patients who were already on a basal insulin who switched to a new basal insulin – presumably, patients who were well-controlled on Lantus or Levemir would not have switched in the first place (unless forced to by formularies) and thus would not have been included in the study. That said, DELIVER 2 provides valuable information that Toujeo can improve hypoglycemia (and lower A1c) for type 2 diabetes patients in a real-world setting.

• This data further supports findings from randomized, controlled clinical trials indicating that Toujeo – and next-generation basal insulins overall – are a major step above first-generation analogs, particularly with regards to their hypoglycemia benefits as well as the greater stability that so many patients find associated with the drugs. We would have loved to see a separate analysis of outcomes following switches to Toujeo vs. switches to Tresiba only. The comparison between Toujeo and Tresiba for hypoglycemia in a real-world setting is especially intriguing, given that Tresiba has already demonstrated significant hypoglycemia reductions versus Lantus in the SWITCH program and is on track for the inclusion of a hypoglycemia benefit on its FDA label. Much discussion has surrounded the comparative efficacy of Toujeo and Tresiba, but to date the two have not been studied head-to-head (though a head-to-head PK/PD trial is ongoing).
• The DECLARE 2 results complement Sanofi’s ongoing program of three randomized, prospective real-life clinical studies with Toujeo, titled ACHIEVE CONTROL, REACH CONTROL, and REGAIN CONTROL. The Real Life Study clinical trial program involves over 4,500 participants with type 2 diabetes who are starting basal insulin treatment with Toujeo or switching to Toujeo from another basal insulin. In addition to the primary endpoint of A1c, the studies will also evaluate hypoglycemia, patient satisfaction, and, interestingly, impact on healthcare resource utilization. The ACHIEVE CONTROL and REACH CONTROL studies involves insulin-naïve people with poorly-controlled type 2 diabetes in the US (n=3,270) and Europe (n=600) respectively, while the REGAIN CONTROL study involves individuals with type 2 diabetes who are poorly controlled on basal insulin (n=600). We’re so glad to see Sanofi investing in “real-world” clinical trials – while randomized controlled trials will always be invaluable, we believe that real-world trials can provide a lot of immensely useful information for practical clinical decision-making.

The Sugar Tax: Can We Really Legislate Weight Loss?

The Economist’s Perspective

Shu Wen Ng, PhD (UNC, Chapel Hill, NC)

According to UNC’s Dr. Shu Wen Ng, the sugar-sweetened beverage industry has reported a ~40% reduction in sales in Philadelphia since the soda tax passed in 2016. She briefly mentioned this statistic in Q&A, while noting that, overall, “the jury’s still out” on the success of the soda tax in Philadelphia. The tax only passed this past November and it’s unclear how the 40% drop compares to ongoing trends in falling soda consumption in the US overall – has the Philadelphia soda tax accelerated the decline in consumption and how does this decline compare to other parts of the country? The 40% figure is of course preliminary and based on companies’ self-reports, but it seems high in considering that Mexico’s one peso/liter excise tax on sugar-sweetened beverages, which only reduced purchases by 7.6% in its first two years (2014-2015) – which was a very impressive feat in and of itself! Regarding the Mexican soda tax, a subject of Dr. Ng’s research, she attributed much of its success to the fact that a significant portion of the tax has been passed on to consumers on average. Dr. Ng noted some heterogeneity, however, on the basis of the type of beverage (there was more pass-through for carbonated drinks than for fruit juices because soda companies face less competition in the carbonated beverages market), package size (pass-through was lower for larger packages, which is concerning because it was already cheaper per unit to buy big bottles), and by region of the country (there was only 7% pass-through in one southern domain). Despite this heterogeneity, Dr. Ng reported that average ml/capita/day purchased declined 6% (-12 ml/cap/d) since 2014. Even more promising, the decline was more pronounced (9%; -19 ml/cap/d) among lower socioeconomic households, the population most vulnerable to type 2 diabetes and obesity. The effect could be accelerating as well – Dr. Ng noted that these changes were more pronounced in year two than in year one. Impressively, a national sales drop of 9.6% in 2016 has correlated with a 7% increase in bottled water sales. Clearly, the tax is minimizing consumption in the way that was intended, but – as was pointed out in Q&A, the jury remains out on the impact of substitutes (like artificially-sweetened diet sodas) on long-term health outcomes. 

• Designing a tax is no easy feat – Dr. Ng rattled off 12 different factors that policy makers must consider: (i) Industry’s market power (oligopolies vs. perfect competition); (ii) Excise vs. sales tax (where is the tax “seen”?); (iii) Type – Ad valorem (%) vs. Specific ($/ml or $/g of sugar); (iv) Based on foods, nutrients, or both?; (v) Level – how high to truly have an impact?; (vi) Elasticities of demand (who bears the cost?); (vii) Scope of coverage – national? State? Local? (consider potential leakage across boundaries); (viii) Implementation – who collects and enforces?; (ix) Use of revenue – earmark for specific purposes or no?; (x) Impact on employment?; (xi) Long-term vs. short-term; and (xii) timing for the best political/societal acceptance. That’s a lot consider!
  
  • With regards to use of revenue, Dr. Ng commended Berkeley’s earmarking strategy. In 2017, revenues will go to (from most funding to least funding): Berkeley Unified School District ($637,500), Healthy Black Families ($245,874), YMCA Central Bay Area ($151,360), Berkeley Youth Alternatives ($125,000), Lifelong Medical Care ($125,000), and The Ecology Center ($115,266) In this way, revenue raised from persisting sugary beverage consumption is poured back into the community, where it can supplement education, family services, and healthier lifestyles.
  • There has been no net change in employment in Mexico since the tax was implemented. Further, simulations suggest there will ultimately be a slight increase in employment in California and Illinois, thanks to these taxes. While there will be a slight decline of employment in the beverage industry, this the models say this will be more than compensated for by gains in government.
• A tiered sugary beverage tax system will debut in the UK in April 2018. Beverages that contain more than 8 grams of sugar/100 ml will be taxed at 24 pence/liter, while those with 5-8 grams of sugar/100 ml will be taxed at 18 pence/liter. The benefit of tiered taxation is that it incentivizes reformulations, which are already reportedly occurring. However, with only two tiers, companies are only really encouraged to bring sugar content just below 5 grams and just below 8 grams. A better design, she added, would include tiers within the existing tiers, to incentivize continued lowering of sugar content.
• Prior to the start of the lecture, an audience poll revealed that a strong majority think a soda tax is a good idea. However, only half of the full room thought that a soda tax would reduce obesity. We assume they answered “yes” to the first question because of promises of additional revenue streams for the government and reduction of other metabolic conditions. This is not an uncommon view – many activists for the soda tax acknowledge that a tax of a few cents is unlikely to substantially change behavior and consumption in and of itself, but argue that the increased awareness of the pitfalls of sugary beverage consumption, coupled with new health and education efforts funded by the revenue of the tax, may be able to elevate the national conversation about the dangers of soda.
• In 2016, Cook County (including Chicago), Philadelphia, Boulder (CO), Albany (CA), Oakland, and our very own home base of San Francisco voted to pass per-ounce taxes on sugar-sweetened beverages. Mexico (2014), Chile (2014), and Berkeley, CA (2015) pioneered the movement, implementing similar taxes in the past couple of years. We are hopeful that many additional states will follow suit in the coming years, especially as more outcomes data become available.
  
  • Even “The Wealth of Nations” author Adam Smith thought sugar should be taxed: "Sugar, rum, and tobacco, are commodities which are nowhere necessaries of life, [but] which are ... objects of almost universal consumption, and which are therefore extremely proper subjects of taxation."

Q: We have good evidence for the link between sugar intake and diabetes/obesity etc. But what are they using instead of sugar, and are we reducing or shifting the problem? Artificial sweeteners – there’s no long term safety data. My theory is to stick with the devil you know.

A: True, the unintended consequences are unknown. We are seeing more products with combinations of regular sugars and artificial sweeteners. What that means in the long run – science is still out. We’re still primed for sweetness, so even if sugar is replaced with artificial sweetener, we still crave sweetness. We don’t know the effects of them yet though. We need to make sure we have a good idea of what these implications will be. From a public health standpoint, recommendations have been that replacing sugars with these is an effective way to lower weight and dental cavities. Personally, I’m not sure I’d go that far.

Q: On the point about excise vs. sales tax – the soda is just on shelf, with a number next to it. So do the consumers really know how much of the price is due to tax?

A: In Philadelphia, it was framed as a grocery tax. They wanted to make consumers angry. They indicated on the shelf and on the receipt the additional cost due to the sugar sweetened beverage tax. That has made people upset, and hopefully consume less. The jury is still out, but industry has reported a 40% reduction in sales.

The Epidemiologist’s Perspective

Michael Long (George Washington University, Washington, DC)

In a literature meta-analysis, Mr. Michael Long et al. estimated that a national sugar drink tax would, in one year, reach 307 million lives (the US population), cost only $47.6 million, prevent 576,000 cases of obesity, and save ~$31 healthcare dollars per dollar spent. The CHOICES project group published these findings in Health Affairs in 2015. Mr. Long conceded that these calculations do have limitations – actual BMI and obesity levels depend on accurate estimates of industry and consumer response, we don’t know whether people will behave differently when they know that a tax is discouraging consumption of sugar-sweetened beverages, and sugary drinks still cause harm even if people don’t carry excess weight. Even so, countless researchers have attempted to ascertain whether or not people would change their behavior in response to higher prices and whether there’s an association between lower consumption and weight loss, but this analysis puts into plain numbers what the government could stand to gain if it instated such a tax – policy makers respond well to dollar incentives.

• The group’s analysis only took observational change-in-change studies and RCTs into account [a change-in-change study looks at baseline consumption/weight and compares to post-tax consumption/weight.] Cross-sectional studies were excluded because they are particularly prone to correlative conclusions. Studies tracking changes in intermediate behaviors (i.e. calorie consumption, total energy intake) were also excluded because they rely on hard-to-track metrics. Looking at existing excise taxes is also wrought with pitfalls: They have shown small effect sizes and have confounds, and most historical taxes aren’t specifically on sugar-sweetened beverages.
• While we were looking for the room, an ENDO volunteer ironically told us “it’s just past the soda machine on your right.” Sigh. It turns out that Mr. Long had received the same instructions. He used this example to illustrate how far we have to come to truly change our built environment.

Questions and Answers

Q: I’m from Australia, and every time I come here, I’m struck by the amount of sweet crap. Even the bread tastes awful because it’s so full of sugar. What percentage of the epidemic is due to sugar-sweetened beverages vs. the incorporation of fructose in everything, and a fall in protein intake? Is this just the tip of the iceberg?

A: Half of total sugar supply in US is in beverages. That’s why we’re starting there. I don’t think dropping sugar consumption in the US would drop the obesity rate immediately. Think about the children though –we’ve set up a generation that will have a very tough time reversing obesity.

Q: What about artificially-sweetened sugary drinks?

A: Philadelphia is the only place in the US to include those in their tax. I think the literature shows that it’s like a nicotine patch, but evidence in trials suggests people lose weight when they switch. The long-term endocrine effects are unclear, however.

Q: What about high-tax coffee drinks or Gatorade? How have those been handled?

A: Any prepared product with added sugar, that would be taxed. But not if you add sugar to your own coffee.

Q: What happened to cigarette consumption when taxes increased? Shouldn’t we template our efforts in the footsteps of the tobacco campaign?

A: Most people in the field have used that as a template. I think those taxes worked better in young people because they have higher price elasticity than adults. There was a stronger response to price in sugary beverages because there are substitutes. But yes, it’s not just aiming to reduce consumption. Why aren’t there messages everywhere showing that sugar causes diabetes, tooth decay…I think the effort shouldn’t just be single-pronged.

Comment: Maybe there could be differential taxes. Schools pay more, for example.

A: Schools have actually gotten rid of sugary drinks pretty widely. They’ve done a pretty good job.

Q: Is there any evidence to suggest that the tax would differentially effect people based on socioeconomic status?

A: You are seeing that in Mexico actually – qualitatively, people are trying to get around the tax. But think about it – tap water is cheaper, but industry doesn’t want to talk about that. Ensuring a secure and clean water supply is the best way to not exacerbate income inequality.

Corporate Symposia

The Revolution in Cholesterol Management: Putting PCSK9 Inhibitors into Practice (Sponsored by Amgen)

Sergio Fazio, MD (Oregon Health and Science University, Portland, OR); John Guyton, MD (Duke University Medical Center, Durham, NC); Harold Edward Bays, MD (Louisville Metabolic and Atherosclerosis Research Center, KY); Eliot Brinton, MD (Utah Foundation for Biomedical Research, Salt Lake City, UT); Michael Overstreet (FH Foundation, Atlanta, GA)

Just a couple weeks following presentation of FOURIER results for Amgen’s PCSK9 inhibitor Repatha (evolocumab), Louisville’s Dr. Harold Bays offered commentary suggesting that a longer outcomes trial would reveal even greater benefit and cautioning that we must refrain from claiming a cardioprotective class effect at this time. Notably, the 20% risk reduction for the secondary endpoint in FOURIER (three-point MACE consisting of non-fatal MI, non-fatal stroke, and CV death) was driven by statistically significant risk reduction for MI (HR=0.73; p<0.001) and stroke (HR=0.79; p=0.01). The hazard ratio for CV death was 1.05 in favor of placebo, but this association did not reach statistical significance. If Dr. Bays’ predictions for a longer study are correct, evolocumab might show risk reduction for CV death with longer-term data collection – the median follow-up in FOURIER was only 26 months (~2.2 years), and an accumulated anti-atherosclerotic effect would presumably mean less CV mortality. In reference to Sanofi/Regeneron’s Praluent (alirocumab), the other PCSK9 inhibitor currently on the market, Dr. Bays underscored that not all LDL-lowering drugs improve CV outcomes. According to Dr. Bays, the CV benefit of the PCSK9 inhibitor class as a whole has not been established, and we must wait patiently for Sanofi/Regeneron’s CVOT ODYSSEY Outcomes to report (expected early next year). Indeed, the ODYSSEY Outcomes trial is continuing as planned following a scheduled interim analysis last November – this move doesn’t indicate positive, neutral, or negative CV results per se, but it was somewhat disappointing given the tremendous LDL-lowering efficacy of Praluent in phase 3 trials and given Sanofi management’s optimism that the CVOT could be halted early due to overwhelming CV efficacy. That said, while we understand that clinical trial purists will argue that we must wait for the data before drawing any conclusions, we do believe that most cardiologists view Repatha and Praluent as essentially interchangeable in clinical practice and would assume that the cardioprotection demonstrated for Repatha would apply to Praluent as well. There’s certainly less questions about a potential class effect or mechanism of action for the PCSK9 inhibitor class compared to SGLT-2 inhibitors and GLP-1 agonists (which, after all, weren’t developed explicitly to reduce cardiovascular events). We’re very optimistic that ODYSSEY Outcomes will add evidence for a PCSK9 cardioprotective class effect come early 2018, especially since, reassuringly, the Spire-2 CVOT for Pfizer’s now-discontinued bococizumab also demonstrated CV benefit for another PCSK9 inhibitor agent. For more from this Amgen-sponsored breakfast symposium, including perspective on statins and PCSK9 inhibitors from a patient advocate with familial hypercholesterolemia, see our detailed discussion and commentary below.

• FH patient representative Mr. Michael Overstreet (FH Foundation, Atlanta, GA) advocated for earlier diagnosis and proactive treatment – “I’ve learned to run faster.” After recognizing that FH was the catalyst for premature mortality in his father, grandfather, and uncle, Mr. Overstreet swiftly started statin therapy. He emphasized the under-diagnosis of FH worldwide, which is disappointing because the condition is manageable if caught and treated early. This certainly applies to diabetes as well, and we loved hearing this view directly from a patient. The patient voice is so valuable at scientific meetings, and we’re glad a room full of endocrinologists received this message.
• Mr. Overstreet shared that he’s been trying to get on a PCSK9 inhibitor for 2.5 months, with a lot of back-and-forth between his doctor and his insurance company. This largely matches the story we’ve heard on PCSK9 inhibitor reimbursement, in that it’s very challenging to get a prior authorization: According to a National Lipid Association survey presented at ACC 2017, requests for PCSK9 coverage are denied 96% of the time. Findings of CV efficacy will ideally compel more payers to sign-on and reimburse Repatha – and perhaps Praluent as well – though we anticipate that Amgen will have to invest in education and marketing efforts around the FOURIER results to see meaningful change in the reimbursement landscape.

How Sweet It Is: New Research in Type 1 Diabetes

Predicting and Preventing Type 1 Diabetes

Jay Skyler, MD (University of Miami, FL)

The great Dr. Jay Skyler, hailing from the University of Miami, gave an excellent overview of the overall disappointing collection of studies in type 1 diabetes primary prevention, secondary prevention, and intervention. A large majority of studies in each domain have had limited to no effect, though some immunomodulatory approaches in cases of new onset type 1 diabetes have shown a transient or potential effect. Dr. Skyler specifically was pessimistic about the TRIGR study, evaluating casein hydrolysate formula as a type 1 diabetes preventive measure  (on the basis that cow’s milk-based formula might be a potential environmental trigger for type 1), despite a modest effect observed in a previous study from Finland. Dr. Skyler was more optimistic about combination immunotherapy – an approach under investigation in his group’s DIPIT trial, which is expected to begin enrolling later this year – may have some promise. Dr. Skyler has previously asserted that other antigen-based or immunomodulatory therapies “bring in the infantry, but leave the artillery,” whereas DIPIT and related studies engage multiple immune mechanisms and could therefore prove more robust. Preventing and reversing type 1 diabetes have proven difficult, but we are encouraged by the breadth of ongoing trials.  See below for tables containing Dr. Skyler’s near-unabridged list of type 1 prevention and intervention trials – our type 1 diabetes cure and prevention competitive landscape also contains of an overview of several ongoing efforts.

Human T1d Primary Prevention Studies

Human T1d Secondary Prevention Studies

** Post-hoc analysis identified subgroup with benefit

Ongoing and Planned Prevention Studies

Immunomodulatory New Onset Studies

Other Recent Onset Studies

Treatment Options for Obesity

Pharmacotherapy for Obesity

Daniel Bessesen, MD (Denver Health Medical Center)

“We need a statin for weight loss – a drug that is well-tolerated, cheap, efficacious, and has hard endpoints and clinical data.” Speaking to an packed room (impressive for the very last session of the meeting!) Denver Health Medical Center’s renowned Dr. Daniel Bessesen lamented that too few patients are on obesity pharmacotherapy currently and was none-too-optimistic about the current options in the pipeline. He believes that there is a perception issue driving the low frequency of physician prescriptions for obesity drugs. Providers believe that medications are dangerous, that they don’t work, and even that obesity patients should be treated differently from, say, diabetes patients. Dr. Bessesen was stunned (“oh my god!”) as he recited a quote from the great Dr. Gordon Schiff: “promoting and prescribing drugs to treat obesity does a disservice to our patients, society, and ourselves...” To fan the flames, these drugs are frequently not covered by insurance. Dr. Bessesen doesn’t have much confidence in the pipeline, but sees a bright spot in certain combination therapies, such as canagliflozin/phenetermine, dapagliflozin/exenatide, and pramlintide/phentermine (along with dual/triple agonists). And before signing off, he made it clear that while he advocates for the prescription of these drugs, supplementing them with behavior modification will always be more effective than the pharmacological intervention alone. See below for his comments on existing weight loss drugs.

• Dr. Bessesen offered his thoughts on six common weight loss drugs: Phentermine, Orlistat, Lorcasarin, Phentermine/Topiramate ER, Naltrexone SR/Bupropion SR, and Liraglutide. “It’s true for all drugs,” he commented, “that you should stop therapy if you observe <5% weight loss at 3 months.”
  
  • Phentermine. Phentermine is the cheapest, most widely prescribed weight loss medication, comprising 74% of all scripts. While it is only FDA approved for 3 months, some physicians, such as Dr. Ken Fujioka, are using it longer term off-label. 
  • Orlistat (Xenical). “Orlistat should at least be on your list and considered at some point. It is the safest medicine, so safe that it comes in 60 mg OTC form.”
  • Lorcasarin (Belviq). Belviq is the most well-tolerated, with minimal side effects of headache, dizziness, and nausea. In addition, it is safe for patients with CVD risk because there is no increase in pulse/blood pressure, unlike with its predecessor, Fenfluramine. Some physicians like to prescribe Belviq with phentermine, but Dr. Bessesen “would not do that at this point” due to a lack of data on safety and efficacy.
  • Phentermine/topiramate ER (Qsymia). This is the most effective weight loss medication (10-12% weight loss). Dr. Bessesen had one note of caution: “There is a tendency to want to prescribe this off-label using generic phentermine and topiramate, but the dosing flexibility is not quite equivalent and the ER version is tolerated better. I would personally prescribe the brand name if possible.”
  • Naltrexone SR/bupropion SR (Contrave). Might be useful in those with concomitant depression or food reward-related issues.
  • Liraglutide (Saxenda). Increases pulse initially, which excludes it from CV risk situations in the Endo guidelines. However, the positive LEADER CVOT (though notably for the lower 1.2 mg and 1.8 mg doses for diabetes) shows that the guidelines may need some modification – “the high pulse may not tell us everything we need to know about CV risk.”

Product Theaters

New Clinical Trial Data for Jardiance (empagliflozin) Tablets

Carol Wysham, MD (University of Washington, Spokane, WA)

In a Lilly/BI product theater on SGLT-2 inhibitor Jardiance (empagliflozin), Dr. Carol Wysham highlighted the EMPA-REG OUTCOME results and characterized the 38% risk reduction for CV death as a major breakthrough for the diabetes field. “I’ve been practicing for quite some time, I’ve even used first-generation sulfonylureas in my practice,” she began, “and this outcomes data on Jardiance fundamentally changes our ability to treat people with diabetes.” The FDA approval of a new Jardiance indication for the reduction of CV death (the first such label claim for a diabetes medicine) and the ADA’s endorsement of empagliflozin for patients with type 2 diabetes and established CV disease in its 2017 Standards of Care officially moves this breakthrough from the research world into the real world. Dr. Wysham also alluded to how inclusion of EMPA-REG OUTCOME data on the Jardiance label invites more overlap between the work of cardiologists and the work of endos. She suggested that cardiologists love number-needed-to-treat metrics (NNT), and these measures are something she’s now citing more often as a clinical endocrinologist – the NNT for Jardiance is 46 for a median 3.1 years to prevent one CV death. In our view, greater collaboration between endos and cardiologists will be an exceptionally good move, if it happens. CV complications remain a leading cause of morbidity and mortality for a diabetes patient population. People with diabetes face an 8x risk for MI and a 6.7x risk for stroke vs. the general population, not to mention 6.7 years reduced life expectancy, which grows to 11.2 years if the patient also has a history of MI and to 15.7 years if the patient has a history of both MI and stroke. These numbers, included in Dr. Wysham’s presentation, are all too high – we only hope that with the emergence of cardioprotective diabetes therapies, all sorts of providers (from PCPs to endos to cardiologists) will be better armed to reduce CV risk in their patients.

Meet the Professor Sessions

Diagnosis and Management of Diabetic Neuropathies

Rodica Busui, MD (University of Michigan, Ann Arbor, MI)

Dr. Rodica Busui, co-author of the ADA’s recent position statement on neuropathy in diabetes, took the stage to review different types of neuropathy, the pathophysiology of this complication, and best practices for diagnosis and treatment. She discussed diabetic peripheral neuropathy (DPN) alongside cardiovascular autonomic neuropathy (CAN), both of which progress distal to central. Small, unmyelinated nerve fibers are the most susceptible to damage: These cells are responsible for nociception (or the body’s response to harmful stimuli), and pain is thus one of the early symptoms reported for neuropathy. Dr. Busui described this pain as reminiscent of electric shock, burning, or a stabbing sensation. Large, myelinated fibers could still succumb to neuropathy in more advanced cases of the complication: These cells are responsible for pressure and balance, and later symptoms thus include numbness, tingling, and unsteady gait. While electrophysiological instruments to probe for neuropathy can be expensive – and the associated tests can be time-consuming – Dr. Busui pointed to several lower-tech methods of diagnosis. A simple knee jerk can test reflexes, while a pinprick would test pain sensitivity. These assessments take <five minutes, a fact Dr. Busui used to advocate for vigilance and earlier diagnosis of neuropathy in diabetes so that the complication can be curtailed. The latter half of her talk was dedicated to treatment options, headlined by early, aggressive glucose-lowering therapy. The famous DCCT trial is enough to show how intensive glycemic control greatly reduces risk for microvascular complications, including neuropathy. This session was a terrific primer on neuropathy, and we were glad to hear such a comprehensive perspective on the complication from a leading expert like Dr. Busui.

Where is the Sweet Tooth? Mechanisms Governing Palatable Intake and Weight Gain

Diet-Induced Modifications of the Mesolimbic System: Implications for Obesity

Carrie R Ferrario, PhD (University of Michigan, Ann Arbor, MI)

University of Michigan’s Dr. Carrie Ferrario presented experiments that may explain why some individuals are more likely to respond to environmental food cues with food-seeking behavior. The research is based on a curious phenotype – a rat model in which some individuals are resistant to obesity, while others are prone. Unsurprisingly, there are behavioral differences between the subgroups. For example, the obesity-prone bunch shows stronger cue-triggered food seeking (a conditioned stimulus indicating that food will come in ~15 seconds induces more foraging behavior). With the knowledge that consumption of sugar alters the nucleus accumbens (NAc; a brain region associated with reward) AMPAR-mediated transmission and receptor composition,  Dr. Ferrario’s group studies whether the obese-prone and resistant rats have altered AMPARs in the NAc following 10 days on a junk-food diet consisting of peanut butter, nesquik, cookies, and other goodies. 14 days after the diet ended, the rats were sacrificed and brain slices including the NAc were collected. Electrophysiological analysis revealed that obesity-prone rats display a significant upregulation in calcium-permeable AMPARs (CP-AMPARs; receptors that lack the GluA2 subunit), despite starting from the same baseline. Importantly, all rats ate the same amount of junk-food and gained the same amount of weight, but the obese-prone animals simply had more plastic synapses. In a follow-up experiment, after the rats were once again trained to seek food in response to a conditioned stimulus, either vehicle or a selective CP-AMPAR blocker (NASPM) was injected into the NAc. The NASPM injection was sufficient to completely abolish the food-seeking in the obese-prone rats, while the resistant rats didn’t exhibit significant foraging to begin with. In addition, immunohistochemistry confirmed that after the cue food-seeking training, obese-prone rats had increased GluA1 expression, and a slight decrease in GluA2 expression, while GluA1 was constant and GluA2 decreased slightly in the resistant rats. Thus, this unique subset of rats were hardwired to redesign their synapses in responses to food cues, putting them at higher risk of obesity. Though at the early stages, the potential application of this research in humans, at least philosophically, is poignant – some people are at risk to become obese, and environmental cues (e.g. advertisements) are more likely to incite them to consume food. The implications are two-fold: First, this research elucidates a pathway that may represent an intervention in these individuals, and second, we need to be careful about the environmental cues we expose people to.  

Case Management Forums

Care of Post Bariatric Surgery Patients: The Good, the Bad, and the Ugly

Drs. Maria Collazo-Clavell and Todd Kellogg of the Mayo Clinic (Rochester, MN) provided a candid overview of post-operative management of bariatric surgery patients – outlining what they termed “the good, the bad, and the ugly” of this increasingly popular obesity treatment. The duo emphasized that bariatric surgery is, of course, not a scalable solution to the obesity epidemic, but provides an option for patients who have been unsuccessful at losing weight via lifestyle modification or pharmacotherapy alone. Of course, surgery is invasive, often prohibitively costly, and not without risks. We continue to view bariatric surgery with cautious optimism for some patients and left this informative symposium with an appreciation for the necessity of shared decision making between healthcare providers and patients in deciding the best course of obesity management.

• The Good: Bariatric surgery is the only treatment proven to produce lasting weight loss in people with severe obesity for whom diet, exercise, and medication are not enough. Furthermore, the STAMPEDE trial (three year results published in March 2014; five year results presented at ACC in April 2016) compellingly demonstrated that gastric bypass and sleeve procedures were more effective than medical treatment of diabetes in managing A1c and BMI, including BMIs under 35 kg/m². This evidence is so compelling that ADA’s most recent Standards of Medical Care in Diabetes treatment guidelines were recently revised to recommend that metabolic surgery be “considered” as a diabetes therapy for individuals with a BMI between 30 and 34.9, and for Asians with BMI as low as 27.5 with inadequately controlled hyperglycemia.
• The Bad: Bariatric surgery is well-known for its risk of long-term nutritional deficiencies. Though patients are advised to take vitamin supplementation, deficiencies in vitamin B12, vitamin D, and iron are particularly common (15%) post-surgery, with the deficiency tending to worsen over time as adherence to the vitamin regimen wanes. Dr. Kellogg noted, however, that patients are often vitamin deficient prior to the surgery due to poor dietary quality, so sometimes levels of essential vitamins actually increase post-surgery because of the ensuing dietary and lifestyle changes.  increase their levels of essential vitamins. Complications of bariatric surgery are rare (“surgery is safe and getting safer,” Dr. Kellogg remarked) but abdominal pain after surgery is the biggest complaint. This is often due to ingestion of certain, difficult-to-tolerate foods too early after surgery (tough meats, fats and sugars, citrus fruit membranes, nuts), but in very rare cases could signify a stricture (3%) or leak (0.2-0.5%) which requires referral back to the surgeon.
• The Ugly: Increased risk of alcoholism or other substance abuse disorders is a little-known effect of bariatric surgery. Accumulating evidence suggests that bariatric surgery alters the way the body metabolizes alcohol, making it act more rapidly and potently, but with a quick offset – the classic pharmacokinetic profile of an addictive substance. 

--by Abigail Dove, Helen Gao, Brian Levine, Payal Marathe, Adam Brown, and Kelly Close