ADA Day 3 Highlights – Emphasis on Add-On Therapies and Better Technology
By Jeemin Kwon, Brian Levine, Payal Marathe, and Adam Brown
Using time-in-range to assess risk for complications, Omnipod Horizon Closed Loop, Bihormonal closed loop (insulin + Pramlintide), SGLT-2 oral drug for type 1, and more!
Having been at ADA 2018 for three days now, we’ve observed two standout areas that consistently excite conference attendees: updates on diabetes technology and new add-on therapies for people with type 1 diabetes. Read more below!
Click to jump:
The outcomes beyond A1C movement scored a big win – new analysis of fingerstick data from the landmark DCCT trial showed a high correlation between time-in-range (defined as 70-180 mg/dl, taken from seven-point blood glucose monitor profiles) and development of eye, kidney, and nerve disease. Specifically, each 10% drop in time-in-range was linked to a 61% increased risk for eye disease, and 53% increased risk for a marker of kidney disease. It’s possible that if CGM had been used for this study, the correlation would have been even stronger. This clever analysis lends further credence to the idea that people with diabetes and their providers should focus on maximizing time-in-range, not solely focusing on A1C. If curious about more reasons to focus on outcomes beyond A1C, click here.
Though small (n = 11 adults), this five-day, free-living hotel study of Insulet’s Omnipod Horizon hybrid closed loop system had promising results. (For background on automated insulin delivery and hybrid-closed loop systems in development, click here.) Compared to standard therapy (multiple daily injections or insulin pump), Horizon drove a 2.7 hour increase in daily time-in-range (70-180 mg/dl). Notably, average glucose (from CGM readings) was not meaningfully different between the groups. However, the Horizon closed loop did significantly smooth out overnight blood sugar variability (26% coefficient of variation on closed loop versus 36% on standard therapy). The average meal size was 53 grams of carbs (largest meal was 200 grams of carbs!), and the study participants frequently exercised, suggesting that the system is well-equipped to handle real-world challenges. For example, Stanford’s Dr. Bruce Buckingham shared that one participant wanted to run a few miles at a 6.5-minute pace – the study team didn’t have anyone that could run that fast while carrying the tablet that housed the closed loop algorithm, so they had someone follow him on a bike! Notably, the system also performed very well in children (ages 6-12) and adolescents (ages 12-18). It is unclear when a larger trial for FDA submission will be planned, but it is likely to be in 2019.
Currently available hybrid closed loops only deliver and adjust insulin, but there are several bihormonal systems being studied, including one that delivers insulin+pramlintide together. Pramlintide is a molecule that mimics the effects of amylin, which is secreted from pancreatic beta cells along with insulin, slowing stomach activity and promoting satiety (feeling full). This 24-hour, in-clinic study enrolled 27 type 1 adults who were divided into three groups, all on closed loop: rapid insulin alone, rapid insulin + pramlintide, and Humulin R (regular insulin) + pramlintide. Time-in-range (70-180 mg/dl) was 86% on rapid insulin + pramlintide, 74% on rapid insulin alone, and 68% on Humulin R + pramlintide. This translates to nearly three more hours in range with insulin + pramlintide than rapid insulin alone, and four more hours in range than Humulin R + pramlintide. There were no meaningful differences in reported side effects (nausea, headache, vomiting, etc.), nor hypoglycemic events. Notably, the positive time-in-range results were attributed to significantly improved post-meal blood sugar responses. It’s worth noting that this study was funded by JDRF and run by a group at McGill, but the algorithm used is actually one that Lilly has licensed for its in-development hybrid closed loop system. The positive results certainly make a strong case for a system that delivers insulin and pramlintide simultaneously.
SGLT-2 Potential in T1D Confirmed: Farxiga and Zynquista Lower A1C, Body Weight, and Insulin Doses; DKA Risk Remains
DEPICT-2 was a continuation of studies on SGLT-2 inhibitor Farxiga use in type 1 diabetes. Similar to the DEPICT-1 study completed last year, these results showed that compared to placebo, participants taking Farxiga saw a greater A1C drop by 0.42%, more weight loss (3-4% of body weight, about 4-7 pounds), and an 11% reduction in total daily insulin dose after 24 weeks. In particular, the presenters emphasized the absence of severe hypoglycemia. Like the earlier study, more patients experienced DKA (a serious diabetes complication) when taking Farxiga compared to placebo – 2-3% vs. 0%.
We also learned about 52-week data from inTandem1, which compared two doses of dual SGLT-1/SGLT-2 inhibitor Zynquista (sotagliflozin) to placebo in people with type 1. Compared to placebo, those on the lower dose saw a 0.36% greater drop in A1C, and those on the higher dose saw a 0.41% greater drop. Weight loss was another key benefit – those on the lower dose lost 4% (~7 pounds) of their body weight, and those on the higher dose lost 5% (~10 pounds). With regards to reduced insulin doses, basal insulin was reduced by 4% in both dosing groups, and bolus insulin was reduced by 4% in the low-dose group and 8% in the higher-dose group. Perhaps the most compelling aspect of it all is the 1.3 - 2.8 hour/day increase in time-in-range (70-180 mg/dl) achieved with Zynquista. Similar to previous findings, there were more reports of DKA: 4.2% of participants on the higher dose and 3.4% on the lower dose, compared to 0.4% on placebo.
Certainly, DKA is a risk that needs to be addressed with appropriate healthcare provider and patient education; the good news is that there is agreement that the DKA is avoidable, and that the benefits of this drug make it a conversation worth having.
New results from a year-long trial studying Victoza in type 1 (run independently of Novo Nordisk) suggest that the GLP-1 agonist may be a candidate for expanded add-on therapies for people with type 1. Compared to placebo (a sugar pill), study participants saw a 0.6% greater drop in A1C and lost 6 more pounds over one year. Though a small study (46 participants) and no results to show on reduced insulin needs, these results are enough to justify further investigation. Some historical background: Novo Nordisk (Victoza’s manufacturer) had decided not to move forward with Victoza as an add-on therapy for type 1 following lukewarm ADJUNCT ONE results, and said at the time it would revisit this decision when oral Victoza was in development. Some called it “interesting” that this study found more positive results – we said, of course it did, because the dosing was approached differently and CGM was used! Overall, it’s our sense that the path was more specifically designed for type 1, as well. Dr. Dandona seemed very excited in the discussion we got to have with him following our learning about the results.