Closed-loop insulin pump home study shows promising results
by Adam Burrack, PhD
Scientific professionals with diverse skill sets are working at various levels to improve the lives of people with type 1 diabetes. While basic scientists continue to work on the “why” of T1D onset and transplant surgeons work on the “how” of beta cell transplantation to cure T1D, engineers and clinicians (including physicians, nurses, dietitians, and pharmacists) continue to work on the “now” of improving insulin pump therapy and clinical diabetes management. All of the above – and others – are priority research funding areas for the JDRF. Today I will describe recent improvements to the closed-loop insulin pump, a method intended to decrease the frequency and severity of overnight low blood sugar events, both in the US and in Europe. Closed-loop pumps combine a continuous glucose sensor – which communicates with the pump – with programming in the pump to stop delivery of basal insulin when blood sugar levels dip below a pre-set value, typically 65 or 70 mg/dl. In this way, closed-loop pumps can help people with diabetes – especially children – avoid severe overnight low blood sugar events.
In a recent paper in the New England Journal of Medicine, a closed-loop insulin pump showed superior efficacy over 12 weeks of at-home use to sensor-augmented insulin pump as measured by patient HbA1c levels. For this study each patient was their own control: each participant spent 12 weeks on the closed-loop system and 12 weeks on the control therapy, sensor-augmented pump therapy. Researchers quantified total time spent within the normal BG range during each 12-week period for each subject. For children in the study, time at night within the normal BG range increased compared to control treatment – a good outcome and a primary goal of the study. Adults in the study demonstrated a decrease in HbA1c – also a good outcome and another key end-point. Of note, 12 weeks is probably the minimum amount of time to see effects on HbA1c, since the lifespan of red blood cells is 90-120 days, the study period was only 84 days, and glycation is an irreversible chemical bond on red blood cells. In other words, these are promising results for a study designed to test the minimum amount of time to observe any effect. The authors may have observed an even greater effect by delaying analysis of HbA1c to the 120-150 day range. However, these results show a benefit to this therapy, so I can understand the ‘rush’ to publish the results.
A paper in Lancet that was published earlier this year by a group in England, demonstrated that closed-loop therapy yielded more spent in the normal BG range overnight, through decreased time spent either below or above the normal range. In other words, the Lancet paper and the New England Journal of Medicine paper reach very similar conclusions, but studied distinct patient populations. Of note, Medtronic helped to fund the research in the NEJM paper, whereas the JDRF and Diabetes UK were the primary funding agencies for the research conducted in the Lancet manuscript.
The NEJM study above is a follow-up to a 2013 publication in Lancet describing initial results of the low-suspend feature, the fore-runner of the closed-loop system. This proof-of-principle study, for which over 200 participants were recruited, established that the low-suspend feature decreases frequency and duration of nocturnal low BG events, while not increasing HbA1c. This second point is key, because it was not clear at the outset what effect decreasing overnight hypoglycemic events would have on the overall picture of diabetes management. It was possibly going in that spending more time in the normal BG range overnight would have increased HbA1c – which is NOT what this study found. This study was also funded in part by Medtronic, and was carried out in part at the Park Nicollet Diabetes Center in suburban Minneapolis, Minnesota, under the direction of a past President of the American Diabetes Association, Richard Bergenstal. The above work is a great example of a productive collaboration between industry and clinicians in attempts to get a more effective insulin pump system to market. The 2015 NEJM paper shows distinct improvements in the system, leading to better patient outcomes.
To conclude: combining the insulin pump with a BG level sensor appears to be a safe and effective way to both (1) decrease frequency/severity of hypoglycemia, and (2) maintain or improve HbA1c levels. Long-term, these improvements could greatly enhance quality of life for people living with T1D, as well as greatly improving peace of mind for caregivers, loved ones, and other “type 3 diabetics” who are emotionally affected by T1D.
However, there is a reality check embedded within these high-level publications: a closed-loop insulin pump is an improved treatment for T1D, not a cure. No insulin pump, closed-loop, or dual-hormone system represents a cure for the underlying pathology that caused T1D. An islet transplant that was not destroyed by autoreactive T cells in the recipient would represent a biologic cure for T1D. This and only this scenario would allow “insulin independence” for the person with T1D. This is a fundamental difference in goals between basic scientists (cure-focused), physicians (patient care-focused), and engineers (focused on improvements to existing technology) working to improve quality of life for people with T1D. While physicians work on the “now”, and engineers work on “what’s next”, basic scientists are continually working on the blue-sky scenarios, trying to change how the immune system responds to beta cells – to prevent destruction either before disease occurs, or following transplantation of islets.