Diabeloop has made rapid progress since its founding in 2015 and has achieved an important milestone with receipt of CE marking in Europe. This positions this closed-loop system for commercial release in 2019.
“ With this certification, we’re getting closer to the market launch and will move towards a new funding round to support our international scaling and our ambitious R&D roadmap. We hope to do all that by the second quarter of 2019 ” added Marc Julien, Diabeloop co-CEO and CFO.
We interviewed the Diabeloop team by email.
Q: What is unique about your DBLG1 System such that T1D people might want to switch to it?
DBLG1 is auto-adaptive (everyone’s diabetes is different) thanks to machine learning, and it is customizable so that everyone can tell the system what his/her preferences are (average vs. hypos/rescue carbs). Other features invented by people living with T1D, include Zen mode (temporarily no-hypo mode) or Confidential mode (no data uploaded to the server at all, for when you want to have your privacy).
The system has been tested on 150 patients overall in clinical tests, representing over 6100 days, 80% being at-home time.
The 1st arm of the SP7 clinical study presented in both scientific and press session at ADA earlier this year showed that we decreased mean glucose 13mg/dL while at the same time decreasing time in hypo by two-thirds.
Quoting from our release in June — “The results showed that the percentage of time spent in the target blood glucose range of 70-180 mg/dl was 69.3 percent for patients using the DBLG1 system compared to 56.6 percent for patients using an open-loop system. While the average glycemic level was lower for patients with the DBLG1 system compared to the one for patients with standard care, the lowered glucose level did not lead to more hypoglycemic events. The percentage of time spent in hypoglycemia (< 70 mg/dl) was about half the amount with the DBLG1 than with the usual, open-loop system (2 percent and 4.5 percent, respectively).”
Q: What insights led your team to develop it the way you did?
We recognized the extreme complexity and variability in Type 1 diabetes, for every single individual, due to factors that are virtually impossible to measure. These include physiology and/or lifestyle, glycemia/insulin ratio that is different for each individual and subject to evolution, and other parameters such as sport, stress, coffee intake, fat intake, insulin infusion site, minor infections etc.).
Dealing with this complexity was our primary goal.
This required us to develop two algorithms:
- A self-learning Artifical Intelligence
- An “expert” mode which reproduces the logic of a doctor or expert patient and takes over when the AI is deemed less reliable
Algorithms alone, however, aren’t everything. Integrating them into the User Experience so that they improve each user’s life is very complicated and has been our focus for the past years.
Q: What is your timing for selling the DBLG1 system in Europe?
A gradual launch is expected to start in 2019 in Europe. We have had quite a bit of interest coming from different European countries. Beyond France, which is our “home country”, primarily countries will be where there is an appetite for innovative medical devices and fluid reimbursement mechanisms.
Q: What is your timing for selling the DBLG1 system in the US and Canada, and will you do this yourself or via a North American partner?
We have just launched a second financing round in order to support ongoing R&D work, European commercial launch, but also FDA submission and preliminary market introduction in the US.
Q: How do you envision charging for the DBLG1 system and its disposables?
Pricing has not yet been set. It will be fine-tuned to adapt to each country. We are currently working on this for our initial markets.