It never rains, but it pours, as they like to say in the western parts of the British Isles. And that seems true for Artificial Pancreas approvals in Europe too, as the second commercial APS system, or Hybrid Closed Loop, as we should call it, clears a regulatory hurdle.
You may never have heard of them, but Diabeloop is a small French company out of Grenoble that had been quietly working on a looping system with, amongst others Kaleido and Cellnovo, and has now got it through the European Medicines Agency.
Who are Diabeloop?
They are, as I’ve already mentioned, a small company based in Grenoble, with roughly 50 employees. Their MD and co-founder, Dr Guillaume Charpentier, is also the founder of the Centre de Recherche pour l’Intensification de traitement du Diabète. Diabeloop was founded as a way to take his research into a commercial world.
They have been partnering with Dexcom and Cellnovo to create DBLG1, the heart of their closed loop. And importantly, they received their CE mark on 8th November, allowing them to move to the next level of both funding and marketing the system to patients and healthcare services.
The initial release that is being undertaken in early 2019 will use the Dexcom G6 and the Kaleido patch-pump, however they are working on integration with other devices.
The system uses a locked down mobile phone at the core of its infrastructure, which will be a familiar set-up to many who have participated in various global closed loop trials. Their website does say that you cannot install your own apps on it.
What they say:
DBLG1 is a self learning medical device that puts AI at the service of the patient
The pricing has not yet been set, but the company is looking to work with the healthcare systems around Europe (and presumably globally) to arrange reimbursement or paid for provision.
How does the system perform?
After the 12-week study across 33 users, undertaken 2017-2018, Diabeloop presented very positive results:
- In closed loop, the time spent in the target, between 70 and 180 mg/dl, was 69.3% against 56.6% open, or 3 hours more per 24 hours.
- Hypoglycemia (<70mg/dl) was 2% in closed loop versus 4.5% in open mode, ie 36 minutes less in 24 hours.
- Nocturnal hypoglycaemia occurred in 3.9% of cases in open loop versus 1.3% in the Diabeloop® system, a risk reduction of three.
- The mean blood glucose level was 168.5 mg/dl in the open loop versus 156 mg/dl in the closed loop.
Of course, we don’t know what the target was in these trials.
As usual with closed loop trials, nearly all the participants were very supportive of the quality of life benefits.
it also begs the question that if an EMA approval is provided for an APS system with only 33 people using it for 12 weeks, with the data we have available from DIY systems, what could be submitted?
What didn’t I know about Diabeloop?
You may not know that Diabeloop have paid close attention to the DIY community and have done a lot of work to review the codebases that are publicly available and test components of their system against those.
From what I understand, they have modified the way their system works where the Open Source solutions offer a better mechanism for achieving similar outcomes, which is a first.
Are they one to watch?
Almost certainly. As the first hybrid closed loop to offer a patch pump, there are a significant number of people who will be very interested in the outcomes of their next round of funding. It will be interesting to see the progress of this David, against the Goliath that is Medtronic.
One thing is for certain though. The performance they showed in their EMA approval study certainly looks as though it’s in line with those of other similar systems, and with a patch pump as the primary roll out device, they look as though they have a great opportunity to make a mark in this market.