Memorandum

Cambridge-Developed CamAPS FX AID Smartphone App Receives CE-Mark, Launches in UK – March 16, 2020

Dexcom G6 + Dana RS pump + CamAPS FX Android app; CE-Mark includes pregnancy and children ≥1-year-old; priced at £70-£80/month (~$86-$98)

Just this morning, University of Cambridge spinoff CamDiab announced CE-Marking and UK launch of the CamAPS FX automated insulin delivery (AID) Android app for people with type 1 diabetes. CamAPS FX is the world’s first available closed loop smartphone app and works with Dexcom G6, SOOIL’s Dana Diabecare RS pump, and compatible Android phones. The app is already available in the UK’s Amazon Appstore with pricing ranging from £70/month (~$86/month) for an annual subscription to £80/month (~$98/month) on a month-to-month basis. Notably, the CE-Marking also makes CamAPS FX the first AID system available for pregnant women and children down to one year old. For now, CamAPS FX availability is limited to patients in 11 NHS clinics, where certified trainers are available. CamDiab is currently working on adding connectivity with other pumps and expanding to other countries. On the US front, CamDiab is planning to submit to the FDA in “2020” under the class II, iController pathway.

CamAPS FX is a hybrid closed loop system, with automated basal rates and manual meal bolusing. At startup, the user gives body weight and total daily insulin dose. The algorithm is “highly personalized” (meaning it is designed to adjust insulin according to the users’ behaviors and needs) and targets a glucose level of ~105 mg/dl (5.8 mmol/l) by adjusting insulin delivery every 8-12 minutes. CamAPS FX works with Android phones compatible with Dexcom G6. The CGM data is available on the smartphone app in real-time (see screenshots below) and the phone also communicates with the Dana RS pump intermittently (every 8-12 minutes) to send insulin delivery commands. When the phone is out of range, the pump reverts to pre-programmed delivery settings. According to an email with Cambridge’s Dr. Roman Hovorka, the effect of CamAPS FX on the smartphone’s battery is “not excessive” and that a Samsung phone was able to run CamAPS FX for four days when other use was “minimized.”

The app also includes two alternate modes, “Boost” and “Ease Off.” “Boost” is designed for times of higher insulin need (e.g., illness, increased food intake) – the app also notes “Boost” can be used for 20-30 minutes before a meal to reduce post-prandial spikes. In contrast, “Ease Off” is designed for times of lower insulin need, e.g., exercise. We really like the lingo. CamAPS FX also includes a bolus calculator and the ability to calibrate their G6 sensor via the app. The bolus calculator allows users to either carb count or choose from one of four preset meal sizes (small, medium, large, and very large). Lastly, data can be automatically uploaded to Diasend/Glooko every 5-10 minutes and text-message alerts can be sent to followers when alarms are generated. In the past, we’d heard from Dr. Hovorka that integration with Dexcom Clarity could be coming later this year.

With launch in the UK, CamAPS FX is the first commercially available closed loop system with smartphone control. Other pump and algorithm organizations all have plans to have some form of smartphone control in the next couple years: Tandem (FDA submission for smartphone bolusing “this summer”), Insulet (Omnipod Horizon launch with smartphone control in “early 2021”), Tidepool (submission for Loop as soon as this year), Medtronic (via Tidepool), and potentially Lilly. We imagine virtually all will be delayed by COVID-19 to some degree.

The CamAPS FX algorithm, developed by Dr. Hovorka at the University of Cambridge, is quite well-studied in an impressively wide variety of populations. The Cambridge algorithm first began studies in 2006. Since then, CamDiab claims “more than 15 clinical trials involving over 300 participants.” These studies have included the often-excluded populations of young children and pregnant women.

Screenshots

 

Compilation of Past Coverage

Date – Title

Update

February 20, 2020 – Closed loop snapshot of a participant in the AiDAPT (n=124 women type 1s) trial

  • Closer look of one the participants’ closed-loop stats, showing a 84% Time in Range

  • Parallel-arm, randomized controlled trial comparing CamAPS FX with open loop (Dexcom G6, pump, or MDI)

November 2019 – Email Update from Dr. Roman Hovorka

  • Aiming for initial launch of CamAPS FX system in the UK for “early 2020.”

  • CE-marking “virtually done” with final paperwork from the age of 1 year and older pending

February 21, 2019 – Dr. Roman Hovorka Announces Plans at ATTD 2019 to Commercialize the CamAPS FX AID System

  • Announced formal plans of commercializing a new AID system called CamAPS FX, comprised of the Cambridge MPC algorithm, Dexcom G6, and Dana R/RS pumps equipped with data streaming to Glooko

  • Closed loop data from a 12-week randomized controlled parallel trial (n=86 type 1s) in an October publication from The Lancet

  • Closed loop group (Enlite 3 sensor, modified Medtronic 640G pump, and Cambridge algorithm on Android phone) achieved 65% Time in Range (~2.6 hours more in range) than the 54% control group (sensor-augmented therapy)

November 9, 2018 – Dr. Hovorka on AID in the Type 2 Market

  • Advocated for the potential of closed loop at diagnosis in type 2 populations to quickly normalize glucose levels, give beta cells a risk, and put the patient into remission

 

November 8, 2018 – Dr. Hovorka Presents Data (n=40) on AID in the Hospital

  • New study completed in patients on parenteral/enteral nutrition who were on 15 days of full closed loop (subcutaneous CGM/pump) in the hospital

 

 

October 3, 2018 – Cambridge 12 -week randomized trial comparing hybrid closed loop (n=46) to sensor-augmented pump (n-40) in patients with baseline A1c of 8.2%-8.3
%

  • Home study including adults, adolescents, and children (6+ years) using the Cambridge MPC algorithm on an Android phone, MiniMed 640G, and Enlite 3 sensor

  • +3 hours/day Time in Range improvement (52% to 65%) and A1c drop of 0.6% on closed loop vs. 0.1% on SAP

  • Results also published in The Lancet with positive commentary from Yale’s Dr. Jennifer Sher

June 2018 at ADA - 15-day RCT Data (n=136) of Navigator 2 transmitter/receiver, Dana R insulin pump, and tablet housing the algorithm

  • Participants who received closed loop therapy (n=70) spent 5.8 more in-range (100-180 mg/dl; 65.8% vs. 41.5%) and 6.2 fewer hours > 180 mg/dl (23.6% vs. 49.5%) per day compared to control (n=66)

  • Positive results attributed to variations in insulin delivery (adjusted every 12 minutes)

  • Results simultaneously published in NEJM

May 14, 2018 – BMJ meta-analysis of 40 randomized controlled trials dating back to 2014 (n=1,027) comparing closed-loop outcomes in the outpatient setting to open-loop pump therapy in type 1 diabetes

  • Time in Range improved by 9.6 percentage points, translating to 2.3 more hours in range per day with closed loop

  • Time spent in hypoglycemia and hyperglycemia also improved, with time <70 mg/dl decreasing by 1.5%-points with closed loop therapy (-21 minutes/day) and time >180 mg/dl decreasing by 8.5%-points (-2 hours/day)

  • Original BMJ publication

March 2018 at Diabetes UK – Review on the Automated Insulin Delivery Competitive Landscape

  • Medtronic’s MiniMed 670G was the only system on the market and available in the US

  • Insulet, Bigfoot, Inreda Diabetic, Tandem, Boston University, University of Virginia/Type Zero, and the University of Cambridge all had AID pursuits

June 2017 at ADA – Review of Upcoming Closed Loop Studies

 

 

--by Albert Cai, Ani Gururaj, and Kelly Close