Kate Hawke, MBBS, FRACP: No financial relationships to disclose
Abstract: Background Commercial automated insulin delivery (C-AID) pump systems have demonstrated improved outcomes in type 1 diabetes (T1D), but have potential drawbacks including cost, delays to innovation related to regulatory processes, and device ecosystem constraints. Given these limitations, unregulated open-source automated insulin delivery (OS-AID) systems are used by some people with diabetes. The benefits of OS-AID have been demonstrated in several observational studies and in randomized trials against non-AID controls. To date there are no randomized trials comparing OS-AID and C-AID, nor comparing different OS-AID systems with one another. When supported in a clinic setting, OS-AID is referred to as supported OS-AID (SOS-AID).
Methods This protocol is in development and has not yet been submitted for Research Ethics Board approval.
The SOAR-UP (Supported Open-source Automated insulin delivery Randomized mUlti-arm Platform) trial is a 90-day non-inferiority, open-label, randomized platform trial assessing multiple C-AID and SOS-AID systems. Participants naïve to automated insulin delivery will be allocated randomly to either SOS-AID or C-AID. Each participant will retain their existing pre-study hardware (smartphone and insulin pump), and be allocated randomly among the SOS-AID/C-AID systems available for that hardware, see Figure 1. They will use the allocated system for 90 days with standardised installation procedures and regular clinician support in an experienced centre. Those who do not wish to be randomized may participate in a separate observational arm for C-AID.
The primary outcome measure is the sensor glucose percentage time-in-range (3.9-10mmol/L) at end-of-study (days 76 through 90) in the randomized SOS-AID and C-AID groups. Secondary outcomes include other glycemic measures, quality-of-life measures, adverse events and system performance. Participants with T1D aged 2-80 years using continuous glucose monitoring will be recruited from a single centre in Canada.
The C-AID systems planned for inclusion in this study (Omnipod 5 and mylife CamAPS FX) await Health Canada approval. The SOS-AID systems planned for inclusion (Loop, iAPS and AAPS) are currently in clinical use at the study centre in the context of a detailed consent form and waiver.
Expected results based on available data (see Table 1) include a mean time-in-range of >70% for all SOS-AID groups and the C-AID Omnipod 5. Adverse events and system performance are expected to show safety of SOS-AID systems.
Conclusion The SOAR-UP trial will compare the efficacy and safety of SOS-AID with C-AID systems. The results will be of assistance to people with diabetes and clinicians in their assessment of available therapies.
