(2.13) Impact of Digital Twin Intervention on Microalbuminuria Progression in Participants Enrolled for Type 2 Diabetes Remission: Outcomes of RCCT At 1 Year

We evaluated the effectiveness of the digital twin (DT) technology to improve HbA1c and improve albuminuria, in patients enrolled to achieve remission of T2DM. DT platform uses AI and Internet of Things, to integrate multi-dimensional data to give precision nutrition and health recommendations via the mobile app.

Methods:

We evaluated the changes in the microalbuminuria in the DT group (n=208) as compared to the standard of care (SC) (n=81). Remission was defined as HbA1C levels less than 6.5% without medication for over 3 months. The DT uses a machine learning algorithm to integrate clinical and sensor data to predict personal glucose response. Patients were connected to continuous glucose monitoring for machine learning to enable prediction and self-recorded dietary intake using the mobile app.

Results:

Of the 319 people with T2DM, 208 subjects completed more than six months of DT. 72.5% (151/208) sustained remission in one year. The mean age was 46 years (±9.7, 95% CI 45 to 47). In DT group, mean HbA1c % at baseline was 9 (±1.8, 95% CI 8.7 to 9.2), which decreased to 6 (±0.6, 95% CI 5.9 to 6), p< 0.0001. In SC group, mean HbA1c % at baseline was 8.5 (±1.8, 95% CI 8.1 to 8.9), which decreased to 8.2 (±1.5, 95% CI 7.9 to 8.5), p=0.31 ns. The difference in the change in HbA1c in DT -3 (±1.8) was significant as compared to change in SC -0.2 (±1.2), p< 0.0001. DT group showed significant improvements, 92% (n=92) of participants had normal albuminuria at D360, an increase from 79% (n=164) at D0. In SC group the grade of albuminuria remained unchanged. In DT group macroalbuminuria decreased to 1 % (n=3) from 4 % (n=9). In DT group, mean albuminuria (mg/g) at baseline was 36.5 (±98, 95% CI 23.1 to 49.9), which significantly decreased to 17 (±58.4, 95% CI 9.4 to 25.4), p= 0.0166. In SC group, mean albuminuria (mg/g) at baseline was 30.3 (±60.1, 95% CI 17 to 43.6), which increased to 56.5 (±271, 95% CI -3.3 to 116), p=0.39 ns. The difference in the change in DT -19 (±90) was significant as compared to change in SC 26.2 (±243), p=0.02.

Discussion/Conclusion:

Digital Twin technology is more effective than standard care in reducing HbA1c and mitigating microalbuminuria progression, with the potential to improve the renal outcomes in patients at risk of progression to overt nephropathy.