Ketone monitoring approaches for DKA risk mitigation in people with T1D on adjunctive SGLT2 inhibitors (karma)

The National Institute of Diabetes and Digestive and Kidney Diseases awarded a $7.432 million Cooperative Agreement (U01DK143384) to Adventhealth Orlando on September 11, 2025, under the Diabetes, Digestive, and Kidney Diseases Extramural Research program (CFDA 93.847).

The award funds development and testing of a risk mitigation strategy using continuous ketone monitoring (CKM) to enable safer use of SGLT2 inhibitors in patients with Type 1 Diabetes. SGLT2 inhibitors reduce cardiovascular and kidney disease risk but carry a sixfold increase in diabetic ketoacidosis (DKA) risk in Type 1 Diabetes patients—a barrier to FDA approval as adjunctive therapy. The study will recruit a cohort of 75 Type 1 Diabetes patients to address three specific aims: determining how adjunctive SGLT2 inhibitor use affects development of hyperglycemia, ketosis, and ketoacidosis during controlled inpatient insulin withdrawal testing compared to insulin monotherapy; assessing insulin, carbohydrate, and fluid requirements to reverse ketosis and refining recovery protocols; and developing and validating a pragmatic CKM-based risk mitigation strategy. Work performs in Orlando, Florida, with period of performance through August 31, 2028.

University of Maryland, Baltimore serves as a sub-recipient, with Dr. Simeon Taylor providing strategic leadership and data analysis collaboration alongside Dr. Richard Pratley, the contact principal investigator. Taylor and Pratley will conduct weekly virtual meetings and approximately one to two annual in-person meetings, collaborate on clinical data analysis when available, co-author manuscripts, and monitor emerging literature on SGLT2 inhibitors.

Abstract

People with type 1 diabetes (T1D) are at increased risk of cardiovascular disease (CVD), heart failure (HF) and chronic kidney disease (CKD). SGLT2 inhibitors have been shown to decrease risk of CVD, HF, and to slow the progression of CKD in people with type 2 diabetes. However, the FDA has not approved the use of SGLT2 inhibitors as adjunctive treatments for T1D because of a ~6-fold increase in the risk of diabetic ketoacidosis (DKA). In trials of SGLT2 inhibitors in people with T1D, self-monitored plasma beta-hydroxybutyrate (bohb) levels were predictive of DKA but this type of monitoring did not prevent ketoacidosis. Recent advances have enabled continuous monitoring of bohb levels in interstitial fluid - raising the possibility that continuous ketone monitoring (CKM) could effectively mitigate the increased risk of DKA in SGLT2 inhibitor treated patients with T1D. The overarching goal of this proposal is to develop and test a pragmatic risk mitigation strategy using CKM that will enable safer use of SGLT2 inhibitors in patients with T1D. To accomplish this, we will recruit and study a diverse cohort of 75 patients with T1D to address 3 specific aims. In aim 1A we will determine how adjunctive SGLT2 inhibitor use affects the development of hyperglycemia, ketosis and ketoacidosis during insulin withdrawal testing (IWT) in a highly controlled, inpatient setting in people with T1D on insulin monotherapy to that on insulin + adjunctive therapy with an SGLT2 inhibitor. In aim 1B we will assess insulin, carbohydrate and fluid requirements to reverse ketosis during these conditions to refine recovery protocols that reverse elevated ketones during these conditions. We hypothesize that adjunctive SGLT2 inhibitor use will increase the rate of development and maximal levels of ketosis and ketoacidosis while blunting increases in glucose during an insulin withdrawal test. In aim 2A we will test the effectiveness of a CKM enabled DKA risk mitigation strategy that accounts for the increased risk for ketosis and ketoacidosis with SGLT2 inhibitors. We will use a sequential, 2- phase study design to compare the number and severity of CKM-assessed ketone events that occur during a generalized DKA risk mitigation plan VS. A personalized DKA plan implemented for people with T1D on SGLT2 inhibitors. In aim 2B we will actively update 'decision tree algorithms' for broad risk mitigation to incorporate new learnings obtained during the conduct of the study using an adaptive design. We hypothesize that a strategy accounting for the physiological effects of SGLT2 inhibitors on the development and resolution of ketosis will be more effective at reducing the number and severity of ketosis events. In aim 3 we will explore intrinsic and extrinsic factors associated with ketone levels and responses to SGLT2 inhibition among people with T1D in both free living conditions and during IWT by conducting secondary analyses of data generated in aims 1 and 2 to determine possible predictors of the number and severity of ketone events and responses to SGLT2 inhibition. The results of this study will lead to an improved risk mitigation strategy that will enable persons with T1D to safely use SGLT2 inhibitors to decrease risk for CVD, HF and CKD, thus improving long-term outcomes.