(DCP048) IMPACT OF GLP-1 RECEPTOR AGONISTS COMPARED TO BASAL INSULIN START ON METABOLIC TARGETS IN PATIENTS LIVING WITH TYPE 2 DIABETES AND CHRONIC KIDNEY DISEASE (IMPACT GLP-1 CKD)

Background: The leading cause of chronic kidney disease (CKD) is type 2 diabetes (T2D). In Canada, 1 out of 10 Canadians has CKD, representing about 4 million people. The aim of this study is to compare the effectiveness of adding a GLP-1 receptor agonist (GLP-1RA) compared with adding basal insulin amongst a propensity-score match group of people affected with T2D and CKD, already treated with an SGLT2i and not currently reaching target glycemic control.

METHODS AND RESULTS: Sociodemographic and clinical variables were analysed from the LMC Diabetes Registry, which represents the accumulated health records of about 60 endocrinologists. 548 participants with T2D and CKD, on an SGLT2i for at least 6 months, eGFR ≥ 15, and HbA1C ≥ 7.5% were started on either a GLP-1RA or basal insulin between January 2019 and September 2022. Using a logistic regression model, 174 participants initiating GLP-1RA were propensity score matched to 174 participants initiating basal insulin (table).

A significantly greater reduction in HbA1C from baseline to follow-up (26-to-52 weeks) was observed for participants in the GLP-1RA group compared to those in the basal insulin group (–0.78±0.10% vs. –0.44±0.09%, p< 0.001, figure). Urinary albumin to creatinine ratio was significantly lower at follow-up for participants in the GLP-1RA group compared to the basal insulin group (14.8±3.0 vs. 23.0±3.9mg/mmol, p=0.013). There were no significant differences observed at follow-up for eGFR or body mass index, although there were a large proportion of missing variables for body weight measurement due to the COVID-19 pandemic leading to a predominance of virtual healthcare visits in Canada during the study period for the entire cohort (49.4% missing). Once-weekly subcutaneous semaglutide was initiated in 84.1%, dulaglutide in 8.0%, liraglutide in 4.6%, and once-daily oral semaglutide in 2.3% of participants in the GLP-1RA group. Glargine U-100 was initiated in 48.9%, degludec U-100 or U-200 in 37.9%, glargine U-300 in 11.5%, and detemir in 1.7% of participants in the basal insulin group. Self-reported hypoglycemic events during follow-up were significantly lower with GLP-1RA compared to basal insulin (0.07±0.03 vs. 0.39±0.09 events/week, p< 0.001). 14.4% of participants in GLP-1RA compared with 25.3% in the basal insulin group discontinued therapy during the follow-up period (p=0.015).

Conclusion: This study demonstrates significant reductions in HbA1C and urinary microalbumin, with fewer hypoglycemic events, and better adherence to therapy amongst a propensity-score matched group of participants initiating a GLP-1RA with T2D and CKD on a background of SGLT2i therapy compared to a group initiating basal insulin.