Early Add-On Combination of GLP-1 Receptor Agonist and SGLT2 Inhibitor in People With Cardiovascular-Kidney-Metabolic Stage 2-3

Overview

This observational target-trial emulation study uses routinely collected electronic health record data from the TriNetX US Collaborative Network to compare early add-on treatment strategies in adults with obesity, type 2 diabetes, and cardiovascular-kidney-metabolic stage 2-3. The study evaluates outcomes among people who start a GLP-1 receptor agonist or an SGLT2 inhibitor and then, within 90 days, either add the alternate therapy, add a DPP-4 inhibitor or sulfonylurea, or do not receive early add-on treatment. The primary outcome is all-cause mortality within 36 months. Secondary outcomes include major adverse cardiovascular events and major adverse kidney events. Nonrandom treatment selection is addressed using propensity-score matching to estimate comparative risks and treatment effects.

This is a retrospective observational study designed as a target-trial emulation using routinely collected electronic health record data from the TriNetX US Collaborative Network. The study aims to compare 36-month risks of all-cause mortality and major cardiorenal outcomes among adults with obesity, type 2 diabetes, and cardiovascular-kidney-metabolic stage 2-3 who initiate a GLP-1 receptor agonist or an SGLT2 inhibitor and subsequently follow different early treatment intensification strategies. Primary estimands are relative hazards and absolute risk differences at 36 months. Eligible participants are adults aged 20 years or older with obesity and type 2 diabetes who meet cardiovascular-kidney-metabolic stage 2-3 criteria. Eligibility, exposures, and outcomes are operationalized using TriNetX electronic health record data, including ICD-10-CM diagnosis codes, body mass index of at least 27 kg/m2, hemoglobin A1c of at least 6.5%, and medication dispensing records. Exclusion criteria include prior use of relevant drug classes within 6 months, major cardiovascular disease or advanced kidney disease within 12 months, major cardiovascular or renal events within 6 months, and any history of non-type 2 diabetes, HIV, bariatric surgery, or solid-organ transplantation. Three mutually exclusive early treatment intensification strategies are compared: early add-on SGLT2 inhibitor therapy, early add-on DPP-4 inhibitor or sulfonylurea therapy, and no early add-on therapy. For add-on strategies, the index date is the date of add-on initiation within 90 days after background therapy initiation. For the no early add-on strategy, the index date is the 90-day landmark after background therapy initiation. Because treatment assignment is not randomized, nonrandom treatment selection is addressed using multinomial propensity-score matching with 1:1 nearest-neighbor matching without replacement and a caliper of 0.2 times the standard deviation of the logit propensity score. Post-match covariate balance is evaluated using standardized mean differences, and any residual imbalance is further adjusted for in the outcome models. Follow-up begins at the index date and continues until the earliest of the outcome of interest, death, last recorded encounter, or January 31, 2026, with intention-to-treat and per-protocol analyses performed. The primary outcome is all-cause mortality within 36 months. Secondary outcomes include major adverse cardiovascular events, defined as cardiovascular arrest, myocardial infarction, or stroke, and major adverse kidney events, defined as end-stage kidney disease, dialysis dependence or initiation, kidney failure, or death. Comparative effects are estimated over 36 months for pairwise early add-on comparisons, with prespecified subgroup analyses by cardiovascular-kidney-metabolic stage, obesity severity, baseline kidney function, and baseline cardiovascular comorbidity.