Patient-Level Multidimensional Response Phenotypes in Obesity-Associated Type 2 Diabetes: A 12-Month Real-World...

Background/Objectives: Type 2 diabetes mellitus (T2DM) has been evaluated through a glucocentric model centered on hyperglycemia, glycated haemoglobin (HbA1c) targets, and complications. However, obesity-associated T2DM reflects a broader metabolic ecosystem in which glycemic control interacts with adiposity, inflammation, muscle function, renal stability, and treatment mechanisms. Therapeutic response may be better captured through patient-level responder phenotypes than through mean changes in isolated outcomes. We aimed to determine, in adults with obesity-associated T2DM maintained on stable 12-month metformin-based regimens, the distribution of concordant versus discordant glycemic-adiposity response phenotypes and the frequency of high multidimensional response, and to describe how these phenotypes were distributed across treatment classes. Methods: This secondary patient-level analysis used a real-world cohort from a Bucharest tertiary diabetes center. Of 291 screened adults, 166 completed 12-month follow-up without treatment change. Regimens included metformin alone or combined with sulfonylurea, DPP-4 inhibitor, SGLT2 inhibitor, GLP-1 receptor agonist, or insulin. Responses were defined as HbA1c reduction ≥ 0.5 percentage points, fat mass reduction ≥ 5%, hs-CRP reduction ≥ 20%, and handgrip strength decline ≤ 1 kg. Primary phenotypes and a 0-4 multidimensional score were derived. Results: Mean age was 59.7 ± 13.0 years, BMI 36.3 ± 4.4 kg/m², and baseline HbA1c 8.9 ± 1.0%. Glycemic-adiposity, glycemic-only, adiposity-only, and low/non-response phenotypes occurred in 49 (29.5%), 88 (53.0%), 9 (5.4%), and 20 (12.0%) patients, respectively. A high multidimensional response occurred in 78 (47.0%), and complete response in 25 (15.1%). GLP-1 receptor agonists showed the most consistent concordant glycemic-adiposity response, whereas insulin was predominantly glycemic-only. Conclusions: In obesity-associated T2DM, therapeutic response was heterogeneous and frequently discordant across metabolic domains. Patient-level responder phenotyping may complement conventional HbA1c-centered evaluation by identifying response quality across glycemic, adiposity, inflammatory, and functional dimensions. This approach may support more individualized diagnostic interpretation and treatment optimization in real-world diabetes care.