Aim: Pooled efficacy studies suggest that glycaemic responses to dipeptidyl-peptidase 4 inhibitors in type 2 diabetes are greatest in Asians, who may also respond better to alpha-glucosidase inhibitors. We assessed the glycaemic impact of sitagliptin by race in the Trial Evaluating Cardiovascular Outcomes with Sitagliptin (TECOS), and whether this was enhanced in Asians with concomitant acarbose therapy. Materials and methods: TECOS enrolled 14 671 patients with type 2 diabetes, cardiovascular disease and HbA1c of 48–64 mmol/mol (6.5%-8.0%), and randomized them, double-blind, to sitagliptin or placebo. There were 3265 patients (22.3%) from Asian countries. Background glucose-lowering therapies were unaltered for the first 4 months post randomization unless clinically essential, facilitating comparison of sitagliptin-associated effects in self-identified East Asian, Other (South) Asian, White Caucasian, Hispanic, Black and Indigenous groups. Results: Median baseline HbA1c by race was 54 to 57 mmol/mol (7.1%-7.4%). Mean 4-month reduction in placebo-adjusted HbA1c was greatest in East Asians (−6.6 mmol/mol [−0.60%] vs ≤6.0 mmol/mol [≤0.55%] in other groups), with significantly greater reduction vs the 2 largest groups (White Caucasians, Other Asians; P <.0001) after adjustment for covariates. After the first 4 months, East and Other Asians were more likely to initiate additional oral therapy (metformin and/or sulfonylureas) than insulin vs White Caucasians (P <.0001). Acarbose use increased in the Asian patients, but no glycaemic interaction with allocated study medication was observed (adjusted P =.12). Conclusions: The greatest initial reduction in HbA1c with sitagliptin in the TECOS population was in East Asians. No enhanced glycaemic effect was seen when sitagliptin was given with acarbose.