Summary: In a post hoc subgroup analysis of the double-blind SUSTAIN China randomised trial (total n=868; Chinese-only n=605), the glycaemic efficacy of once-weekly subcutaneous semaglutide 0.5 mg and 1.0 mg versus sitagliptin 100 mg was not significantly modified by most baseline characteristics, and attainment of HbA1c below 7.0% was independent of baseline HbA1c (p>0.05). Reductions in glucose variability and greater HOMA-β ratios favoured semaglutide (p<0.05).
PICO Summary
| Element | Detail |
|---|---|
| Population | Adults with type 2 diabetes from the SUSTAIN China phase 3a double-blind randomised trial; post hoc analysis of the total population (n=868) and the Chinese-only subset (n=605); China. |
| Intervention | Once-weekly subcutaneous semaglutide 0.5 mg and 1.0 mg. |
| Comparison | Sitagliptin 100 mg once daily. |
| Outcome | Treatment effects on HbA1c and body weight were not significantly affected by most baseline characteristics (age, sex, BMI, baseline HbA1c, diabetes duration, HOMA-β tertile) in either population. Attainment of HbA1c below 7.0% (53 mmol/mol) was not affected by baseline HbA1c (p>0.05). Less glucose variability (standard deviation of seven-point self-monitored plasma glucose and derived time-in-range) and greater HOMA-β ratios to baseline at end of treatment favoured semaglutide 0.5 mg and 1.0 mg versus sitagliptin (p<0.05). As a post hoc subgroup analysis, these comparisons are exploratory and associational; no adjusted effect estimates with 95% CIs, ARR, or NNT were reported in the abstract. |
Expert Commentary
This is a post hoc subgroup analysis of an already completed randomised trial, so its proper reading is one of consistency rather than fresh proof of efficacy. The reassuring signal is that the glycaemic advantage of once-weekly semaglutide over sitagliptin appeared stable across age, sex, body mass index, baseline HbA1c, diabetes duration, and β-cell function, which suggests the drug behaves predictably in a broad Chinese cohort. The secondary observations of reduced glucose variability and improved HOMA-β ratios are biologically coherent for an incretin agent but should be treated as hypothesis-generating, not as established endpoints. The dominant limitation is that subgroup analyses of this kind are underpowered for interaction testing, so an absence of significant effect modification is weak evidence of true uniformity and may simply reflect limited statistical power. The industry origin of the work is material: several authors are employees of the manufacturer and the trial was sponsored by it, which warrants cautious interpretation of consistently favourable framing. Can I use this with my patients? Yes, in a limited sense, for an East Asian adult with type 2 diabetes for whom semaglutide is already being considered, this analysis offers modest reassurance that baseline phenotype is unlikely to undermine the response. It does not justify expanding indications or overstating glycaemic gains. Confirmatory prespecified subgroup work would strengthen these conclusions.
References
Ji L, Lu Y, Shen Z, Hu P, Liu W, Zhang Q, Shi B. Impact of baseline characteristics on the efficacy of once-weekly subcutaneous semaglutide among participants with type 2 diabetes: A post hoc analysis of SUSTAIN China. Diabetes Obes Metab. 2024;26(11):5312-5324. doi:10.1111/dom.15888
