Genotype-derived interpretive categories of drug response in 908 pharmacogenomic reports from Chinese patients with type 2 diabetes: A retrospective single-centre study

Abstract

Abstract Aims To characterize the pharmacogenomic landscape in a Chinese Type 2 Diabetes Mellitus (T2DM) cohort and develop a systematic framework to translate raw genotype data into predicted clinical phenotypes. Methods In this retrospective study of 908 Chinese patients, we used a novel 'Interpretation Rulebook' to derive eight predicted phenotypes from seven SNPs. These classifications are genotype-based, not measured clinical outcomes. A weighted polygenic risk score (PRS) for T2DM was constructed and evaluated for its ability to discriminate the genotype-derived risk classification. Results Allele frequencies were consistent with East Asian populations. Predefined genetic variants were the primary drivers of their corresponding phenotype classifications, for instance, IRS1 rs1801278 for insulin resistance risk (Adjusted OR = 6450.06) and SLCO1B1 rs4149056 for predicted repaglinide efficacy (OR = 0.00009). The T2DM PRS demonstrated excellent discriminatory power for the genotype-derived risk classification (AUC = 0.990). Conclusions This study establishes a transparent and validated framework for standardizing real-world pharmacogenomic data. The high-performance PRS confirms its internal logic, providing a robust blueprint for clinical decision support and implementing pre-emptive pharmacogenomic testing. While acknowledging the need for prospective validation against measured outcomes, this work addresses the translational gap in interpreting heterogeneous genetic data.

The authors' abstract, as published at the source. PLoS ONE, 2026 · DOI ↗