CDK4/6 inhibitors are standard-of-care for HR+/HER2- advanced breast cancer, but grade 3/4 neutropenia occurs in up to 60% of patients. Germline polymorphisms in ABCB1, encoding the P-glycoprotein efflux transporter, have been investigated as predictors of CDK4/6 inhibitor-induced neutropenia, with conflicting results across populations. No meta-analysis has previously pooled these findings. PubMed, Scopus, and Web of Science were systematically searched. Studies reporting genotype-stratified grade 3/4 neutropenia in CDK4/6 inhibitor-treated patients were eligible. Four SNPs were analyzed: ABCB1 rs1128503, ABCB1 rs1045642, ERCC1 rs11615, and ERCC1 rs3212986. Odds ratios were pooled using Mantel-Haenszel weighting with random-effects (rs1128503, rs11615) or fixed-effect (rs1045642, rs3212986) models under dominant genetic models, stratified by ancestry. The protocol was registered in PROSPERO (CRD420261379298). Four studies (1,138 patients) were included. No SNP showed a significant overall association. However, significant ancestry-dependent heterogeneity was detected for two ABCB1 SNPs. For rs1045642, T-carrier status was significantly associated with increased neutropenia in European/Caucasian patients (pooled OR 1.62, 95% CI 1.05-2.52, p = 0.03, I2 = 0%) but decreased risk in East Asians (OR 0.31, 95% CI 0.13-0.74, p = 0.009; subgroup difference p = 0.0009). For rs1128503, a concordant European trend was observed (OR 1.87, 95% CI 0.87-4.00, p = 0.11) with reversed direction in North African/Middle Eastern patients (OR 0.33, 95% CI 0.12-0.92, p = 0.03; subgroup difference p = 0.02). For ERCC1 rs11615, the inclusion of Wang 2024 nullified the previously borderline European signal (pooled OR 0.99, 95% CI 0.42-2.33, p = 0.98, I2 = 75%), with subgroup differences no longer significant (p = 0.17). ERCC1 rs3212986 showed no association (OR 1.07, 95% CI 0.53-2.17, I2 = 0%). These preliminary findings suggest that ABCB1 rs1045642 T-carrier status may be associated with CDK4/6 inhibitor-induced neutropenia in European/Caucasian patients, with a concordant trend for rs1128503 findings. Both associations appear ancestry-dependent, although the limited number of studies and single-study ancestry subgroups preclude definitive conclusions. ERCC1 rs11615, previously borderline in a single study, was not confirmed upon independent replication. Prospective validation in adequately powered, ancestry-stratified cohorts incorporating ABCB1 haplotype analysis and pharmacokinetic sampling is warranted. identifier CRD420261379298.
山东省济南市章丘区文博路2号
齐鲁师范学院 genelibs生信实验室
山东省济南市高新区舜华路750号
大学科技园北区F座4单元2楼
电话: 0531-88819269