A Simulation Study to Compare the Treatment Effect of Tamoxifen by CYP2D6 Genotypes and Third-Generation Aromatase Inhibitors
Some prospective, randomized clinical trials, including ATAC and BIG 1–98, demonstrated superior treatment effect of third-generation aromatase inhibitors (AIs) versus tamoxifen in postoperative therapy for patients with breast cancer. In retrospective genotyping analyses of the 2 studies using tumor samples, no difference in the treatment effect of tamoxifen was observed by CYP2D6 genotypes. However, those analyses did not consider loss of heterozygosity that could have occurred when genotyping using tumor tissue. The present simulation study aimed to comparatively evaluate the treatment effect of tamoxifen versus AIs of anastrozole and letrozole by CYP2D6 genotypes. A meta-analysis was conducted to estimate disease-free survival (DFS) hazard ratios of CYP2D6 genotypes representing extensive metabolizers (EMs), HRW/W,TAM, versus intermediate metabolizers (IMs)/poor metabolizers (PMs), HRV/W,TAM, using previous study results in which genotypes were determined using blood samples. Based on known allele frequencies, the CYP2D6 genotype distribution of participants in ATAC and BIG 1–98 trials were simulated. Subsequently, DFS HRs of AIs versus tamoxifen by CYP2D6 genotypes (HRAI/TAM,W for EMs, HRAI/TAM,V for IMs/PMs) were estimated via regression analyses using NONMEM, based on the simulated genotype distributions, HRV/W,TAM, and HRs, of AIs versus tamoxifen (HRAI/TAM) reported in the ATAC and BIG 1–98 trials. Median HRAI/TAM,V (95% prediction interval [PI]) was 0.43 (0.23–0.79) and 0.40 (0.22–0.73) for the ATAC and BIG 1–98 trials, respectively. However, the corresponding HRAI/TAM,W values were 0.97 (0.84–1.11) and 0.91 (0.77–1.08), respectively. These results suggest that in patients with the CYP2D6 genotype representing EMs, the treatment effect of tamoxifen is comparable to that of AIs.