Computational Treatment Simulations to Assess the Need for Personalized Tamoxifen Dosing in Breast Cancer Patients of Different Biogeographical Groups

Abstract

Tamoxifen is used worldwide to treat estrogen receptor-positive breast cancer. It is extensively metabolized, and minimum steady-state concentrations of its metabolite endoxifen (C_{SS,min ENDX}) >5.97 ng/mL have been associated with favorable outcome. Endoxifen formation is mediated by the enzyme CYP2D6, and impaired CYP2D6 function has been associated with lower C_SS,min _ENDX. In the Women's Healthy Eating and Living (WHEL) study proposing the target concentration, 20% of patients showed subtarget C_{SS,min ENDX} at tamoxifen standard dosing. CYP2D6 allele frequencies vary largely between populations, and as 87% of the patients in the WHEL study were White, little is known about the risk for subtarget C_{SS,min ENDX} in other populations. Applying pharmacokinetic simulations, this study investigated the risk for subtarget C_{SS,min ENDX} at tamoxifen standard dosing and the need for dose individualization in nine different biogeographical groups with distinct CYP2D6 allele frequencies. The high variability in CYP2D6 allele frequencies amongst the biogeographical groups resulted in an up to three-fold difference in the percentages of patients with subtarget C_{SS,min ENDX}. Based on their CYP2D6 allele frequencies, East Asian breast cancer patients were identified as the population for which personalized, model-informed precision dosing would be most beneficial (28% of patients with subtarget C_{SS,min ENDX}).

Keywords: CYP2D6; breast cancer; genotype; individualized dosing; model-informed precision dosing; personalized dosing; polymorphism; tamoxifen.

Figure 1

Populations included in the respective biogeographical groups previously defined [28]. As in [28], African American/Afro-Caribbean and Latino groups are not shown in this figure, as the map gives an overview of the genetic ancestor-based borders of geographic groups, and this concept cannot be applied to the two admixed groups. Figure recreated from [28] using https://mapchart.net/ (accessed on 12 January 2021).

Figure 2

CYP2D6 genotype-predicted phenotype frequencies in the nine virtual populations (n = 10,000 patients) representing the biogeographical groups defined in [28]. gNM, gIM, and gPM: genotype-predicted normal, intermediate, and poor metabolizers, respectively.

Figure 3

Frequencies of breast cancer patients not attaining C_{SS min, ENDX} upon six months of simulated tamoxifen standard dosing in the nine virtual populations (n = 10,000 patients, each) representing the biogeographical groups defined in [28]. Red dashed line: reported frequency of patients with subtarget C_{SS min, ENDX} in the WHEL study (20%) [9]. C_{SS,min ENDX}: Endoxifen minimum concentrations at steady state; WHEL: Women’s healthy eating and living.