Population pharmacokinetics and exposure-response of trastuzumab emtansine in advanced breast cancer previously treated with ≥2 HER2-targeted regimens

Abstract

Aims: We conducted population pharmacokinetic (PopPK) and exposure-response analyses for trastuzumab emtansine (T-DM1), to assess the need for T-DM1 dose optimization in patients with low exposure by using TH3RESA [A Study of Trastuzumab Emtansine in Comparison With Treatment of Physician's Choice in Patients With human epidermal growth factor receptor 2 (HER2)-positive Breast Cancer Who Have Received at Least Two Prior Regimens of HER2-directed Therapy] study data (NCT01419197). The randomized phase III TH3RESA study investigated T-DM1 vs. treatment of physician's choice (TPC) in patients with heavily pretreated HER2-positive advanced breast cancer.

Methods: We compared a historical T-DM1 PopPK model with T-DM1 pharmacokinetics in TH3RESA and performed exposure-response analyses using model-predicted cycle 1 maximum concentration (Cmax ), cycle 1 minimum concentration (Cmin ) and area under the concentration-time curve at steady state (AUCss ). Kaplan-Meier analyses [overall survival (OS), progression-free survival (PFS)] and logistic regression [overall response rate (ORR), safety] were stratified by T-DM1 exposure metrics. Survival hazard ratios (HRs) in the lowest exposure quartile (Q1) of cycle 1 Cmin were compared with matched TPC-treated patients.

Results: T-DM1 concentrations in TH3RESA were described well by the historical PopPK model. Patients with higher cycle 1 Cmin and AUCss exhibited numerically longer median OS and PFS and higher ORR than patients with lower exposure. Exposure-response relationships were less evident for cycle 1 Cmax . No relationship between exposure and safety was identified. HRs for the comparison of T-DM1-treated patients in the Q1 subgroup with matched TPC-treated patients were 0.96 [95% confidence interval (CI) 0.63, 1.47] for OS and 0.92 (95% CI 0.64, 1.32) for PFS.

Conclusions: Exposure-response relationships for efficacy were inconsistent across exposure metrics. HRs for survival in patients in the lowest T-DM1 exposure quartile vs. matched TPC-treated patients suggest that, compared with TCP, the approved T-DM1 dose is unlikely to be detrimental to patients with low exposure.

Keywords: exposure-response (E-R); metastatic breast cancer; population pharmacokinetics (PopPK); trastuzumab emtansine (T-DM1).

© 2017 The British Pharmacological Society.

Figures

Figure 1

Goodness‐of‐fit plots for the PopPK analysis of TH3RESA. (A) DV vs. PRED; (B) DV vs. IPRED; (C) CWRES vs. PRED; and (D) CWRES vs. time. CWRES, conditional‐weighted residual; DV, observed; IPRED, individual population‐predicted T‐DM1 serum concentrations; PopPK, population pharmacokinetic; PRED, population‐predicted T‐DM1 serum concentrations; TH3RESA, A Study of Trastuzumab Emtansine in Comparison With Treatment of Physician's Choice in Patients With human epidermal growth factor receptor 2 (HER2)‐positive Breast Cancer Who Have Received at Least Two Prior Regimens of HER2‐directed Therapy

Figure 2

Kaplan–Meier curves for (A) OS and (B) PFS in the pharmacokinetic‐evaluable population of T‐DM1‐treated patients (n = 351) by quartile of model‐predicted T‐DM1 exposure metric and the efficacy‐evaluable population of TPC‐treated patients (n = 185). AUCss, area under the concentration–time curve at steady state; Cmax, maximum concentration; Cmin, minimum concentration; OS, overall survival; PFS, progression‐free survival; Q, quartile; T‐DM1, trastuzumab emtansine; TPC, treatment of physician's choice

Figure 3

HRs for each T‐DM1 exposure quartile vs. TPC before and after adjusting for covariates in the final Cox proportional hazard models of OS and PFS. (A) model‐predicted Cmin at cycle 1. (B) AUCss. AUCss, area under the concentration–time curve at steady state; Cmin, minimum concentration; HR, hazard ratio; OS, overall survival; PFS, progression‐free survival; Q, quartile; TPC, treatment of physician's choice

Figure 4

Logistic regression analysis of the relationship between ORR and (A) Cmin at cycle 1 and (B) AUCss. Vertical ticks at each individual patient's exposure value represent whether the individual did (at 1) or did not (at 0) have an objective response. Solid points represent the mean ORR for each T‐DM1 exposure quartile (red) or the TPC arm (blue). Error bars represent ±2 standard errors of the mean. Centred curves and shaded areas represent predicted values and 95% CIs of model‐predicted response probability after covariate adjustment (ECOG). AUCss, area under the concentration–time curve at steady state; CI, confidence interval; Cmin, minimum concentration; ECOG, Eastern Cooperative Oncology Group performance status; ORR, objective response rate; OS, overall survival; PFS, progression‐free survival; T‐DM1, trastuzumab emtansine; TPC, treatment of physician's choice