(W-123) Population pharmacokinetic analysis of Dato-DXd, a TROP2-targeting ADC, in NSCLC and HR-positive, HER2-negative breast cancer patients

Zoey Tang, PhD – Associate Director, AstraZeneca; KyoungSoo Lim, MD, PhD. – Director, AstraZeneca; Yu Jiang, PhD – Associate Director, AstraZeneca; David Dai, PhD – Associate Director, AstraZeneca; Haitao Yang, PhD – Director, AstraZeneca; Raju Cheerla, BPharm, MSc – Principal SAS Programmer, AstraZeneca; Srinivas Bachina, PhD – Director, AstraZeneca; Cecil Chen, PhD – Head of CPQP Clinical Immunogenicity, AstraZeneca; Yuzhuo Pan, PhD – Director, Daiichi Sankyo; Ying Hong, PhD – Senior Director, Daiichi Sankyo; Song Ren, PhD – Senior Director, AstraZeneca; Alex Phipps, PhD – Senior Director, AstraZeneca; Diansong Zhou, PhD – Senior Director, AstraZeneca

Objectives: Dato-DXd, a TROP2-targeting antibody-drug conjugate, has demonstrated clinically meaningful efficacy and a manageable safety profile in multiple tumour types1. A population pharmacokinetic (PopPK) model describing PK profiles of Dato-DXd and the released drug (DXd) was developed based on data from the TROPION-PanTumor01, TROPION-Lung05, and TROPION-Lung01 studies, including 644 NSCLC patients and 85 breast cancer patients2. Data from patients in TROPION-Breast01 was added to the previously developed PopPK model to examine the impact of intrinsic and extrinsic covariates on Dato-DXd and DXd PK, and to evaluate if Dato-DXd dose adjustment is needed for HR-positive, HER2-negative breast cancer patients.

Methods: The previously-developed model served as a starting point for the model refinement. From the full covariate model, backward elimination was conducted to find a statistically parsimonious model. After backward elimination, prespecified covariates of interest (e.g., tumor type) were tested by including them in a step-wise addition (forward selection). The adequacy of the final model was evaluated by prediction-corrected visual prediction check (pcVPC) and non-parametric bootstrap method. The impact of covariates of interest were investigated in post-hoc analyses

Results: PK data from 1081 patients, including 437 breast cancer patients (40% of the total patients), were included in the PopPK analysis dataset. Consistent with prior results, Dato DXd PK was well-characterized by a two compartmental distribution model with parallel Dato DXd linear clearance (CLlinDatoDXd) and Dato DXd nonlinear Michaelis Menten clearance (CLnonlinDatoDXd) from the central compartment. DXd PK was well-characterized by a one compartment model with first order elimination kinetics from the central compartment. The release of DXd from intact Dato-DXd was equal to the total (linear + nonlinear) elimination rate of Dato DXd. The payload:intact ratio (PIR) decreased with time both within the dosing cycle and between cycles (cycle 1 vs later cycles). The evaluated covariates of clinical interest were not associated with a clinically significant impact on Dato-DXd or DXd PK. At 6 mg/kg, the difference in Dato-DXd AUC at the steady state (cycle 3 AUC, or AUC3) between lung cancer and breast cancer patients was 1.51%. DXd AUC3 of breast cancer patients was 1.74% higher than lung cancer patients. Neither difference was considered to be clinically relevant.

Conclusions: The current PopPK analysis suggested no Dato-DXd dose adjustment is needed for HR-positive, HER2-negative breast cancer patients, further justifying the Dato-DXd 6 mg/kg Q3W dose in this patient population.

Citations: 1Lu, Yasong, et al. CPT: Pharmacometrics & Systems Pharmacology 13.1 (2024): 23-28 2PAGE 2024