Population pharmacokinetic analysis of datopotamab deruxtecan (Dato-DXd), a TROP2-targeting antibody-drug conjugate
Sophie Peigné (1)*, Yuzhuo Pan (2), Sofia Friberg Hietala (1), Anna M. Mc Laughlin (1), Naoyuki Tajima (3), Deise Uema (4), Hong Zebger-Gong (5), Zoey Tang (6), Diansong Zhou (7), Tushar Garimella (2), Malaz Abutarif (2), Ying Hong (2)#
(1) Pharmetheus AB, Uppsala, Sweden (2) Quantitative Clinical Pharmacology, Daiichi Sankyo, Inc., Basking Ridge, New Jersey, USA (3) Quantitative Clinical Pharmacology, Daiichi Sankyo, Tokyo, Japan (4) Oncology Clinical Development, Daiichi Sankyo, Inc., Basking Ridge, New Jersey, USA (5) Oncology Clinical Development, Daiichi Sankyo Europe GmbH, Munich, Germany (6) Clinical Pharmacology and Quantitative Pharmacology, AstraZeneca, Gaithersburg Maryland, USA (7) Clinical Pharmacology and Quantitative Pharmacology, AstraZeneca, Boston Massachusetts, USA *Presenting author # Corresponding author
Introduction/objectives
Datopotamab deruxtecan (Dato-DXd) is a trophoblast cell surface antigen 2 (TROP2)-directed antibody-drug conjugate (ADC) in clinical development for the treatment of advanced or metastatic non-small cell lung cancer (NSCLC) and other tumor types. The released drug, MAAA-1181a (DXd), inhibits DNA topoisomerase I leading to the inhibition of cell replication and promotes apoptosis of the target tumor cells.
The aim of the present analysis was to establish a population PK model describing the exposure to both Dato-DXd and DXd in patients with NSCLC or other solid tumor malignancies, with the goal of supporting the dose optimization of Dato-DXd for future patient populations.
Methods
The population modeling analyses included 9036 Dato-DXd and 9012 DXd PK plasma samples collected from 729 subjects receiving Dato-DXd doses ranging from 0.27 to 10 mg/kg Q3W in three clinical studies (Phase 3 Study TROPION-Lung01, Phase 2 Study TROPION-Lung05, and Phase 1 Study TROPION-PanTumor01).
First, a model for Dato-DXd was established using a legacy model as the starting point [1]. This model was updated to describe DXd by using individual post hoc PK parameters from the Dato-DXd model as input.
Potential covariate-parameter relationships were evaluated using the stepwise covariate model building procedure (SCM) with adaptive scope reduction denoted SCM+ [2]. The evaluated covariates included age, body weight, albumin, sex, race, country/region, creatinine clearance, hepatic function, ECOG, tumor size, last prior line of therapy including immuno-oncology agent (LPIO), number of prior lines of therapy (NPLT), tumor type, histology, tumor membrane TROP2 protein expression, actionable genomic alterations status (AGA), anti-drug antibody status (ADA), drug product, and smoking status. The impact of covariates on model parameters and on derived secondary exposure metrics (Cmax3, maximum concentration in cycle 3; AUC3, area under the concentration-time curve in cycle 3) was illustrated using forest plots. The final models were used to simulate Dato-DXd and DXd plasma concentration-time data after multiple doses of Dato-DXd at the body weight-based dosing of 6.0 mg/kg Q3W across the relevant body weight range.
Results
The PK of Dato-DXd was best described by a two-compartment model with both linear and nonlinear elimination. Body weight was added as a mechanistic covariate on Dato-DXd linear clearance (CLlinDatoDXd) and Dato-DXd volume of distribution (VcDatoDXd and VpDatoDXd). When estimated, the exponent was close to 0.75 for CLlinDatoDXd, close to 0 for Dato-DXd intercompartmental clearance (QDatoDXd) and close to 0.5 for volumes. As VpDatoDXd did not exceed 3 L, it indicated that the distribution was restricted to the plasma compartment and that the bi-phasic behavior was linked to the interaction with TROP2 target. Thus, a model using a fixed allometric exponent of 0.75 on CLlinDatoDXd, and estimated exponents on volumes was chosen. The PK of DXd was best described by a one-compartment model with a linear clearance (CLDXd), with the release of DXd from Dato-DXd equal to the total elimination rate of Dato-DXd. The release of DXd was found to be time dependent within and between cycles, as shown previously for other ADCs [3,4]. The extreme values of the body weight distribution and albumin values lower than 19 g/L resulted in Cmax3 and AUC3 outside the bioequivalence (BE) boundary of 0.8-1.25 for both entities. Moreover, hepatic function measured by high total bilirubin and aspartate aminotransferase led to DXd exposure outside the BE boundary.
Remaining covariates had no marked effect on Dato-DXd and DXd exposures. Simulations from the final models supported a body weight – normalized dosing regimen is appropriate for Dato-DXd.
Conclusion
The population PK model adequately described PK data of Dato-DXd and DXd. The linear clearance was the major elimination pathway for Dato-DXd doses of 4 mg/kg and above, whereas nonlinear clearance was necessary for satisfactory description of PK at doses of less than 4 mg/kg. Body weight was the most influential covariate on Dato-DXd exposures. Other covariates have no clinically meaningful impact on the PK of Dato-DXd and DXd. The model-based simulation revealed the proposed dose of 6 mg/kg Q3W resulted in similar Dato-DXd and DXd exposures across the body weight groups. Dose adjustments based on tested covariates or for special populations are not recommended.
References:
[1] Lu Y, Liang S, Hong Y, Tajima N, Patel K, Li H, Wada DR, Greenberg J, Petrich A, Zebger-Gong H, Shuster D, Vaddady P. Application of the model-informed drug development paradigm to datopotamab deruxtecan dose selection for late-stage development. CPT Pharmacometrics Syst Pharmacol. 2024 Jan;13(1):23-28. doi: 10.1002/psp4.13058. Epub 2023 Nov 1. PMID: 37915242; PMCID: PMC10787203.
[2] Svensson R, Jonsson EN. Efficient and relevant stepwise covariate model building for pharmacometrics.CPT Pharmacometrics Syst Pharmacol. 2022 July;00:1-13.
[3] Li H, Han T, Hunder N, Jang G, Zhao B. Population Pharmacokinetics of Brentuximab Vedotin in Patients With CD30-Expressing Hematologic Malignancies. J Clin Pharmacol. 2017 March;57:1148-58.
[4] Li C, Chen S, Chen Y, Girish S, Kaagedal M, Lu D, et al. Impact of Physiologically Based Pharmacokinetics, Population Pharmacokinetics and Pharmacokinetics/Pharmacodynamics in the Development of Antibody-Drug Conjugates. J Clin Pharmacol. 2020 October;60:S105-19.
