Han Liu (1), Ana-Marija Grisic (2), Sreenath Madathil Krishnan (1), Siv Jönsson (1), Lena Friberg (1), Pascal Girard (3), Karthik Venkatakrishnan (4), Yulia Vugmeyster (4), Akash Khandelwal* (2), Mats Karlsson* (1)
(1) Department of Pharmacy, Uppsala University, Sweden (2) Merck Healthcare KGaA, Darmstadt, Germany (3) Merck Institute of Pharmacometrics, Lausanne, Switzerland, an affiliate of Merck KGaA (4) EMD Serono Research & Development Institute, Inc., Billerica, MA, USA, an affiliate of Merck KGaA *shared senior authorship
Objectives:
Bintrafusp alfa (BA) is a first-in-class bifunctional fusion protein targeting transforming growth factor-β and programmed death-ligand 1. This analysis aimed to (1) characterize relationships between pharmacokinetics (PK), tumor size (TS) and clinical endpoints in second-line (2L) non-small cell lung cancer (NSCLC) patients receiving BA, by extending the previously developed multistate modeling framework for pharmacometric analyses [1-2]; and (2) learn about the dose-response relationships.
Methods:
Data originated from the 2L NSCLC expansion cohort of a phase I trial (NCT02517398) with 80 patients randomized 1:1 to receive 500 mg or 1200 mg Q2W [3]. Patient demographics (n=5), lab values (n=3), disease characteristics (n=9) and premedication (n=3) at baseline were evaluated as covariates. The plasma concentration and the sum of longest diameters (SLD) of target lesions informed the PK-TS model [4]. The tumor response category per RECIST 1.1 and survival informed the multistate model with six states:
S1 – Stable disease
S2 – Response (partial or complete)
S3 – Progression (progressive disease)
S4 – Unknown (lost to follow-up for TS and response category evaluation due to treatment discontinuation without preceding progression, but followed-up for survival)
S5 – Dropout (lost to follow-up for survival but not right-censored nor died during the follow-up)
S6 – Death.
All patients were defined to have stable disease (S1) at treatment initiation, and could stay stable (S1), respond (S2) or progress (S3). Patients with response (S2) could progress (S3). State of patients who discontinued treatment without progression (S1/S2) would become unknown (S4). From the current state (S1/S2/S3/S4), patients could drop out from survival follow-up (S5) or die (S6). Each transition hazard λij from state i to j was estimated with parametric hazard functions (exponential or Weibull).
Predicted SLD was tested as a covariate on clearance to explore the relation between disease status and elimination capacity as described for other immune-oncology agents [5,6]. The effect of dose, clearance and covariates were explored on both TS growth and shrinkage. The predictors prospectively tested on transition hazards were dose, clearance, tumor metrics (change in SLD from baseline, nadir and previous SLD; tumor growth and shrinkage rate), covariates, time to progression and time in response.
Virtual patients (n=5000) were resampled from the analysis population (n=80), stratified on dose. Clinical trials (n=500) were simulated by sampling parameters from the posterior distribution obtained with Sampling Importance Resampling [7].
Results:
The final models were 2-compartment linear PK model, and TS model with zero-order tumor growth rate, first-order tumor shrinkage rate, and resistance constant. No association between clearance and tumor dynamics was found, and dose and covariates had no significant effect on TS profiles. The hazards of progression (λ13) and death upon progression (λ36) decreased with time, while the others were constant. The only significant predictor was metastases at baseline on λ13 (HR=4.18, Yes vs. No).
Although not statistically significant, dose (1200 vs. 500 mg Q2W) was retained on all transitions as a predictor to assess its impact on patients’ benefit. The parameter estimates suggested a trend of higher probability of response, as well as lower probability of progression and death upon progression, in patients receiving 1200 vs. 500 mg Q2W, illustrated by simulations:
The median (95% CI) of simulated overall response rate (ORR) was 20% (10%, 32%) and 27% (15%, 42%), PFS was 1.4 (1.3, 1.5) and 2.6 (1.3, 3.7) months, and OS was 11.0 (7.4, 16.5) and 13.7 (9.5, 17.8) months for 500 mg and 1200 mg, respectively. This is in line with the observed ORR of 15% and 28%, median PFS of 1.4 and 2.7, and median OS of 9.5 and 12.2 months for the two doses. The average HRs (1200 vs. 500 mg Q2W) for PFS and OS over 500 simulated trials were 0.83 and 0.86, respectively.
Conclusions:
The developed model adequately described the PK, SLD, competing events and jointly the PFS and OS. The effect of dose on transitions was evaluated independently and combined for the comparison of clinical efficacy outcomes between doses via simulation. Overall tendency of better efficacy was observed for the 1200 mg vs. the 500 mg Q2W regimen, supporting the RP2D selection reported previously [8].
References:
[1] Krishnan SM, Friberg LE, Bruno R, Beyer U, Jin JY, Karlsson MO. Multistate model for pharmacometric analyses of overall survival in HER2-negative breast cancer patients treated with docetaxel. CPT Pharmacometrics Syst Pharmacol. (2021)
[2] Krishnan SM, Friberg LE, Karlsson MO. Multistate model for pharmacometric analyses of overall survival in anticancer treatments. PAGE 29 (2021) Abstr 9641 [www.page-meeting.org/?abstract=9641]
[3] Paz-Ares LG, Kim TM, Vicente Baz D, Felip E, Lee DH, Lee KH, et al. Results from a second-line (2L) NSCLC cohort treated with M7824 (MSB0011359C), a bifunctional fusion protein targeting TGF-β and PD-L1. J Clin Oncol. (2018)
[4] Claret L, Girard P, Hoff PM, Van Cutsem E, Zuideveld KP, Jorga K, et al. Model-based prediction of phase III overall survival in colorectal cancer on the basis of phase II tumor dynamics. J Clin Oncol. (2009)
[5] Wilkins JJ, Brockhaus B, Dai H, Vugmeyster Y, White JT, Brar S, et al. Time-Varying Clearance and Impact of Disease State on the Pharmacokinetics of Avelumab in Merkel Cell Carcinoma and Urothelial Carcinoma. CPT Pharmacometrics Syst Pharmacol. (2019)
[6] Liu C, Yu J, Li H, Liu J, Xu Y, Song P, et al. Association of time-varying clearance of nivolumab with disease dynamics and its implications on exposure response analysis. Clin Pharmacol Ther. (2017)
[7] Dosne AG, Bergstrand M, Karlsson MO. An automated sampling importance resampling procedure for estimating parameter uncertainty. J Pharmacokinet Pharmacodyn. (2017)
[8] Vugmeyster Y, Wilkins J, Koenig A, El Bawab S, Dussault I, Ojalvo LS, De Banerjee S, Klopp-Schulze L, Khandelwal A. Selection of the recommended phase 2 dose for bintrafusp alfa, a bifunctional fusion protein targeting TGF-β and PD-L1. Clin Pharmacol Ther. 108:566-574 (2020).
Reference: PAGE 30 (2022) Abstr 10021 [www.page-meeting.org/?abstract=10021]
Poster: Drug/Disease Modelling - Oncology