A model-based meta-analysis of immune-related adverse events during immune checkpoint inhibitors treatment for NSCLC
Renwei Zhang, Daming Kong, Rong Chen, Yuchen Guo, Weizhe Jian, Mengyi Han, Tianyan Zhou
Beijing Key Laboratory of Molecular Pharmaceutic and New Drug Delivery System, Department of Pharmaceutics, School of Pharmaceutical Sciences, Peking University, Beijing, China
Introduction: Immune checkpoint inhibitors (ICIs) have become a landmark therapy over recent years benefitting from their remarkable clinical efficacy in non-small cell lung cancer (NSCLC)[1]. However, ICIs can cause immune-related adverse events (irAEs) that may lead to treatment discontinuation even life threat[2].
Objectives: To identify the influential factors of irAE rates in NSCLC patients treated with ICIs using model-base meta-analysis (MBMA).
Methods: A systematic literature search was conducted for not only randomized clinical trials (RCTs) but also dose escalation trials, single-arm trials, trials with nonrandomized design and real-world studies (RWSs) reporting at least one of 7 clinical regularly-used ICIs in NSCLC patients (nivolumab, pembrolizumab, atezolizumab, durvalumab, avelumab, ipilimumab and tremelimumab) on Pubmed, Embase and Cochrane Library up to April 30, 2021. Trial-level irAE rates were meta-analytically pooled using a random-effects model. The published population pharmacokinetic models were used to simulate the average steady-state plasma concentrations (Cav) at various dosing regimens for each drug[3-10]. In order to combine irAE data from different ICIs acting on the same target receptor, ICI exposure was normalized by dividing the simulated Cav by drug concentrations at 50% inhibition (IC50)[11-15]. A logit-transformed meta-regression model including 3 types of ICIs (Programmed cell death-1, PD-1; programmed cell death ligand-1, PD-L1; cytotoxic T lymphocyte-associated antigen-4, CTLA-4) was applied to characterize the exposure-dependence of irAE rates, and was subsequently used as the base model to identify covariates. Last, using the developed model, we simulated the probability of irAEs considering a wide range of covariate combinations to provide clinicians with a visually prediction of the incidence of irAEs under diverse treatment scenarios or dosing regimens.
Results: A total of 120 articles across 81 clinical studies and 129 ICI treatment cohorts involving 19322 NSCLC patients were included in this study. For ICI monotherapy, CTLA-4 inhibitors were associated with the highest incidence of irAEs (43.96% and 12.56% for any grade and grade ≥3 irAEs, respectively), followed by PD-1 inhibitors (21.15% and 3.68% for any grade and grade ≥3 irAEs, respectively), while PD-L1 inhibitors were associated with the lowest irAE rates (16.62% and 2.23% for any grade and grade ≥3 irAEs, respectively). Only CTLA-4 inhibitors exhibited a statistically significant exposure dependence of irAE rates when used alone or combined with PD-1/PD-L1 inhibitors (with coefficients of 0.0013 and 0.0016 for any grade and grade ≥3 irAEs, respectively), whereas PD-1/PD-L1 inhibitors showed no exposure dependence. Additionally, patients receiving ICIs as first-line therapy had a higher irAE rates compared with those receiving ICIs as second-line or later therapy (odds ratio, OR=1.61 and 1.87 for any grade and grade ≥3 irAEs, respectively). Treatment combining ICIs with chemotherapy or target therapy would increase the irAE rates (OR=2.48 and 2.41 for any grade and grade ≥3 irAEs, respectively).
Conclusions: A comprehensive MBMA was conducted to quantify exposure dependence and covariate effects of irAEs in NSCLC patients treated by ICIs. This study provides quantitative evidence and a quantitative reference for the reasonable clinical application of ICIs in NSCLC patients from the perspective of irAEs.
References:
[1] Hsu ML et al. Thorac Surg Clin. 2020;30(2):187-198.
[2] Martins F et al. Nat Rev Clin Oncol. 2019;16(9):563-580.
[3] Shulgin B et al. Onco Targets Ther. 2020;9(1):1748982.
[4] Bajaj G et al. CPT Pharmacometrics Syst Pharmacol. 2017;6(1):58-66.
[5] Ahamadi M et al. CPT Pharmacometrics Syst Pharmacol. 2017;6(1):49-57.
[6] Shemesh CS et al. J Immunother Cancer. 2019;7(1):314.
[7] Wilkins JJ et al. CPT Pharmacometrics Syst Pharmacol. 2019;8(6):415-427.
[8] Ogasawara K et al. Clin Pharmacokinet. 2020;59(2):217-227.
[9] Feng Y et al. Br J Clin Pharmacol. 2014;78(1):106-117.
[10] Wang E et al. J Clin Pharmacol. 2014;54(10):1108-1116.
[11] de Sousa Linhares A et al. Sci Rep. 2019;9(1):11472.
[12] Wang C et al. Cancer Immunol Res. 2014;2(9):846-856.
[13] https://clinicaltrials.gov/show/NCT04434560
[14] Goldman JW et al. J Clin Oncol 2017; 35(15 Supplement 1):9093.
[15] Tremelimumab. Drugs R D. 2010;10(2):123-132.