Assil Merlaud1,2, Marion Kerioui1,3, Jérémie Guedj1, René Bruno4, Alyse Lin5, Steve Dang5, Benjamin Wu5, Phyllis Chan5, Jin Y. Jin5, Pascal Chanu2,6, Julie Bertrand1
1Université Paris Cité, IAME, Inserm, F-75018, 2Institut Roche, 3MRC Biostatistics Unit, University of Cambridge, 4R&B Pharmacometrics, 5Clinical Pharmacology, 6Clinical Pharmacology, Genentech-Roche
Context: In oncology clinical trials, the primary clinical endpoint is often overall survival (OS), requiring prolonged study durations. Tumor size measurements, summarized as the Sum of Longest Diameters (SLD) of target lesions according to RECIST¹, are collected over time as a short-term marker of disease progression and treatment response. Nevertheless, this marker fails to account for the heterogeneity in lesion dynamics, and the impact of their organ locations. Indeed, in metastatic urothelial carcinoma (mUC)² and metastatic colorectal cancer (mCRC) trials³, lesion location has been shown to differentially influence OS, with liver progression as the most important prognostic factor. However, a formal comparison of organ-specific impact across cancers has yet to be conducted. Objectives: – To develop a meta-regression4 of multilevel non-linear joint models of lesions trajectories and survival5 6 – To investigate whether the association between organ-level tumor dynamics and OS is consistent across cancer types on data from seven atezolizumab (ATZ) trials in four indications Methods: The association was investigated in non-small cell lung cancer (NSCLC)7 8 ? ¹°, small cell lung cancer (SCLC)¹¹, triple negative breast cancer (TNBC) ¹², and mUC ¹³. A Stein model¹4 was fitted to describe the lesion dynamics associated with two levels of random effects (patient and lesion level), while accounting for organ-specific tumor dynamics patterns through fixed covariates effects, categorized as lung, lymph nodes, adrenal gland, liver and all remaining locations as “other”4 5. We assumed an impact of the sum of the current tumor size of the lesions per organ (TS) on the survival baseline risk function. For each clinical trial, the longitudinal and survival parameters were simultaneously estimated. Finally, the inter-trial variability was quantified using a random-effect meta-analysis on the estimates of the joint model parameters focusing on the organ-specific TS-OS association parameters. All analyses were conducted using a Bayesian framework via Stan¹5. Results: Survival data from 2,310 patients enrolled in 7 Phase II/III clinical trials were analyzed, corresponding to 5,573 lesions distributed across 3,959 organs and measured longitudinally, resulting in 31,329 observations. The median baseline tumor size of lung lesions was larger in SCLC patients and NSCLC compared to those with TNBC and mUC. In contrast, baseline tumor sizes of liver and lymph node lesions remained consistent across cancer types. The analysis of the tumor dynamics shrinkage parameters revealed no organ-specific trends, across studies and cancers. However, clinical trials evaluating second-line treatments in NSCLC consistently demonstrated slower tumor shrinkage. In patients with NSCLC receiving ATZ as second-line treatment, the median shrinkage half-life was 12 months for liver lesions, compared to 1 month in patients treated with a first-line regimen of ATZ, carboplatin, and nab-paclitaxel. Within each study, an increase in tumor size was consistently associated with a significant reduction in OS, regardless of lesion location. The liver TS-OS association parameters had the highest impact on OS, in comparison to the other locations for 6 studies out of 7. A lack of power in that 7th study, might explain why its TS-OS association parameter for the liver was lower. Through the meta-regression, a low between-study standard deviation of 0.0037mm-1 was estimated, associated with a pooled estimate of the liver TS-OS association parameters of 0.0095mm-1, meaning that a 10mm increase of a liver lesion leads to a 9% increase of the instantaneous risk of death. A greater between-study heterogeneity was observed for the rest of the organ-specific TS-OS association parameter estimates. Conclusion: This work reinforces the notion that liver tumor size is associated with a poor prognosis of OS, in several studies, cancer types, or treatment regimens. This could potentially help develop metrics that do not depend on the cancer type, the line of therapy or the treatment modality (single agent or combination) to potentially predict the treatment effect on OS. However, some variability persists for the other organ-specific TS-OS and could be better quantified by modeling lesion dynamics across all trials simultaneously¹6. Integrating all sources of variability into a single model is a complex challenge that would require methodological advancements and innovative solutions.
(1) Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer Oxf Engl 1990. 2009;45(2):228-247. doi:10.1016/j.ejca.2008.10.026 (2) Kerioui M, Desmée S, Mercier F, et al. Assessing the impact of organ-specific lesion dynamics on survival in patients with recurrent urothelial carcinoma treated with atezolizumab or chemotherapy. ESMO Open. 2022;7(1):100346. doi:10.1016/j.esmoop.2021.100346 (3) Chen C, Feng YS, Wang Z, Gupta M, Xu XS, Yan X. Organ-specific tumor dynamics predict survival of patients with metastatic colorectal cancer. Eur J Cancer Oxf Engl 1990. 2024;207:114147. doi:10.1016/j.ejca.2024.114147 (4) Sudell M, Tudur Smith C, Gueyffier F, Kolamunnage-Dona R. Investigation of 2-stage meta-analysis methods for joint longitudinal and time-to-event data through simulation and real data application. Stat Med. 2018;37(8):1227-1244. doi:10.1002/sim.7585 (5) Kerioui M, Beaulieu M, Desmée S, et al. Nonlinear multilevel joint model for individual lesion kinetics and survival to characterize intra-individual heterogeneity in patients with advanced cancer. Biometrics. 2023;79(4):3752-3763. doi:10.1111/biom.13912 (6) Kerioui M, Bertrand J, Desmée S, et al. Assessing the Increased Variability in Individual Lesion Kinetics During Immunotherapy: Does It Exist, and Does It Matter? JCO Precis Oncol. 2023;7:e2200368. doi:10.1200/PO.22.00368 (7) Rittmeyer A, Barlesi F, Waterkamp D, et al. Atezolizumab versus docetaxel in patients with previously treated non-small-cell lung cancer (OAK): a phase 3, open-label, multicentre randomised controlled trial. Lancet Lond Engl. 2017;389(10066):255-265. doi:10.1016/S0140-6736(16)32517-X (8) Fehrenbacher L, Spira A, Ballinger M, et al. Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised controlled trial. The Lancet. 2016;387(10030):1837-1846. doi:10.1016/S0140-6736(16)00587-0 (9) Jotte R, Cappuzzo F, Vynnychenko I, et al. Atezolizumab in Combination With Carboplatin and Nab-Paclitaxel in Advanced Squamous NSCLC (IMpower131): Results From a Randomized Phase III Trial. J Thorac Oncol Off Publ Int Assoc Study Lung Cancer. 2020;15(8):1351-1360. doi:10.1016/j.jtho.2020.03.028 (10) Nishio M, Barlesi F, West H, et al. Atezolizumab Plus Chemotherapy for First-Line Treatment of Nonsquamous NSCLC: Results From the Randomized Phase 3 IMpower132 Trial. J Thorac Oncol. 2021;16(4):653-664. doi:10.1016/j.jtho.2020.11.025 (11) Horn L, Mansfield AS, Szczesna A, et al. First-Line Atezolizumab plus Chemotherapy in Extensive-Stage Small-Cell Lung Cancer. N Engl J Med. 2018;379(23):2220-2229. doi:10.1056/NEJMoa1809064 (12) Schmid P, Adams S, Rugo HS, et al. Atezolizumab and Nab-Paclitaxel in Advanced Triple-Negative Breast Cancer. N Engl J Med. 2018;379(22):2108-2121. doi:10.1056/NEJMoa1809615 (13) Powles T, Durán I, van der Heijden MS, et al. Atezolizumab versus chemotherapy in patients with platinum-treated locally advanced or metastatic urothelial carcinoma (IMvigor211): a multicentre, open-label, phase 3 randomised controlled trial. The Lancet. 2018;391(10122):748-757. doi:10.1016/S0140-6736(17)33297-X (14) Stein WD, Gulley JL, Schlom J, et al. Tumor regression and growth rates determined in five intramural NCI prostate cancer trials. The growth rate constant as an indicator of therapeutic efficacy. Clin Cancer Res Off J Am Assoc Cancer Res. 2011;17(4):907. doi:10.1158/1078-0432.CCR-10-1762 (15) Carpenter B, Gelman A, Hoffman MD, et al. Stan: A Probabilistic Programming Language. J Stat Softw. 2017;76:1. doi:10.18637/jss.v076.i01 (16) Sudell M, Kolamunnage-Dona R, Gueyffier F, Tudur Smith C. Investigation of one-stage meta-analysis methods for joint longitudinal and time-to-event data through simulation and real data application. Stat Med. 2019;38(2):247-268. doi:10.1002/sim.7961
Reference: PAGE 33 (2025) Abstr 11571 [www.page-meeting.org/?abstract=11571]
Poster: Drug/Disease Modelling - Oncology