Abstract 1449P
Background
PRO-based symptom scores are routinely employed in time to deterioration (TTD) analyses. In contrast to PFS, the recurrent nature of PRO symptom deterioration may result in endpoints with “transient” terminal event times. The TTD framework forces a single-event solution onto a recurrent event problem. Recurrent event survival models remedy this limitation. Thus, we present a joint survival model (JM) linking recurrent PRO-based deterioration to PFS assessed by investigator (progression or death).
Methods
RATIONALE-305 (NCT03777657) was a randomized phase 3 trial of tislelizumab (TIS) + chemotherapy (chemo) vs. placebo (PBO) + chemo as first-line treatment for patients (pts) with locally advanced, unresectable, or metastatic gastric/gastroesophageal junction (GEJ) adenocarcinoma. EORTC QLQ-C30 and QLQ-STO22 symptom endpoints were modeled. Osoba’s 10pt threshold was used to define PRO-based deterioration events from treatment cycles 2 to 38. Emphasis in abstract is placed on recurrent UGI deterioration (R UGI-D) events modeled with frailty cox models and PFS modeled via Cox proportional hazards models within a JM. The PFS model adjusted the R UGI-D model for informative missing PRO data. R UGI-D frailties predicted PFS. Pts were censored without R UGI-D events by Cycle 38 or time of disease progression or if they remained progression free to end of study.
Results
Of the 896 pts, 514 (57.4%), 254 (28.3%), 83 (9.3%), 26 (2.9%), 16 (1.8%), and 3 (0.3%) experienced zero (censored), one, two, three, four, and five R UGI-D events, respectively. Higher rates of R UGI-D events predicted increased risk of progression/death (HR 3.38, P<0.00001), irrespective of treatment. In contrast to PBO + chemo, TIS + chemo was not associated with significantly increased risk of R UGI-D events but was associated with a 27% reduction in risk of progression/death (HR 0.73, P=0.007). Similar findings were observed between recurrent deterioration in appetite, fatigue, and pain/discomfort and PFS.
Conclusions
This analysis demonstrated that recurrent PRO deterioration was strongly predictive of PFS in pts with advanced gastric/GEJ cancer.
Clinical trial identification
NCT03777657.
Editorial acknowledgement
Medical writing support, under the direction of the authors, was provided by Jason C. Allaire, PhD of Generativity Solutions Group. Editorial and submission support was provided by Envision Pharma Inc.
Legal entity responsible for the study
BeiGene, Ltd.
Funding
BeiGene, Ltd.
Disclosure
M. Moehler: Financial Interests, Personal, Speaker, Consultant, Advisor: Bayer, MSD, Merck Serono, Amgen, Taiho Pharmaceutical, Pfizer, Roche, Lilly, Servier, BeiGene, BMS, AstraZeneca, Astellas, Dragonfly, Novartis; Non-Financial Interests, Personal, Non remunerated activity: Amgen, Roche/Genentech, Merck Serono, MSD Oncology, BMS, AstraZeneca/MedImmune, Servier, Pierre Fabre, Sanofi, Falk Foundation, Transcenta, Daiichi Sankyo, Astellas, Nordic; Financial Interests, Personal, Research Grant: Amgen, Leap Therapeutics, Merck Serono, MSD. D. Serrano: Financial Interests, Personal, Research Funding: AbbVie, Alcon, BeiGene, BioMarin, Boehringer Ingelheim, Insmed, Janssen, Sanofi, Scilex, Seagen. H. Xu, T. Sheng, B. Barnes: Financial Interests, Personal, Full or part-time Employment: BeiGene. L. Li: Financial Interests, Personal, Full or part-time Employment: BeiGene; Financial Interests, Personal, Stocks/Shares: BeiGene; Financial Interests, Personal, Other, Travel: BeiGene. G. Barnes: Financial Interests, Personal, Full or part-time Employment: BeiGene; Financial Interests, Personal, Stocks/Shares: BeiGene. D. Oh: Financial Interests, Personal, Advisory Board: AstraZeneca, Novartis, Genentech/Roche, Merck Serono, Bayer, Taiho, Aslan, Halozyme, Zymeworks, BMS/Celgene, BeiGene, Basilea, Turning Point, Yuhan, Arcus Biosciences, IQVIA, MSD, LG Chem, Astellas, AbbVie, J-Pharma, Mirati Therapeutics, Eutilex, Moderna, Idience; Financial Interests, Institutional, Research Grant: AstraZeneca, Novartis, Array, Eli Lilly, Servier, BeiGene, MSD, Handok.
Resources from the same session
1505P - Exploratory biomarker analysis of the NEONAX trial for response prediction to perioperative (PO) and adjuvant (A) Gem-nabPac chemotherapy (CTX) in resectable PDAC patients (rPDAC Pts)
Presenter: Anton Lahusen
Session: Poster session 18
1506P - ctDNA testing based on deep NGS predicts treatment efficacy and prognosis in advanced pancreatic cancer patients
Presenter: Tingting You
Session: Poster session 18
1507P - Revolutionizing clinical trials: Harnessing artificial intelligence for optimized phase III studies in pancreatic cancer
Presenter: Marie-Edith Bonneterre
Session: Poster session 18
1509P - Cytokines and ductal pancreatic adenocarcinoma: Exploring their relationship with prognosis and molecular subtypes
Presenter: Laura Gutierrez Sainz
Session: Poster session 18
1510P - Exploration of circulating free bacterial DNA as a biomarker for therapy response in metastatic and locally advanced pancreatic cancer patients treated with systemic chemotherapy
Presenter: Sai Surendran
Session: Poster session 18
1511P - The influence of intratumoral bacteria and peri-operative antibiotic treatment on the outcome of resected pancreatic cancer patients treated with adjuvant gemcitabine-based chemotherapy
Presenter: Steffen Ormanns
Session: Poster session 18
1512P - The role of pancreaticoduodenectomy for octogenarians with pancreatic cancer
Presenter: Keita Sonoda
Session: Poster session 18
1513P - Impact of symptom and functional burden on survival in pancreatic ductal adenocarcinoma (PDAC)
Presenter: Omali Pitiyarachchi
Session: Poster session 18
1514P - A prospective, single-center, randomized, controlled, evaluator-blinded, feasibility clinical trial for efficacy and safety of a non-tissue destructive compact focused ultrasound system with neoadjuvant FOLFIRINOX in locally advanced or borderline resectable pancreatic cancer
Presenter: Jae Young Lee
Session: Poster session 18