Abstract 287P
Background
OS is the most robust endpoint in randomized clinical trials (RCTs) in advanced BTC. PFS is increasingly used as a primary endpoint in RCTs but has not been validated as a surrogate endpoint.
Methods
We applied a two-stage meta-analytic approach, which requires evaluating association both at the trial- and patient-level. For the first condition, we performed a systemic review and searched Pubmed/MEDLINE, Embase, Cochrane library, clinicaltrials.gov and conference proceedings from database inception to October 2023 for phase II-III RCTs in BTC testing systemic agents in the first- or second-line setting. We used a weighted linear regression to measure the correlation of OS and PFS based on trial size. For the second condition, we analyzed patient-level data from two RCTs and three real-world datasets and estimated correlation via the iterative multiple imputation method. The protocol is registered with PROSPERO, CRD42023398279.
Results
For condition 1, we included 41 studies, involving 88 treatment arms and 7817 patients. The median follow-up was 10.85 months. 29/12 (70.7%/29.3%) were phase II/III RCTs, respectively; 27 (65.9%) were first-line and 2 (4.9%) allowed for cross-over. The coefficient of determination (R2) of the model was 0.71, and further improved to 0.77 in a pre-planned analysis excluding trials with a cross-over design. Additional sensitivity analysis and leave-one-out cross-validation further confirmed these results. For condition 2, we analyzed four separate cohorts comprising 1994 patients. We pooled patient data from the NIFTY and FIReFOX trials (n=271, median follow-up 33 months). The correlation ρimi was 0.73 (95% CI 0.67-0.79). This was confirmed in a real-world dataset compromising 322 second-line patients, showing a ρimi of 0.81 (95% CI 0.78-0.83). In the first-line setting, we analyzed a real-world dataset of patients treated with first-line chemotherapy (n=773, median follow-up 32 months) and cisplatin-gemcitabine-durvalumab (n=628, median follow-up 8.4 months), showing a ρimi of 0.83 (95% CI 0.8-0.85) and 0.85 (95% CI 0.79-0.89), respectively.
Conclusions
PFS is a valid surrogate endpoint for OS in advanced BTC according to the ReSEEM and BSES criteria.
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
F. Castet: Financial Interests, Personal, Invited Speaker: Eisai, Roche. L. Rimassa: Financial Interests, Personal, Advisory Board, Consulting and advisory role: AbbVie, AstraZeneca, Basilea, Bayer, Elevar Therapeutics, Exelixis, Genenta, Hengrui, Incyte, Ipsen, IQVIA, Jazz Pharmaceuticals, MSD, Nerviano Medical Sciences, Roche, Servier, Taiho Oncology; Financial Interests, Personal, Invited Speaker, Lecture fees: AstraZeneca, Bayer, BMS, Incyte, Ipsen, Roche, Servier; Financial Interests, Personal, Other, Travel expenses: AstraZeneca; Financial Interests, Personal, Advisory Board: Zymeworks; Financial Interests, Institutional, Invited Speaker: AstraZeneca, Exelixis, Incyte, Ipsen, Nerviano Medical Sciences, Roche, Servier, Agios, Eisai, FibroGen, Lilly, MSD, Roche, Servier; Financial Interests, Institutional, Invited Speaker, National (Italian) coordinating PI: AstraZeneca, BeiGene, Zymeworks; Financial Interests, Institutional, Funding: Ipsen; Financial Interests, Institutional, Invited Speaker, European PI: AstraZeneca; Non-Financial Interests, Leadership Role, Treasurer: ILCA; Non-Financial Interests, Leadership Role, Co-chair: EORTC GITCG HB/NET Task Force; Non-Financial Interests, Other, Special Expert Clinical Trials Europe: NCI HB Task Force. C. Yoo: Financial Interests, Personal, Invited Speaker: Bayer, Celgene, Eisai, Ipsen, Servier, Roche, Novartis; Financial Interests, Personal, Advisory Board: AstraZeneca; Financial Interests, Personal, Research Grant: AstraZeneca, Bayer, Servier; Financial Interests, Institutional, Research Grant: Genentech. T. Macarulla Mercade: Financial Interests, Personal, Advisory Board: Ability Pharmaceutical, SL, AstraZeneca, Basilea Pharma, Batxer, BioLineRX Ltd., Celgene SLU, Eisai, Ipsen Pharma, Incyte; Financial Interests, Personal, Other, Direct research fund: Servier, Merck, Sharp & Dohme, Novocure, QED Therapeutics Inc., Roche, Sanofi-Aventis, Zymeworks; Financial Interests, Personal, Invited Speaker: Lilly, Janssen; Financial Interests, Institutional, Research Grant: Amc Medical Research, Armo Biosciences, Basilea, Biokeralty Research Institute, Merck Sharp & Dohme, Oncomed Pharmaceuticals, QED Therapeutics, VCN Biosciences, AbbVie Farmaceútica, Ability Pharmaceuticals, Agios, Amgen, Aslan, AstraZeneca, Bayer, BeiGene, Biolinerx, Blueprint Medicines, Boston Biomedical, Bristol Myers Squibb (BMS), Cantargia, Celgene, Eisai, Erytech Pharma, F. Hoffmann-La Roche, FibroGen, Genentech, Hallozyme, Immunomedics, Incyte, Ipsen, Lab. Menarini, Lilly, Loxo Oncology, MedImmune, Merimarck, Millenim, Nelum, Novartis, Novocure, Pfizer, Pharmacyclics, Roche, Zymeworks; Non-Financial Interests, Member: American Society of Clinical Oncology - ASCO, “Sociedad Española de Oncología Médica” – SEOM, Sociedad Europea de Oncología Médica - ESMO; Other, Editorial Board: GI Annals og¡f Oncology. All other authors have declared no conflicts of interest.