ESMO Congress 2026

FTD/TPI plus bevacizumab belongs to the "cytotoxic + anti-VEGF monoclonal antibody" combination strategy, and regimens built on a bevacizumab backbone already have formally published phase III randomized controlled evidence in other tumor types: IMagyn050 (GOG 3015/ENGOT-OV39) was a multicenter, placebo-controlled, double-blind phase III trial evaluating the addition of bevacizumab 15 mg/kg and atezolizumab to paclitaxel plus carboplatin in newly diagnosed FIGO stage III or IV ovarian cancer, with investigator-assessed PFS and OS as co-primary endpoints ^[1]. This background shows that the bevacizumab-plus-chemotherapy platform has a mature evaluation paradigm in large phase III trials, whereas the current report concerns the dosing schedule within the same combination in later-line mCRC.

The study reported at the meeting was ESMO26 LBA31 PRABITAS: a pragmatic phase III non-inferiority trial comparing biweekly FTD/TPI + bevacizumab versus conventional FTD/TPI + bevacizumab in refractory metastatic colorectal cancer (mCRC). OS was 11.3 vs 10.3 months, HR 0.93, 95% CI 0.80–1.08, with the upper bound of the confidence interval below the non-inferiority margin of 1.25; PFS was 3.9 vs 3.9 months, HR 0.93, 95% CI 0.81–1.06. For safety, grade ≥3 neutropenia was 13% vs 39%, febrile neutropenia 1% vs 5%, and grade ≥2 anorexia 15% vs 26%. The conclusion was that the biweekly schedule achieved non-inferior OS with less toxicity, and the results were simultaneously published in The Lancet.

Relative to existing knowledge, PRABITAS advances the toxicity-reduction step of "keeping the drugs, changing only the schedule." Whereas IMagyn050 layered the new agent atezolizumab onto the bevacizumab-plus-chemotherapy platform for phase III testing ^[1], PRABITAS did not alter drug composition; instead, while maintaining non-inferior OS and PFS, it reduced grade ≥3 neutropenia from 39% to 13%, febrile neutropenia from 5% to 1%, and grade ≥2 anorexia from 26% to 15%, directly answering the practical question of whether the same efficacy can be delivered more safely in later-line mCRC, with direct implications for outpatient chemotherapy and tolerability management.

As for limitations, this was a non-inferiority design in refractory mCRC, with OS of about 11.3 months and PFS of only 3.9 months, an efficacy level consistent with palliative later-line treatment; details such as sample size, stratification, and follow-up duration were not disclosed in the on-site posts and should be confirmed in the simultaneously published Lancet paper.