ESMO Congress 2026

The evidence base for the paclitaxel backbone has long been established: in HER2-overexpressing metastatic breast cancer, paclitaxel-containing chemotherapy plus trastuzumab prolonged time to disease progression versus chemotherapy alone ^[1]; carboplatin-paclitaxel served as the standard chemotherapy comparator in advanced NSCLC ^[2][4], and adding bevacizumab raised median OS from 10.3 to 12.3 months ^[6]; nab-paclitaxel plus gemcitabine improved OS in a phase 3 trial in metastatic pancreatic cancer ^[3]; weekly carboplatin-paclitaxel with concurrent radiotherapy is also a standard neoadjuvant regimen in esophageal cancer ^[5].

At this meeting, the paclitaxel readouts centered on two main stories. One is the Tata Memorial phase III trial (n=475) in technically unresectable oral cancer: weekly paclitaxel-carboplatin plus oral metronomic therapy (PCm) versus DCF, with 2-year OS of 38.9% vs 29.4% and non-inferior OS, along with higher response rates, greater conversion to resectability, significantly less toxicity, outpatient delivery without a central line, described as potentially more accessible and cost effective. The other is the randomized phase 2/3 COMPANION-002 (n=168, 2:1): the DLL4×VEGF-A bispecific tovecimig plus paclitaxel versus paclitaxel alone as second-line therapy for advanced biliary tract cancer, with BICR-assessed ORR of 17.1% vs 5.3% (p=0.031), median PFS 4.7 vs 2.6 months (HR 0.44, p<0.0001), median OS 8.9 vs 9.4 months, OS HR 1.05 (p=0.78), and 54% crossover; the study was simultaneously published in JCO. The randomized AIUR trial of compression with normal-sized surgical gloves to prevent paclitaxel-induced peripheral neuropathy in breast cancer was also presented.

The main-site news did not overlap with these presentations, instead adding formulation, regulatory, and pipeline context. 3SBIO's oral paclitaxel solution (柏瑞素) was approved by NMPA on 2026-09-29 for recurrent or metastatic HER2-negative breast cancer, extending from its September 2024 approval in second-line advanced gastric cancer. In the randomized 2b trial of VCN-01 plus gemcitabine and nab-paclitaxel in first-line metastatic pancreatic ductal adenocarcinoma, OS was 10.8 vs 8.6 months (HR 0.57, P=0.055) and PFS 7.0 vs 4.6 months (HR 0.55, P=0.011). Two registrations were also noted: chidamide plus benmelstobart, nab-paclitaxel and SBRT as second-line BTC therapy (n=41, planned start 2026-11-01), and amantadine to prevent paclitaxel-induced neuropathy (n=80, start 2026-12-01), the latter echoing the AIUR glove approach on the same symptom-management theme.

The advances: in oral cancer, PCm achieved non-inferior OS with a lower toxicity burden and an outpatient model, offering a more accessible path in resource-limited settings; in second-line BTC, it shows a DLL4/VEGF-A bispecific can move ORR and PFS on a paclitaxel backbone, echoing the earlier addition of bevacizumab to carboplatin-paclitaxel ^[6], though without an OS gain, and some attendees considered it not yet a new standard, potentially to be explored in the first line. The limitations are equally clear: OS in COMPANION-002 is hard to interpret given 54% crossover, with grade ≥3 hypertension at 44%; PFS and subgroup details of the oral cancer trial were not disclosed, and AIUR results remain to be confirmed.