ESMO Congress 2026

Durvalumab, an anti-PD-L1 antibody, has established phase 3 evidence across multiple tumor types. In unresectable stage III non-small-cell lung cancer, the PACIFIC trial randomized 713 patients without progression after concurrent chemoradiotherapy 2:1 to consolidation durvalumab (10 mg/kg every 2 weeks for up to 12 months) or placebo; one primary endpoint, progression-free survival by blinded independent central review, was significantly improved, and a later analysis confirmed overall survival benefit ^[1][2]. In first-line extensive-stage small-cell lung cancer, CASPIAN enrolled treatment-naive patients at 209 sites across 23 countries and compared, 1:1:1, durvalumab plus platinum-etoposide, a triple regimen adding tremelimumab, and chemotherapy alone ^[4]. In first-line unresectable hepatocellular carcinoma, HIMALAYA enrolled 1171 patients: tremelimumab 300 mg as a single dose plus durvalumab 1500 mg every 4 weeks (STRIDE) achieved median overall survival of 16.43 months versus 13.77 months with sorafenib, and durvalumab monotherapy 16.56 months ^[6]; a review also lists this combination as a regimen with superior overall survival versus sorafenib ^[5]. At the class level, fatal toxic effects of immune checkpoint inhibitors are rare but real ^[3].

The actual durvalumab report at this meeting is the POP-DURVA study: in stage I TNBC, a single dose of durvalumab without chemotherapy was evaluated for clinical and immunological effects (n=71). Key numbers: pCR 1.7%, objective response 11.6%, tumor shrinkage 41.3%, and conversion to an inflamed phenotype in 33% of initially immune-negative tumors. The on-site interpretation is that a single dose of immunotherapy induced substantial immune activation, but pathological response was very limited; the study's main value lies in identifying potential biomarkers of immunotherapy sensitivity rather than demonstrating the efficacy of a chemo-free approach, and results after only 15 days cannot establish the efficacy of longer-term monotherapy. A separate on-site discussion also involved durvalumab: in perioperative MIBC trials, durvalumab worked while pembrolizumab and nivolumab failed, and no one knows why — an open question with no new data attached.

In terms of evidence advancement, POP-DURVA sits at a different level from the prior phase 3 trials: PACIFIC, CASPIAN and HIMALAYA answered whether consolidation after chemoradiotherapy or combinations with chemotherapy or a CTLA-4 antibody improve survival ^[1][2][4][6], whereas POP-DURVA probes the magnitude of immune activation itself under a single-dose, chemo-free, 15-day window design — the 41.3% tumor shrinkage and 33% phenotype conversion show that immunity was indeed activated, while the 1.7% pCR shows that this activation barely translates into pathological response in the short term, and that gap is precisely the entry point for biomarker research. The limitations are equally clear: single-dose administration, 15-day observation, a small sample of n=71, and restriction to the early-stage stage I TNBC population mean efficacy conclusions cannot be extrapolated, nor can the study explain the divergent outcomes of PD-L1 versus PD-1 inhibitors in perioperative bladder cancer; the feasibility of a chemo-free strategy remains to be confirmed.