ESMO Congress 2026

The class standing of nivolumab as a PD-1 inhibitor was established by a series of published phase 3 studies: in previously treated advanced nonsquamous NSCLC, median overall survival was 12.2 months versus 9.4 months with docetaxel (HR 0.73) ^[1]; in the squamous population, 9.2 versus 6.0 months (HR 0.59) ^[2]; in previously untreated advanced melanoma, nivolumab plus ipilimumab achieved a median PFS of 11.5 months versus 2.9 months with ipilimumab alone ^[3], and in the BRAF wild-type population one-year overall survival was 72.9% versus 42.1% with dacarbazine ^[5]; in previously treated advanced renal-cell carcinoma it was 25.0 versus 19.6 months ^[4], and in platinum-pretreated squamous-cell carcinoma of the head and neck 7.5 versus 5.1 months ^[6].

At this ESMO26 there are two major readouts for this drug, one positive and one negative. The long-term NICHE-2 follow-up reported by Chalabi: 111 patients with locally advanced dMMR colon cancer received only 6 weeks of neoadjuvant nivolumab/ipilimumab followed by surgery, with zero recurrences at 5 years of follow-up and 100% overall survival, and many of these were high-risk stage III disease. The final results of the phase III ENERGIZE (CA017-078) reported by Necchi: in muscle-invasive bladder cancer (N=653), perioperative nivolumab plus gemcitabine/cisplatin versus chemotherapy alone showed a pCR of 39.1% versus 29.6% but the statistical threshold was not met, EFS HR 0.94 (p=0.67), OS HR 0.88 descriptive only, and grade ≥3 adverse events 73% versus 56%; ctDNA clearance was strongly prognostic for EFS and OS, but neither baseline ctDNA status nor clearance identified patients who derived greater benefit from nivolumab.

There are also several other data points involving the drug at this meeting: in a phase 3 study of HBI-8000 plus nivolumab versus nivolumab alone in advanced melanoma, mPFS was 11.7 versus 7.4 months (HR 0.73) and ORR 41.4% versus 32.3% (p=0.0569), but grade 3/4 treatment-related AEs were 54.8% versus 17.9% with 4 treatment-related deaths in the HBI-8000 arm; in the dMMR mCRPC subgroup of INSPIRE treated with nivolumab plus ipilimumab, ORR was 75%, PSA50 86%, and median PFS 32.7 months, while grade ≥3 treatment-related AEs occurred in 48% of the overall INSPIRE population; in EVOLUTION (ANZUP2001), 177Lu-PSMA-617 plus nivolumab/ipilimumab versus the radioligand alone showed 12-month rPFS of 47% versus 23% and median rPFS 12.0 versus 8.5 months, at the cost of grade 3-4 AEs of 75% versus 29% and treatment-related myocarditis of 7% in the combination arm, with trial recruitment stopped early. Postow will also present a poster on October 24 on real-world outcomes of neoadjuvant nivolumab plus relatlimab or nivolumab plus ipilimumab in resectable stage III melanoma, with no data disclosed yet.

Overall, NICHE-2 extends the benefit of short-course dual checkpoint neoadjuvant therapy in dMMR colon cancer from pathological response to 5-year zero recurrence and 100% overall survival, whereas CA017-078 shows that adding nivolumab to perioperative chemotherapy in unselected muscle-invasive bladder cancer failed to improve EFS, with the numerical pCR gain not translating into survival benefit and toxicity clearly increased. The limitations are equally clear: NICHE-2 is single-arm with no control group; the OS of CA017-078 is descriptive; the ORR difference for the HBI-8000 combination did not reach significance and OS is not disclosed; and the EVOLUTION and other Lu-PSMA combination data are cross-trial descriptive comparisons, not evidence that any combination is superior to VISION.