Toripalimab plus chemotherapy as perioperative treatment for resectable stage II or III non-small cell lung cancer (NSCLC): Final analysis of the phase 3 Neotorch study

doi:10.1016/j.annonc.2026.10.094

Neotorch

Paciente / Población Intervención / Exposición Comparación Desenlace

In patients with resectable stage II or III non-small cell lung cancer, perioperative toripalimab plus chemotherapy improved event-free survival, major pathological response and overall survival compared with placebo plus chemotherapy.

N

501pacientes

Diseño

Phase 3, 1:1 randomised, placebo-controlled trial in adults with resectable stage II or III NSCLC (non-squamous patients with EGFR or ALK alterations excluded) comparing toripalimab with placebo, each given with platinum-based chemotherapy for 3 cycles before and 1 cycle after surgery and then as monotherapy for up to 13 cycles (NCT04158440).

Desenlace

Co-primary endpoints: investigator-assessed event-free survival (INV-EFS) and major pathological response rate by blinded independent pathologic review (BIPR), in the stage III and stage II or III populations.

ResultadoStage II or III population. Median INV-EFS not reached (95% CI 51.3-NE) with toripalimab vs 20.2 months (95% CI 14.3-24.8) with placebo (HR 0.52; 95% CI 0.391-0.677; P < 0.0001). BIPR major pathological response rate 47.2% (95% CI 40.9-53.6) vs 9.2% (95% CI 5.9-13.4) (P < 0.0001). Median OS not reached (95% CI NE-NE) vs not reached (95% CI 58.6-NE) (HR 0.66; 95% CI 0.486-0.894; P = 0.0072). BIPR pathologic complete response 25.6% (95% CI 20.3-31.5) vs 1.2% (95% CI 0.2-3.5) (P < 0.0001). Most common adverse events: anaemia 66.4% vs 64.9%, neutropenia 58.0% vs 53.4%, leukopenia 56.8% vs 50.6%, decreased appetite 30.0% vs 15.1%.

Li Z, Wu L, Wang W, et al. Annals of Oncology. 2026. 10.1016/j.annonc.2026.10.094

Discusión y crítica

This placebo-controlled phase 3 trial reports a benefit in event-free survival and pathological response that extends to overall survival in the stage II or III population, at a median follow-up of 48.2 months. Median OS was not reached in either arm, so the size of the survival benefit is not yet clear, and the abstract gives no separate results for the 97 stage II patients or for biomarker subgroups. The adverse event table lists any-grade events only, and the abstract leaves out grade 3 or higher toxicity, rates of surgery and completion, and subsequent therapy. Because the design pairs a neoadjuvant phase with an adjuvant phase, it cannot separate how much each contributes. The study was funded by Shanghai Junshi Biosciences.