RESUMO
BACKGROUND: In preliminary findings from the recurrent or metastatic cervical cancer cohort of CheckMate 358, nivolumab showed durable anti-tumour responses, and the combination of nivolumab plus ipilimumab showed promising clinical activity. Here, we report long-term outcomes from this cohort. METHODS: CheckMate 358 was a phase 1-2, open-label, multicohort trial. The metastatic cervical cancer cohort enrolled patients from 30 hospitals and cancer centres across ten countries. Female patients aged 18 years or older with a histologically confirmed diagnosis of squamous cell carcinoma of the cervix with recurrent or metastatic disease, an Eastern Cooperative Oncology Group performance status of 0 or 1, and up to two previous systemic therapies were enrolled into the nivolumab 240 mg every 2 weeks group, the randomised groups (nivolumab 3 mg/kg every 2 weeks plus ipilimumab 1 mg/kg every 6 weeks [NIVO3 plus IPI1] or nivolumab 1 mg/kg every 3 weeks plus ipilimumab 3 mg/kg every 3 weeks for four cycles then nivolumab 240 mg every 2 weeks [NIVO1 plus IPI3]), or the NIVO1 plus IPI3 expansion group. All doses were given intravenously. Patients were randomly assigned (1:1) to NIVO3 plus IPI1 or NIVO1 plus IPI3 via an interactive voice response system. Treatment continued until disease progression, unacceptable toxicity, or consent withdrawal, or for up to 24 months. The primary endpoint was investigator-assessed objective response rate. Anti-tumour activity and safety were analysed in all treated patients. This study is registered with ClinicalTrials.gov (NCT02488759) and is now completed. FINDINGS: Between October, 2015, and March, 2020, 193 patients were recruited in the recurrent or metastatic cervical cancer cohort of CheckMate 358, of whom 176 were treated. 19 patients received nivolumab monotherapy, 45 received NIVO3 plus IPI1, and 112 received NIVO1 plus IPI3 (45 in the randomised group and 67 in the expansion group). Median follow-up times were 19·9 months (IQR 8·2-44·8) with nivolumab, 12·6 months (7·8-37·1) with NIVO3 plus IPI1, and 16·7 months (7·2-27·5) with pooled NIVO1 plus IPI3. Objective response rates were 26% (95% CI 9-51; five of 19 patients) with nivolumab, 31% (18-47; 14 of 45 patients) with NIVO3 plus IPI1, 40% (26-56; 18 of 45 patients) with randomised NIVO1 plus IPI3, and 38% (29-48; 43 of 112 patients) with pooled NIVO1 plus IPI3. The most common grade 3-4 treatment-related adverse events were diarrhoea, hepatic cytolysis, hyponatraemia, pneumonitis, and syncope (one [5%] patient each; nivolumab group), diarrhoea, increased gamma-glutamyl transferase, increased lipase, and vomiting (two [4%] patients each; NIVO3 plus IPI1 group), and increased lipase (nine [8%] patients) and anaemia (seven [6%] patients; pooled NIVO1 plus IPI3 group). Serious treatment-related adverse events were reported in three (16%) patients in the nivolumab group, 12 (27%) patients in the NIVO3 plus IPI1 group, and 47 (42%) patients in the pooled NIVO1 plus IPI3 group. There was one treatment-related death due to immune-mediated colitis in the NIVO1 plus IPI3 group. INTERPRETATION: Nivolumab monotherapy and nivolumab plus ipilimumab combination therapy showed promise in the CheckMate 358 study as potential treatment options for recurrent or metastatic cervical cancer. Future randomised controlled trials of nivolumab plus ipilimumab or other dual immunotherapy regimens are warranted to confirm treatment benefit in this patient population. FUNDING: Bristol Myers Squibb and Ono Pharmaceutical.