C-POST (cemiplimab after surgery and radiotherapy for high-risk skin squamous cell carcinoma)

Almost everyone with a cutaneous squamous cell carcinoma is cured by an operation. The group C-POST was built for is the small minority who are not: those with extracapsular nodal spread, three or more involved nodes, bone invasion, in-transit metastases or perineural invasion, who have already had surgery and radiotherapy and whose remaining risk is high. Until 2025 they were simply watched.

Four hundred and fifteen patients were randomised to cemiplimab (209) or placebo (206), given at 350 mg every three weeks for twelve weeks and then 700 mg every six weeks for up to a further 36 weeks, 48 weeks in all. At a median follow-up of 24 months, disease-free survival strongly favoured cemiplimab: 24 events against 65, a hazard ratio for recurrence or death of 0.32 (95 per cent confidence interval 0.20 to 0.51, p<0.001). Estimated 24-month disease-free survival was 87.1 per cent (80.3 to 91.6) against 64.1 per cent (55.9 to 71.1). Both components moved: locoregional recurrence in 9 against 40 patients (hazard ratio 0.20, 0.09 to 0.40) and distant recurrence in 10 against 26 (hazard ratio 0.35, 0.17 to 0.72).

The cost is real but modest against that effect. Grade 3 or higher adverse events occurred in 23.9 per cent of the cemiplimab group against 14.2 per cent of the placebo group, and 9.8 per cent stopped because of adverse events against 1.5 per cent.

Overall survival was not the primary endpoint and is not yet mature. The comparison worth holding beside this trial is TROG 05.01, whose radiotherapy-alone arm gave 83 per cent freedom from locoregional relapse at five years in a partly overlapping population: adjuvant immunotherapy is being added on top of local treatment that already works well, which is why the absolute gain, 23 percentage points at two years, is larger than most adjuvant trials in any cancer.