Immune-checkpoint inhibition (ICI) is an approved treatment approach for both unresectable and completely resected advanced-stage melanoma; however, patients with completely resected stage IV disease have been under-represented in clinical trials to date. In particular, the efficacy of dual PD-1 and CTLA4 ICI in this subgroup is unclear. Now, data from the phase II IMMUNED trial address these knowledge gaps.

In IMMUNED, 167 patients with stage IV cutaneous melanoma and no evidence of disease after surgery and/or radiotherapy were randomly assigned (1:1:1) to receive adjuvant nivolumab plus ipilimumab (nivo+ipi), nivolumab alone (nivo) — both at the approved dosages — or placebo. At a median follow-up duration of 28.4 months, the median relapse-free survival (RFS) duration was not reached in the nivo+ipi group (HR 0.23, 97.5% CI 0.12–0.45; P < 0.0001) and was 12.4 months in the nivo group (HR 0.56, 97.5% CI 0.33–0.94; P = 0.011), versus 6.4 months with placebo. At 1 year and 2 years, RFS was 75% and 70%, respectively, in the nivo+ipi group, with a plateau on the Kaplan–Meyer curve out to 42 months. These rates were 52% and 42% with nivo alone, and 32% and 14% with placebo. The benefits of nivo+ipi and nivo alone over placebo were similar across all pre-specified subgroups; tumour PD-L1 expression <5%, BRAF mutation or CNS involvement did not seem to diminish efficacy.

“Our findings clearly demonstrate greatly improved disease control with ICI than with observation alone. Furthermore, our study provides strong evidence that dual ICI enables much better disease control than PD-1 ICI as monotherapy,” principal investigator Dirk Schadendorf summarizes. Indeed, in an exploratory analysis of RFS with nivo+ipi versus nivo, the HR was 0.40 (97.5% CI 0.20–0.79). “This effect was seen despite the fact that 50% of patients in the nivo+ipi group only received ≤2 doses of the combination (owing to toxicities). Importantly, this raises questions regarding how intensely and for how long we need to treat patients, particularly in the adjuvant setting,” adds Schadendorf. The investigators plan to examine these aspects, as well as predictive biomarkers, to optimize patient management.

The rates of treatment-related grade 3–4 adverse events were 71% with nivo+ipi versus 27% with nivo, most commonly increased levels of serum liver enzymes, amylase and/or lipase, and autoimmune disorders (including hepatitis); the treatment discontinuation rates were 62% versus 13%. However, no deaths were deemed treatment related.

our study provides strong evidence that dual ICI enables much better disease control than PD-1 ICI as monotherapy

These results support the use of single-agent ICI in the adjuvant treatment of resected stage IV melanoma. The data might lead to a shift towards dual ICI in this setting.