Advances in immunotherapy for melanoma

Anti-PD-1/PD-L1 plus anti-CTLA-4

Preclinical murine studies verified the hypothesis that, in light of their distinct mechanisms, combining CTLA-4 and PD-1 blockade could augment antitumor activity beyond that of either strategy alone. Combination therapy increased the degree of tumor response and was associated with greater numbers of effector T cells and less Tregs in the TME in murine models involving syngeneic implants of either colon cancers or melanoma [50]. A phase I trial of nivolumab plus ipilimumab in patients with advanced melanoma demonstrated an ORR of 43 % and 1- and 2-year OS rates of 85 % and 79 %, respectively [51, 52]. The rate of grade 3–4 treatment-related adverse events was substantially higher (60 %) compared to the rates previously seen with anti-CTLA-4 or anti-PD-1 monotherapy. However, these events were similar to those seen with ipilimumab monotherapy and were also manageable with early institution of high-dose corticosteroids or other immune modulatory agents.

Subsequently, two randomized studies (Checkmate-069 and Checkmate-067) were conducted to compare combined immunotherapy with nivolumab plus ipilimumab to immune checkpoint inhibitor monotherapy. The Checkmate-069, a double-blind phase II trial, randomized patients to ipilimumab 3 mg/kg plus nivolumab 1 mg/kg or placebo every 3 weeks, followed by nivolumab 3 mg/kg or placebo every 2 weeks until disease progression or toxicity requiring study withdrawal [53]. In patients with BRAF-wildtype tumors, the ORR was 61 % in the group that received nivolumab plus ipilimumab, compared to 11 % in the ipilimumab plus placebo group. Median PFS was 4.4 months in the ipilimumab monotherapy group, whereas median PFS was not reached for the combination group at the time of analysis. There were 16 patients (22 %) with complete responses in the combination group, and none in the ipilimumab-monotherapy group. A subset of patients with BRAF mutant tumors were observed to have similar ORR and PFS to those in the larger study, suggesting that tumor BRAF status has no effect on response to checkpoint inhibitor therapy. This favorable data clearly established that combination therapy produced superior antitumor efficacy to ipilimumab in patients with BRAF-wildtype melanoma and led to the FDA approval of the combination for this patient population in October 2015.

As mentioned earlier, the Checkmate-067 trial was a three-arm, double blind, phase III trial that randomized patients to nivolumab 3 mg/kg every 2 weeks or nivolumab 1 mg/kg every 3 weeks plus ipilimumab 3 mg/kg every 3 weeks for four doses, followed by nivolumab 3 mg/kg every 2 weeks or ipilimumab 3 mg/kg every 3 weeks for four doses [37]. While the study was not preplanned for a statistical comparison between nivolumab plus ipilimumab versus nivolumab monotherapy, the data provides insight into how these two strategies compare to each other. The ORR was 57.8 % in patients receiving the combination therapy compared to 43.7 % in patients receiving nivolumab monotherapy. Response was independent of tumor BRAF mutational status. At the time of publication, OS data had not yet matured; however, overall tumor shrinkage was greater (51.9 % vs. 34.5 %) and median PFS was longer in those patients receiving the combination compared to nivolumab monotherapy (11.5 months vs. 6.5 months; HR, 0.74; 95 % confidence interval, 0.60–0.92).

The results from the Checkmate-069 and -067 studies establish that the combination produces impressive antitumor activity and suggests that all of the antitumor effects of immunotherapy are not subsumed in the activity of single agent PD-1 blockade. However, the combination of nivolumab?+?ipilimumab also produces a clear increase in severe treatment-related adverse events. In Checkmate-069, the nivolumab plus ipilimumab group had a grade 3–5 adverse event rate of 54 % compared to a rate of 24 % observed in the ipilimumab-alone group [53]. In Checkmate-067, grade 3–4 adverse events were seen at a rate of 55 % in the nivolumab plus ipilimumab group, compared to 16 % in the nivolumab group and 27 % in the ipilimumab group [37]. While there were three reported deaths in the combination therapy group that were attributable to checkpoint inhibitor therapy in the phase II trial [53], there were none in the phase III trial. Similar to checkpoint inhibitor monotherapy, timely recognition of irAEs and treatment with immunomodulators can control these side effects in most patients receiving the combination. More importantly, stopping treatment does not preclude derivation of benefit from treatment. While 36 % of patients had treatment discontinuation on Checkmate-069, 67 % of these patients had an objective response that continued on past discontinuation of therapy [37, 53].

In light of the high toxicity profile of nivolumab plus ipilimumab combination despite its clinical activity, alternative combination strategies are now being explored, including a randomized phase II sequencing trial of nivolumab followed by ipilimumab versus ipilimumab followed by nivolumab in patients with advanced melanoma (Checkmate-064) [54]; the cumulative ORRs at week 25 were 47.7 % and 22.6 %, respectively, suggesting higher clinical activity in patients who receive nivolumab first. Unfortunately, the cumulative rates of grade 3–5 treatment-related adverse events remained high with both sequencing approaches (50 % and 43 %, respectively). Combination of pembrolizumab with a lower dose of ipilimumab (1 mg/kg) is also currently being studied in advanced melanoma patients enrolled in the KEYNOTE-029 trial. Preliminary data showed antitumor activity and perhaps less toxicity [55]. Dose expansion of this combination in patients with melanoma is ongoing with results pending. In a comparable approach, the anti-PD-L1 mAb durvalumab is being combined with the CTLA-4 mAb tremelimumab in a phase I trial (NCT02537418).