Various combinations of ipilimumab with other immune modulating, antiangiogenic or chemotherapeutic, or targeted agents have been reported or are ongoing. Guiding principles for combination treatment designs could be to use drugs that lead to immunogenic cell death [49–51]. Since radiotherapy can also induce immunogenic cell death, the reported observations of abscopal antitumor effects after radiotherapy and ipilimumab have led to a number of clinical studies to further investigate this phenomenon [52, 53].

Three studies regarding the combination of chemotherapy with ipilimumab in melanoma patients have been published thus far:

The second phase II trial regards the combination of pegylated IFN and ipilimumab [60]. In this study in 31 patients, ipilimumab was administered at 3 mg/kg for four doses along with concurrent peginterferon alfa-2b at 1–3 mcg/kg weekly for up to 156 weeks. Among 27 evaluable patients, there were 4 CRs, 8 PRs, 4 SDs, and 13 PDs. Peginterferon alfa-2b added to ipilimumab resulted in a response rate of 40% and was associated with a grade 3 toxicities rate of 45%. With the arrival of anti-PD1, this combination will probably be replaced by exploring additive effects with anti-PD1.

Combinations of BRAF inhibitors and MEK inhibitors with immune checkpoint inhibitors such as anti-CTLA are theoretically attractive, but have in practice proven to be not so simple to develop.

The PD1 protein is another immune checkpoint expressed in many tumor-infiltrating lymphocytes in response to inflammation. It has two ligands, PD-L1 (B7-H1) and PD-L2 (B7-DC). The engagement of PD1 on the lymphocyte surface by PD-L1 on melanoma cells and PDL1 expressing dendritic cells and other components of the tumor infiltrate delivers inhibitory signals downregulating T-cell function [15]. This neutralization of the executive phase of T cells at the tumor site is very powerful, and avoiding this by the use of anti-PD-1 and anti-PDL-1 antibodies has been remarkably successful, both in terms of response rates (30–45% in melanoma and 20–30% in various other tumor types) and in terms of a very favorable toxicity profile in comparison to anti-CTLA4 antibodies (15–17). Immune-related adverse events (irAEs) are rare and less severe than with anti-CTLA4, especially cumbersome events such as colitis and hypophysitis. In only a few years, phase I trial results launched phase III trials leading to rapid approval of nivolumab and pembrolizumab for advanced melanoma [21, 22, 68–75]. Both pembrolizumab and nivolumab have been reported to induce response rates around 30–45% in advanced melanoma patients, even in patients that previously failed to respond to ipilimumab. Responses tend to be very durable, up to 2–3 years. PDL-1 expression in the tumor is a good biomarker for response to monotherapy with either agent, but even in PDL-1-negative patients, it is more effective than chemotherapy or ipilimumab [76]. Anti-PD1 has been shown in all melanoma patients to be superior to chemotherapy or ipilimumab and to be effective in ipilimumab failures and in patients that have failed targeted therapies. Anti-PD1 is the drug of choice in first line for all metastatic melanoma patients, with the exception of bulky rapidly progressive BRAF-mutant melanoma patients [77]. Now that anti-PD1 is positioned in first line for most melanoma patients, it is an interesting observation that ipilimumab administered after progression on anti-PD1 therapy seems to be more effective than anti-PD1 therapy following ipilimumab failure [78]. The role of combination therapy with anti-PD1 and anti-CTLA4 agents will be discussed later.

Regarding side effects and quality of life outcomes, it is a clear observation from the randomized trials that both nivolumab and pembrolizumab have fewer and less severe side effects and irAEs than ipilimumab and that QoL is superior with these drugs than with various chemotherapies or with ipilimumab [79–83]. In particular lower colities, hepatitis, and hypophysitis rates are associated with anti-PD1 treatment than with ipilimumab. Yet a wide variety of irAEs can be seen at low frequencies. Excellent overview articles deal with diagnosis and treatment of these toxicities [84, 85].

Overall the incredible impact of anti-PD1 and anti-PDL1 monoclonal antibodies lies in its broad transversal impact in oncology with now activity demonstrated against a wide panel of neoplasms other than melanoma, including lung cancer, renal cell cancer, bladder cancer, stomach cancer, head and neck cancer, Merkel cell cancer, Hodgkin lymphoma, and many other to come [86].

In July 2017 a press release by BMS declared that an interim analysis of the randomized trial Checkmate-238 had demonstrated superiority regarding the primary endpoint (RFS) in this trial. Patient who had received nivolumab (3mg/Kg every 2 weeks) had demonstrated a significant prolongation of RFS compared to the patients who had received ipilimumab (10mg/Kg). [86]

The rational to combine these two checkpoint inhibitors is that they have different mechanisms of action, with anti-CTLA4 mainly acting at the central level in the lymph node compartment by perpetuating and/or restoring the induction and proliferation of activated T cells and with anti-PD1 mainly acting at the peripheral level at the tumor site by preventing the neutralization of cytotoxic T cells by PDL1 expressing tumor cells and PDL2 expressing plasmacytoid dendritic cells in the tumor infiltrate. The first report in 2013 already indicated that the combination is associated with clearly increased response rates up to 50–60%, with an increased CR rate of around 20% and a clear increase in near-complete responses [87].

A randomized phase II trial, nivolumab plus ipilimumab versus ipilimumab (2:1), was conducted, and the results were recently reported in 2016 [23]. One hundred forty-two patients were randomized, assigning 95 patients to nivolumab plus ipilimumab and 47 to ipilimumab alone. At a median follow-up of 24.5 months, the 2-year overall survival was 63.8% for the combination therapy and 53.6 for those assigned to ipilimumab alone. Treatment-related grade 3–4 adverse events were reported in 51 (54%) of 94 patients who received nivolumab plus ipilimumab compared with 9 (20%) of 46 patients who received ipilimumab alone. Serious grade 3–4 treatment-related adverse events were reported in 36% of patients who received nivolumab plus ipilimumab compared with 9% of patients who received ipilimumab alone [23]. A randomized phase III trial in 945 treatment-naïve patients, ipilimumab versus nivolumab versus ipi + nivo (1:1:1), was also launched, and the first results on the PFS end point were reported in 2015 [24]. Median PFS was superior for ipi + nivo (11.5 months) versus nivolumab alone (6.9 months) versus ipilimumab alone (2.9 months) (p < 0.001). PD-L1 expression (≥5%) played a very important role in this trial in the sense that in patients with tumors positive for PD-L1, the median PFS was the same (14 months) for patients treated with nivolumab alone or with the combination. But in patients with PD-L1-negative tumors, PFS was longer for ipi + nivo (11.2 months) vs nivolumab alone (5.3 months). Combination therapy was the most toxic with grade 3 or 4 occurred in 55% versus 16.3% for nivolumab versus 27.3% for ipilimumab. Overall results are expected to be reported in 2017. It seems highly unlikely that the combination ipi + nivo will be superior to nivolumab alone in patients with PD-L1 positive tumors. It is not excluded that in patients with PD-L1-negative tumors, the combination will outperform nivolumab monotherapy.