ASCO: Immunotherapy for Advanced Melanoma

Building on the landmark studies of the immunotherapeutic agent ipilimumab just a few years back-and on tantalizing early data presented at ASCO 2012 on immunotherapy agents targeting another pathway, the programmed death 1 (PD1) pathway-ASCO 2013 saw the presentation of truly impressive data on two PD1 blockers, as well as noteworthy studies of other immunotherapeutic approaches to advanced melanoma.

There are a variety of pathways by which tumors can evade the body’s natural immune response. The first of these to be discovered, the CTLA-4 pathway, suppresses T-cell activation; blockade of this pathway is the mechanism of action of the first successful immunotherapy for melanoma, ipilimumab (Yervoy). The PD1/PD ligand (PDL) pathway contributes to functional T-cell exhaustion, and blockade of the PD1/PDL pathway can revive exhausted T cells.

Mario Sznol, MD

At ASCO 2012, phase I data on the anti-PD1 antibody nivolumab in a variety of tumor types were presented. This year, Dr. Mario Sznol of the Yale Cancer Center presented overall survival and long-term safety and response data from the melanoma patients in this study, providing a fuller picture of how nivolumab performs in the setting of advanced melanoma. The results were notable. First, there was a high rate of response: the objective response rate in the trial was 31% across all doses, and was 41% in the patients who received the highest dose (the dose that will be used in a subsequent phase III trial). Moreover, the responses were seen sooner than is typically the case with the CTLA-4 blocker ipilimumab: 45% of patients exhibited a response at the first checkpoint (8 weeks after the start of treatment). And in addition, responses appear to be durable. The overall survival rate was 62% at 1 year, 44% at 2 years, and 40% at 3 years, results that were achieved with a grade 3/4 adverse event rate of 21% and no treatment-related deaths. While follow-up has not been long enough yet to tell just how the long-term response rate with nivolumab will compare with that of ipilimumab, Dr. Sznol remarked that he thinks nivolumab may ultimately produce more durable responses. It is this prolongation of survival at 1 year, 2 years, and later that makes immunotherapy different from other approaches, noted Dr. Walter Urba, the discussant at the Clinical Science Symposium on the PD1/PDL1 target in melanoma.

Jeffrey Weber, MD, PhD

The question of whether nivolumab is effective in patients who have failed prior treatment with another immunotherapy (ie, ipilimumab) was addressed by results presented by Dr. Jeffrey Weber of the Moffitt Cancer Center. Weber and colleagues tested nivolumab, along with a multi-peptide vaccine (as an immune monitoring tool), in three cohorts of ipilimumab-naive patients and in three cohorts of ipilimumab-refractory patients. The clinical efficacy data in the ipilimumab-refractory patients was comparable to previously published data on nivolumab, and Dr. Weber noted that responses were generally of very long duration. The drug was also well tolerated, with only a few grade 3/4 adverse events. Also of note, one patient who failed nivolumab was subsequently given ipilimumab off-protocol and had a nice response, leading Dr. Weber to remark that the opposite sequence of agents also appears to be effective.

Another question regarding PD1 pathway inhibition that was addressed at ASCO this year was whether a combination immunotherapy regimen-of a PD1 blocker together with a CTLA-4 blocker-would prove more effective than blockade of either pathway alone, and whether the associated toxicities would be tolerable. There is a strong rationale for assuming that this would be the case: while CTLA-4 dampens an early stage of T-cell activation, PD1 interferes further down the line; also, it had been shown that an anti–CTLA-4 and anti-PD1 combination was potently active in mice. Results of a phase I study of nivolumab plus ipilimumab, presented by Dr. Jedd Wolchok of Memorial Sloan-Kettering, provided evidence that this combination is also effective in humans. In one cohort of the trial, patients received both drugs concurrently; in two sequential-therapy cohorts, patients who had received prior ipilimumab went on to receive nivolumab. Forty percent of all patients had an objective response, and Dr. Wolchok noted that most responding patients had deep tumor regression. The responses were generally more rapid than is seen with ipilimumab alone, with many patients showing 80% or greater tumor shrinkage at the time of the first CT check (at 12 weeks). The response rates were also higher than those that have been achieved with ipilimumab or nivolumab alone. While adverse events were more frequent and occurred earlier with the concurrent regimen than is typically seen with monotherapy, they were generally manageable; no treatment-related deaths were reported.

Antoni Ribas, MD, PhD

Also presented at ASCO this year were impressive results from a phase I study of another anti-PD1 antibody-lambrolizumab. Some patients in the study were ipilimumab-nave, and some had received prior ipilimumab treatment. The overall response rate in this trial, in both the ipilimumab-naive and previously treated patients, and for all doses and schedules, was greater than 35%. The median duration of response had not yet been reached. And as with the other PD1 blocker, nivolumab, responses were generally seen sooner than with ipilimumab: Dr. Antoni Ribas of UCLA’s Jonsson Comprehensive Cancer Center reported that most patients showed at least a partial response at the first CT check at 12 weeks. Moreover, he expressed confidence that there would likely be better results with time, since stable disease can turn into a partial response, and a partial response into a complete response-and since the majority of patients are continuing on therapy. The adverse event profile in the study was manageable, with only a 10% rate of drug-related adverse events.

Other immunotherapy studies of note that were presented this year included the phase III OPTiM trial of the oncolytic virus talimogene laherparepvec (T-VEC) vs granulocyte colony–stimulating factor (GM-CSF; sargramostim [Leukine]) in patients with metastatic melanoma. The interim results of this trial showed that the durable response rate in the T-VEC–treated patients was significantly better than that in patients treated with GM-CSF (16.3% vs 2.1%), with only 2% of patients experiencing grade 3/4 adverse events. Overall survival data are not yet mature, but an interim analysis is trending toward significantly better overall survival in the T-VEC arm. Also noteworthy this year was a study of GM-CSF plus higher-dose ipilimumab, which showed that the addition of GM-CSF increased survival compared with ipilimumab alone (1-year survival of 68.9% vs 52.9%, respectively)-as well as reducing the rate of adverse events seen with ipilimumab therapy.

The discussant at the Clinical Science Symposium on PD1/PDL1 pathway immunotherapy, Dr. Walter Urba of the Providence Cancer Center (Portland, OR), summed up the stunning progress that has been made in immunotherapy for melanoma with the bold statement that “immunotherapy is now the first-line treatment for metastatic melanoma regardless of mutation status.”