番茄社区

Expert Critique

FROM THE ASCO Reading Room

Amit Reddy, MBBS Postdoctoral Fellow Mitchell Cancer Institute, University of South Alabama

Treatment with nivolumab, an anti-programmed death-1 (PD-1) monoclonal antibody for monotherapy as well as in combination therapy with ipilimumab has been investigated in recent trials, with good benefits for patients with mucosal melanoma, an aggressive and rare melanoma subtype that is resistant to traditional drug regimens. For example, as described here, Alexander Shoushtari et al are currently performing a pooled retrospective analysis of multiple prospective trials to get the best estimate for therapy activity; this is believed to be the first manuscript to include activity of combined nivolumab plus ipilimumab for this subtype.

The study includes 889 patients with a variety of melanomas who received single or combination therapy. Those with mucosal melanomas had an objective response rate (ORR) of 23% with nivolumab monotherapy and 37% with the combination. In addition, in the patients receiving monotherapy, the median progression-free survival (PFS) was 3 months for those with mucosal melanoma and 6.2 months for those with cutaneous melanoma. The PFS with combined therapy was 5.9 and 11.7 months, respectively.

Still, Shoushtari warned that there were many small subsets that were analyzed in a non-specified way. About 25% of the mucosal melanomas that expressed PD-ligand 1 (PD-L1) in 5% or more of cells had ORRs of 53% with monotherapy and 60% with combined therapy; in contrast, about three quarters of the mucosal melanoma patients with PD-L1 with less than 5% expression had ORRs of 12% with monotherapy and 33% with the combination.

There are many factors that account for the poorer response rate, he said -- for example, the majority of mucosal melanomas have PD-L1 expression levels below 5%, and those cases do not respond well to monotherapy. In contrast, about half of the patients with cutaneous melanoma had PD-L1 expression levels of 5% of more. He said that doctors can now offer PD-1 based therapy to patients with mucosal melanoma, because the response rates are reasonable and patients have good therapeutic responses lasting months to years.

Full Critique

Nivolumab alone and in combination with ipilimumab appears to offer benefits for patients with mucosal melanoma, an aggressive subtype that is largely resistant to traditional therapies.

A pooled retrospective analysis of multiple prospective trials "may be the best estimate we get of the activity of programmed death-1 [PD-1] monotherapy in this disease subtype," one of the study's co-authors, , of Memorial Sloan Kettering Cancer Center, told 番茄社区. "It is also the first manuscript to document the activity of combined nivolumab plus ipilimumab in this disease subtype."

In the study, published in the , Shoushtari and colleagues followed up on evidence of activity from , retrospective analyses, and single case reports. The new study is a pooled analysis of data from 889 patients with a variety of melanomas who received nivolumab alone or combined with ipilimumab in six clinical trials, including phase III trials; 86 patients had mucosal melanoma and 665 had cutaneous melanoma. Data were also pooled for patients who received nivolumab combined with ipilimumab, including 35 patients with mucosal melanoma and 326 with cutaneous melanoma.

For those with mucosal melanomas, the objective response rate (ORR) with PD-1 monotherapy with nivolumab was 23%, which increased to 37% when nivolumab was combined with ipilimumab.

"These responses appear to have similar durability as those seen in cutaneous melanomas," Shoushtari said. "Given the heterogeneity of the patients involved in a retrospective pooled analysis, it is not advisable to directly compare these responses to prior trials of cutaneous melanoma. However, it provides important data to guide physicians in selecting therapy for patients with mucosal melanomas."

Among patients who received nivolumab monotherapy, median progression-free survival (PFS) was 3 months for those with mucosal melanoma and 6.2 months for those with cutaneous melanoma. Median PFS in patients treated with nivolumab combined with ipilimumab was 5.9 and 11.7 months, respectively.

The incidence of grade 3 or 4 treatment-related adverse events was 8.1% for mucosal melanoma and 12.5% for cutaneous melanoma with nivolumab monotherapy and was higher (40.0% mucosal melanoma, 54.9% cutaneous melanoma) for combination therapy.

An exploratory analysis by programmed death-ligand 1 (PD-L1) status was "interesting," Shoushtari said, but there are many caveats because of the small subsets analyzed in a non-prespecified way. "We found that the roughly one-quarter of mucosal melanomas that express PD-L1 in 5% or more of cells had ORRs of 53% and 60% to monotherapy and combination therapy, respectively. In contrast, in the three-quarters of mucosal melanomas with PD-L1 expression less than 5%, the ORR was 12% with monotherapy and 33% with combination therapy."

Multiple factors may account for the poorer response to therapy in mucosal melanoma versus cutaneous melanoma, he said: "Presuming these response rates represent true, robust differences, one hypothesis is that the majority [about 75%] of mucosal melanomas have PD-L1 expression levels by immunohistochemistry below 5%. Those cases do not tend to respond as frequently to PD-1 monotherapy as those with PD-L1 expression levels 5% or greater." In contrast, almost half of cutaneous melanomas (46%) in the analysis had PD-L1 expression levels at or above 5%.

Overall survival data should mature in concert with the Phase III nivolumab-plus-ipilimumab survival data, which are expected sometime in 2017, Shoushtari noted.

The mechanisms underlying response to PD-1 therapy in mucosal melanoma are still unknown. "If we try to extrapolate from in cutaneous melanoma, a subset of mucosal melanomas may have higher concentrations of CD8/PD-1 or PD-L1+ infiltration, and these tumors utilize adaptive immune resistance via PD-1/PD-L1 signaling to evade immunity."

Other mucosal melanomas likely employ other mechanisms of evading the immune system by actively excluding T cells. "One would expect that the former subgroup would be the one that responded better to PD-1+/- CTLA-4 blockade than the latter. Until more detailed studies are performed, however, no patients should be excluded from PD-1 based therapy, even on the basis of PD-L1 results."

He added that the results cast doubt on a once widely held hypothesis that melanomas need to have high mutational burdens to respond to these immunotherapy agents. "Mucosal melanomas have many-fold lower rates of mutations than sun-exposed cutaneous melanomas, but they still elicit immunologic reactions," he said, noting that a multicenter trial of nivolumab plus ipilimumab in mucosal and acral melanomas is now being planned, with correlative studies aimed at understanding the mechanisms of response and resistance.

In some key ways, mucosal melanoma treated with PD-1 blockade shares features with cutaneous melanoma, Shoushtari noted: "Clinicians can offer PD-1 based therapy to patients with mucosal melanoma with the knowledge that response rates are reasonable. Patients who respond tend to have durable responses lasting many months to even years. These therapies have adverse-event profiles in patients with mucosal melanoma similar to those with cutaneous melanoma documented in the larger melanoma trials. Our PD-L1 analysis is another reminder, however, that there are significant molecular differences between mucosal and cutaneous melanomas."