Single-Agent PD-1 Blockade Is “Treatment of Choice” for Desmoplastic Melanoma

Patients with inoperable metastatic desmoplastic melanoma (DM) should receive anti–PD-1 monotherapy up front, based on findings from the SWOG phase II S1512 trial—the first prospective assessment of PD-1 blockade's utility in this disease subtype.

Strongly associated with sun exposure, DM is characterized by “a unique pathology of spindle-shaped cells separated by collagen fibers,” said Zeynep Eroglu, MD, of Moffitt Cancer Center in Tampa, FL. A rare subtype, comprising roughly 4% of melanoma diagnoses in the United States, it often originates in the head and neck, and is more common in older men.

DM is also typically resistant to radiation and—where feasible—surgery, noted trial investigator Antoni Ribas, MD, PhD, of the University of California, Los Angeles. “I've had patients relapse repeatedly, needing multiple skin grafts and reconstructions over the years” that can be deeply disfiguring.

After pembrolizumab (Keytruda; Merck) was greenlighted as a first-line treatment for metastatic melanoma, Ribas began noticing “really good responses in one or two patients” with DM. Curious, he queried various colleagues about similar anecdotal observations. The researchers compiled enough data for a 60-patient retrospective analysis (Nature 2018;553:347–50), highlighting a complete response rate (CRR) of 32%. Buoyed, “we decided to move ahead with a prospective study, even though we knew that [DM] being ultrarare, this would take a while.” From late 2017 to 2021, he and the group, including Eroglu, enrolled 27 patients across nine states nationwide.

Kari Kendra, MD, PhD, of The Ohio State University in Columbus, reported S1512’s results during the American Association for Cancer Research Annual Meeting 2023 in Orlando, FL, April 14–19. All 27 patients were evaluable: The overall response rate (ORR) to pembrolizumab monotherapy was 89%, including a CRR of 33%. At 2 years, progression-free survival (PFS) was 74% and overall survival was 89%.

By contrast, Eroglu pointed out, in the general melanoma population, the ORR and CRR with anti–PD-1 monotherapy tend “to be around 40% and 10%,” respectively, and the 2-year PFS is approximately 37%.

Kendra also spotlighted one patient “with prominent scalp lesions that were no longer visible by 6 weeks; 4 months later, his hair was even growing back in the area of disease,” as well as CT scans of another patient's lung metastases, “which regressed beautifully.”

“We've gotten some really nice before-and-after images,” Ribas observed. “This is one of the most responsive tumor subtypes, akin to MSI [microsatellite instability]-high colorectal cancer.”

Combination therapies such as nivolumab, another PD-1 agent, with the LAG-3 inhibitor relatlimab (jointly marketed as Opdualag; Bristol Myers Squibb) have become the norm for inoperable or metastatic melanoma, Kendra said. For DM, however, “single-agent PD-1 is sufficient and should be the treatment of choice.”

It's good to reduce unnecessary therapy where feasible, Eroglu agreed. Neoadjuvant PD-1 blockade should be considered for patients with operable DM, she added, given favorable pathologic complete response data from a separate cohort of S1512 reported last year (J Clin Oncol 40, 2022 [suppl 16; abstr 9502]).

Trial biopsies revealed a high tumor mutational burden (TMB)—the median was 49.7 mutations per megabase—and NF1 alterations in 67% of cases, “pretty consistent with what's known” about DM, Kendra said. Still, exactly why it's so therapeutically sensitive is unclear.

To Ribas, TMB “doesn't explain everything.” In fact, DM “being full of collagen, with no tertiary lymphoid structures, goes against a lot of the dogma on immunotherapy responsiveness.” His lab is currently diving deeper into the disease's underlying biology.

“I used to dread this [DM] diagnosis,” Ribas remarked. “Now, to see patients doing so well, often within days, is remarkable. It's not by chance—there's something here, and we should strive to find and study more such exceptional responders” to immune checkpoint inhibition. –Alissa Poh