Data from the phase 2 SWOG S1512 trial showed a high pCR and durable survival outcomes with neoadjuvant pembrolizumab in resectable desmoplastic melanoma.
Neoadjuvant therapy with pembrolizumab (Keytruda) led to pathologic complete responses (pCRs) in 71% (95% CI, 51%-87%; P < .001) of patients with resectable desmoplastic melanoma, meeting the primary end point of cohort A of the phase 2 SWOG S1512 trial (NCT02775851), according to findings published in Nature Cancer.1
Additionally, the benefit of neoadjuvant pembrolizumab was largely comparable across all subgroups, irrespective of age, sex, performance status, lactate dehydrogenase level, disease status, and desmoplastic histologic subtype.
“The high pCR rate supports the role of systemic treatment for localized desmoplastic melanoma before surgical intervention, after consideration of the potential toxicities induced by neoadjuvant anti–PD-1 therapy. pCRs were demonstrated when pembrolizumab was administered to patients either at initial diagnosis or at the time of locally recurrent disease,” Kari L. Kendra, MD, PhD, lead study author and professor of internal medicine and chair of the Melanoma Disease Specific Committee at The Ohio State University Comprehensive Cancer Center––James in Columbus, and coauthors wrote in the publication.
Desmoplastic melanoma is a rare form of melanoma that is traditionally amelanotic and stems from sun-exposed areas. Although it is one of the most genetically mutated cancers, it lacks common drivers of cutaneous melanoma, such as BRAF and NRAS mutations.
Standard treatment for localized disease consists of surgical excision with or without radiation and adjuvant PD-1 inhibition with nivolumab (Opdivo) or pembrolizumab if stage IIB or higher. However, the full extent of lesions is often not entirely captured clinically or by imaging, leading to additional surgeries and excisions, which may result in clinical defects. Moreover, surgical morbidity presents a challenge for older and frail individuals.
Historically, it was thought that advanced disease was not responsive to most systemic therapy, but a retrospective analysis identified high response rates with PD-1 inhibition in patients with pure and mixed histological subtypes of desmoplastic melanoma.2
When explaining the study rationale, the authors added that, “A high response rate in the neoadjuvant setting could change the course of therapy for individuals with locally advanced disease, with improved local and systemic results. If substantial tumor regression was achieved, the use of radical surgery and radiation therapy may be reduced along with surgery-related morbidities, resulting in a higher quality of life for patients.”1
SWOG S1512 was a prospective trial comprising 2 cohorts: patients with locally advanced disease receiving neoadjuvant pembrolizumab (cohort A) and patients with advanced desmoplastic melanoma deemed ineligible for resection (cohort B).
Between July 2017 and May 2021, 30 patients with resectable, stage I to III desmoplastic melanoma were enrolled across 10 sites in the United States. The study’s primary end point was pCR rate. Secondary end points were clinical response rate, overall survival (OS), relapse-free survival (RFS), and toxicity. A total of 28 patients were included in the analysis after 1 refused protocol therapy and another was deemed ineligible following pathologic assessment.
The median age was 75 years (range, 37-91). Most patients were male (75%) and White (96%), which is consistent with the primary demographic characteristics for the disease. The most common primary disease site was the head and neck area (68%). Most patients had primary (82%) as opposed to recurrent (18%) disease at enrollment, and 5 patients had positive nodes.
Patients received three 200-mg infusions of pembrolizumab every 3 weeks before surgery at week 9. Those who did not achieve a clinical response were permitted to undergo an optional fourth cycle of neoadjuvant therapy and up to 15 cycles of adjuvant therapy. Most patients (89%) received the intended 3 cycles of neoadjuvant pembrolizumab; 1 patient discontinued treatment due to colitis, and 2 others received the fourth dose of neoadjuvant pembrolizumab. Of the 28 patients who started neoadjuvant therapy, 27 underwent resection. The remaining patient elected not to undergo surgery and did not have a pCR with pembrolizumab.
The median follow-up was 42 months (range, 35-50). The median time from start of therapy to surgery was 80 days (range, 52-135). None of the 28 patients received adjuvant pembrolizumab.
Twenty-five patients were evaluable for the secondary end point of clinical response. At the 9-week assessment, the overall clinical response rate was 48% (95% CI, 28%-69%), which included a complete response rate of 16% and a partial response rate of 32%. Stable disease occurred in 36% of patients and progressive disease in 4%.
The 3-year RFS rate among patients who underwent surgical resection (n = 27) was 74% (95% CI, 51%-87%), and the median RFS has not been reached. The 3-year OS rate was 87% (95% CI, 65%-96%), and the median OS has not been reached.
Four deaths occurred among the 28 patients evaluable for survival, 3 of which were unrelated to the disease and one from an unknown cause. The estimated 3-year melanoma-specific survival, which was evaluated as a post hoc end point, was 95% (95% CI, 80%-100%).
Any-grade treatment-related adverse effects (AEs) occurred in 79% of patients, the most common of which included grade 1/2 fatigue (43%), maculopapular rash (21%), and diarrhea (14%). Other reported events included arthralgia (7%), headache (7%), decreased lymphocyte count (7%), and myalgia (7%).
“Neoadjuvant pembrolizumab was generally well tolerated, with the type of AEs consistent with the known toxicity of this regimen. Furthermore, none of the patients became surgically unresectable,” the authors wrote.
Grade 3 mucositis and immune-mediated colitis occurred in 7% (n = 2) of patients. There were no serious AEs and only 1 case requiring treatment discontinuation. Immune-related AEs also occurred (any grade, 43%; grade 3, 4%); none were deemed serious or led to death.
“These data provide evidence that 3 doses of neoadjuvant single-agent anti–PD-1 therapy could be considered before surgery for patients with resectable desmoplastic melanoma,” the authors concluded.
