scout
News|Articles|January 30, 2026

Neoadjuvant Pembrolizumab Achieves High pCR Rates in Desmoplastic Melanoma

Author(s)Jonah Feldman
Fact checked by: Paige Britt
Listen
0:00 / 0:00

Key Takeaways

  • Neoadjuvant pembrolizumab achieved a 71% pCR rate in desmoplastic melanoma, surpassing expectations for neoadjuvant anti–PD-1 treatment.
  • The 3-year relapse-free survival rate was 74%, and the overall survival rate was 87%, with no melanoma-related deaths.
SHOW MORE

Neoadjuvant pembrolizumab showed promising high response rates and melanoma-specific survival outcomes in desmoplastic melanoma.

Neoadjuvant Pembrolizumab Achieves High pCR Rates in Desmoplastic Melanoma

Preoperative pembrolizumab (Keytruda) led to deep and durable responses in patients with resectable desmoplastic melanoma, according to findings published in Nature Cancer.1

The findings from the multicenter phase 2 SWOG S1512 trial (NCT02775851) revealed that treatment with immune checkpoint inhibition resulted in a pathologic complete response (pCR) rate of 71% in a rare melanoma subtype that is difficult to fully manage with surgery. At 3 years of follow-up, no patients had died because of melanoma or adverse events (AEs).

Key Clinical Findings

Cohort A of the SWOG S1512 trial evaluated neoadjuvant pembrolizumab in patients with surgically resectable desmoplastic melanoma. Among 28 treated patients, 71% achieved a pCR following 3 preoperative doses of pembrolizumab at 200 mg every 3 weeks, exceeding the prespecified efficacy threshold and representing one of the highest pCR rates reported for melanoma in the neoadjuvant setting.¹ The null hypothesis of 25% pCR rate was the expected outcome for neoadjuvant single-agent anti–PD-1 treatment in non-desmoplastic cutaneous melanoma.

At 9-week clinical assessment, 4 of the 25 evaluable patients (16%) had a complete response and 8 (32%) had a partial response; 21 of 25 (84%) who had assessable disease had no clinical progression during the neoadjuvant treatment. Only 1 of the 28 patients did not receive surgical resection; another discontinued pembrolizumab but did have their tumor resected.

The 3-year relapse-free survival rate was 74% and the 3-year overall survival (OS)rate was 87%, with the median not reached for either survival end point. There were 6 relapse events, 3 of which were recurrent melanoma and 3 were death events. Of 4 patients who died, 3 were of unrelated causes and 1 was of unknown cause, this being the only event counted in the 3-year melanoma survival rate of 95%.

Investigating Genomic Profile

Desmoplastic melanoma is a rare histologic variant characterized by dense fibrous stroma, deep local invasion, and frequent neurotropism. Surgical management is often challenging due to ill-defined tumor margins, and local recurrence rates remain high despite wide excision and adjuvant radiation. Although immune checkpoint inhibitors have revolutionized advanced melanoma treatment, data supporting their use in the neoadjuvant setting for desmoplastic melanoma have been limited.¹

Genomic analysis demonstrated that desmoplastic melanomas harbored a high tumor mutational burden (TMB), consistent with prior observations in this melanoma subtype, and there was no significant different in TMB between those who did or did not have pCR. Recurrent ultraviolet radiation–associated mutational signatures were prevalent, supporting an immunogenic tumor profile. Responders tended to exhibit enriched immune-related gene expression, including pathways involved in antigen presentation and interferon signaling, whereas nonresponding tumors showed relative attenuation of these immune signatures. No single genomic alteration was uniquely predictive of response.

Safety Considerations

Treatment was well tolerated, with a low incidence of grade 3 immune-related AEs and no unexpected perioperative complications. Twenty-five patients received the scheduled 3 cycles of neoadjuvant therapy, with 1 discontinuing after cycle 1 because of immune-related colitis.

Pembrolizumab was generally well tolerated in the neoadjuvant setting. Immune-related AEs were consistent with known PD-1 inhibitor toxicity profiles and were manageable with standard interventions. No increase in perioperative morbidity was observed, alleviating concerns that preoperative immune activation might complicate wound healing or surgical recovery.¹

Study Design and Methods

SWOG S1512 was a single-arm, multicenter phase 2 trial enrolling patients with resectable stage II–III desmoplastic melanoma as well as advanced melanoma in a separate cohort, where it showed an objective response rate of 89%.2 Participants in Cohort A received 3 cycles of pembrolizumab prior to definitive surgical resection. Tumor biopsies were obtained at baseline and during treatment when feasible. The primary end point was pCR, defined as no viable tumor cells in the resected specimen. Secondary end points included safety, event-free survival, and OS.¹

The median patient age was 75 (range, 37-91) and 75% were male patients. The most common site of disease was the head or neck area in 68% of patients. The majority (82%) had desmoplastic melanoma at primary diagnosis vs 18% at disease recurrence. None of the patients received adjuvant pembrolizumab after surgery.

Clinical Implications

These findings support further investigation of neoadjuvant immunotherapy in randomized trials and raise the possibility of expanding this approach to other melanoma subtypes with similar immunobiologic characteristics. This strategy represents a promising advance in a historically difficult-to-treat melanoma subtype and highlights the growing role of preoperative immunotherapy in solid tumor oncology.

REFERENCES

1. Kendra KL, Bellasea SL, Eroglu Z, et al. Neoadjuvant PD-1 blockade in surgically resectable desmoplastic melanoma: cohort A of the phase 2 SWOG S1512 trial. Nat Cancer. Published online January 29 2026. doi:10.1038/s43018-025-01113-y

2. Kendra KL, Bellasea SL, Eroglu Z, et al. Anti-PD-1 therapy in unresectable desmoplastic melanoma: the phase 2 SWOG S1512 trial. Nat Med. 2025;31(11):3668-3674. doi:10.1038/s41591-025-03875-5

Newsletter

Stay up to date on practice-changing data in community practice.


Latest CME