A. F. M. Salem Jr1, M. Chen1, J. McQuade1, R. Amaria1, D. Swanson1, M. D. Williams2, A. Farooqi1, A. J. Bishop1, E. Y. Hanna3, B. A. Guadagnolo1, S. Y. Su3, and D. Mitra1; 1The University of Texas MD Anderson Cancer Center, Houston, TX, 2MD Anderson Cancer Center, Houston, TX, 3Department of Head and Neck Surgery, The University of Texas MD Anderson Cancer Center, Houston, TX
Purpose/Objective(s):Sinonasal mucosal melanoma (SMM) is the most common type of mucosal melanoma (MM), an aggressive and rare entity representing <1% of melanomas. We evaluated the outcomes for SMM patients (pts) treated with neoadjuvant ipilimumab and nivolumab (Ipi/Nivo).Materials/
Methods: Using our institutional database, we identified 17 pts treated with neoadjuvant intent Ipi/Nivo between 2016-2023. Descriptive statistics were used to summarize data and survival analyses were conducted by the Kaplan-Meier method. Results: Median age was 72 years (IQR 60-73) with 59% males (n=10). The disease was confined to the nasal cavity (29%, n=5) or with sinus (47%, n=8) and/or skull base (24%, n=4) involvement. One patient (6%) presented with nodal disease. Of the 15 pts whose pathologic features at diagnosis were known, 67% had ulceration (n=10), 40% had lymphovascular invasion (n=6), and 20% had melanoma in-situ (n=3).The median tumor size was 32 mm (IQR 27-35). All pts received at least 1 cycle of Ipi/Nivo (range 1-4). By RECIST 1.1 criteria, 3 pts (18%) had a radiographic complete response (rCR) and 1 had a partial response (rPR) (6%). Five patients (29%) had stable disease (rSD) and 8 (47%) had progressive disease. Thus, the overall response rate was 24%, and the clinical benefit rate (rCR +rPR + rSD) was 53% Three pts did not undergo surgical resection after Ipi/Nivo: 2 opted for surveillance following a rCR and one developed distant metastasis. One patient had poor tolerance of Ipi/Nivo, and subsequently received radiotherapy (RT) and carboplatin/paclitaxel prior to resection. Of the 13 pts who proceeded to surgery after Ipi/Nivo alone, pathologic response was evaluated for 11: 2 had pCR (18%), one had 80% viable tumor at one focus and 0% viable at a second, and 8 were non-responders with >50% viable tumor (73%). Pathologic response was congruent with radiographic response except for 1 patient who had rSD but pCR. Of the 2 pts without quantified pathologic response, one had rSD and one had rPR prior to resection. Overall, 5 pts (29%) had evidence of pathologic or radiographic response to Ipi/Nivo. Eleven pts (79%) received adjuvant radiation therapy (RT) and 4 received adjuvant nivolumab (29%). Median follow-up was 28 months from the start of treatment (QR 20-36). Pts who had evidence of response to neoadjuvant Ipi/Nivo had significantly better progression-free survival (2-yr PFS 80% vs. 17%, p=0.02). Of the 11 pts receiving adjuvant RT, only one had local progression. Conclusion: Consistent with prior reports for MM, we observed about a third of patients respond to neoadjuvant Ipi/Nivo, and correspondingly had significantly better outcomes. However, strategies to further improve outcomes for this aggressive malignancy are urgently needed; thus, we are currently evaluating the addition of preoperative RT to neoadjuvant Ipi/Nivo in the “PRISM” clinical trial at our institution (NCT05546827).
