Radiation and Immune Checkpoint Inhibitor Therapy in Mucosal Melanoma

Abstract

Mucosal melanoma is a rare subtype of melanoma with a dismal prognosis for which novel approaches are needed. Older series report variable rates of local control following radiation. Immune checkpoint inhibitors (ICI) are used but response rates are limited, and resistance can develop after initial response. We aimed to evaluate the use of radiation therapy (RT) in combination with ICI in recent clinical practice. We reviewed cases with histologically confirmed mucosal melanoma who were treated with RT (palliative or definitive intent) from 2012-19. Rates of local control (LC), distant metastasis (DM), and overall survival (OS) were calculated with the Kaplan-Meier method and measured from the time of RT. We identified 24 patients; 9 (38%) who received RT with definitive intent (7 following surgery and 2 as definitive therapy for unresectable disease), 2 for local recurrence (LR), and 13 for metastatic disease. Sites irradiated were 8 (33%) head and neck, 8 (33%) pelvis, 6 (25%) brain, 2 (8%) spine. Median dose was 66 Gy (range 36-70) in 6-35 fractions (fx) for patients treated with definitive intent, 45 Gy (40-50) in 10-20 fx for those treated for LR, and 25 Gy (8-37.5) in 1-15 fx for those treated for metastatic disease. Technique was IMRT in 8 cases (33%), SRS/SRT in 5 (21%), conformal in 5 (21%), 2D in 4 (17%), SBRT in 2 (8%). Twenty-three of 24 patients were treated with ICI: 12 patients (52%) were treated with PD-1/PD-L1 inhibitors, 2 (9%) with CTLA4 inhibitors, 8 (35%) with simultaneous CTLA4 and PD-1 therapy, and 1 (4%) CTLA4 followed by PD-1 inhibitor. ICI was delivered >3 months prior to RT onset in 5 cases (22%). ICI was concurrent with RT in 14 cases (61%). Four patients (17%) received ICI >3 months after completing their first RT treatment, and of those, 3 had subsequent RT treatments during or after receiving ICI. The median follow-up was 11.6 months from time of RT. Median OS was 15 months with median duration of LC 40 months among surviving patients. 1-year LC was 72% (73% in definitive plus LR cases), OS was 57%. Among 7 patients with local recurrence: 3 had sinus/nasopharyngeal melanomas treated after surgery, 3 received RT to brain metastases, and 1 received RT to a recurrent pelvic primary. 1-year LC was 80% for patients receiving ICI prior to RT, 68% for ICI concurrent with RT, and 75% for ICI after RT (p = 0.82). Among 6 cases treated with definitive RT and concurrent ICI, 1-year LC was 67%, 1 had a complete response to ICI, 2 stable disease, and 3 progressive disease. Of those 6 patients, 2 developed colitis, 1 pneumonitis, and 1 hypothyroidism. Among these patients, 1-year distant control was 17%. In a recent series of patients treated with RT for mucosal melanoma, we found that the timing and type of ICI is heterogenous. LC is favorable but DM remains common. Larger studies will be necessary to evaluate the effectiveness of combination therapy with ICI and RT, particularly in the definitive or postoperative setting.