Background:
Adjuvant (adj) radiotherapy (RT) halves the risk of locoregional (LR) recurrence in patients with high risk stage III melanoma after lymphadenectomy (CLND), however its role in the adj immunotherapy (IO) era without CLND is unknown.
Methods:
Patients with resected stage III melanoma who received adj IO and recurred with resectable LR only disease were studied. After resection of this 1st recurrence, adj RT may or may not have been administered.
Results:
71 patients were included. Prior to adj IO, median age was 60y, 59% male, 56% BRAFm, 61% stage IIIC, 52% underwent CLND, 17% had in-transit (IT) only disease. Adj IO included: 90% PD1, 8% ipilimumab-nivolumab (IN) and 1% nivolumab or IN (blinded on trial); median duration of therapy 5 months. 21(30%) had high risk stage III disease at diagnosis.
Median time to 1st recurrence was 7 months; 49 (69%) recurred during adj IO. At 1st recurrence, 55 (77%) had IIIC and 31 (44%) commenced systemic therapy. 24 (34%) received adj RT after resection of 1st recurrence.
Adj RT was associated with a reduced risk of any 2nd recurrence (7/24, 29% vs 26/47, 55%, p=0.03) and LR 2nd recurrence (2/24, 8% vs 17/47, 36%, p=0.012). Of note, 70% of patients who did not receive adj RT at 1st recurrence had IT only disease, and though this did not significantly affect rate of 2nd recurrence (p=0.19), this likely reflects an inherent selection bias in this study.
Median follow-up was 22 months. Median RFS to 2nd recurrence was not reached and 19 months for those who did and did not receive adj RT at 1st recurrence, respectively (p=0.047).
Conclusions:
Whilst adj RT appears to reduce 2nd recurrences, this may have been influenced by an unavoidable selection bias in the data, particularly an imbalance in the percentage of patients with IT disease.