Aim: The risk of developing melanoma, the natural history of this disease, and the response to therapy with biological reagents may be determined, in part, by a patient's human leukocyte antigen (HLA) phenotype. In order to study this, the relationship between HLA type and clinical response to therapy with interleukin-2 (IL-2) was evaluated.
Methods: We retrospectively determined the HLA phenotype of 82 patients with metastatic melanoma who were treated with IL-2-based therapy. Fresh or frozen lymphocytes were serologically typed by standard lymphocytotoxicity techniques (NIH or Amos modified). The treatment regimens included IL-2 alone or with tumour infiltrating lymphocytes, lymphokine activated killer cells, cyclophosphamide, interferon-alpha, and tumour necrosis factor-alpha. Initially, the relationship between clinical response and each HLA antigen was evaluated by performing a two-tailed Fisher exact test. Associations with a P-value less than or equal to 0.10 without adjustment for multiple comparisons were considered worthy of further study. Independent confirmation of these apparent associations was obtained by studying the relationship between HLA phenotype and patient survival using Cox proportional hazards models.
Results: In the initial screening, a statistically significant association between clinical response and the expression of HLA-DQ1 was observed (unadjusted P2 = 0.0017). HLA-DQ1 was also independently associated with prolonged survival (P2 = 0.026). This positive association with survival was evident both for patients who responded to therapy and those who did not respond, as defined by > 50% tumour regression.
Conclusions: Among patients with metastatic melanoma, HLA-DQ1 appears to be associated with clinical response to therapy using IL-2. This apparent association is confirmed by the observation that HLA-DQ1 is independently associated with prolonged survival in this group of patients.