The median overall survival of the vaccinated patients was 19.2 months, calculated from the day of leukapheresis instead of from diagnosis of metastasis, as is done in unselected case series. Overall Selleckchem BLU9931 survival from date of diagnosis of metastatic disease in our dendritic cell vaccinated patients was 30.3 months. According to the American Joint Committee on Cancer Staging Manual, median overall survival is 17 months for M1a, 9 months for M1b, and 4.5 months for M1c.43 Our patients showed a median overall survival of 29 months for M1a, 22.5 months for M1b, and 6 months for M1c. No large difference in overall survival was seen in patients who received prior therapy for metastatic disease to treatment-naïve

patients. Comparing our results on survival with other published series, the observed median overall

survival of 19.2 months in dendritic VX770 cell-vaccinated patients not only exceeded the overall survival as reported in studies using systemic treatment (range, 5.2 to 15.3 months), but also the overall survival in almost all studies in more selected metastatic uveal melanoma patients treated with local therapies of the liver (range, 5.2 to 24 months), such as surgical resection of liver metastasis, hepatic artery chemoembolization, and hepatic artery infusion chemotherapy.17 These invasive therapies excluded patients with extrahepatic metastasis and high World Health Organization performance status, that is, have more strict inclusion criteria, and consequently included patients with more favorable prognostic factors. Further comparison with

a cohort of patients with a similar proportion of pretreated patients (12 of 20 patients) and selection criteria, treated with treosulfan and gemcitabine, showed a similar median overall survival (19.2 vs 17 months).44 Although our results do not allow definite conclusions about clinical outcome, the immunologic responses, previously shown to correlate with clinical outcome,28 and the observed long overall survival in our cohort of metastatic uveal melanoma patients seem promising. Additionally, the minimal toxicity associated many with dendritic cell vaccination compares favorably with other treatment methods. As to metastatic patients, the high tumor burden may hamper the induction of effective immune responses, creating a suppressive tumor microenvironment by the secretion of suppressive cytokines and attraction of regulatory T cells.45 Robust immunologic responses on dendritic cell vaccination are induced more frequently in patients with no evidence of disease (72%) (manuscript in preparation) compared with patients with macroscopic tumor burden (32%).28 On the basis of the association of tumor-specific T cells and improved clinical outcome, this suggests that dendritic cell-based vaccination may have a more pronounced role in an adjuvant setting and should be initiated at an early stage after tumor resection.