To evaluate the role of hyperthermia combined with chemotherapy in the loco-regional treatment of tumors, a retrospective analysis was done with 228 limb melanoma patients treated with hyperthermic antiblastic perfusion (HAP). A series of treatment- and tumor-related prognostic factors was analyzed to establish their influence on tumor response, loco-regional control, and survival. Concerning tumor response, the logistic model showed that the number of lesions and the minimal tumor temperature (min T) maintained their individual predictive values (P < 0.000001 and P = 0.04, respectively). For loco-regional control, only the number of lesions had a significant predictive value. No direct correlation was found between the treatment-related variables and loco-regional control. However, the 5-year survival rate was significantly higher for patients who achieved a complete response (CR) (51.5%, P = 0.0033) as compared to those who did not (33.3%), providing indirect evidence of the role of the treatment. Multivariate analysis showed that both disease-free and overall survival are strongly influenced by numerous clinical variables and the min T always maintained its significance. When analyzing the subgroup of 119 patients evaluable for tumor response, the Cox model selected the tumor response as the dominant factor for both disease-free and overall survival. These data seem to demonstrate that the optimization of treatment parameters is crucial in determining the CR rate, which, in turn, positively affects the disease outcome. HAP is the treatment of choice for recurrent limb melanoma, and hyperthermia plays an important role in exploiting the efficacy of this technique.