Isolated limb infusion for malignant melanoma: predictors of response and outcome.

Purpose: Isolated limb infusion (ILI) is an alternative to isolated limb perfusion (ILP) for the treatment of unresectable limb melanoma recurrence. The aims of this study were to determine the response rates of unresectable local and/or in-transit melanoma of the upper or lower limb to ILI and to identify factors predictive of survival.
Methodology: A prospective database identified 74 patients (35 male and 39 female) with local and/or in-transit melanoma recurrence without metastatic disease who underwent hyperthermic ILI with melphalan at a single institution between January 1996 and December 2008. Three patients could not be evaluated for response. Median follow-up was 34 months.
Results: Of the 74 patients, the majority had N2c disease (57/74, 67%), while 17/74 (33%) patients had N3 disease. Median maximum temperature achieved was 38.1 degrees C and median tourniquet time was 32.5 min. Wieberdink III/IV complications occurred following 7/74 (10%) ILI and were associated with higher limb volumes and higher total melphalan dose. Complete response (CR) was seen following 17/71 (24%) ILI and the partial response rate was 30% (22/71). The median duration of CR was 43 months. Univariable analyses found that limb volume >8.0 l and maximum limb temperature >38.5 degrees C were the only independent factors predictive for a CR following ILI. Multivariate analyses identified CR and positive lymph nodes as the only independent prognostic factors for melanoma-specific survival.
Conclusions: Patients who obtain CR have significantly improved survival compared with nonresponders. The challenge remains to improve CR rates and prospectively identify responders.