Factors predictive of acute regional toxicity after isolated limb infusion with melphalan and actinomycin D in melanoma patients.

Introduction: Isolated limb infusion (ILI) with cytotoxic drugs is a low-flow isolated limb perfusion (ILP) performed via percutaneous catheters without oxygenation to treat metastatic melanoma confined to a limb. Response rates and duration of response following ILI are similar to those after ILP. Previously we have shown that more significant limb toxicity is not associated with a higher response rate or improved patient outcome. In this study we sought to determine factors predicting toxicity following ILI.
Methods: From our prospective database 185 patients with advanced metastatic melanoma of the limb treated with a single ILI between 1992 and 2007 were identified. In all patients a cytotoxic combination of melphalan and actinomycin D was used. Drug circulation time was 20-30 min under mild hyperthermic conditions (38-39 degrees C). Limb toxicity was assessed using the Wieberdink scale.
Results: The average patient age was 74 years (range 29-93 years) and 62% were female. Most patients (134/185) had MD Anderson stage III disease (satellites and in-transit metastases). Toxicity grade I (no reaction) occurred in 3 patients, grade II (slight erythema and edema) in 105 patients, grade III (considerable erythema and edema +/- blistering) in 72 patients, and grade IV (threatened or actual compartment syndrome) in 5 patients. No patient developed grade V toxicity (requiring amputation). On univariate analysis high peak and high final melphalan concentrations were found to be predictive factors for grade III/IV limb toxicity as well as the area under the curve of the melphalan concentration. Surprisingly, a greater rise in the CO(2) level during the procedure was associated with lower toxicity in the univariate analysis. Increased serum creatine phosphokinase (CK) postoperatively was related to higher toxicity score. In the multivariate analysis high final melphalan concentration and shorter tourniquet time were independent predictive risk factors for developing grade III/IV limb toxicity.
Conclusions: ILI is a safe alternative to the more invasive and laborious ILP technique to treat melanoma confined to a limb. Regional acute toxicity following ILI is mild to moderate in most patients. Based on the predictive factors found in this series, altering melphalan dose and tourniquet time may allow further reductions in post-ILI toxicity without compromising effectiveness.