Hypoxic abdominal stop-flow perfusion in the treatment of advanced pancreatic cancer: a phase II evaluation/trial

Abstract

In the past decade, some authors have reported objective responses and prolonged median survival times using hypoxic abdominal perfusion (HAP) for the treatment of advanced pancreatic cancer. However, these promising results have not been confirmed by others, making it difficult to define the effectiveness of this loco-regional chemotherapy. The aim of this study, therefore, was to evaluate the response rate, time to disease progression and overall survival following HAP treatment of 22 consecutive patients with advanced pancreatic tumors. Within the period from 1999 to 2003, 22 patients with histological diagnosis of unresectable stage III/IV pancreatic cancer, not responsive to systemic chemotherapy, were treated with mitomycin C 30mg/m(2) and cisplatin 60mg/m(2) by HAP (stop flow technique). Immediately after perfusion, hemofiltration was performed to reduce systemic side toxic effects. Responses were assessed by CT-scan 30days from the end of treatment. Minor or partial responses were confirmed by a second CT-scan 4weeks later. Following 26 treatment cycles no death or technical complications were recorded; four patients (18.2%) achieved a partial response, 2 (9.1%) a minimal response and 13 (59.1%) stable disease. The remaining 3 patients (13.6%) showed progression of the disease. The median time to disease progression was 3 months (range 1-10). The median survival time from the start of regional chemotherapy was 6 months (range 1.9-16), with a 1-year survival rate of 9%. Our data show that HAP is a relatively effective second-line treatment for advanced stage pancreatic cancer with a low complication rate. We do not concur with the opinion of others that HAP is an inactive treatment approach. However, taking into account the invasiveness of this procedure, and associated morbidity and cost, HAP would not appear to be preferable to less invasive loco-regional chemotherapeutic alternatives.