Background: Locoregional intra-arterial (i.a.) chemotherapy may provide high levels of cytostatic concentrations within the tumour and, simultaneously, a low rate of systemic side effects compared with systemic administration of anti-neoplastic drugs. In addition, this may lead to an increase of tumour response rate and prolongation of survival time. The aim of the study was (1) to evaluate the benefit of an i.a. infusion of cytostatic drugs via the coeliac trunk on tumour response rate and survival time, (2) to elucidate problems and risks, and finally (3) to achieve an improvement of overall therapeutic management in pancreatic carcinoma.
Patients and methods: In 22 patients (12 female; 10 male; mean age 57.1 years) with locally advanced pancreatic carcinoma, which was confirmed by histopathology, i.a. chemotherapy was administered. Through a catheter, which was inserted via the femoral artery by the Seldinger technique and placed with the tip in the coeliac trunk, two different drug combinations were given. Group A ( n=12) were given a bolus injection of a mixture (chemo-occlusion) consisting of amilomer (Spherex) and epirubicin (Farmorubicin) followed by short-time infusion of folic acid and 24-h infusion of 5-FU. Group B ( n=10) were given treatment over 5 days: mitoxantrone (Novantrone, day [d] 1), 5-FU and folic acid (Haemato-folin, d 2-4), and cisplatin (d 5). Treatment was repeated in both groups every 4 weeks. Tumour response was assessed by computed tomography every 8 weeks.
Results: In group A, there was one complete and one partial remission, resulting in a remission rate of 16.6%. Two patients showed stable disease, while in two-thirds of the patients ( n=8), progressive disease was found. Median survival time was 3 months; 1-year survival rate was 33.3% (4 of 12 patients). In group B, again, one complete and one partial remission were observed (remission rate 20%). In three cases, stable disease, and in 50% of patients ( n=5), progressive disease, were documented. Median survival was 7.0 months; 1-year survival rate was 20% (2 of 10 patients). If both groups were compared, there was no difference in survival. In addition, no prolongation of survival time was found in comparison with patients of a historical study group treated with established systemic chemotherapy using gemcitabine monotherapy ( n=28; median survival time 9 months). Though a tendency for poorer outcome of i.a. chemotherapy was seen when the Kaplan-Meier curves of survival were compared, this difference was not statistically significant (log rank test, P=0.08).
Conclusion: Despite conceptual and pharmacokinetic advantages of locoregional i.a. chemotherapy, better outcome with regard to tumour response rate and survival time could not be found. I.a. chemotherapy is, therefore, still an experimental treatment option in pancreatic carcinoma and can, currently, not be recommended for routine use.