5-FLUOROURACIL PLUS INTERFERON ALPHA-2A COMPARED TO 5-FLUOROURACIL ALONE IN THE TREATMENT OF ADVANCED COLON-CARCINOMA - A MULTICENTRIC RANDOMIZED STUDY

Biochemical modulation is one of the most interesting fields in cancer chemotherapy. Interferon-alpha (IFN alpha) is a cytokine that is able to influence the pharmacodynamics of 5-fluorouracil (5FU) through a n umber of mechanisms. With the aim of confirming some data emerging fro m the literature, we initiated a multicentric randomized study compari ng the combination of 5FU and IFN alpha-2a with 5FU alone in the treat ment of advanced or metastatic colon cancer. A group of 205 colon canc er patients (104 in the 5FU aim and 101 in the 5FU + IFN alpha-2a arm) were included in the final intention-to-treat analysis. Rectal cancer s were not considered eligible. All patients had measurable disease, w ere aged 75 years or less, had a Karnofsky index of at least 60 and ha d good bone marrow, renal, liver and cardiac functions. No previous ch emo-immunotherapy was allowed. The treatment was 750 mg/m(2) 5FU (4 h i.v. infusion) on days 1-5 and then i.v. bolus weekly, starting from d ay 12, with or without IFN alpha-2a given s.c. three times weekly (sta rting dose 3 x 10(6) IU rising to 9 x 10(6) IU, if tolerated). Patient s were treated until progression or, if responsive. for a maximum of 4 8 weeks and then observed for a period of 2 years. The primary end-poi nt of the study was objective clinical response (OR); secondary parame ters were time to progression, overall survival, and time to death aft er progression. WHO criteria were used for both clinical response and toxicity measurements. Dose reduction was planned a priori in the even t of significant toxicity due to 5FU, IFN alpha-2a or both. Associatio n between primary and secondary end-points and treatment was studied b y univariate and multivariate analysis. Altogether, 47 patients achiev ed a documented response. A 25% OR was observed in the combination arm while a 21% OR was seen in the 5FU ann; this difference is not statis tically significant (P = 0.6). Patients with a small tumour burden (be low 5 cm(2)) showed a higher probability of response in both arms. Pat ients in the experimental arm had a higher but not statistically signi ficant cumulative progression-free probability. Median survival was 47 .1 weeks overall, while it was 43.7 and 48.5 weeks in the control and experimental arms, respectively. The combination tvas clearly more tox ic than 5FU alone, leukopenia being the most frequent side-effect in t he experimental arm and nausea and vomiting in the control arm. In con clusion these results are quite disappointing and 5FU + IFN alpha-2a c an not be considered a standard treatment for advanced colon cancer.