The impact of treatment factors on local control in T2‐T3 anal carcinomas treated by radiotherapy with or without chemotherapy

BACKGROUND This study was conducted to investigate the influence of therapeutic parameters on local control (LC) in the sphincter‐conserving treatment of T2‐T3 anal carcinoma. METHODS From 1976 to 1993, 137 patients with anal carcinoma classified as T2 (85 patients) or T3 (52 patients) were treated curatively by radiotherapy (RT) alone (54 patients) or by concomitant chemotherapy and RT (83 patients). RT was delivered in two sequences, with a median gap of 46 days between the sequences. The first sequence was delivered at a median dose of 39.6 gray (Gy) using megavoltage photon beams. Boost treatment consisted of either 192Ir implantation or external beam RT (median dose, 20 Gy). Chemotherapy started on Day 1 and generally consisted of 1 cycle of mitomycin C (10 mg/m2) and a 5‐day infusion of 5‐fluorouracil (600‐800 mg/m2/day). For surviving patients, median follow‐up was 83 months. Univariate and multivariate analyses were performed to determine therapeutic parameters affecting LC after adjustment for clinical factors. RESULTS The 5‐year actuarial LC was 76%. Factors associated with poorer LC (univariate) were as follows: age < 66 years (LC was 67% with the factor vs. 85% without), male gender (65% vs. 81%), tumor extent > 1/3 canal circumference (67% vs. 90%), lymph node involvement (64% vs. 81%), use of external beam boost (62% vs. 79%), and overall treatment time (OTT) ≥ 75 days (69% vs. 85%). In multivariate analysis, no therapeutic parameters remained significant when adjusted for significant clinical factors, although OTT was of borderline significance (P = 0.09). CONCLUSIONS The results of this multivariate analysis suggest that therapeutic factors have a less marked effect on LC compared with clinical parameters; the only factor that appeared to have some effect was OTT. Efforts to improve LC in patients with poor prognoses should concentrate on optimizing OTT and the chemotherapeutic aspects of treatment (in other words, attempts should be made to provide more effective agents and optimize scheduling). Cancer 1997; 79:2329‐35. © 1997 American Cancer Society. The results of this multivariate analysis suggest that therapeutic factors have a less marked effect on local control compared with clinical parameters; the only factor that appeared to have some effect was overall treatment time.