Impact of clinical and therapeutic factors on major late complications after radiotherapy with or without concomitant chemotherapy for anal carcinoma

Purpose: To investigate factors potentially influencing major late morbidity after sphincter-conserving treatment for anal carcinoma. Methods and Materials: Grade 3-4 complications were retrospectively analysed in 144 evaluable patients (pts), 55 pts after split-course radiotherapy (RT), and 89 after concomitant chemo-RT. First sequence RT delivered a median dose of 39.6 Gy using megavoltage photon beams, Boost treatment used either Ir-192 implantation or external beam RT (median dose 20 Gy), Chemotherapy started on day 1 and in 83% of pts consisted of Mitomycin-C (10 mg/m(2)) and a 5-day infusion of 5-fluorourcil (600-800 mg/m(2)/day), Uni-and multivariate analyses tested the association of following factors with complication rate: age, gender, stage, anatomic tumor extent, type of biopsy, external RT technique (dose, fraction size, field arrangement), boost type (brachytherapy vs, external), brachytherapy dose and dose rate, overall treatment time, and addition of chemotherapy. Results: Five-year actuarial complication rate was 16%, Two variables were significantly associated with complication rate: anatomic tumor extent (canal or margin vs. both +/- rectum; 10 vs, 31% complications, p = 0.0004) and first sequence prescribed dose (<39.6 Gy vs. greater than or equal to 39.6 Gy; 7 vs, 23% complications, p = 0.012), confirmed as independent factors by Cox analysis, Grade 4 anal morbidity correlated significantly with prior local excision, All six bone complications were observed in pts treated by chemo-RT using large pelvic fields, five occurring in pts older than 66. Conclusion: Pts with tumors involving more than one anatomic subsite or treated with the higher first sequence RT dose are at greater risk of major complications. Prior tumor excision and combined modality therapy in older pts appear to favor major anal and bone complications, respectively. (C) 1997 Elsevier Science Inc.