Prospective Evaluation of IMRT for Anal and Perianal Cancer: Early Patterns of Failure

IMRT and concurrent chemotherapy is standard for anal and perianal cancer at our center. An REB-approved prospective cohort study was conducted to evaluate toxicity, quality of life and clinical outcomes. Updated results and early patterns of failure are reported. From June 2008-Nov 2010, patients with anal or perianal cancer treated with IMRT were enrolled. IMRT dose was 36 Gy/20f for elective and 54 Gy/30f-63 Gy/35f for gross targets using standardized contouring, dose prescriptions, and dose-volume criteria. Chemotherapy (weeks 1 and 5) consisted of 5-FU 1g/m2/d (max 1.5g/d), d1-4 and MMC 10 mg/m2, d1. Toxicity was graded using NCI-CTC v3. Overall survival (OS), disease-free survival (DFS) and colostomy-free survival (CFS) were estimated using the Kaplan-Meier method. Locoregional failure (LRF) was estimated using the cumulative incidence method. To analyze patterns of failure, LRF were contoured on the planning CT after registration with the first diagnostic scan showing recurrence. QOL was assessed using the EORTC QLQ-C30 and CR29 questionnaires. Fifty-eight patients were enrolled: median FU 17 months; median age 56; 30 female (52%); 11 HIV positive (19%); 5 Stage I (9%), 32 Stage II (55%), and 21 Stage III (36%). Median dose was 63 Gy to the primary tumor and 58.5 Gy to gross nodes. Twenty-six (45%) patients required a treatment break (1-15d, median 8d) due to acute toxicity, mainly dermatitis (23/26, 88%). Acute Grade 3+ toxicities included skin 46%, hematologic 38%, GI 9%, and GU 0%. Eighteen-month OS, DFS, CFS and LRF rates were 89%, 80%, 88% and 13%, respectively. Four patients with anal cancer experienced local failure (LF) after 63 Gy: 2 (T3N0 and T3N2) had refractory disease; 1 (T3N0) had partial response; and 1 (T4N2) developed in-target (IT) LF 10 months post-treatment. Two patients developed nodal failure after 36 Gy: 1 (T3N1 perianal cancer) developed IT contralateral inguinal nodal recurrence 9 months post-treatment; and the other (T3N0 anal cancer) developed extensive IT inguinal and iliac nodal recurrences, but the most superior common iliac node was a marginal failure. There were no IT failures in lymph node metastases diagnosed prior to treatment. The median contoured GTV volume of patients with LRF was 116 cc. On univariate analysis, tumor size >5 cm was prognostic for higher LRF (29% vs. 0%, p = 0.004). Global QOL, skin, pain and defecation scores were clinically and statistically significantly worse at end of treatment compared to baseline, but returned to baseline 3 months post-treatment. IMRT produces favorable acute Grade 3+ hematologic and GI toxicities, yet still frequent dermatitis requiring treatment breaks, without persistent negative impact on QOL. We report good 18-month OS, DFS and CFS. Bulky primary tumors have higher LRF despite 63 Gy.