(P096) Preoperative Radiation Dose Escalation for Rectal Cancer Improves Tumor Downstaging Without Significant Increase in Toxicity: A Matched-Pair Analysis

David Weksberg, C.V. Patel, K.V. Kattepogu, Jonathan D. Grant, I.J. Park, Cathy Eng, Marc E. Delclos, Barry Feig, Chris H. Crane, J.M. Skibber, Prajnan Das, Miguel A. Rodriguez-Bigas, Bruce D. Minsky, Sunil Krishnan, George Chang; UT MD Anderson Cancer Center

PURPOSE: With large studies demonstrating improved sphincter preservation and local control following neoadjuvant chemoradiation therapy (NCRT) for rectal cancer, this approach has been established as standard in locally advanced or node-positive disease. While a tantalizing minority of patients enjoys pathologic complete response (pCR) at the time of surgery, ~50% of lesions fail to be downstaged by treatment. We sought to explore a potential dose-response relationship with tumor downstaging after NCRT. To this end, we conducted a case-control analysis of 152 patients treated preoperatively, with and without a concomitant boost.

METHODS: From 1995–2003, 76 patients were enrolled on an institutional review board (IRB)-approved phase II protocol examining the feasibility of adding a concomitant boost to NCRT. Patients received venous infusion 5-fluorouracil (5-FU) and 52.5 Gy in 5 weeks-42.5 Gy at 1.8 Gy per fraction daily, plus a concomitant boost of 1.5 Gy per day in the final week of treatment. Using a case-control approach, 76 additional patients were identified who received low-dose NCRT (5-FU plus 45 Gy in 1.8-Gy fractions), matching for type of surgery performed, tumor stage (T-stage), and clinical nodal metastases (N-stage). For all 152 patients, chart review was undertaken with respect to clinicopathologic parameters and treatment outcomes. Radiation toxicity and surgical complications were assessed in a subset of 63 pairs with complete toxicity data available. McNemar’s chi-square test and Kaplan-Meier analyses were used as appropriate.

RESULTS: The high-dose (HD) and low-dose (LD) cohorts were well matched; in addition to the selection criteria, there were no significant differences in gender (HD 64.5% male vs LD 65.8% male) or age (HD 57.1 yr vs LD 55.9 yr). For a subset of patients, information regarding circumferential involvement (50% of patients) and tumor length was available (65% of patients); there were no significant differences in these criteria. A statistically significant difference in distance from the anal verge was noted (HD 4.7 cm vs LD 5.7 cm; P < .03).

The rate of tumor downstaging was substantially improved in the HD cohort, with 76% of patients found to have reduction in T-stage vs 51% in the LD arm (P < .01). However, the rates of pCR did not differ significantly (HD 17.1% vs LD 15.8%). Toxicity data reveal that T-downstaging did not appear to come at a cost of substantially increased toxicity. While ~90% of patients experienced some radiation toxicity, the incidence of grade 3 or higher toxicities was low (HD 9% vs LD 3%), as was the rate of wound complication (HD 8% vs LD 7%), with no statistically significant differences between the cohorts.

CONCLUSIONS: Our results suggest that dose escalation in NCRT is feasible without significant increase in radiation toxicity or surgical complication. While pCR rates did not differ between HD and LD NCRT groups, the increased T-downstaging seen in the HD arm supports a dose-response relationship with tumor downstaging and suggests further exploration of dose escalation as a tool to improve sphincter preservation rates in appropriately selected patients.