In our series, all patients were treated with doses from a minimum of 15 Gy up to 18.5 Gy and no neuropathy was observed. Three patients (19%) had Grade 3 complication as reported by the physician during followup visit: one had urethral stenosis and two others had fistulas (rectovaginal and ureter). Nonetheless, it is not easy to separate treatment-related local complications in patients with recurrent learn more tumors previously treated with surgery and radiation therapy. Previously published data from our institution described the most common type of toxicity: wound (24%), ureter (23%), and bladder (20%) complication in patients with recurrent colorectal cancer who received

HDR-IORT without DP (14). Despite the use of HDR-IORT, local failure can occur in up to 50% of patients [2] and [8]. Resection margin status has been shown to be the primary predictor of local failure. In a prior study from our institution, patients with R1 or R2 resections had a median time to local failure of 5-Fluoracil 38 vs. 63 months for patients with an R0 resection (15). Although negative

margins can be obtained microscopically in a second radical resection, in previously irradiated patients, clear margins are difficult to achieve even by the most experienced surgeons in high-volume cancer centers. The cohort of patients receiving IORT-HDR-DP was particularly high risk for positive margins as all patients had recurrent disease and previous EBRT. Yet, despite positive microscopic margins in 25% of our patients, the 2-year LC was excellent (80%), suggesting that IORT was effective as an adjuvant treatment. Given the small cohort of patients and its retrospective nature, we cannot draw definitive conclusions related to survival outcomes. Also, Verteporfin cell line owing to the lack of a control group, we cannot evaluate the real impact of HDR-IORT-DP in LC compared with regular HDR-IORT without DP. The largest single-institution experience in IOERT on recurrent colorectal cancer (n = 607) from the Mayo Clinic showed a 3-year local and distant relapse incidence of 23% and 49%, respectively (2). In their series, 37% of

the resections were R1. Interestingly, despite comparable LC rates to the Mayo Clinic series, the DM rate (69%) was higher in our cohort, potentially demonstrating more advanced disease at the time of surgery or more aggressive tumor biology in our group of patients who had tumors of other sites in addition to colorectal cancers. Local recurrence after previous EBRT also seems to be an unfavorable factor. The Mayo Clinic series reported 5-year survival of 20% in patients with recurrent colorectal cancer without prior radiation vs. 12% in previously irradiated patients (16). In our study of previously irradiated patients, the 2-year actuarial OS was 20%. DM was the major problem in our series because about two-thirds of the patients developed DM and died of disease.