Patterns of morphologic alteration in residual rectal carcinoma following preoperative chemoradiation and their association with long-term outcome.

Preoperative radiation (RT) and chemotherapy improve outcome in patients with locally advanced rectal adenocarcinoma and, therefore, have been used increasingly in patient management. The histopathologic alterations in postirradiated rectal adenocarcinoma and their prognostic significance have not been fully characterized. In this study, detailed analyses of morphologic alterations of stromal and tumor cells were performed in a series of 66 posttreatment rectal carcinomas, and the pathologic findings were correlated with long-term outcome. All tumors were locally advanced, with a bulky and/or tethered tumor or endorectal ultrasound or magnetic resonance imaging evidence of T3-4 and / or N1 disease. All patients were treated at one institution with preoperative RT to the pelvis (at least 4500 cGy) with or without concurrent 5-fluorouracil (5-FU)-based chemotherapy 4 to 7 weeks prior to surgical resection. Pathologic assessment showed some treatment response in all patients. Nine patients (13.4%) had complete response, and 8 (11.9%) had near-complete response (> 95% of the tumor replaced by fibroinflammatory tissue). Salient morphologic features included marked fibrosis with or without prominent inflammatory cells replacing neoplastic glands; lack of active tumor necrosis; increased mucin production and mucin pools; marked cytoplasmic eosinophilia, often in combination with marked nuclear atypia but without active mitoses in tumor cells showing treatment effect; endocrine tumor phenotype; and retention of mucosal adenoma in the presence of tumor regression within the bowel wall. With a median follow-up of 69 months, the estimated 5-year recurrence-free survival (RFS) for the entire group was 79%. By univariate analysis, the residual tumor stage (P < 0.05) and reduction of pretreatment T stage (P = 0.002) significantly correlated with RFS, as did pN stage (P = 0.002) and lymphovascular invasion (P = 0.008). The extent of treatment response did not correlate with RFS (P = 0.4). However, patients with a treatment response > or = 95% seemed to fare better than those with a treatment response < 95% (marginally significant difference in RFS, P = 0.057). Univariate and multivariate analyses identified the following morphologic patterns that were significantly associated with a reduced RFS independent of other risk factors: a fibrotic-type stromal response with minimal inflammatory infiltrates (P = 0.001) and absence of surface ulceration (P = 0.026). Our study represents the first detailed morphologic assessment of rectal carcinomas that have been subjected to long course preoperative RT and chemotherapy. Our results demonstrate distinct morphologic features in treated rectal carcinomas that are prognostically relevant.

Duke Scholars

Author Harvey Gorden Moore III Surgical Oncology