Résumé
To retrospectively determine whether magnetic resonance (MR) volumetry of rectal cancer is a reproducible method for predicting disease-free survival (DFS) in patients with locally advanced low or midrectal tumors who undergo combined chemotherapy and radiation therapy (CRT) before total mesorectal excision.
The institutional review board does not require approval for the use of patient data obtained for an observational retrospective study. Fifty-eight patients were included in the study; 42 patients had low-lying tumors. Two radiologists independently measured tumor volumes before and after CRT with use of semiautomated software. The radiologists were blinded to the clinical information for each patient. The tumor volume reduction ratio, circumferential resection margin, T stage, and occurrence of downstaging were compared with the histopathologic response and DFS. The threshold of tumor volume reduction for predicting DFS was assessed with receiver operating characteristic curve analysis. DFS was estimated with the Kaplan-Meier method and compared between groups with the log-rank test.
The interobserver correlation coefficient between the two radiologists was 0.87 (95% confidence interval [CI]: 0.76, 0.93) for pre-CRT volumetry and 0.81 (95% CI: 0.74, 0.90) for post-CRT volumetry. A tumor volume reduction of at least 70% was significantly associated with good histologic regression (tumor regression grade [TRG], 3 or 4) (P <.0001) compared with a volume reduction rate of less than 70%. DFS was studied in 51 patients. The mean follow-up of survivors at the time of analysis was 52 months ± 20 (standard deviation). Patients with a volume reduction ratio of at least 70% had a higher DFS (P <.0001). Tumor volume reduction was an independent prognostic parameter in multivariate analysis for DFS (P = .003; 95% CI: 0.01, 0.4).
The results demonstrate that volumetric measurements are reliable markers of rectal cancer prognosis, enabling the prediction of DFS and TRG. The cutoff of 70% is an easy parameter to use as a surrogate for clinical response to predict both TRG and outcome.