OBJECTIVETo evaluate the accuracy and clinical significance of clinical complete response (cCR) after neoadjuvant themoradiotherapy for locally advanced rectal cancer.

METHODSLocally advanced rectal cancer patients who received neoadjuvant chemoradiotherapy following radical resection were retrospectively assessed for tumor response during 2005 to 2014 from the database of colorectal cancer. The concomitant preoperative chemoradiation consisted of 50 Gy radiation, fractionated within 5 weeks and 5-FU combined with oxaliplatin. Endorectal ultrasound and MRI were applied to preoperative staging, and postoperative gross pathologic inspection was retrospectively employed to evaluate the status of clinical complete response(cCR).

RESULTSA total of 227 patients undergoing radical surgery were enrolled in the study. Complete pathological response (ypT0N0, pCR) was found in 40 patients(17.6%) by postoperative pathologic examination while the rate of node involved in ypT0 patients was 11.1%. The preoperative rectal MRI was more sensitive to correlate ypT0 than endorectal ultrasound and gross pathologic inspection(60.0% vs 19.4% and 17.8%), but the accuracy and specificity showed no significant differences among these three tests. Multivariate Logistic regression analysis revealed preoperative MRI evaluation of cT0 might predict ypT0 independently(OR=4.975, 95% CI: 1.073 to 23.067, P=0.040).

CONCLUSIONIt is difficult to diagnose the primary tumor to be a cCR status based on preoperative MRI, EUS, or ulceration of rectal mucosa, and further to predict pCR. Preoperative MRI is more sensitive. The strategy of "wait and see" for cCR patients after neoadjuvant chemoradiation should be seriously considered in the decision-making before surgery.