Predictive value of rectal MRI variables for pathological complete response in locally advanced rectal cancer following neoadjuvant chemoradiotherapy.

PURPOSE: To evaluate the predictive value of MRI-determined variables for pathological complete response (pCR) in locally advanced rectal cancer (LARC) patients following neoadjuvant chemoradiotherapy (NCRT).

METHODS

Clinical data were collected from patients who received NCRT between January 2019 and 2022. Patients with rectal adenocarcinoma, cT3-4N0, or TanyN1-2 were included. pCR was defined pT0N0. Patients were divided into pCR and non-pCR group. Logistic regression analysis was performed to identify factors associated with pCR. A nomogram model was constructed to validate its predictive ability and accuracy.

RESULTS

A total of 585 patients were identified, with 144 (24.6%) in the pCR group and 441 (75.4%) in the non-pCR group. Patients with mrT2-3 (OR 6.41, P < 0.001), mrN0 (OR 2.17, P < 0.001), circumferential occupation range < 1/2 cycles (OR 2.11, P < 0.001), tumor vertical diameter < 36 mm (OR 2.10, P < 0.001), negative mesorectal fascia (OR 3.21, P < 0.001), and extramural vascular invasion (OR 5.68, P < 0.001) were more likely to achieve higher pCR rates. Logistic regression analysis revealed that mrT2-3 (OR 3.50, P < 0.001), tumor vertical diameter < 36 mm (OR 2.57, P < 0.001), and negative extramural vascular invasion (OR 4.03, P < 0.001) were independent protective factors for pCR. A nomogram was developed to predict pCR, achieving a C-index of 0.778.

CONCLUSION

Patients with mrT2-3, tumor vertical diameter < 36 mm, and negative extramural vascular invasion are more likely to achieve pCR after NCRT.