Correlation between magnetic resonance imaging and definitive histological response in adenocarcinoma of middle and low rectum after neoadjuvant treatment - 02/07/25

Doi : 10.1016/j.jviscsurg.2025.04.006

Jeanne Vertier ^a,⁎ , Rémi Grange ^b, François Casteillo ^c, Clément Costanza ^d, Loïc Campion ^e, Bertrand Le Roy ^a, Laura Ornella Perotto ^a
^a University Hospital, Department of Digestive Oncology and Hepatobiliary Surgery, Saint-Étienne, France
^b University Hospital, Department of Radiology, Saint-Étienne, France
^c University Hospital, Department of Anatomy and Pathological Cytology, Saint-Étienne, France
^d Clinique Mutualiste, Department of Digestive Surgery, Saint-Étienne, France
^e Institut de Cancérologie de l’Ouest, Nantes, France

^⁎Corresponding author at: Department of Digestive Oncology and Hepatobiliary Surgery, Saint-Étienne University Hospital, avenue Albert Raymond, 42270 Saint-Priest-en-Jarez, France.Department of Digestive Oncology and Hepatobiliary Surgery, Saint-Étienne University Hospitalavenue Albert RaymondSaint-Priest-en-Jarez42270France

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Wednesday 02 July 2025
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Highlights

•	TRG score is a marker that predicts survival outcomes for good and poor responders after NAT.
•	There is 63.2% of concordance between mrTRG and pTRG.
•	MRI has a good NPV (81.4%) and therefore MRI is more accurate with poor tumor regression.
•	mrTRG cannot be as surrogate for pTRG in identifying patient with complete pathological response after neoadjuvant treatment in locally advanced rectal cancer.

Le texte complet de cet article est disponible en PDF.

Summary

Background

Data Rectal preservation strategies are being developed for small tumors in complete or good response on magnetic resonance imaging (MRI) after neoadjuvant treatment. Therefore, correlation between tumor regression on MRI and definitive histology is not clearly known. The aim of the present study is to show this correlation to see if MRI can be a reliable tool to propose a rectal preservation strategy.

Methods

All patients over 18 years of age with non-metastatic adenocarcinoma of the lower or middle rectum who have received neoadjuvant treatment followed by a re-assessment MRI prior to surgery, between 2015 and 2023 were retrospectively included. Tumor regression on MRI was defined using mrTumor Regression Grade (mrTRG) classification. Histological tumor regression grade (pTRG) was defined according to the Mandard classification. The statistical relationship between pTRG and mrTRG was determined by univariate logistic regression, with calculation of the odds ratio.

Results

76 patients were included. Most of the patients (57%) received chemoradiotherapy based on CAP50 and 26% received total neoadjuvant treatment. We found 63% concordance between mrTRG and pTRG. Moreover, among the 37% of patients for whom mrTRG and pTRG were not concordant, MRI overestimated the histological response in 71% of cases. MRI has a NPV of 81% (95% CI 73%–90%). Concordance of mrTRG and pTRG is significantly associated with mrT (P=0.026), mrTRG (P=0.002), endoscopic stenosing aspect (P=0.034) and respect of fascia recti on MRI (P=0.021).

Conclusion

In conclusion, this analysis reveals 63% concordance between mrTRG and pTRG. Moreover, MRI has a NPV of 81% and therefore MRI is more accurate with poor tumor regression. Thus, mrTRG must be used in association with other clinical or endoscopic outcomes to assess a rectal preservation strategy.

Le texte complet de cet article est disponible en PDF.

Keywords : Rectal cancer, Tumor regression, MRI, Histology, Neoadjuvant treatment, Rectal preservation

Abbreviations : CRT, MRI, mrTRG, NAT, pCR, pTRG, TNM, TNT, TRG