Imaging Assessment of the Response to Neoadjuvant Treatment in Rectal Cancer in Relation to Postoperative Pathological Outcomes.

Abstract

Background and objective: Rectal cancer remains a global health challenge with significant morbidity and mortality. Neoadjuvant therapy (NAT) is critical for downstaging tumors and improving surgical outcomes. This study aimed to compare the performance of magnetic resonance imaging (MRI) and computed tomography (CT) in evaluating tumor response to NAT in rectal cancer patients, focusing on imaging accuracy and correlation with histopathological findings.

Materials and methods: This retrospective study analyzed 201 patients diagnosed with rectal cancer who underwent NAT followed by surgery. Imaging evaluations were conducted pre-NAT and post-NAT using MRI or CT. Tumor responses were categorized into complete response, downstaging, stable disease, and upstaging. Statistical analyses included Chi-square, Mann-Whitney U, and Z-tests to assess differences in imaging performance and concordance with pathological outcomes.

Results: Both imaging modalities identified significant downstaging post-NAT, with CT demonstrating a higher sensitivity for detecting complete responses (12.5% vs. 6.61% for MRI) and better local staging accuracy. CT showed superior performance in systemic metastasis detection but had limitations in distinguishing fibrosis from residual tumors. Overall diagnostic accuracy was 42.29%, with MRI having a statistically significant edge in detecting favorable tumor responses (p=0.043). The combined use of CT and MRI provided complementary insights, enhancing clinical decision making.

Conclusions: MRI excels in local tumor evaluation post-NAT, offering detailed imaging crucial for surgical planning and conservative strategies like "watch-and-wait" in complete responders. CT remains indispensable for systemic staging and metastasis detection. Integrating both modalities within a standardized protocol can optimize staging accuracy and treatment outcomes in rectal cancer.