EP714 - Low Tumor Location and Circumferential Morphology Are Associated with Radiolographic–Pathologic Discordance after Total Neoadjuvant Therapy in Rectal Cancer

Purpose/Background: Current management of locally advanced rectal cancer requires a multidisciplinary approach that includes radiographic staging, endoscopic evaluation, and potential surgical resection following Total Neoadjuvant Therapy (TNT). Despite this, discrepancies arise between pre-operative MRI staging and final pathology. We aimed to identify clinical factors associated with radiographic-pathologic discordance at time of surgery for patients with rectal cancer treated with TNT.

Methods/Interventions: We queried our institutional database for patients ≥18 years old from 2011-2024 with rectal adenocarcinoma who received TNT. Exclusion criteria included ≥90 days between post-TNT MRI and surgery, absence of post-TNT endoscopy, and treatment with immunotherapy. Patients were considered concordant when post-TNT MRI and pathology showed the same T and N stage, and discordant when any discrepancy in T or N stage existed. We used point biserial correlation and Fisher's exact test to examine associations between clinical variables and concordance. For patients with residual polyp tissue on post-TNT endoscopy, we analyzed the characteristics of the lesion associated with concordance as a secondary endpoint.

Results/Outcomes: The final analysis included 48 patients. 36 (75%) were male. The median age was 55.5 (IQR 48.25-63.75). The majority of tumors were initially staged as T3 (n=28, 58.3%) and T4 (n=17, 35.4%) with location in the mid (n=24, 50%) and lower (n=22, 45.8%) rectum. Most patients received induction TNT (n=45, 93.8%). 6 (12.5%) patients underwent abdominoperineal resection (APR) whereas 42 (87.5%) underwent low anterior resection (LAR). 13 patients had persistent circumferential tumors post-TNT. The median time from end of treatment to MRI and end of treatment to surgery was 35 days (IQR 26.5-47.75) and 75.5 days (IQR 68.25-93), respectively. The median time from MRI to surgery was 40 days (IQR 29-55.25).



33 (68.8%) of tumors were pathologic-radiologic discordant, of which 7 (14.6%) had complete pathologic response. 22 (66.7%) were isolated T stage discordance, 7 (21%) had N stage discordance, and 4 (12%) had both T and N stage pathologic discordance with post-treatment, pre-operative MRI. Of the 26 patients with T stage discordance, 13 (50%) were upstaged on pre-operative MRI compared to surgical pathology and 13 (50%) were downstaged. Of the 11 patients with N stage discordance, 4 (36.4%) were upstaged on pre-operative MRI compared to surgical pathology and 7 (63.6%) were downstaged.



There was no significant association between discordance and the time from pre-operative MRI to surgery, time from end of TNT to MRI, or time from end of TNT to surgery. Low tumors close to the anal verge (AV) were associated with N stage discordance by pre-operative MRI (p=0.04). Among the 39 patients who had residual polyp tissue on post-TNT endoscopy, circumferential lesions after TNT were associated with N stage discordance (OR = 4.7, 95% confidence interval 1.10-5.64, p=0.039).

Conclusion/Discussion: Tumors closer to the AV and persistent circumferential lesions were associated with radiologic-pathologic discordance. We posit that low, bulky tumors may obscure lymph node evaluation with post-TNT MRI and should be approached with caution in surgical planning. Further studies are needed to identify risk factors for radiographic-pathologic discordance to improve outcomes in this disease.