EPD106 - The Ohio Valley Society of Colon & Rectal Surgeons Award Best ePoster: Restaging MRI for Mesorectal Lymph Node Diagnosis After Neoadjuvant Therapy: Is It Still Reliable?

ePoster of Distinction Presenter(s)

• 

  Ibrahim A. Gomaa, M.D

  Research Fellow
  Mayo Clinic
  Mayo Clinic
  ROCHESTER, MN, United States

  All disclosed relationships have been reviewed and managed prior to the start of this activity. Any relevant relationships will be listed below. This individual has no financial relationships to disclose

Author(s)

• 

  Sara A. Aboelmaaty, MD

  Research Fellow
  Mayo clinic
  Mayo Clinic
  Rochester, MN, United States

  Disclosure information not submitted.
• DC

  Dovile Cerkauskaite, MD

  Research Fellow
  Mayo Clinic
  Rochester, Minnesota, United States

  Disclosure information not submitted.
• NM

  Nicholas P. McKenna, M.D.

  Assistant Professor of Surgery
  Mayo Clinic
  Rochester, Minnesota, United States

  Disclosure information not submitted.
• 

  William R. Perry, MBChB MPH FRACS FACS FASCRS (he/him/his)

  Associate Professor of Surgery
  Mayo Clinic
  Mayo Clinic
  Rochester, Minnesota, United States

  All disclosed relationships have been reviewed and managed prior to the start of this activity. Any relevant relationships will be listed below. This individual has no financial relationships to disclose
• 

  Eric Joseph Dozois, Professor of Surgery

  Chair, Colon and Rectal Surgery
  Mayo Clinic
  Rochester, Minnesota, United States

  All disclosed relationships have been reviewed and managed prior to the start of this activity. Any relevant relationships will be listed below. This individual has no relevant financial relationships.
• 

  Kellie L. Mathis, M.D. (she/her/hers)

  Professor of Surgery
  Mayo Clinic
  Mayo Clinic
  Oronoco, Minnesota, United States

  Disclosure information not submitted.

Senior Author(s)

• 

  David w. Larson, MD MBA

  consultant
  Mayo Clinic
  Mayo Clinic
  rochester, MN, United States

  All disclosed relationships have been reviewed and managed prior to the start of this activity. Any relevant relationships will be listed below. This individual has no financial relationships to disclose

Purpose/Background: Restaging magnetic resonance imaging (MRI) is routinely performed to assess mesorectal lymph node (LN) status after neoadjuvant therapy in locally advanced rectal cancer (LARC). However, diagnostic performance remains uncertain, particularly when used to support non-operative management decisions. Therefore, this study evaluated the diagnostic performance and concordance of restaging MRI for mesorectal nodes versus pathology.

Methods/Interventions: Patients diagnosed with stage II and III rectal cancer between 2014 and 2023 who underwent baseline MRI staging, neoadjuvant therapy, restaging MRI, and proceeded to surgical resection with pathological LN evaluation were included. Diagnostic performance (sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), and Cohen's κ) was calculated using pathology as the reference standard. Subgroup analyses were prespecified by clinical stage, and receiver operating characteristic (ROC) analysis evaluated restaging LN size (mm) as a predictor of pathological LN positivity. Prespecified subgroup analyses were conducted by clinical stage, EMVI status, tumor level, and cN2 disease.

Results/Outcomes: Of 414 eligible patients (Stage II = 78; Stage III = 336), 10/78 (12.8%) clinical stage II and 97/336 (28.9%) clinical stage III cases were pathologically node positive. On restaging MRI, LN positivity was reported in 3/78 (3.8%) clinical stage II and 113/336 (33.6%) clinical stage III cases (Table 1). Among Stage III, median restaging LN size was larger in pathologically positive versus negative disease (6.0 mm [IQR: 4.0-7.3] vs. 4.0 mm [IQR: 3.0-5.0]), p = 0.0026. Using LN size alone in Stage III, ROC AUC was 0.722 (95% CI 0.656-0.787; p< 0.0001). The optimal cutoff (Youden's J = 4.0 mm) yielded sensitivity=0.667, specificity=0.729, and NPV=0.857. (Figure 1)
In selected subgroups, using size thresholds achieved high rule-out performance: lower rectum (AUC=0.76; 3.0 mm cutoff, NPV=90%), EMVI-negative (AUC=0.81; 4.0 mm cutoff, NPV=90%), and cN2 (AUC=0.77; 3.0 mm cutoff, NPV=92%). Restaging MRI failed to identify 43% of pathologically positive cases.

Conclusion/Discussion: Restaging MRI using LN size alone provides moderate discrimination for pathological nodal status after neoadjuvant therapy. In stage III, a 4-mm cutoff provides a balanced sensitivity and specificity (NPV=0.86), while achieving NPV ≥90% is limited to selected subgroups (e.g., EMVI-negative, lower rectum, cN2). Thus, LN size may support a rule-out in carefully chosen patients, but it is insufficient for universal nodal clearance; external validation and multidisciplinary confirmation are warranted. Notably, the radiologist restaging failed to identify 43% of node-positive cases, underscoring the limitations of restaging MRI as a standalone tool for post-neoadjuvant decision-making.

Table1. Demographics and Preop characteristics
Figure1: ROC Restaging LN size [mm] predicting pathologic node positivity in Stage III
The AUC is 0.722 [95% CI 0.656-0.787], Youden's J threshold = 4.0 mm.