Concordance between clinical and pathologic assessment of T and N stages of rectal adenocarcinoma patients who underwent surgery without neoadjuvant therapy: A National Cancer Database analysis.

Abstract

Clinical assessment of T and N stages in rectal cancer is important to guide decision-making. The present study aimed to assess the accuracy of the clinical T and N staging of rectal cancer compared to the pathological staging and their overall agreement in a large cohort of patients. This retrospective study used data from the National Cancer Database (NCDB) between 2004 and 2017. Patients with non-metastatic rectal adenocarcinoma who did not receive neoadjuvant therapy were reviewed and the clinical T and N stages were compared to their pathologic counterparts. The overall concordance between clinical and pathologic assessments was calculated using Kappa coefficient. The study included 8929 patients (57.3% male) with a mean age of 64 years. Clinical T stage and N stage were identical to pathologic stages in 70.3% and 77.6% of patients, respectively. Sensitivity and specificity of the clinical assessment of N stage was 35.2% and 95.5%, respectively. Concordance between the clinical and pathologic stages was moderate for the T stage (kappa=0.575) and fair for the N stage (kappa=0.346). Pathologic T4 stage (OR: 2.12, p<0.0001), poorly differentiated adenocarcinoma (OR: 1.45, p=0.026), lymphovascular invasion (OR: 4.5, p<0.001), and longer time from diagnosis to first treatment (OR=0.996, p=0.046) were the independent predictors of N stage discrepancy. There was a moderate agreement between the clinical and pathologic T stages and a fair agreement between the clinical and pathologic N stages. The clinical assessment of the N stage was highly specific yet had low sensitivity.