Are We Overestimating Rectal Cancer Nodal Involvement?

Abstract

<h3>Purpose/Objective(s)</h3> Per current guidelines, rectal cancer is staged with pelvic MRI (or EUS if MRI contraindicated) to determine T- and N-stage. Those with ≥ clinical stage T3 (cT3) or N1 (cN1) are recommended to neoadjuvant treatment (NT). The specificity of MRI for N-staging of rectal cancer is ∼ 70% - this is lower than the specificity of MRI for T-staging. Thus, for those cancers for which NT is recommended based on N-staging, there is the potential that up to 30% of patients are over-staged. The current retrospective study aims to quantify the percentage of clinically lymph node positive (cN+) rectal cancers treated with NT that have pathologic complete nodal response (ypN0) with no treatment effect noted at surgical resection. <h3>Materials/Methods</h3> Cases of cN+ rectal cancer treated at our COC-Accredited INCP institutions between 2015 - 2020 were identified. Data were collected with regard to staging modality, clinical T- and N-stage, size of largest suspicious lymph node (LN) on MRI, NT-type (total neoadjuvant therapy (TNT) with capecitabine/RT (CRT), TNT with short course RT (SCRT), or neoadjuvant radiation or chemoradiation alone), ypN-stage, and treatment effect. We compared proportions of ypN0 cases by cT- and cN-stage, size of largest suspicious LN on MRI, and NT-type using multivariable logistic regression. <h3>Results</h3> A total of 64 cases of cN+ rectal cancer were identified. Fifty-three were staged by MRI, 2 by EUS, 2 by PET, and 7 by CT. Primary tumor staging was as follows: cTx = 2, cT1 = 1, cT2 = 10, cT3 = 36, cT4 = 15. Clinical N-stage was as follows: cN1 = 41, cN2 = 23. For 13 patients (20%), the recommendation for NT was based entirely on MRI N+ staging. Fifty-seven patients were treated with TNT: All with FOLFOX followed by 44 with CRT and 13 with SCRT. Seven patients were treated with neoadjuvant CRT or SCRT without neoadjuvant FOLFOX. Pathologic results showed residual ypN2 disease in 4 cases, ypN1 disease in 19 cases and ypN0 disease in 42 cases (65%) with treatment effect (n = 5) or possible treatment effect (n = 3) noted in 8 of those cases. ypN0 disease was seen in 57% (n = 4) receiving no TNT, 46% (n = 6) receiving TNT with SCRT, and 72% (n = 32) receiving TNT with CRT. Of the 13 patients with cTx-T2 tumors who received NT based on cN+ status, 10 (77%) had ypN0 disease and 1 had signs of treatment effect in a LN. On multivariable logistic regression, cN-stage (cN1 vs cN2) predicated for ypN+ disease with a hazard ratio of 5.15 for cN2 disease (<i>p</i> = 0.0076). cT-stage, size of largest suspicious LN on MRI and NT-type did not predict for yp nodal status. <h3>Conclusion</h3> In patients with cN+ rectal cancer (staged predominately by MRI), more than half of patients had no pathologic residual nodal disease and no treatment effect noted in the LN. While this finding is in line with the results of major randomized trials such as RAPIDO, it confirms a 65% rate of ypN0 disease following neoadjuvant treatment for cN+ rectal cancer in the setting of a 70% specificity of MRI for rectal cancer nodal staging. cN2 disease predicts for residual pathologic LN disease.