ULTRASTAGING OF NON SENTINEL LYMPH NODES (SLNS) DOES NOT BENEFIT STAGING OF COLORECTAL CANCER (CRCA)
Publishing Number: 277
S. Saha, D. Wiese, M. Nolff, A. Dan, E. Schochet, R. Aravapalli, S. Jaswani, S. Arora, J. Badin, McLaren Regional Medical Center, Flint, MI
Introduction SLN mapping accurately stages many solid tumors including CRca. SLNs are 3 - 5 times more likely to harbor metastatic disease when ultrastaged by multilevel microsections and immunohistochemistry (IHC) as compared to non-SLNs examined by standard pathological methods. It is unknown whether such ultrastaging of the non-SLNs would lead to higher incidence of +ve nodes. Hence, we retrospectively analyzed all non-SLNs by multilevel microsections and IHC identical to the SLNs in 136 consecutive patients (pts) with CRca undergoing SLN mapping to determine its impact on nodal staging. Methods All pts underwent intraoperative subserosal injection of 1% lymphazurin around the tumor. The first 1 to 4 blue stained nodes were marked as SLNs, followed by a standard oncologic resection. All SLNs were examined by 4 sections for H&E at 20 to 40 micron intervals and 1 section for cytokeratin (AE-1/AE-3). All non-SLNs were initially examined by a single H&E section and initial staging designations were made according to AJCC criteria. For this study, all non-SLNs were subsequently re-examined with 3 additional H&E sections and one for IHC by a senior pathologist blinded to previous results. Results There were 106 pts with colon and 30 pts with rectal cancer (median age = 70yrs; M=64 : F=72). The mapping was successful in 99% of pts. A total of 1852 (13.62/pt) nodes were identified, of which 319 (17.2%) were SLNs and 1533 (82.7%) were non-SLNs. Metastases (mets) were detected in 21.3% of SLNs and 6.9% of non-SLNs (p<0.0001). The exclusive site of nodal metastases was detected in 8.8% (28/319) of SLNs, in contrast to 0.6% (9/1533) of non-SLNs (p<0.0001). Skip mets were found in 4.5% of pts. After additional ultrastaging of all non-SLNs, only 0.56% (8/1427) of initially -ve non-SLNs became +ve for mets in 8 different pts. Of these 8 pts, 7 already had +ve SLNs, hence no change of AJCC staging occurred. Only in 1 out of 136 pts, 1 non-SLN out of 39 initially negative nodes (including 3 SLNs) was found to contain a cluster of tumor cells seen only by IHC. Thus, such ultrastaging of 1427 non-SLNs changed the AJCC staging from stage II to III only in 0.7% of pts. The accuracy of identifying the presence or absence of nodal mets by SLN mapping was 96%. Conclusion SLN mapping is highly successful and accurate in staging CRca. The chance of finding additional mets by ultrastaging of all non-SLNs is extremely low (<1%), hence of little benefit. Therefore, ultrastaging restricted to SLNs alone will assure accurate staging of CRca.