Reply to Y.-T. Lan et al

Abstract

We appreciate the thoughtful comments of Lan et al from Taiwan. We are pleased that they feel the Surveillance, Epidemiology, and End Results (SEER) outcomes data for 130,762 patients with colon cancer provide powerful evidence to support the revised staging system for colorectal cancer in the seventh edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual. The letter by Lan et al centers on the finding that a small percentage of patients with Tis colon cancers in the SEER analysis had involved lymph nodes (114 TisN lesions in 5,935 patients with TisN0-N lesions or 1.9%). Lan et al mention the widely held concept that lymphatic vessels are not found in the mucosal layer and therefore lymph node metastases should not occur but note recent immunohistochemical data showing the presence of lymphatic channels extending to colonic mucosa in neoplastic and inflammatory conditions. In Lan’s institution, involved nodes were found in only one of 118 patients with Tis lesions who had a regional resection, and on pathologic review, the lesion was in fact T2N , not TisN , based on extension of the primary lesion into the superficial muscularis propria. Accordingly, Lan et al suggest that slides of all TisN patients in the SEER colon analysis should be re-reviewed, if possible, to be certain they were limited to the mucosa. Given the possibility of involved nodes in patients with intramucosal colorectal cancers, Lan et al also note that for patients with presumed Tis lesions, further investigation is warranted to define the high-risk group before considering local treatment options such as endoscopic mucosal or submucosal resection. Although we did not comment on patients with Tis lesions in our article, we agree with Lan et al that the finding of involved nodes in patients with intramucosal colorectal cancers (Tis) is both unexpected and uncommon (1.9% in SEER colon analysis and 2.1% in SEER rectal analysis [75 TisN lesions in 3,614 patients with TisN0-N rectal cancer). Even though it would be of interest to review the slides of all patients with TisN lesions in the SEER colon and rectal analyses, unfortunately we cannot attempt such. Many of the blocks are now unavailable for study, so a review of available material would be unavoidably biased. Given the extremely low risk of nodal involvement in patients with Tis lesions, it is reasonable to consider local treatment options in such patients, unless high-risk pathologic findings are identified when the lesion is biopsied.