Lymphatic tumor emboli detected by D2-40 immunostaining can more accurately predict lymph-node metastasis

Resected specimens of superficial squamous cell carcinoma of the esophagus (SSCCE) underwent D2-40 immunostaining to accurately assess lymphatic tumor emboli (LY) and to analyze correlations between LY and lymph node metastasis (N). This present study was designed to determine the accuracy of LY grade for predicting the risk of N.

We studied 75 patients with SSCCE who underwent surgical resection of their tumors. Resected specimens were sliced into continuous sections at 5 mm intervals. Intramucosal cancers are classified into three groups (m1, m2, m3), and submucosal cancers are also divided into three groups (sm1, sm2, sm3). The numbers of LY present in lymphatic ducts on D2-40 immunostaining, venous tumor emboli (V) on CD34 immunostaining, and lymphatic tumor emboli (ly) and V on hematoxylin-eosin staining (HE) and elastica van Gieson staining (EVG) were counted for each case. The presence of lymphatic tumor emboli was graded according to the total number of LY per case as follows: 0, LY0; 1 to 2, LY1; 3 to 9, LY2; and 10 or more, LY3.

All m1 and m2 cases were LY- and N- Lymphatic tumor emboli were present in 54% of m3 cases, 70% of sm1 cases, 54% of sm2 cases, and 75% of sm3 cases. Determination of N was positive in 18% of m3 cases, 47% of sm1 cases, 36% of sm2 cases, and 62% of sm3 cases. The frequency of LY significantly correlated with the number of N (p < 0.0001). Multiple regression analysis showed that only LY and V significantly correlated with N. When the detection rate of N was compared between LY and ly, LY was superior to ly in terms of specificity, accuracy, positive predictive value, and false positive rate. As for LY grade, N was positive in 39.1% of LY1 cases, 81.8% of LY2 cases, and 100% of LY3 cases. Even in LY-, N was positive in one sm1 case and in two sm2 cases. In the sm1 case, the depth of invasion was 350 μm from the lower margin of the muscularis mucosae.

Evaluation of lymphatic invasion on the basis of LY is more accurate for the prediction of N than conventional techniques, and LY grade strongly correlates with N. In patients with SSCCE, mucosal cancers (m1, m2, and m3) and submucosal cancers with a depth of invasion of ≤ 200 μm from the lower margin of the muscularis mucosae on endoscopic mucosal resection have a low risk of N if the number of LY is 0. Endoscopic mucosal resection alone can provide good treatment outcomes in such patients.