Prediction of lymph node metastasis with binary logistic regression in gastric carcinoma / 中华胃肠外科杂志

ABSTRACT

<p><b>OBJECTIVE</b>To investigate more specific markers to predict the lymph node metastasis in gastric carcinoma.</p><p><b>METHODS</b>The expression of heparanase mRNA was detected by reverse transcription polymerase chain reaction (RT-PCR) in 43 cases with gastric cancer. The expressions of CD44V6, MMP-7, nm23 and syndecan-1 protein were examined by streptavidin-peroxidase (SP) two-step method. Clinicopathological features influencing lymphatic metastasis such as age,sex,tumor size,tumor location, Borrmann classification, histological type, differentiation and serosal infiltration were also analyzed.</p><p><b>RESULTS</b>Twenty-seven cases (62.8%) in 43 gastric cancer patients had lymphatic metastasis. The incidence of metastatic lymph nodes was(36.3 +/- 30.8)%. The median incidence was 19%. Univariate analysis showed that tumor size, serosal infiltration, expressions of heparanase mRNA, CD44V6, nm23 and syndecan-1 protein were risk factors for lymph node metastasis in gastric carcinoma. Multivariate analysis showed expressions of nm23 and syndecan-1 protein, serosal infiltration were independent factors for lymph node metastasis.</p><p><b>CONCLUSION</b>Gastric cancer with serosal infiltration, positive expressions of nm23 and syndecan-1 has greater possibility of lymph node metastasis.</p>