Analise comparativa da eficácia do método imuno-histoquímico (d2-40 e cd34) e coloração de rotina (he) para detecção de invação vascular linfática e sanguinea no carcinoma gástrico: correlação com outros indicadores prognósticos

Abstract

BACKGROUND: Worldwide, gastric carcinoma (GC) is the second largest cause of cancer-related death. The TNM staging system for GC is widely used and it provides important prognostic information, especially with regards to the lymph node status. The presence of lymphatic invasion (LVI) detected on the primary tumor is the strongest risk factor for lymph node metastasis and is known as an independent prognostic factor in patients with GC. Because the evaluation of LVI is subjective with conventional Hematoxylin and Eosin (H&E) staining, it can lead to inaccurate false-positive and false-negative results. D2-40 is a specific monoclonal antibody for the lymphatic endothelium, being useful to identify LVI and distinguish lymphatic from blood vessel invasion (BVI) by neoplastic cells. AIMS: The aim of this study is to evaluate and compare LVI and BVI detected by H&E and immunohistochemistry (IHC) in a series of GC, relating them to morphologic features of the primary tumor and to the presence of lymph node metastasis. METHODS: IHC using D2-40 (lymphatic endothelial marker) and CD34 (pan-endothelial marker) was performed to study LVI and BVI in surgical specimens of a consecutive series of 95 primary GC. The results of the IHC study were compared with the detection of LVI and BVI by H&E, and related to other prognostic factors. The morphologic features of tumors and the presence of LVI and BVI were also related to the presence of lymph node metastasis. RESULTS: The detection of LVI was considerably higher than BVI. Sixty-one cases (64,2%) were positive for LVI when evaluated by H&E, whether 66 (69,5%) were detected by IHC. From the 61 cases with LVI by H&E, 8 were false-positives by IHC. The IHC detected LVI in 13 cases that were negative by H&E. BVI was identified in 38 cases (40,0%) by H&E and in just 24 (25,2%) by IHC. The IHC study identified 24 cases of false-positives and 10 cases of false-negatives for BVI. The comparison of LVI detected by both methods was not statistically significant. There was a significant difference between BVI detected by H&E and IHC (p=0,02). The average Kappa value determined for LVI was moderate (Kappa=0,50) and low for BVI (Kappa=0,20). The correlation between lymph node metastasis and the presence of LVI and BVI was statistically significant when evaluated by H&E (p=0,001, p=0,013), and IHC (p=0,001, p=0,019). The presence of lymph node metastasis was associated with the depth of tumor invasion when compared advanced to early GC (p=0,001), mixed-type of Lauréns classification when compared to the diffuse-type (p=0,017) and organ invasion (esophagus and duodenum) (p=0,026). The morphologic features associated with LVI were: location of tumor at distal third of the stomach (p=0,039), Borrmanns macroscopic type (p=0,001), organ invasion (p=0,003) and depth of tumor invasion (p=0,001). The presence of BVI is related only to the depth of tumor invasion (p=0,003). CONCLUSION: LVI was more easily identified in the D2-40 immunostained sections than in the H&E stained sections. The presence of LVI and BVI is associated with the presence of lymph node metastasis. The presence of LVI is associated to others morphological features of poor prognosis, as advanced stage of disease, mixed-type of Lauréns classification when compared to the diffuse-type and organ invasion (esophagus and duodenum), probably reflecting a potential for lymphatic metastatic spread and aggressive behavior. Our results show that the IHC method with these two markers is superior to the conventional H&E staining for detecting LVI and BVI in GC. Immunohistochemical identification of lymphatic and blood vessels in GC is useful to increase the accuracy of diagnosing the presence lymph or blood vessel invasion and to predict lymph node metastasis.