| The aim of our study are as follows: to compare lymphovascularinvasion(LVI)detected by HE versus that was marked byimmunohistochenistry using the D2-40, CD31, CD34 antibodies, toinvestigate the value of detection of LVI and identification of lymph vesseland blood vessel, to explore the distribution, morphological characteristicsof lymph vessel, blood vessel and LVI, the relationship between LVI andpatient age, tumor size, axillary lymph node metastasis, grading andmolecular classification. Thereby, it can provide strong basis for clinicaltherapy, neoadjuvant chemotherapy choice and studying the lymph nodemetastases.We selected the invasive breast cancer surgically treated at the AirForce General Hospital of Chinese PLA between January 2007 andDecember 2010. And we deleted the cases whose clinical data wereincomplete or received radiotherapy or chemotherapy before surgery. Thereare 110 cases which meet the conditions and are confirmed by pathology.We selected 10 cases of benign breast tumor as control.All the cases were stained with HE and immunohistochemistry usingthe PR, ER, HER-2, CK5/6, EGFR, D2-40, CD34 and P63 antibodies.Fluorescence in situ hybridization(FISH)for HER2 gene amplificationwas done on the cases with HER-2(++)by immunohistochemisty.Histologic grading was evaluated on the basis of tubule formation, nuclearmorphology and mitotic score. According to immunohistochemisty, weclassified the breast cnacer in four types, as Luminal A, Luminal B, HER-2 positive tumors, Basal-like type. Statistical analysis was performed with theSPSS17.0. The significance was set at a P value＜0.05.D2-40 was specific maker for lymph vessel whereas it did not reactwith the endothelium of blood vessels Including capillaries, veins andarteries. D2-40 is not expressed by fiber stroma. But it is expressed bybreast myoepitheliium surrounding benign ducts and lobules as well asmyoepithelium surrounding in carcinoma in situ. In addition to markingblood vessels, CD31, CD34 also reacted with basement membranesurrounding carcinoma in situ and fiber stroma. In breast cancer, D2-40positive lymph vessel was most frequently seen in the peritumoral, lessfrequently in the central zone. CD31, CD34 positible blood vessels wereseen in the peritumoral and the central zone. The distribution of bloodvessel made no difference. The lumen react with D2-40 are irregular,thin-walled lined by a single layer flattened endothelium with nosupporting smooth muscle. They were not filled with red cells. The lymphvessels in the intratumoral were small, narrow and occlusive. The lymphvessel located peritumorally were usually large and expansive. The bloodvessels in intratumoral were offen narrow and in peritumoral were smallbut expansive. In the benign breast tumor, lymphovascular were usuallylarge and expansive. When CD31, CD34 marked the blood vessels,specially CD34, it was difficulty to identify blood vessels for nonspecificstaining of the stroma. So we should combine immunohistochemical andmorphology when we identified blood vessels. By immunohistochemisty,lymphovascular invasion was most frequently seen in the peritumoral, lessin the intratumoural. LVTE were found in 40 cases, of which 28peritumoural, 11 intratumoural, 1 bouth intra-and peritumoural wereinclued. BVI were found in 23 cases, of which 14 peritumoural, 7intratumoural, 2 bouth intra-and peritumoural were inclued. The form oflymphovascular tumor emboli were a single or clusters of tumor cellslocated in the lumen. LVI can be connected with wall or not. It was not frequently seen in the form of tumor cell filled the lumen. The positive rateof LVTE and BVI by HE staining were 15.5%, and 9.1% respectively.Positive rate of LVTE and BVI by immunohistochemisty were 36.36%,20.91. Immunohistochemisty improved the rate of LVTE and BVI by20.91%, 11.82% which compared with the HE. LVTE was positivelyrelated to age(χ~2=14.559,P =0.001); but there was no significantcorrelation between BVI and age(χ~2=0.509,P =0.775); Relationship wasobserved between LVTE(χ~2=7.734,P =0.021), BVI(χ~2=0.168,P =0.018)and tumor side. LVTE and BVI were positively related to histologicgrading and node status(χ~2=18.323,P=0.000,χ~2=13.880,P =0.001；χ~2=7.121,P=0.008,χ~2=110,P=0.000)；no relationship was observedbetween LVTE and molecular classification(χ~2=1.428,P=0.699); therewas no significant correlation between BVI and molecular classification(χ~2=4.299,P=0.231).D2-40 was a specific maker for lymph vessel whereas it did not reactwith the endothelium of blood vessels. Forthermore, D2-40 can be appliedto paraffin section and to study metastases of breast cancer. CD31andCD34 have shown to be sensitive and specific marker for blood vessel, andthey can stain with fiber. The sensitivity of the CD34 is better than that ofCD31 but the specificity of CD31 is better than CD34. By D2-40, CD31,CD34, they can be greatly improved the rate and the accurate diagnosis ofLVTE and BVI in breast cancer. Thus, we can accurately evlaluate theprognosis and treatments. LVTE is positively related to age, tumor side,histologic grading and node status. No relationship was seen betweenLVTE, BVI and molecular classification. There was a significantcorrelation between BVI and histologic grade, node status. LVTE and BVIare reliable indicators for prognosing and making treatments. |
