Clinical implications of increased lymph vessel density in the lymphatic metastasis of early-stage invasive cervical carcinoma: a clinical immunohistochemical method study

Patients (n = 41) with cervical carcinoma who were treated between September 2007 and June 2008 were enrolled in the study. The tissue samples were obtained at the time of surgery in the Department of Gynecology, Qilu Hospital, Shandong University. The present study was approved by the ethics committee of the Shandong University. All samples were obtained with medical-ethics approval and all patients gave informed consent. Tissues were fixed with 4% paraformaldehyde and paraffin-embedded for further analysis. The pathological examination validated the fact that no radio- or chemotherapy was received before surgery. All patients with early-stage invasive cervical cancer were staged according to the 2000 International Federation of Gynecology and Obstetrics (FIGO) staging system. Seven patients had a cervical cancer classified as FIGO stage Ia, 14 as FIGO stage Ib, and 20 as FIGO stage IIa. Based on analysis of cellular differentiation, seven cases were HG1, 13 were HG2, and 21 were HG3. Of all the cases, 37 were squamous cell carcinomas and four were adeno-carcinomas. Histological review confirmed that 15 cases showed lymph node metastasis. Lymph node status was determined by routine pathological examination with the surgical specimens. The standard for lymphatic vessel invasion was the detection of cancer cells in the cavity of the lymphatic vessel by light microscopy. By this standard, 35 cases showed lymphatic vessel invasion and six were negative. The age of the patients varied from 26 to 70, with a median value of 42. Of all the patients, 29 were premenopausal and 16 were postmenopausal. Pelvic/paraaortic lymph node dissection was performed in all cases; the mean and median number of lymph nodes per case examined was 24.7 and 24.0, respectively (range 1–82). Lymph node status was determined by routine pathological examination with the surgical specimens. Lymph node metastasis was present in 15 (36.6%) cases; the mean and median number of positive lymph nodes was 2.5 and 2.0, respectively (range 1–7). All tissue specimens and slides were examined by experienced pathologists.

Immunohistochemistry (IHC) was done using the immunohistochemistry SP kit from Jingmei Inc. (Shanghai, China; catalog no. LHK612) according to the manufacturer's instructions. Briefly, serial section slides of 5 μm were obtained from the paraffin-embedded specimens and the paraffin medium was removed. The slides were then re-hydrated by passing through a descending serial alcohol dilutions. Then the slides were placed in antigen retrieval solution (pH 6.0) and heated in a microwave oven for 10 min at 95°C. Next, the slides were incubated in a 3% hydrogen peroxide-methanol solution for 10 min to remove endogenous peroxidase. IHC staining was performed as follows: nonspecific binding was blocked with 10% goat serum; the slide was incubated for 1 h with polyclonal rabbit anti-LYVE-1 antibody (1:80 dilution, Abcam, catalog # ab36993), followed by incubation for 10 min with biotin-labeled secondary antibody; then the slide was incubated for 10 min with horseradish peroxidase-labeled streptavidin. Color was developed using DAB, and the slide was counterstained with hematoxylin. Finally, the slides were mounted and coverslipped with resinene. Negative control slides were stained with PBS instead of LYVE-1 antibody.

Intratumoral and peritumoral lymphatic vessel density (LVD) was determined according to the methods previously described by Weidner et al.[2] and Gombos et al.[3]. Briefly, the following had to be observed in order for lymphatic vessels to be considered positively stained: (1) brown-yellow stained endothelial cells forming tubular structures and situated on their own or bundled, and (2) an open cavity inside the tubular structure. The slides were scanned on a low-power microscope and intra- and peritumoral areas with the highest vessel density, called hot spots, were identified. The number of lymphatic vessels in five high-power fields in the selected areas was counted. LVD was determined as the mean value of vessel counts.

Intratumoral lymphatic vessels were defined as those located within the tumor mass. Peritumoral lymphatic vessels were those located outside of the tumor mass, but within 2 mm from the tumor. The lymphatic vessels in the control samples were defined as those located in the stromal area, within 2 mm of the underlying basement membranes. The slides were evaluated independently by investigators blinded to the clinical data.

Lymphatic vessels in the control uterine cervix are often open and have a regular shape (Figure 1A). Under the high-power light microscope, the positively stained vessels were brown-yellow and the vessel walls were thin. The vessel endothelium was usually a single layer of cells, each with a large nucleus extruding toward the lumen face (Figure 1B). In contrast, most of the lymphatic vessels in cervical carcinoma tissue were localized around the tumor mass in increased numbers. They were often in irregular shape with an enlarged cavity. The lymphatic vessels in the tumor mass were small, collapsed tubular structures that were either isolated or bundled (Figure 1C). At the same time, invasive cancer cells were observed in some peritumoral lymphatic vessels (Figure 1D).

The intratumoral lymphatic vessel density (ILVD) was 7.71 ± 1.12 and peritumoral lymphatic vessel density (PLVD) was 9.10 ± 0.80. PLVD was significantly higher than ILVD (P < 0.01) in the corresponding cervical carcinoma samples. The ILVD and PLVD associated with carcinomas in situ were both significantly higher than the LVD of the normal uterine cervix (2.08 ± 1.00) (P < 0.01).

The clinicopathological features and LVD data are summarized in Table 1. In samples from patients with lymph node metastasis, the ILVD was 8.27 ± 0.80 and the PLVD was 9.67 ± 0.72. Both of these values were significantly higher than the corresponding values in the samples without metastasis (7.38 ± 1.17, 8.77 ± 0.65, respectively; P < 0.05, P < 0.01). Similarly, ILVD and PLVD were found to correlate significantly with lymphatic vessel invasion (P < 0.05 and P < 0.01, respectively). The ILVD and PLVD from cases showing lymphatic vessel invasion were much higher than in those cases without invasion although these differences did not reach statistical significance (8.67 ± 0.82 vs. 7.54 ± 1.09, 9.67 ± 0.72 vs. 8.94 ± 0.73).