Tumor lymphangiogenesis: a novel prognostic indicator for cutaneous melanoma metastasis and survival

Abstract

Malignant melanomas of the skin are distinguished by their propensity for early metastatic spread via lymphatic vessels to regional lymph nodes, and lymph node metastasis is a major determinant for the staging and clinical management of melanoma. However, the importance of tumor-induced lymphangiogenesis for lymphatic melanoma spread has remained unclear. We investigated whether tumor lymphangiogenesis occurs in human malignant melanomas of the skin and whether the extent of tumor lymphangiogenesis may be related to the risk for lymph node metastasis and to patient survival, using double immunostains for the novel lymphatic endothelial marker LYVE-1 and for the panvascular marker CD31. Tumor samples were obtained from clinically and histologically closely matched cases of primary melanomas with early lymph node metastasis (n = 18) and from nonmetastatic melanomas (n = 19). Hot spots of proliferating intratumoral and peritumoral lymphatic vessels were detected in a large number of melanomas. The incidence of intratumoral lymphatics was significantly higher in metastatic melanomas and correlated with poor disease-free survival. Metastatic melanomas had significantly more and larger tumor-associated lymphatic vessels, and a relative lymphatic vessel area of >1.5% was significantly associated with poor disease-free and overall survival. In contrast, no differences in the density of tumor-associated blood vessels were found. Vascular endothelial growth factor and vascular endothelial growth factor-C expression was equally detected in a minority of cases in both groups. Our results reveal tumor lymphangiogenesis as a novel prognostic indicator for the risk of lymph node metastasis in cutaneous melanoma.

Figure 1.

Detection of intratumoral and peritumoral lymphatic vessels in cutaneous malignant melanomas. A: Histology of a thick melanoma (4.5 mm) shows compact masses of frequently pigmented tumor cells. B: Immunofluorescent stain of a serial section of the same tumor for the lymphatic marker LYVE-1 (green) and the panvascular marker CD31 (red) reveals prominent hotspots of high lymphatic vessel density (hotspots are circled by solid line). In contrast, blood vessels are homogeneously distributed throughout the tumor. C: Thin-walled, LYVE-1-positive intratumoral lymphatic vessels with open lumina. D: Histology of a thin melanoma (≤1.5 mm) with dilated peritumoral lymphatics (arrows). E: Immunofluorescent stain of a serial section for LYVE-1 (green) and CD31 (red) reveals lymphatic vessels within (arrowheads) and surrounding (arrows) the tumor border (dotted line). F: LYVE-1-positive intratumoral lymphatic vessels with open lumina. The adjacent blood vessels (asterisks) are LYVE-1-negative. Cell nuclei are counterstained (blue) with Hoechst (C and F). Original magnifications: ×100 (A, B, D, and E); ×400 (C and F).

Figure 2.

Higher frequency of intratumoral lymphangiogenesis in metastatic melanomas. A: Immunofluorescent stain for LYVE-1 (green) depicts thick-walled peritumoral (arrows) and thin-walled intratumoral lymphatics (arrowheads) in a thin melanoma (1.05 mm). The tumor border is indicated by a dotted line. Blood vessels (asterisks) are negative for LYVE-1. B: Higher magnification of intratumoral lymphatics reveals thin-walled, basket-like morphology. C: Double immunostain for LYVE-1 (red) and PCNA (brown) reveals an intratumoral lymphatic vessel with proliferating lymphatic endothelial cells (arrows) and adjacent melanoma cells (arrowhead). D: Significantly increased frequency of detectable intratumoral lymphatic vessels in metastatic melanomas (M; n = 18), as compared with nonmetastatic (NM; n = 19) tumors (mean ± SEM, chi-square test, P = 0.01). E: Detection of melanin-containing tumor cells within an intratumoral lymphatic vessel. H&E stain. F: Immunofluorescent stain of a serial section for LYVE-1 (green) and CD31 (red) confirms that tumor cells are located within a lymphatic vessel. Cell nuclei are counterstained blue with Hoechst (B and F). Original magnifications: ×200 (A); ×400 (B, C, E, F).

Figure 3.

Computer-assisted image analysis of tumor-associated blood vessels and peritumoral lymphatic vessels in nonmetastatic versus metastatic melanomas. A–C: Comparable blood vascular density, average blood vessel size, and relative area occupied by blood vessels in metastatic (M; n = 18) and nonmetastatic (NM; n = 19) malignant melanomas. D–F: Significant increase of lymphatic vascular density, average lymphatic vessel size, and relative area occupied by lymphatics in metastatic malignant melanomas, as compared with nonmetastatic melanomas. Mean ± SEM.

Figure 4.

VEGF-C mRNA expression in cutaneous melanomas. A: Bright-field microscopy shows a dermal melanoma nodule. B: Dark-field microscopy reveals increased signal (white grains) in the tumor area but not the dermis. C: Bright-field microscopy shows increased black grains under a higher magnification from the rectangle in A. D: VEGF-C mRNA expression by peritumoral stromal cells. Arrows indicate areas of strong hybridization signals. Original magnifications: ×200 (A and B); ×600 (C and D).

Figure 5.

VEGF-C protein expression in cutaneous melanomas. A: Immunoperoxidase staining for VEGF-C (red) demonstrates focal expression in the cytoplasm of melanoma cells in one tumor nest but not in the adjacent nests (asterisks). B: Serial section negative control with omission of secondary antibody. C: Immunoperoxidase staining for VEGF-C (red) demonstrates VEGF-C expression by peritumoral fibroblasts (rectangle) at the invasive edge of a metastatic melanoma (top left) in the reticular dermis. D: H&E stain of the invasive tumor edge of a metastasizing cutaneous melanoma reveals mononuclear inflammation (asterisks) near dilated lymphatic vessel (arrow). E: Immunoperoxidase staining of a serial section for LYVE-1 (red) decorates dilated peritumoral lymphatics (arrow) and adjacent mononuclear infiltrate. Blood vessels are LYVE-1-negative (arrowheads). F: Immunoperoxidase staining for VEGF-C (red) stains the cytoplasm of peritumoral mononuclear cells in a granular pattern. Original magnifications: ×200 (A–D); ×400 (E); and ×600 (F).

Figure 6.

Kaplan-Meier analyses of overall and disease-free survival as a function of tumor lymphangiogenesis. A: The distribution of lymphatic vascular area (LVA) in the patient population indicates an overlap in the >1.0 to ≤1.5 range between the nonmetastatic and metastatic melanoma groups. B, C: Overall and disease-free survival curves were stratified by low (≤1.0%), medium (1.0 to ≤1.5%), or high (>1.5%) level of tumor lymphangiogenesis. Statistically significant correlation between the extent of tumor-associated lymphangiogenesis, expressed as percentage of area covered by lymphatic vessels, and reduced disease-free survival and overall survival are noted. D: The disease-free survival analysis revealed that the detection of intratumoral lymphatics was significantly associated with a poorer disease-free survival.