Lymphatic vessels assessment in feline mammary tumours.

Background: The lymphatic vessels play a crucial role in a variety of human cancers since tumour cell lymphatic invasion significantly influences prognosis. It is not known if pre-existing lymphatics are enough for tumour dissemination or de novo development is necessary. VEGFR-3 is an angiogenetic mediator for both lymphatic and blood vessels during embryonic development, and only for lymphatics after birth. VEGF is a mediator of both vasculogenesis and angiogenesis, regulates the growth of lymphatics in various experimental models, and is produced in many solid tumours. CD44 mediates hyaluronic acid (HA)-dependent cell adhesion: besides promoting invasion, this interaction also supports neoangiogenesis that indirectly stimulates tumour cell proliferation. The expression of VEGF-C (Vascular Endothelial Growth Factor-C), its receptor VEGFR-3 and CD44, were studied on feline mammary samples to assess the importance of lymphangiogenesis and lymphangiotrophism in neoplasia.

Methods: Samples were taken from six normal mammary glands (NMG), ten benign (BT) and 32 malignant (MT) tumours. Immunohistochemical laminin/VEGFR-3 double stain, VEGF-C and CD44 stains were applied to 4 mum-thick sections, and their expression evaluated in intratumoral/extratumoral and intramammary/extramammary fields.

Results: All groups revealed a higher number of lymphatics in the extratumoral/extramammary areas. VEGF-C expression in the epithelium paralleled the number of positive vessels in the NMG, BT and MT, whereas VEGF-C higher expression was noted in the intratumoral fields only in infiltrating MT. CD44 score was lower in extratumoral than intratumoral fields in tumours and showed a significant increase in extramammary/extratumoral fields from NMG to MT. Pearson test showed a significant and inversely proportional correlation between CD44 expression and the number of lymphatic vessels with VEGFR-3 in malignant infiltrating tumours.

Conclusion: The number of both VEGFR-3 positive and negative lymphatics in the extratumoral and extramammary stroma was significantly higher than intratumoral and intramammary fields respectively in the NMG, BT and MT. This suggests a scant biological importance of intratumoral lymphatics while their higher number is due to the concentration of existing vessels following compression of the extratumoral stroma in spite of a non demonstrable increase from NMG to MT. The tumour model employed provided no evidence of lymphangiogenesis, and metastasis in the regional lymph node develops following the spread through the pre-existing lymphatic network.