Prognostic value of the expression of Ki-67, CD44 and vascular endothelial growth factor, and microvessel invasion, in renal cell carcinoma.

OBJECTIVE

To determine if use of cell proliferation, cell adhesion, level of angiogenesis-related factors and presence of microscopic vascular invasion (MVI) could better predict the biological behaviour of renal cell carcinoma (RCC), which has a widely variable clinical outcome despite the use of conventional prognostic factors (staging and grading).

MATERIALS AND METHODS

The expression of Ki-67, CD44H and vascular endothelial growth factor (VEGF) were assessed immunohistochemically in formalin-fixed, paraffin-embedded tissues from 48 RCCs, using a Ki-67 labelling index (LI), CD44 LI and level of VEGF expression, respectively. In addition all the pathological slides were reviewed retrospectively for the presence and absence of MVI. The prognostic value of all the variables assessed was then evaluated, and correlated with the usual prognostic variables and cancer-specific survival.

RESULTS

Univariate analysis of cancer-specific survival showed that tumour stage (P < 0.001), tumour size (P = 0.005), metastasis, MVI, Ki-67 LI, CD44H LI and VEGF expression (all P < 0.001) were predictors of tumour-related death. There was a statistical correlation between CD44H LI and each of Ki-67 LI (r = 0.61), expression level of VEGF (r = 0.72) and presence of MVI (r = 0.71). Independent predictors of cancer-specific survival in a multivariate analysis were: in all patients with RCC, the MVI (P = 0.003) and VEGF expression (P = 0.01); in those with no metastases, MVI (P = 0.01); in patients with no MVI, VEGF (P = 0.04); and in patients with MVI, Ki-67 LI (P = 0.003). No independent predictor was identified in patient with metastases.

CONCLUSION

This study suggests that cell proliferation, cell adhesion, the level of VEGF expression and the presence of MVI represent a complex tumour-host interaction that may favour the progression of RCC. Cell proliferation, CD44H and VEGF expression appear to be powerful markers for identifying patients with an adverse prognosis.