RESULTSEighteen patients had either synchronous or metachronous metastases and 11 died during the follow-up. Caveolin-1 intensity was significantly correlated with tumour size ( P = 0.005), TNM stage ( P = 0.028), M stage ( P = 0.012), grade ( P = 0.015), and metastasis (synchronous or metachronous; P < 0.001). The caveolin-1 proportion ( P = 0.037) and MVD ( P = 0.011) were significantly correlated with metastasis. MVD was correlated with caveolin-1 intensity ( r = 0.385, P = 0.001) and caveolin-1 proportion ( r = 0.388, P = 0.001). There was no difference in the expression of caveolin-1 and MVD between primary and metastatic sites. The survival of patients with higher caveolin-1 intensity was significantly worse than that of patients with lower caveolin-1 intensity. Multivariate analyses indicated that only M-stage was an independent prognostic factor for cancerspecific survival and caveolin-1 expression was not an independent factor. CONCLUSIONSIncreased expression of caveolin-1 and MVD is associated with metastasis and a worse prognosis in clear cell RCC. Caveolin-1 expression is correlated with MVD. These results suggest that caveolin-1 may be important in the progression of clear cell RCC and angiogenesis may be affected by caveolin-1 during the progression of RCC. KEYWORDScaveolin-1, metastasis, microvessel density, prognosis, renal cell carcinoma OBJECTIVETo investigate the relationship of caveolin-1 expression and microvessel density (MVD), a reflection of angiogenesis, with metastasis and prognosis in patients with clear cell renal cell carcinoma (RCC). PATIENTS AND METHODSFormalin-fixed, paraffin-embedded tissue sections of clear cell RCC from 67 patients who had undergone radical nephrectomy were stained immunohistochemically with specific antibodies against caveolin-1 and CD34. Caveolin-1 immunostaining was semi-quantitatively estimated based on the proportion (percentage of positive cells) and intensity. MVD was determined with CD34-stained slides. The expression pattern of caveolin-1 and MVD was compared with the clinicopathological variables.

No abstract

No abstract

No abstract

No abstract