Tumour suppressor gene (CDKNA2) status on chromosome 9p in resected renal tissue improves prognosis of localised kidney cancer

Abstract

Background: Genetic alterations on chromosome 9p, including inactivation of the tumour suppressor gene, CDKN2A, result in cellular proliferation and growth of tumours. Our aim was to use microsatellite analysis and fluorescence in situ hybridization (FISH) to characterise the architecture of this region.

Results: Seventy-five out of 77 clear cell renal cell cancers (tumour/normal pairs) were interpretable for LOH analysis on chromosome 9p (two tumours were excluded, as all five primers were uninformative). Twenty out of 75 (26.6%) tumours showed LOH in at least one of the five primers employed. Most allelic deletions were detected, telomeric to the CDKN2A region at D9S916, with 11 out of 52 informative tumours (21%) displaying LOH. The LOH in the coding region of CDKN2A, at D9S974 and D9S942, was associated with a higher pT-stage (p = 0.004) and metastasis (p = 0.006, both markers). The rate of chromosome 9p deletion in ccRCC was 44% (35/80 cases) according to FISH. Somatic copy number loss of chromosome 9p was associated with a larger tumour size (p = 0.002), higher pathological tumour stage (p = 0.021), presence of tumour necrosis (p = 0.019) and microvascular invasion (p = 0.032). The cases with copy number loss, loss of heterozygosity and copy number neutral (n = 42) were at a higher risk of cancer-specific death when compared to tumours in category D (n = 32) (Log-rank: p = 0.001). Seventeen patients with localised ccRCC developed recurrence, and fourteen of those showed either LOH or somatic copy number loss at CDKN2A (Log-rank: p = 0.005). Multivariate analysis showed that LOH or copy number loss at CDKN2A retained its independent prognostic effect, improving the predictive accuracy of stage and SSIGN score by concordance Index C from 0.823 to 0.878 (p = 0.001).

Materials and Methods: Cytogenetics data, microsatellite analysis and FISH were acquired for a cohort of patients undergoing resection for clinically localised renal cancer between January 2001 and December 2005. Five microsatellite markers (D9S916, D9S1814, D9S974, D9S942 and D9S171) assessed loss of heterogeneity (LOH) using DNA samples and in the same cohort FISH analysis was accomplished on tissue microarray slides. The FISH data were scored by two observers blinded to the histological data of the patients. Cytogenetic aberrations were correlated with histological and clinical outcomes by univariate and multivariate analyses using different prognostic models. Disease specific and recurrence free survival based on cytogenetic changes were assessed by Kaplan Meier methods.

Conclusions: A comprehensive cytogenetic analysis using microsatellite analysis and FISH of the CDKN2A region on chromosome 9p improves the predictive accuracy of known prognostic factors in clinically localised renal cell carcinoma undergoing surgical resection.