Application of tissue microarray technique in the evaluation of chromosomal abnormalities of clear cell renal cell carcinoma

Background Renal cell carcinoma (RCC) accounts for approximately 3% of all adult malignancies. More than 40% of RCC patients die of the disease, and the vast majority of them are represented by clear-cell renal cell carcinoma (ccRCC). Initiation and progression of ccRCC is due to genetic alterations. Genes potentially involved in kidney cancer include the von Hippel-Lindau gene on chromosome 3p; specifically, loss of the short arm of chromosome 3 represents the main genetic aberration in ccRCC and has to be considered as one of the primary events in the development of this neoplasia. Conversely there are only few studies in which 3p deletions are considered as a potential prognostic marker in ccRCC. Validation of prognostic markers in multiple clinical specimens has been usually performed by traditional histopathological techniques, which are time consuming, labour intensive and economically costly. The tissue microarray (TMA) is a new histopathologic technique which allows the rapid and cost effective validation of novel markers in multiple pathological tissue specimens. TMAs seem to be very promising for the application of in situ hybridization techniques, including FISH analyses. Aim of our study is to investigate the gene status of 3p in 122 ccRCC using tissue microarrays with FISH and correlate this data with follow up of the patients. Material and methods For this study we have selected 122 ccRCC tumor specimens, obtained from radical or partial nephrectomies by the Department of Pathology of the University of Verona. In the critical review, we defined Fuhrman nuclear grade and TNM staging at diagnosis. Tissues from 122 histology paraffin blocks have been accurately arrayed into newly created 13 TMA paraffin blocks. Consecutive, 4 µm thick sections were obtained by each TMA paraffin block, for hematoxylin and eosin staining and for immunohistochemistry (AE1/AE3, CD10, CAM5.2, CK7 and vimentin) and FISH analyses. FISH was performed with a mixture of centromeric-satellite DNA probes for chromosome 3 and subtelomeric probes for 3p25. To evaluate the presence of 3p deletion, we scored the number of nuclear signals for both subtelomeric and centromeric probes, and calculated the ratio between 3p/CEP3; we considered as 3p deleted all the neoplasms with ratios 3p/CEP3 below 0.7, according to previous experiences. Results: The age of patients ranged from 30 to 86 years with an average age of about 60 years, 31 patients were female, 91 male; Fuhrman grade was distributed among grade 1 (6 cases), grade 2 (52 cases), grade 3 (52 cases) and grade 4 (12 cases). The tumor size ranged from 2 cm to 19 cm. The TNM staging at diagnosis showed an uneven distribution of cases (57 cases with pT1 stage, 25 with pT2,10 with pT3a, 25 with pT3b, and 5 with pT4). Lymph node involvement at diagnosis was observed in 5 cases (1 case with pT3a stage, 3 with pT3b, 1 with pT4). Metastatic spread at diagnosis was found in 17 of the 122 examined cases ( 2 cases with pT1 stage, 2 with pT2, 3 with pT3a, 8 with pT3b, and 2 with pT4). Follow-up data were available for all 122 selected patients: 36 patients died of disease, 8 patients died for unrelated causes, 77 patients were alive with no evidence of disease, and 1 patient was alive with disease progression. The follow-up data were based on a time-lapse ranging from a minimum of 2 months to a maximum of 168 months. According to the criteria exposed in the Methods section, related to the number of nuclei to be scored for each tumor sample, informative cases were limited to 110. For the remaining 12 cases, we could not reach the adequate number of nuclei to evaluate for centromeric and locus-specific signals, due to the presence of focal hemorrhagic or necrotic areas; nevertheless, the number of nuclei which could be evaluated in these cases ranged from 60 to 85. Conversely, all the non neoplastic renal samples could be scored, due to the integrity and the well represented tubular component of kidney parenchyma. Among the 110 evaluable cases, we could be able to identify 3p losses in 78 tumor cases (71%), with ratios 3p/CEP3 ranging from 0.46 to 0.68 (mean: 0.55); the remaining 32 cases (29%) showed ratios ranging from 0.72 to 1.1 (mean: 0.92), and were considered to be not deleted. The analysis of non neoplastic renal parenchyma showed ratios 3p/CEP3 consistently above the cut-off value of 0.7, ranging from 0.95 to 1.2. Correlating the results of the FISH analysis for 3p25 gene status with stage, cases showing 3p deletion were predominantly low stage, with 78% included in pT1-T2 categories, whereas 22% were high stage, included in pT3-T4 categories; conversely, non deleted cases were mostly high stage (66%) with only one third at low stage (34%). With regard to the correlation between Fuhrman grade and 3p deletion, we could not find significant variations in the distribution of grade; in contrast, non deleted cases showed a prevalence of high grade (g3: 47%, g4: 25%). About patients’ follow-up, from among the 78 cases with 3p deletion, 62% were alive with no evidence of disease, compared with only 14% who died of disease. For the 32 cases without 3p deletion, 71% died of cancer, and 21% are alive with no evidence of disease. Out of a total of 110 cases, 56% of patients who were alive with no evidence of disease, showed deletion of 3p, whereas 20% of patients who died of disease had no evidence of 3p deletion. Finally, the 16 cases with metastatic disease, 5 of which with lymph node involvement, 6 cases showed 3p deletion, while 10 cases had no deletion. Discussion: Cytogenetic analyses have emerged as a powerful tool for diagnosis and classification of ccRCC. FISH genetic studies could greatly benefit from TMA procedures. TMA- technology are ideally suitable for genomic based diagnostic and drug target discovery because allows rapid visualization of molecular markers in thousands of tissue specimens, with considerable diminution in time consuming activities and reduction of more than 90% the amount of probes to be utilized. Several reports have established that loss of the short arm of chromosome 3 has to be considered as one of the primary events in the development of ccRCC, and a specific “hallmark” for this histotype; nevertheless, there are only few studies in which 3p deletions are considered as a potential prognostic marker in ccRCC. Recently, Klatte et al. showed that 3p/VHL loss was associated with improved survival and decreased risk of death in 282 patients affected by ccRCC. In conclusion, our study: 1. Confirms that TMA is a reliable technique for genetic investigations on formalin fixed, paraffin embedded neoplastic specimens by FISH. 2. Is in keeping with previous experiences, to confirm that loss of 3p is a specific, genetic marker for ccRCC. 3. Is in agreement with previous observations of a potential prognostic role of 3p deletion in ccRCC.