708 THE HIGH-RESOLUTION WHOLE GENOME PROFILING OF RENAL CELL CARCINOMA BY SINGLE NUCLEOTIDE POLYMORPHISM ARRAYS

708 THE HIGH-RESOLUTION WHOLE GENOME PROFILING OF RENAL CELL CARCINOMA BY SINGLE NUCLEOTIDE POLYMORPHISM ARRAYS

O7 KIDNEY TUMOURS: BASIC RESEARCH Friday, 20 March, 12.15-13.45, Room A2 705 706 Aberrant methylation of KS1- a candidate tumor suppressor in...

513KB Sizes 0 Downloads 50 Views

O7 KIDNEY TUMOURS: BASIC RESEARCH Friday, 20 March, 12.15-13.45, Room A2

705



706

Aberrant methylation of KS1- a candidate tumor suppressor in renal cell carcinoma and its relationship to clinicopathological features

CA9 gene single nucleotide polymorphisms predict prognosis and treatment response of metastatic renal cell carcinoma

Zhang Q..1, Ying J.M.2, Poon F.F.3, Tao Q.3, Jin J.1, Beijing-Hong Kong Joint Research Center for Urology Oncology

De Martino M.1, Klatte T.1, Seligson D.B.2, Larochelle J.C.1, Shuch B.1, Caliliw R.R.1, Li Z.1, Kabbinavar F.F.3, Pantuck A.J.1, Belldegrun A.S.1

Peking University, Institution of Urology, Dept. of Urology, Beijing, China, 2Cancer Hospital/institute, PUMC & CAM, Dept. of Pathology, Beijing, China, 3Hong Kong Cancer Institute and Li Ka Shing Institute of Health Sciences, Dept. of Clinical Oncology, Hong Kong, China

1

1

Introduction & Objectives: Identification of tumor suppressor genes (TSG) silenced by methylation uncovers mechanism of tumorigenesis and identifies new epigenetic tumor markers for early cancer detection. Aberrant methylation of KS1, a recently identified TSG located at 16q23, has been reported in several tumors. To gain insight into the role of epigenetic silencing of KS1 in the tumorigenesis of renal cell carcinoma (RCC), we investigated the methylation status of KS1 and its relationship with clinicopathologic features of RCC. Material & Methods: Genomic DNA and/or total RNA were extracted from six RCC cell lines and 81 primary samples of RCC with their corresponding normal renal tissues. Expression of KS1 was analyzed by semiquantitative reverse-transcription PCR. Promoter methylation status was analyzed by methylation-specific PCR and bisulfite genomic sequencing. Finally, monolayer colony formation assay was performed to examine the effect of KS1 re-expression on the tumor cell clonogenicity of RCC. Results: KS1 down-regulation and methylation were detected in all six RCC cell lines. Treatment with 5-aza-2’-deoxycytidine resulted in KS1 demethylation and reexpression, indicating methylation directly mediates its silencing. Moreover, aberrant methylation was detected in 42.0% (34/81) of primary RCC tumors. In contrast, only 3 of the 53 (5.7%) non-malignant renal tissues had methylation. KS1 methylation status was significantly associated with TNM classification (p=0.014, chi-square) and Grade stages (p=0.02, Chi-square) of RCC patients. Furthermore, ectopic expression of KS1 in silenced tumor cells resulted in substantial inhibition of tumor cell clonogenicity.

University of California-Los Angeles, Dept. of Urology, Los Angeles, United States of America, 2University of California-Los Angeles, Dept. of Pathology, Los Angeles, United States of America, 3University of California-Los Angeles, Dept. of Medicine, Los Angeles, United States of America Introduction & Objectives: Carbonic anhydrase 9 gene (CA9) is located in a prognostically relevant chromosomal area on chromosome 9p and is encoding for one of the most significant protein markers in metastatic renal cell carcinoma (MRCC), CAIX. In contrast to CAIX protein, however, no efforts have been made to date to study CA9 gene in metastatic RCC. Here, we test the hypotheses that single nucleotide polymorphisms (SNPs) and mutations of the CA9 gene are associated with CAIX expression, response to immunotherapy and survival. Material & Methods: Genomic DNA was extracted from frozen tumor samples of 54 patients with clear cell MRCC. All exons of the CA9 gene were PCR-amplified and sequenced. The antibody M75 was used to evaluate CAIX protein expression immunohistochemically. Statistical associations of CA9 gene status and CAIX protein expression with response to IL-2 based immunotherapy and overall survival were assessed with chi-square tests, t-tests, Kaplan-Meier survival estimates and Cox proportional hazards regression models. Results: CA9 reference SNP (rs) 2071676 was found in 59%, rs12553173 in 15%, rs3829078 in 11% and rs1048638 in 33% of the patients. The deletion c.376del393 was observed in two patients. CAIX expression was high (>85%) in 65% of the patients. None of the SNPs was significantly associated with CAIX expression. Patients with the C allele variant of rs12553173 had improved median survival (27.3 vs. 13.6 months, p=0.0431) and a greater likelihood of response to IL-2 (57% vs. 22%, p=0.081) Likewise, high CAIX expression was associated with longer median survival (25.5 vs. 8.5 months, p<0.0001) and a greater IL-2 response rate (37% vs. 8%, p=0.070). In a multivariate Cox model, both C allele variant of CA9 SNP rs12553173 and CAIX expression were retained as independent prognostic factors.

Conclusions: KS1 is a candidate TSG that plays an important role in the development and progression of RCC. Tumor-specific methylation of KS1 might serve as a biomarker for early tumor detection and prognosis prediction.

Conclusions: CA9 SNPs are frequently found in patients with MRCC. The C allele variant of rs12553173 is associated with improved overall survival and a greater likelihood of response to IL-2. CA9 rs12553173 and CAIX are both independent prognostic factors of overall survival and complementary in predicting prognosis of MRCC.





707

Tissue micro array reveals novel prognostic markers for clear cell renal cell carcinoma Kroeze S.G.C.1, Van Melick H.H.1, Vermaat J.S.2, Van Diest P.J.3, Voest E.E.2, Bosch J.L.H.R.1, Jans J.J.M.1 1 University Medical Center Utrecht, Dept. of Urology, Utrecht, The Netherlands, 2University Medical Center Utrecht, Dept. of Medical Oncology, Utrecht, The Netherlands, 3University Medical Center Utrecht, Dept. of Pathology, Utrecht, The Netherlands

Introduction & Objectives: Half of all clear cell renal cell carcinoma (CC RCC) patients will present with or develop metastases, associated with low survival rates. For optimal efficacy of systemic therapy for RCC the risk for progressive disease needs to be defined. Identification of new molecular markers that predict prognosis of RCC is therefore necessary. The hypoxia pathway plays an important role in the development and treatment of RCC. Our aim was to determine the prognostic significance of hypoxia marker expression in CC RCC. Material & Methods: A renal cancer tissue micro array containing 139 RCC specimens was constructed. Expression of the markers Hif-1α, Hif-2α, prolyl hydroxylase domain (PHD) 1, 2 and 3, factor inhibiting Hif (FIH) and VHL was determined by immunohistochemical staining. Expression of markers was scored by one pathologist (PvD) and subsequently CC RCC tumours were categorized according to their level of expression. Marker expression was correlated to cancer specific survival (CSS) using Kaplan-Meier plots followed by Log rank analysis. Marker independence was determined by Multivariate Cox-regression analysis.

708

The high-resolution whole genome profiling of renal cell carcinoma by single nucleotide polymorphism arrays Yokomizo A.1, Yamamoto K.2, Tada Y.1, Kuroiwa K.1, Eto M.1, Tatsugami K.1, Naito S.1 Graduate School of Medical Science, Kyushu University, Dept. of Urology, Fukuoka, Japan, 2Medical Institute of Bioregulation, Kyushu University, Dept. of Molecular Genetics, Fukuoka, Japan

1

Introduction & Objectives: The VHL gene is highly mutated or aberrant in clear cell renal cell carcinoma (RCC), but not in papillary RCC or chromophobe RCC. A variety of chromosomal aberrations has been reported in RCC, however, few studies have attempted to analyze on a genomewide and high density scale on different histopathological subtype of RCC. In this study, Human CNV370-Duo DNA Analysis BeadChip, which covers the 370K single nucleotide polymorphism (SNP) markers, was applied to detect the histopathological specific genetic alterations among the primary clear cell, papillary and chromophobe RCC. Material & Methods: Human CNV370-Duo DNA Analysis BeadChip (Illumina Corporation) was applied to detect the genomic changes in a panel of 22 primary clear cell RCCs, 7 papillary RCCs and 8 chromophobe RCCs. This system has an advantage in the detection of copy-neutral LOH as well as higher density, which allows discovery of more small aberrations than previous lower density SNP arrays. According to the SNP database, candidate genes located within that position can be indicated.

Results: The mean age was 62 years, with a male:female ratio of 60:40. Pathological stage of the tumours was T1:40%, T2:14%, T3:41%, T4:4%. Metastatic disease was present in 33% of the patients. Median (inter-quartile range) CSS was 3.55 (1.74-13.58) years. Low expression of VHL and nuclear FIH were significantly associated with shortened overall survival (p=0.004, p=0.005). Hif-1α, Hif-2α, PHD1,2 and 3 did not show significance in relation to survival. FIH and VHL expression was independent of Motzer criteria and metastases (p<0.001).

Results: In clear cell RCC, the most common genetic loss was identified in 3p, which was found in 95% (21 samples / 22 samples) of the tumors, suggesting that 3p loss can be a first step in clear cell carcinogenesis. Other frequent changes were losses of 1p (23%), 3q (46%), 8p (32%), 8q (22%), 9p (27%), 9q (27%), 18q (23%) and gain of 15q (32%), 7p (27%), 7q (27%), 1q (23%). Also, the copy neutral LOH, that is undetectable in a prior LOH study, was identified in 30 area (3.4% of all chromosomes analyzed). Furthermore, microamplifications and microdeletions less than 1Mb were detected in 4 and 18 regions respectively. The microdeletion of 10q23.31 was overlapped in two tumors and 5 genes including PTEN were identified as candidate tumor suppressor gene in SNP based genomic database. While in papillary RCC, the most frequent chromosomal losses (43%) were observed in 3p and 3q, followed by twenty nine percent of losses on 1p, 1q, 11q, 18q, 22p and 22q. The higher frequency of chromosomal gains were observed in 20q (57%), 12q (43%) and 20q (43%). The microamplifications and microdeletions less than 1Mb were detected in 9 and 6 regions respectively. Among them, the database search on SNPs in microdeleted area of 8q24.23 and 10q21.3 identified only one candidate tumor suppressor gene. On the other hand, hemi chromosomal loss of chromosome 1, 2, 6, 8, 10, 13 and 17 was specific alterations in chromophobe RCC. Additionally, the only 5 microdeletions and no microamplifications were observed in chromophobe RCC.

Conclusions: This study shows that while Hif-1α, Hif-2α, PHD1,2 and 3 have no significant relation to survival, low expression of VHL and FIH are associated with unfavourable survival and can be used as independent significant prognostic markers for CC RCC.

Conclusions: The genomic profiles of papillary and chromophobe RCC were totally different from that of clear cell RCC, suggesting that different molecular mechanisms can be related to the carcinogenesis and tumor progression in each histopathological subtype of RCC.

Eur Urol Suppl 2009;8(4):297