Genetic Subtyping of Renal Cell Carcinoma by ... - Springer Link

Genetic Subtyping of Renal Cell Carcinoma by Comparative Genomic Hybridization Kerstin Junker, Gregor Weirich, Mahul B. Amin, Petr Moravek, Winfried Hindermann, Joerg Schubert K. Junker (I~) Department of Urology, Friedrich-Schiller University, 07743 Jena, Germany

Abstract The prognosis of renal cell carcinoma (RCC) varies dependent on histologic tumor subtypes. However, differentiation between RCC types may sometimes be difficult on histologic grounds alone. Furthermore, the prognostic value of histologic parameters for the individual prognosis is limited. Additional information on the molecular level seems necessary to obtain more certainty in diagnostic and prognostic evaluation. By investigating genetic alterations in different RCC subtypes, we sought to obtain a genotype-phenotype correlation. Eighty-two clear-cell, 53 papillary, 23 chromophobe RCCs and 26 renal oncocytomas were investigated. Comparative genomic hybridization (CGH) was performed on DNA from paraffin-embedded tissue samples. DNA was isolated from tumor areas by microdissection and amplified by degenerated oligonucleotide primed polymerase chain reaction (DOP-PCR). CGH was performed according to standard protocols. We were able to detect specific alterations in each RCC subtype: clear cell RCC showed -3p, +5/5q, -8p, -9, -14, -18; papillary (chromophilic) RCC gains of chromosomes 7, 17, 16,3, 12; chromophobe RCC loss of chromosomes 1, 2, 6, 10, 13, 17, 21; renal oncocytomas loss of chromosomes VIp and 14. Furthermore, for clear cell RCC, it was possible to define alterations which are associated with metastatic disease: loss of 9, 10, 14. Our results demonstrate that each RCC subtype is characterized by distinct genetic alterations. The definition of genetic alterations seems helpful for a tumor typing especially when morphology is equivocal. Therefore, genetic analyses represent a powerful diagnostic and prognostic tool for RCC.

Introduction Tumors of the kidney account for 3% of all human neoplasms. The majority of these tumors are renal cell carcinomas (RCCs). RCCs are characterized by

H. Allgayer et al. (eds.), Molecular Staging of Cancer © Springer-Verlag Berlin Heidelberg 2003

170

Kerstin Junker et al.

a

b

Fig. 1a-d. CGH profiles for each RCC tumor type. a Clear cell RCC with loss on chromosomes 3 and 14 and gain on chromosome 5. b Papillary (chromophilic) RCC with gains of chromosomes 7 and 17 as well as loss of the Y chromosome. c Chromophobe RCC with loss of chromosomes 1, 2, 6, 10, 13, 17 and 21. d Renal oncocytoma with loss of chromosomes 1 and 14

Genetic Subtyping of Renal Cell Carcinoma by Comparative Genomic Hybridization

(

d

171

172


high resistance to chemo-, radio-, and immunotherapy. Metastatic disease represents the major prognostic factor in RCC patients. At the moment, no parameters are available for individual prognostic evaluation including response to therapy. The classification system introduced by Thoenes et al. in 1986 led to a practicable clinically relevant subtyping of RCC [1]. The system is based on histologic and histochemical features. At present, three benign and six malignant renal tumor types are included in the Thoenes classification, which was the basis for the WHO classification of RCC in 1998 (UICC; Fig. 1) [2]. The UICC system has partly been corroborated by immunohistochemical and genetic investigations, which have shown that morphologically different RCC types are based on different genetic alterations. The diagnostic practice of tumor typing relies primarily on morphology using UICC criteria. In some cases, however, tumor histology remains equivocal, and morphology alone cannot predict a metastatic potential. Therefore, additional information on the molecular level is nec.essary to obtain more certainty in diagnostic and prognostic evaluation. We have subjected four major subtypes of renal epithelial neoplasms (clear cell RCC, papillary RCC, chromophobe RCC, renal oncocytoma) to comparative genomic hybridization (CGH) analysis in order to establish genetic fingerprints which may serve as a reliable supplement for RCC diagnosis and patient management.

Materials and Methods DNA was isolated from 5-10 frozen or formalin-fixed paraffin-embedded tissue sections. For CGH, normal DNA was isolated from blood cells collected from normal individuals using a commercial kit (Qiagen). In order to obtain sufficient amounts of tumor DNA for CGH analysis, DNA was amplified according to a modified protocol for DOP-PCR [3]. This protocol employs Sequenase during the first eight cycles of nonspecific PCR, followed by 30 additional cycles under specific conditions using TaqPolymerase (Stoffel fragment). Labeling of tumor DNA and normal DNA was achieved by 20 PCR cycles using biotin-16dUTP and digoxigenin-lldUTP, respectively. One microgram of both tumor DNA and normal DNA was hybridized to 50 Ilg Cot-1 DNA on normal metaphases at 37°C for 48 h. Detection of fluorescent signals was carried out with avidin-FITC (tumor DNA) and anti-digoxigenin-rhodamine (normal DNA). DAPII Antifade was used for chromosome counterstaining. Fifteen metaphases were analyzed in each case using an Axioplan-Microscope (Zeiss, Germany) and a computer system from Metasysterns (Altlussheim, Germany). Chromosomal alterations were defined as shifts to the red (loss of chromosomal region in the tumor DNA) or the green borderline (gain of chromosomal region in the tumor DNA).


173

Table 1. Frequency of genetic alterations in different RCC tumor types Clear cell RCC

Papillary (chromophilic) RCC

Chromophobe RCC

Oncocytomas

Loss of 3p: 97% Loss of 9: 34% Gain of 5: 32% Loss of 14: 25% Loss of 10: 18%

Gain of 7: 84% Gain of 17: 55% Gain of 16: 20% Gain of 12: 18% Gain of 3: 16% loss of y. 16%

loss of 1: 87% Loss of 10: 69% Loss of 6: 56% Loss of 2: 50% loss of 17: 56% Loss of 13: 44% loss of 21 : 38%

loss of l/lp: 79% loss of 14: 16%

Results In total, 174 tumors were analyzed by CGH including 82 clear cell, 53 papillary, 23 chromophobe RCC, and 26 renal oncocytomas" In 161 tumors, genetic alterations were detected .by CGH (93%). Clear cell RCCs were characterized by total or partial loss of chromosomes 3p, 6, 9, 10, 1 and 4 and gain of chromosome 5/5q. In papillary RCC, we frequently found gains of chromosomes 7, 17, 16, 12, and 3 as well as loss of Y chromosome. The typical alteration of chromophobe RCC was combined losses of chromosomes I, 2, 6, 10, 13, 17, and 21. Loss of chromosome 1 and less frequently loss of chromosome 14 occurred in renal oncocytomas. The results are presented in detail in Table 1. A representative CGH profile for each tumor type is given in Fig. 1. Considering staging and metastases, in clear cell RCC, we found an association between progression of disease and losses of chromosomes 9, 10, and 14. Unfortunately, the number of cases was too small for correlation analyses in all other subtypes.

Discussion The VICC classification system for renal tumors (Table 2) supports the concept that each RCC subtype represents an independent tumor entity. Several immunohistochemical and genetic investigations have supported this concept. A tumor-specific genetic fingerprint may be helpful as a diagnostic tool in Table 2. UICC classification system, 1998 (adapted from [9]) Renal cell adenoma

Renal cell carcinoma

Metanephric Papillary Oncocytic

Clear cell Papillary Chromophobe Collecting duct Neuroendocrine Unclassified

174


cases where morphology remains equivocal. Clinical and histological data alone are insufficient for a prediction of individual outcome. There are many data regarding genetic alterations in clear cell RCC, fewer data about papillary RCC, and only some reports about rare tumor types like chromophobe and oncocytic RCC. The aim of our study was to ascertain a standardized genetic fingerprint of each of the four major renal epithelial tumors using CGH. CGH allows for the analysis of chromosomal imbalances, i.e., losses or gains of whole chromosomes or parts of chromosomes. The advantage of CGH is the applicability to archival material; thus artifacts generated by cultured tumor cells can be avoided. Applying CGH to DNA derived from formalin-fixed paraffin-embedded tissues, we were able to demonstrate a specific pattern of genetic alterations for each of the four tumor types. In clear cell RCC, we frequently found deletions on chromosome arm 3p. The high frequency (97%) indicates that loss of 3p is an early event in tumor development of clear cell ReC, independent of tumor stage or grade. This is in concordance with results reported by other groups [4-7]. Loss of chromosomes 9, 10, and 14 as well as gain of chromosome 5/5q were detected with lower frequencies. Alterations of chromosomes 9, 10, and 14 were associated with tumor progression (more frequently in higher stage and grade as well as in metastatic disease). On the other hand, gain of chromosome 5 was more common in clear cell RCC from patients with better outcome. Therefore, CGH-generated genetic data yielded valuable information about individual prognosis in clear cell RCC. The genetic pattern of papillary RCC is completely different from that of clear cell RCC. In papillary tumors, gains of chromosomes 7 and 17 as well as loss of the Y chromosome were frequently observed, whereas gains of chromosomes 16, 12, and 3 were not consistently found. Loss of 3p was not detected. Results are in concordance with published results from other groups [5, 8, 9]. There are only some reports concerning genetic alterations of chromophobe RCC. We detected combinations oflosses of chromosomes 1,2,6,10,13, 17, and 21 as a typical feature of this tumor type. The only frequent genetic alteration of renal oncocytomas was loss of chromosome 1/1 p. Based on loss of chromosome 1 in oncocytomas and chromophobe RCC, Starkel proposed a genetic relationship between both tumors [1]. However, this hypothesis should be corroborated by additional analyses. No alterations were detected in 23 tumors, and in single tumors fingerprints other than those mentioned above were detected. Performing CGH, it is possible to detect only losses and gains of chromosomal regions but not balanced translocations, which have previously been identified for some tumors. Furthermore, deleted or amplified regions can only be detected by CGH if they are larger than 10-20 Mb. Our results demonstrate that clear cell RCC, papillary RCC, chromophobe RCC, and renal oncocytomas are each characterized by a distinct fingerprint of genetic alterations which can be ascertained using CGH. The definition of genetic alterations seems helpful for a reliable RCC subtyping, especially when


175

histopathological features of a given tumor are equivocal. The use of CGH may also help to improve individual prognosis prediction, as was shown for a subset of clear cell RCC.

References 1. Guinan P, Sobin LH, Algaba F, Badellino F, Kameyama S, MacLennan G, Novick A (1997)

2. 3. 4. 5.

6. 7. 8. 9.

TNM staging of renal cell carcinoma: Workgroup No.3. Union International Contre Ie Cancer (urCe) and the American Joint Committee on Cancer (AJCC) Cancer 80:992-933 Storkel, S (1999) [Epithelial tumors of the kidney. Pathological subtyping and cytogenetic correlation]. Urologe A 38:425-432 Chudoba IHT, Senger G, Claussen U, Haas OA (1997) Comparative genomic hybridization using DOP-PCR amplified DNA from a small number of nuclei. Cs Pediat 52:519-521 Kovacs G (1994) The value of molecular genetic analysis in the diagnosis and prognosis of renal cell tumours. World J UroI12:64-68 van den Berg E, van der Hout AH, Oosterhuis JW, Storkel S, Dijkhuizen T, Darn A, Zweers HM, Mensink HJ, Buys CH, de Jong B (1993) Cytogenetic analysis of epithelial renal-cell tumors: relationship with a new histopathological classification. Int J Cancer 55:223-227 Velickovic M, Delahunt B, Grebe SK (1999) Loss of heterozygosity at 3p14.2 in clear cell renal cell carcinoma is an early event and is highly localized to the FHIT gene locus. Cancer Res 59:1323-1326 Bernues M, CasadevallC, Miro R, Caballin MR, Gelabert A, Ejarque MJ, Chechile G, Egozcue J (1998) Analysis of 3p allelic losses in renal cell carcinomas: comparison with cytogenetic results. Cancer Genet Cytogenet 107:121-124 Kattar MM, Grignon DJ, Wallis T, Haas GP, Sakr WA, Pontes JE, Visscher DW (1997) Clinicopathologic and interphase cytogenetic analysis of papillary (chromophilic) renal cell carcinoma. Mod Pathol10:1143-1150 Kovacs G, Fuzesi L, Emanual A, Kung HF (1991) Cytogenetics of papillary renal cell tumors. Genes Chromosomes Cancer 3:249-255