- Email: [email protected]
Inactivation of p16 by CpG hypermethylation in renal cell carcinoma
Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
Original article
Inactivation of p16 by CpG hypermethylation in renal cell carcinoma夡 Marta Vidaurretaa, M. Luisa Maestro, Ph.D.a,*, M. Teresa Sanz-Casla, M.D.a, Carmen Maestro, Ph.D.b, Sara Rafaela, Silvia Veganzonesa, Jesus Moreno, M.D.c, Julia Blanco, M.D.d, Angel Silmi, M.D.c, Manuel Arroyo, M.D.a a
Department of Genomics Laboratory, Hospital Clínico San Carlos, Madrid, Spain b Department of Morphological Science, Complutense University, Madrid, Spain c Department of Urology, Hospital Clínico San Carlos, Madrid, Spain d Department of Pathology, Hospital Clínico San Carlos, Madrid, Spain
Received 11 December 2006; received in revised form 30 January 2007; accepted 30 January 2007
Abstract Objective: Renal carcinoma develops as a consequence of the accumulation of several genetic aberrations. Alterations in the p16 gene have been described in many tumors. Methylation of its promoter in CpG islands is the most common mechanism of inactivation of this gene. The aim of this study was to establish whether p16 gene methylation leads to a loss of the encoded protein in 57 patients with renal carcinoma, and if this aberration has any value in predicting disease progression in these patients. Methods: Gene promoter methylation was determined by deoxyribonucleic acid treated with sodium bisulfite to subsequently amplify methylated and unmethylated regions rich in CpG islands. The p16 protein product was detected for immunohistochemical examination. Results: Hypermethylation of the p16 gene was detected in 22.9% of the patients, none of whom had the protein product. A lack of p16 protein was confirmed in 52.9% of the tumors, indicating another genetic alteration or posttranscriptional modifications preventing the codification of this protein. Through multivariate analysis of overall survival, gene methylation was found to have independent prognostic value: the absence of alteration confers an undefined risk of death. Conclusions: Of the molecular modifications described for renal carcinoma, aberrations in the p16 gene are frequent. In these patients, methylation of the p16 gene promoter seems to afford a protective effect against the risk of death. © 2008 Elsevier Inc. All rights reserved. Keywords: Cancer; Kidney; p16 promoter methylation; p16 protein; Prognosis
1. Introduction Renal carcinoma accounts for close to 3% of all human cancers. As any other tumor, its development occurs as the consequence of the accumulation of several genetic aberrations that also determine disease progression in each patient. The gene p16INK4a, otherwise known as CDKN2, MTS1, INK4a, and CDK4I, has been identified as a cyclin kinasedependent inhibitor acting as a gene suppressor capable of arresting the cell cycle [1]. This gene, located on chromosome 9p21, features 3 exons and codes for a 16 kDa protein 夡 This work is supported by Fundación para la investigación en urología (AEU), Madrid, Spain. * Corresponding author. Tel.: ⫹34-91-3303171; fax: ⫹34-913303280. E-mail address: [email protected] (M.L. Maestro).
1078-1439/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.urolonc.2007.01.018
that inhibits the cyclin kinase D1-CDK4/6 complex responsible for phosphorylating the retinoblastoma protein. In this manner, the cell cycle is stopped in the G1 stage, at least in cells containing functional retinoblastoma protein [2]. Genetic modifications in the 9p21 region are common in human tumors [3]. Several mechanisms whereby the p16 gene is inactivated have been described, including: deletion, promoter region methylation, point mutations (generally non-sense) and mutations in the reading frame; their incidence depending on tumor type [4]. Methylation of the 5= region in CpG islands is a significant transcriptional repression mechanism [5]. In normal cells, CpG islands are not methylated except during chromosome X inactivation. Merlo et al. [6] observed that, although the loss of heterozygosity in 9p21 is 1 of the most frequently identified genetic alterations in human cancer, p16 point mutations in the other allele are relatively rare. The methylation of some
240
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
genes plays an important role in tumorigenesis. De novo methylation of the CpG sequences in the p16 gene promoter region has been detected in approximately 20% of several primary neoplasias [6]. In primary renal cancer, loss of heterozygosity in chromosome 9p has been noted in 20% to 30% of cases [7,8], yet p16 inactivation by homozygous deletion or point mutation is rare [9,10]. The present study was designed to establish whether methylation of the p16 suppressor gene promoter produces inactivation and a consequent lack of p16 protein expression. A second objective was to evaluate the prognostic value of this alteration in patients undergoing surgery for renal carcinoma.
2. Patients and methods 2.1. Patients This prospective cohort study was performed on 57 patients operated on for primary renal carcinoma at the Urology Unit of the Hospital Clínico San Carlos, Madrid, Spain. No patient had received previous chemotherapy or radiotherapy. Informed consent to participate in the study was obtained from each patient. The Hospital Ethics Committee approved the study protocol. 2.2. Methods 2.2a. Sample processing. Tumor samples and corresponding normal tissue samples were obtained from each patient during surgery. The tumor specimen was divided into 2 fragments: 1 for genetic analysis and the other for histopathology. The first of these fragments was immediately introduced in liquid nitrogen and stored at ⫺80°C. Two pathologists independently performed the anatomopathologic examination, especially for this study. It was confirmed that all tumor specimens had at least 80% tumor cells. 2.2b. Histologic examination. The histologic examination was performed in at least 2 zones of each sample, including both tumor and nontumor sites. The tumors were classified according to the Störkel classification scheme of 1997 [11]
as clear cell, papillary, chromophobe, or Bellini duct carcinomas. The degree of cell differentiation was determined according to the classification of Fuhrman et al. [12]. The tumors were staged using the tumor-nodes-metastasis classification proposed by the Union Internationale Contre le Cancer and American Joint Committee on Cancer of 1997 [13]. 2.3. Analytical determinations 2.3a. Methylation status of the p16 gene promoter. Specimens were treated with proteinase K and the deoxyribonucleic acid (DNA) of each tumor specimen extracted using standard phenol-chloroform procedures. The status of p16 gene promoter methylation was determined in 1.0 g samples of DNA, first denaturized with 3.0 M NaOH for 15 minutes at 37°C and then treated with sodium bisulfite at 50°C for 18 hours. Sodium bisulfite converts unmethylated cytosines to uracils and leaves methylated cytosines unchanged. Uracil is replicated as thymine during polymerase chain reaction (PCR), allowing for methylation analysis by designing primers that preferentially anneal to sequences containing either methylated (CpG) or unmethylated (TpG) sites. Sodium bisulfite modified DNA was purified by the Wizard’s resin filter method (Promega Corp., Madison, WI). The samples were incubated at 37°C for 15 minutes with 0.3 M NaOH in 50 l reactions and again purified with DNA QIAamp columns (QIAGEN, Inc., Valencia, CA). Modified DNA was PCR amplified using primers (Table 1) specific for methylated and unmethylated CpG regions of the p16 promoter. The PCR mixture contained PCR buffer, MgCl2 (1.5 mM), triphosphate deoxynucleotides (200 M of each), primers (0.4 M of each), modified DNA (50 g), and 2.5 U Taq polymerase (Roche Diagnostics, Mannheim, Germany) in a reaction volume of 50 l. The PCR amplification conditions were: 5 minutes at 95°C, 35 cycles of 30 seconds at 95°C; 1 minute at 69°C and 1 minute at 72°C; and a final extension at 72°C for 10 minutes. PCR products were loaded onto 2% agarose gels stained with ethidium bromide to ascertain the methylation state in a ultraviolet transilluminator. 2.3b. p16 Immunohistochemistry. Paraffin sections were mounted on poly-lysine or xylene-coated slides, dewaxed,
Table 1 Sequences of the primers used in the PCR amplifications Chromosome
Locus symbol
Primers: sense 5=¡3=
Primer: antisense 5=¡3=
Ta
Size bp
9p
p16-W p16-M p16-U p16-M2 p16-U2
CAGAGGGTGGGGCGGACCGC TTATTAGAGGGTGGGGCGGATCGC TTATTAGAGGGTGGGGTGGATTGT TTATTAGAGGGTGGGGCGGATCGC TTATTAGAGGGTGGGGTGGATTGT
CGGGCCGCGGCCGTGG GACCCCGAACCGCGACCGTAA CAACCCCAAACCACAACCATAA CCACCTAAATCGACCTCCGACCG CCACCTAAATCAACCTCCAACCA
65 65 60 65 60
140 150 151 234 234
p16-W represents the nonmodified DNA primer. p16-M and p16-M2 represent the modified and methylated DNA primers. p16-U and p16-U2 represent the modified and unmethylated DNA primers. Differences between the modified methylated and unmethylated DNA primers are underlined.
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
rehydrated, and incubated for 10 minutes with 3% hydrogen peroxide in methanol in the dark, to block endogenous peroxidase activity. Sections were pretreated with ethylenediaminetetraacetic acid in Tris-buffer in a microwave oven for 30 minutes and blocked with 10% normal serum bovine in Tris-buffer. Subsequently, the sections were incubated with the anti p-16 mouse monoclonal antibodies (F-12; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) diluted 1:200 in Tris-buffer for 2 hours in the dark. Sections were then rinsed in Tris buffer (pH 7.4 –7.6) and finally incubated with Envision® (Dako Denmark A/S, Glostrup, Denmark) 1:2 in Tris buffer as a secondary antibody for 30 minutes in the dark at room temperature. After rinsing, staining was developed with 3,3=-diaminobenzidine tetrahydrochloride (DAB; Dako Denmark A/S), following the manufacturer’s instructions. The sections were dehydrated and coverslipped using Eukit (Sigma-Aldrich, St. Louis, MO) mounting medium. A previously known positive renal carcinoma that strongly expresses p16 protein was used as a positive control. Negative controls did not include the primary antibody, and it was replaced by 0.01 mol/l phosphatebuffered saline, and normal renal tissue was used as normal control. For scoring the p16 staining patterns, we used previously published criteria [14]. Specimens were considered immunohistochemically positive and, therefore, positive for p16 protein expression when cell nuclei showed more than 10% staining, regardless of staining of the cytoplasm. 2.4. Statistical analysis Qualitative variables are provided with their distribution frequencies. Quantitative variables were summarized as their mean, standard deviation, and range. Associations among qualitative variables were evaluated using the Pearson 2 test or Fisher exact test when 25% of expected associations were less than 5. For overall survival, the event was defined as deaths produced as a consequence of the tumor, with live patients or those dying of another cause excluded. Overall survival was calculated as the time elapsed from the date of surgery and the date of death or last follow-up. For disease-free survival, the event was defined as a diagnosis of a locoregional or distant recurrence in patients previously free of disease (i.e., in all those undergoing curative surgery). Disease-free survival was calculated as the time elapsed from the date of surgery to diagnosis of the first recurrence. The functions overall survival and disease-free survival were estimated by the KaplanMeier method and compared between groups using Breslow’s exact test. Data were fitted to a proportional risk regression model, and the assumption of proportionality of risks was confirmed. Adjusted relative risks and their significance were estimated. In each hypothesis contrast, the null hypothesis was rejected when the type I error was less
241
Table 2 Description of the population of 57 patients undergoing surgery for renal carcinoma Variable Sex Male Female Age (yrs) ⬎60 ⱕ60 Tumor size T1 and T2 T3 Lymph nodes N0 N1 Distance metastasis M0 M1 Stage I II III IV Grouped stage I and II III and IV Histologic type Clear cell Papillary cell Chromophobe cell Unclassified Bellini duct Grouped histology Papillary Nonpapillary Fuhrman grade I II III IV
No. (%) 39 (68.4) 18 (31.6) 31 (54.4) 26 (45.6) 35 (61.4) 22 (38.6) 54 (94.7) 3 (5.3) 56 (98.1) 1 (1.8) 32 (56.1) 2 (3.5) 22 (38.6) 1 (1.8) 34 (59.6) 23 (40.4) 38 (66.7) 5 (8.8) 8 (14.0) 2 (3.5) 4 (7.0) 5 (8.8) 52 (91.2) 13 (22.8) 31 (54.4) 12 (21.1) 1 (1.8)
than 0.05. All statistical analyses were performed using SSPS version 11.5 software (SPSS, Inc., Chicago, IL).
3. Results The mean age of our series of 57 patients diagnosed with renal carcinoma was 61.3 ⫾ 12.4 years, with a range of 36 – 83 years. A total of 39 patients (68.4%) were men, and 18 (31.6%) were women. Table 2 shows detailed information on the tumor characteristics. p16 methylation results were obtained for 48 patients because of a lack of specimens from the remaining 9 patients. The proportion of patients with p16 methylation was 22.9% (11 patients), and unmethylated p16 was 78.1% (37). The expression of p16 protein was determined in 34 of the 48 patients, and was positive in 16 (47.1%) and negative in 18 (52.9%). A lack of p16 expression was observed in 100%
242
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
Table 3 p16 methylation according to tumor characteristics in 48 patients undergoing surgery for renal carcinoma Variable
Sex Male Female Age (yrs) ⬎60 ⱕ60 Tumor size T1 and T2 T3 Lymph nodes N0 N1 Distance metastasis M0 M1 Stage I II III IV Grouped stage I and II III and IV Histologic type Clear cell Papillary cell Chromophobe cell Unclassified Bellini duct Grouped histology Papillary Nonpapillary Fuhrman grade I II III IV
No. p16 methylated (%)
No. p16 unmethylated (%)
3.1. Postoperative disease progression: Analysis of overall survival P value
1.00 8 (24.2) 3 (20.0)
25 (75.8) 12 (80.0)
2 (8.0) 9 (39.1)
23 (92.0) 14 (60.9)
7 (25.0) 4 (20.0)
21 (75.0) 16 (80.0)
11 (24.4) 0 (0)
34 (75.6) 3 (100)
11 (23.4) 0 (0)
36 (76.6) 1 (100)
5 (20.0) 2 (100) 4 (20.0) 0 (0)
20 (80.0) 0 (0) 16 (80.0) 1 (100)
7 (25.9) 4 (19.0)
20 (74.1) 17 (81.0)
6 (19.4) 0 (0) 4 (57.1) 0 (0) 1 (25)
25 (80.6) 4 (100) 3 (42.9) 2 (100) 3 (75)
0.01
0.70
1.00
1.00
0.08
0.10
0.50 4 (100) 33 (75) 0.60 4 (36.4) 5 (18.5) 2 (22.2) 0 (0)
Table 4 p16 protein expression according to tumor characteristics in 34 patients undergoing surgery for renal carcinoma Variable
0.70
0 (0) 11 (25)
The median follow-up time of our study was 76 months (6 years), with an interquartile range of 2–117 months. In our patient population, 6-year overall survival was 85%. All our survival analyses are referred to our median length of follow-up. Of the 48 patients, 7 died during follow-up, all as a consequence of the neoplasia. The overall survival of the patients with p16 gene methylation was 100% and of those without this gene alteration was 81% (P ⫽ 0.14). The relative risk of dying for those patients without methylation was undefined (P ⫽ 0.02) with respect to those with the alteration, because no deaths were recorded in the subset of patients with a methylated p16
7 (63.6) 22 (81.5) 7 (77.8) 1 (100)
Statistically significant: P ⬍ 0.05.
of the tumors showing methylation and in 36% of those without this genetic alteration. We also examined the relationship between the gene methylation status and p16 protein and the clinicopathologic variables described in Tables 3 and 4. A significant relationship was observed between patient age and the genetic aberration: 8% of patients older than 60 years had methylation, and 39.1% of those aged ⱕ60 years (P ⫽ 0.01) did not. Although the association with histologic type was not significant, it is noteworthy that all the papillary tumors expressed the p16 protein yet lacked promoter methylation. In 80% of the chromophobe carcinomas, no p16 protein expression was detected. Interestingly, 38.9% and 64.3% of the tumors of Fuhrman stages I and III, respectively, did not express the protein.
Sex Male Female Age (yrs) ⬎60 ⱕ60 Tumor size T1 and T2 T3 Ganglia affected N0 N1 Distance metastasis M0 M1 Stage I II III IV Grouped stage I and II III and IV Histologic type Clear cell Papillary cell Chromophobe cell Unclassified Bellini duct Grouped histology Papillary Nonpapillary Fuhrman grade I II III IV p16 methylation Methylation No methylation
No. p16 IHC (⫹) (%)
No. p16 IHC (⫺) (%)
11 (47.8) 5 (45.5)
12 (52.2) 6 (54.5)
9 (50.0) 7 (43.8)
9 (50.0) 9 (56.3)
11 (55.0) 5 (35.7)
9 (45.0) 9 (64.3)
1 (100) 15 (45.5)
0 (0) 18 (54.5)
16 (47.1) 0 (0)
18 (52.9) 0 (0)
11 (61.1) 0 (0) 5 (35.7)
7 (38.9) 2 (100) 9 (64.3)
11 (55.0) 5 (35.7)
9 (45.0) 9 (64.3)
12 (50.0) 2 (100) 1 (20.0) 1 (33.3)
12 (50.0) 0 (0) 4 (80.0) 2 (66.7)
2 (100) 14 (43.8)
0 (0) 18 (56.3)
4 (44.4) 9 (50.0) 3 (42.9)
57 (55.6) 9 (50.0) 4 (57.1)
0 (0) 16 (64.0)
9 (100) 9 (36.0)
P value 1.00
0.70
0.30
0.40
0.09
0.30
0.10
0.07
0.10
0.001
Statistically significant: P ⬍ 0.05.
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
gene promoter. When a tumor expressed p16, overall survival was 78%, and in the absence of expression, the rate was 93% (P ⫽ 0.17). The relative risk of dying was 2.30 times higher (95% confidence interval [CI] 0.50 –10.53) in the patients with the protein (P ⫽ 0.22), although the relationship was not statistically significant. We also examined overall survival in the patients stratified according to the classic factors used to predict the risk of death in patients with renal carcinoma (histologic type, tumor stage, and differentiation). Neither the methylation state nor protein expression served to establish groups with a different prognosis. In the multivariate analysis, we included clinicopathologic variables (age, sex, tumor stage, histologic type, and cell differentiation) and the molecular alterations examined (gene methylation and protein expression). Only methylation was found to have independent prognostic value: the absence of alteration confers an undefined risk of death (P ⫽ 0.03), independently of the other variables. With regard to expression of the p16 protein, the relative risk of dying was 8.47 times higher in patients who had the protein (95% CI 0.77–92.46) (P ⫽ 0.06). 3.2. Postoperative disease progression: Analysis of disease-free survival In the disease-free survival analysis, 1 patient with a distant metastasis preoperatively was excluded. Diseasefree survival for our study population was 79%. During follow-up, 10 patients had recurrence of their tumor. In patients with p16 gene methylation, disease-free survival was 90%, and in those without the genetic alteration disease-free survival, it was 75% (P ⫽ 0.30). The relative risk of recurrence in those without methylation was 2.78fold higher (95% CI 0.34 –22.22) than in patients with the aberration (P ⫽ 0.27). When the tumors expressed p16, disease-free survival was 73%, and if they lacked the expression of the protein, it was 80% (P ⫽ 0.40). The relative risk of tumor recurrence was 1.56 times greater (95% CI 0.35–7.02) in patients with the protein (P ⫽ 0.55). In the stratified disease-free survival analysis, neither gene methylation nor gene product expression served to establish groups of patients with a different prognosis. Neither did gene methylation nor protein expression had any prognostic value independent of the clinicopathologic variables.
4. Discussion Many genetic aberrations have been observed in renal carcinoma. Analysis of the loss of heterozygosity in several chromosomes of these tumors has indicated their polygenic origin (i.e., multiple genetic alterations are needed for the renal tumor to develop) [15]. The methylation of some genes plays an important role
243
in tumorigenesis as a gene inactivation mechanism. It has been shown that transcriptional gene silencing through methylation of the CpG islands is an alternative inactivation mechanism, which has been described in genes such as the Von Hippel-Lindau (VHL), retinoblastoma (Rb), p15, and p16 [16]. Some renal cell carcinomas show methylation of 5=CpG islands of the VHL gene (19%), and some retinoblastomas present de novo methylation in the Rb gene [17,18]. The incidence of gene methylation in different tumors ranges from 20% to 67%, and can be even higher in cell cultures of the same tumor type. For example, in 90% of colon cancer cells, the p16 gene promoter is methylated. Hypermethylation of the promoter of certain genes differs in prevalence, depending on the type of tumor analyzed, showing a tumor-specific and gene-specific pattern [19]. In our study, the incidence of methylation of the p16 promoter was 22.9% in renal tumors, and a lack of p16 protein expression was observed in 52.9% of these cases. Esteller et al. [19] observed p16 gene hypermethylation in a similar proportion (23%) of renal tumors to us. However, Kawada et al. [20] detected p16 hypermethylation in only 3.3% of the renal cancers they examined, and 5.5% lacked the p16 protein detected immunohistochemically. These investigators concluded that p16 gene inactivation is an infrequent event in primary renal carcinoma in the Japanese population [20]. We consider this to indicate that p16 methylation in renal tumors is also population specific, although further studies are needed to establish prevalences in the different populations. Here, we noted that hypermethylation of the p16 promoter led to transcriptional silencing, which is a common observation in renal cell carcinoma. Kawada et al. [20] also noted a close link between the immunohistochemical expression of p16 and hypermethylation of the gene promoter in renal tumors. This correlation has also been described in other tumors, such as gastrointestinal and hepatic [21–24]. In the present study, in no tumor with p16 methylation was the nuclear expression of the protein detected. Of the 18 tumors immunohistochemically negative for the gene product, gene methylation was noted in 9 (50%). In the remaining 50% of the patients in whom p16 methylation was absent, there must be some other genetic defect responsible for the protein not being encoded. The mechanism of inactivation of the p16 gene in this group of tumors remains unclear because homozygous deletions and point mutations are rare in renal carcinoma [25] and other tumor types [26]. Some investigators [21] suggest that this genetic silencing could be the result of a posttranscriptional modification in tumors lacking gene methylation. We detected no significant relationship between p16 gene methylation status and the expression of the protein and the stage or degree of cell differentiation of the tumors. Nevertheless, it is of interest that none of the papillary tumors showed p16 methylation, yet they all expressed the p16 protein. In this histologic type, we have not detected aberration of the p16 gene. Several investigators [27] argue
244
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
that genetic analysis is an important tool for the diagnosis and prognosis of renal carcinoma, and have suggested different genetic patterns depending on the histologic tumor type. In our patients, p16 methylation was significantly correlated with age and was found to be most frequent in patients younger than 60 years. A link between gene methylation status and age has also been reported for other tumors [28]. With regard to the prognostic value of aberrant methylation of the p16 gene, in our study, patients with p16 hypermethylation had a 6-year overall survival of 100%, and those without hypermethylation had a rate of 81% (P ⫽ 0.14). Although not significant, this difference indicates a clinical tendency in that there were no deaths in the group of patients with the gene aberration. Moreover, patients with p16 promoter hypermethylation had a lower risk of dying than those without this aberration, regardless of the classic prognostic factors described for renal cancer included in the multivariate analysis. To our knowledge, this protective effect has not been previously described in renal carcinoma or any other type of tumor. We have sought an explanation for this in the interpretation of colorectal tumorigenesis, in which the successive genetic aberrations that accumulate at each stage have been widely explored. Results have led to the hypothesis that the colorectal tumor can occur via 2 different mechanisms: 1 of chromosome instability, whereby oncogenes and gene suppressors are consecutively altered; or via the instability of microsatellites (MSI) or mutant phenotype that occurs from the aberration of repair genes, especially hMLH1 and hMSH2 [29,30]. The MSI mechanism has been associated with a less aggressive component. In colorectal cancer, methylation of the p16 gene has been considered as part of the “methylator phenotype,” which includes the methylation of CpG islands described by Toyota et al. [30]. This phenotype defines a group of tumors characterized by the methylation of multiple loci, including the hMLH1 promoter. Several investigators [30,31] have reported that this phenotype is associated with MSI. This association can be explained by the theory that epigenetic changes, which include methylation of the promoters of certain genes such as p16 and hMLH1 blocking their transcription, contribute to a modification of the actions of the repair genes that condition the MSI phenotype [32,33]. It is generally accepted that patients who have colorectal tumors develop via the MSI route have a better prognosis [34,35]. Norrie et al. [21] observed that a lack of p16 expression was related to MSI in colorectal tumors and that this subset of tumors had different clinicopathologic features. Regardless of the mechanism whereby p16 gene aberrations contribute to renal tumorigenesis, the findings of our study indicate that methylation of the gene confers a better prognosis to patients with cancer of the kidney. An interesting next step in this investigation will be to ascertain the mutant phenotype in this patient population, to
check its association with the p16 aberration and its possible prognostic value.
5. Conclusions Our findings indicate that p16 gene aberrations are frequent events among the molecular modifications already described for renal carcinoma and that methylation of the gene promoter is 1 of the most common mechanisms of silencing the p16 gene. Our findings also point to a protective role of p16 gene hypermethylation against the risk of dying that these patients carry.
References [1] Liggett WH, Sidransky D. Role of the p16 tumor suppressor gene in cancer. J Clin Oncol 1998;16:1197–206. [2] Serrano M, Hannon GJ, Beach D. A new regulatory motif in cellcycle control causing specific inhibition of cyclin D/CDK4. Nature 1993;366:704 –7. [3] Nobori P, Miura K, Wu DJ, et al. Deletions of the cyclin-dependent kinase-4 inhibitor gene in multiple human cancers. Nature 1994;368: 753– 6. [4] Herman JG, Merlo A, Mao L, et al. Inactivation of the CDKN2/p16/ MTS1 gene is frequently associated with aberrant DNA methylation in all common human cancers. Cancer Res 1995;55:4525–30. [5] Bird A. The essentials of DNA methylation. Cell 1992;70:5– 8. [6] Merlo A, Herman JG, Mao L, et al. 5= CpG island methylation is associated with transcriptional silencing of the tumour suppressor p16/CDKN2/MTS-1 in human cancers. Nature Med 1995;1:686 –92. [7] Moch H, Presti JC, Sauter G, et al. Genetic aberrations detected by comparative genomic hybridization are associated with clinical outcome in renal cell cancer. Cancer Res 1996;56:27–30. [8] Schullerus D, Herbers J, Chudek J, et al. Loss of heterozygosity at chromosomes 8p, 9p, and 14p is associated with stage and grade of non-papillary renal cell carcinomas. J Pathol 1997;183:151–5. [9] Cairns P, Tokino K, Eby Y, et al. Localization of tumor suppressor loci on chromosome 9 in primary human renal cell carcinomas. Cancer Res 1995;55:224 –7. [10] Kinoshita H, Yamada H, Ogawa O, et al. Contribution of chromosome 9p21–22 deletion to the progression of human renal cell carcinoma. Jpn J Cancer Res 1995;86:795–9. [11] Störkel S, Eble JN, Adlakha K, et al. Classification of renal cell carcinoma: Workgroup No. 1. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer 1997;80:987–9. [12] Fuhrman SA, Lasky LC, Limas C. Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 1982; 6:655– 63. [13] Guinan P, Sobin LH, Algaba F, et al. TNM staging of renal cell carcinoma: Workgroup No. 3. Union International Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer 1997;80:992–3. [14] Gerardts J, Hruban RH, Schutte M, et al. Immunohistochemical p16INK4a analysis of archival tumors with deletion, hypermethylation, or mutation of CDKN2/MTS1 gene. A comparison of four commercial antibodies. Appl Immunohistochem Mol Morphol 2000; 8:71–9. [15] Thrash-Bingham CA, Salazar H, Freed J, et al. Genomic alterations and instabilities in renal cell carcinomas and their relationship to tumor pathology. Cancer Res 1995;55:6189 –95.
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245 [16] Kashiwabara K, Oyama T, Sano T, et al. Correlation between methylation status of the p16/CDKN2 gene and the expression of p16 and Rb proteins in primary non-small cell lung cancers. Int J Cancer 1998;79:215–20. [17] Herman JG, Latif F, Weng Y, et al. Silencing of the VHL tumorsuppressor gene by DNA methylation in renal carcinoma. Proc Natl Acad Sci USA 1994;91:9700 – 4. [18] Sakai T, Toguchida J, Ohtani N, et al. Allele-specific hypermethylation of the retinoblastoma tumor-suppressor gene. Am J Hum Genet 1991;48:880 – 8. [19] Esteller M, Corn PG, Baylin SB, et al. A gene hypermethylation profile of human cancer. Cancer Res 2001;61:3225–9. [20] Kawada Y, Nakamura M, Ishida E, et al. Aberrations of the p14ARF and p16INK4␣ genes in renal cell carcinomas. Jpn J Cancer Res 2001;92:1293–9. [21] Norrie MWA, Hawkins NJ, Todd AV, et al. Inactivation of p16INK4a by CpG hypermethylation is not a frequent event in colorectal cancer. J Surg Oncol 2003;84:143–50. [22] Schneider-Stock R, Boltze C, Lasota J, et al. High prognostic value of p16INK4 alterations in gastrointestinal stromal tumors. J Clin Oncol 2003;21:1688 –97. [23] Qin Y, Liu J, Li B, et al. Association of low p16INK4a and p15INK4b mRNAs expression with their CpG islands methylation with human hepatocellular carcinogenesis. World J Gastroenterol 2004;10:1276 – 80. [24] Kaneto H, Sasaki S, Yamamoto H, et al. Detection of hypermethylation of the p16INK4A gene promoter in chronic hepatitis and cirrhosis associated with hepatitis B or C. Gut 2001;48:372–7. [25] Kinoshita H, Yamada H, Ogawa O, et al. Contribution of chromosome 9p21–22 deletion to the progression of human renal cell carcinoma. Jpn J Cancer Res 1995;86:795–9.
245
[26] Okamoto A, Demetrick DJ, Spillare EA, et al. Mutations and altered expression of p16INK4 in human cancer. Proc Natl Acad Sci USA 1994;91:11045–9. [27] Junker K, Weirich G, Amin MB, et al. Genetic subtyping of renal cell carcinoma by comparative genomic hybridization. Recent Results Cancer Res 2003;162:169 –75. [28] Wiencke JK, Zheng S, Lafuente A, et al. Aberrant methylation of p16INK4a in anatomic and gender-specific subtypes of sporadic colorectal cancer. Cancer Epidemiol Biomarkers Prev 1999;8: 501– 6. [29] Lengauer C, Kinzler K, Vogelstein B. Genetic instability in colorectal cancers. Nature 1997;386:623–7. [30] Toyota M, Ohe-Toyota M, Ahuja N, et al. Distinct genetic profiles in colorectal tumors with or without the CpG island methylator phenotype. Proc Natl Acad Sci USA 2000;97:710 –5. [31] Hawkins N, Norrie M, Cheong K, et al. CpG island methylation in sporadic colorectal cancers and its relationship to microsatellite instability. Gastroenterology 2002;122:1376 – 87. [32] Kane MF, Loda M, Gaida GM, et al. Methylation of hMLH1 promoter correlates with lack of expression of hMLH1 in sporadic colon tumors and mismatch repair-defective human tumor cell lines. Cancer Res 1997;57:808 –11. [33] Ward RL, Cheong K, Ku SL, et al. Adverse prognostic effect of methylation in colorectal cancer is reversed by microsatellite instability. J Clin Oncol 2003;21:3729 –36. [34] Gryfe R, Kim H, Hsieh ET, et al. Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer. N Engl J Med 2000;342:69 –77. [35] Wright CM, Dent OF, Barker M, et al. Prognostic significance of extensive microsatellite instability in sporadic clinicopathological stage C colorectal cancer. Br J Surg 2000;87:1197–202.
Original article
Inactivation of p16 by CpG hypermethylation in renal cell carcinoma夡 Marta Vidaurretaa, M. Luisa Maestro, Ph.D.a,*, M. Teresa Sanz-Casla, M.D.a, Carmen Maestro, Ph.D.b, Sara Rafaela, Silvia Veganzonesa, Jesus Moreno, M.D.c, Julia Blanco, M.D.d, Angel Silmi, M.D.c, Manuel Arroyo, M.D.a a
Department of Genomics Laboratory, Hospital Clínico San Carlos, Madrid, Spain b Department of Morphological Science, Complutense University, Madrid, Spain c Department of Urology, Hospital Clínico San Carlos, Madrid, Spain d Department of Pathology, Hospital Clínico San Carlos, Madrid, Spain
Received 11 December 2006; received in revised form 30 January 2007; accepted 30 January 2007
Abstract Objective: Renal carcinoma develops as a consequence of the accumulation of several genetic aberrations. Alterations in the p16 gene have been described in many tumors. Methylation of its promoter in CpG islands is the most common mechanism of inactivation of this gene. The aim of this study was to establish whether p16 gene methylation leads to a loss of the encoded protein in 57 patients with renal carcinoma, and if this aberration has any value in predicting disease progression in these patients. Methods: Gene promoter methylation was determined by deoxyribonucleic acid treated with sodium bisulfite to subsequently amplify methylated and unmethylated regions rich in CpG islands. The p16 protein product was detected for immunohistochemical examination. Results: Hypermethylation of the p16 gene was detected in 22.9% of the patients, none of whom had the protein product. A lack of p16 protein was confirmed in 52.9% of the tumors, indicating another genetic alteration or posttranscriptional modifications preventing the codification of this protein. Through multivariate analysis of overall survival, gene methylation was found to have independent prognostic value: the absence of alteration confers an undefined risk of death. Conclusions: Of the molecular modifications described for renal carcinoma, aberrations in the p16 gene are frequent. In these patients, methylation of the p16 gene promoter seems to afford a protective effect against the risk of death. © 2008 Elsevier Inc. All rights reserved. Keywords: Cancer; Kidney; p16 promoter methylation; p16 protein; Prognosis
1. Introduction Renal carcinoma accounts for close to 3% of all human cancers. As any other tumor, its development occurs as the consequence of the accumulation of several genetic aberrations that also determine disease progression in each patient. The gene p16INK4a, otherwise known as CDKN2, MTS1, INK4a, and CDK4I, has been identified as a cyclin kinasedependent inhibitor acting as a gene suppressor capable of arresting the cell cycle [1]. This gene, located on chromosome 9p21, features 3 exons and codes for a 16 kDa protein 夡 This work is supported by Fundación para la investigación en urología (AEU), Madrid, Spain. * Corresponding author. Tel.: ⫹34-91-3303171; fax: ⫹34-913303280. E-mail address: [email protected] (M.L. Maestro).
1078-1439/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.urolonc.2007.01.018
that inhibits the cyclin kinase D1-CDK4/6 complex responsible for phosphorylating the retinoblastoma protein. In this manner, the cell cycle is stopped in the G1 stage, at least in cells containing functional retinoblastoma protein [2]. Genetic modifications in the 9p21 region are common in human tumors [3]. Several mechanisms whereby the p16 gene is inactivated have been described, including: deletion, promoter region methylation, point mutations (generally non-sense) and mutations in the reading frame; their incidence depending on tumor type [4]. Methylation of the 5= region in CpG islands is a significant transcriptional repression mechanism [5]. In normal cells, CpG islands are not methylated except during chromosome X inactivation. Merlo et al. [6] observed that, although the loss of heterozygosity in 9p21 is 1 of the most frequently identified genetic alterations in human cancer, p16 point mutations in the other allele are relatively rare. The methylation of some
240
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
genes plays an important role in tumorigenesis. De novo methylation of the CpG sequences in the p16 gene promoter region has been detected in approximately 20% of several primary neoplasias [6]. In primary renal cancer, loss of heterozygosity in chromosome 9p has been noted in 20% to 30% of cases [7,8], yet p16 inactivation by homozygous deletion or point mutation is rare [9,10]. The present study was designed to establish whether methylation of the p16 suppressor gene promoter produces inactivation and a consequent lack of p16 protein expression. A second objective was to evaluate the prognostic value of this alteration in patients undergoing surgery for renal carcinoma.
2. Patients and methods 2.1. Patients This prospective cohort study was performed on 57 patients operated on for primary renal carcinoma at the Urology Unit of the Hospital Clínico San Carlos, Madrid, Spain. No patient had received previous chemotherapy or radiotherapy. Informed consent to participate in the study was obtained from each patient. The Hospital Ethics Committee approved the study protocol. 2.2. Methods 2.2a. Sample processing. Tumor samples and corresponding normal tissue samples were obtained from each patient during surgery. The tumor specimen was divided into 2 fragments: 1 for genetic analysis and the other for histopathology. The first of these fragments was immediately introduced in liquid nitrogen and stored at ⫺80°C. Two pathologists independently performed the anatomopathologic examination, especially for this study. It was confirmed that all tumor specimens had at least 80% tumor cells. 2.2b. Histologic examination. The histologic examination was performed in at least 2 zones of each sample, including both tumor and nontumor sites. The tumors were classified according to the Störkel classification scheme of 1997 [11]
as clear cell, papillary, chromophobe, or Bellini duct carcinomas. The degree of cell differentiation was determined according to the classification of Fuhrman et al. [12]. The tumors were staged using the tumor-nodes-metastasis classification proposed by the Union Internationale Contre le Cancer and American Joint Committee on Cancer of 1997 [13]. 2.3. Analytical determinations 2.3a. Methylation status of the p16 gene promoter. Specimens were treated with proteinase K and the deoxyribonucleic acid (DNA) of each tumor specimen extracted using standard phenol-chloroform procedures. The status of p16 gene promoter methylation was determined in 1.0 g samples of DNA, first denaturized with 3.0 M NaOH for 15 minutes at 37°C and then treated with sodium bisulfite at 50°C for 18 hours. Sodium bisulfite converts unmethylated cytosines to uracils and leaves methylated cytosines unchanged. Uracil is replicated as thymine during polymerase chain reaction (PCR), allowing for methylation analysis by designing primers that preferentially anneal to sequences containing either methylated (CpG) or unmethylated (TpG) sites. Sodium bisulfite modified DNA was purified by the Wizard’s resin filter method (Promega Corp., Madison, WI). The samples were incubated at 37°C for 15 minutes with 0.3 M NaOH in 50 l reactions and again purified with DNA QIAamp columns (QIAGEN, Inc., Valencia, CA). Modified DNA was PCR amplified using primers (Table 1) specific for methylated and unmethylated CpG regions of the p16 promoter. The PCR mixture contained PCR buffer, MgCl2 (1.5 mM), triphosphate deoxynucleotides (200 M of each), primers (0.4 M of each), modified DNA (50 g), and 2.5 U Taq polymerase (Roche Diagnostics, Mannheim, Germany) in a reaction volume of 50 l. The PCR amplification conditions were: 5 minutes at 95°C, 35 cycles of 30 seconds at 95°C; 1 minute at 69°C and 1 minute at 72°C; and a final extension at 72°C for 10 minutes. PCR products were loaded onto 2% agarose gels stained with ethidium bromide to ascertain the methylation state in a ultraviolet transilluminator. 2.3b. p16 Immunohistochemistry. Paraffin sections were mounted on poly-lysine or xylene-coated slides, dewaxed,
Table 1 Sequences of the primers used in the PCR amplifications Chromosome
Locus symbol
Primers: sense 5=¡3=
Primer: antisense 5=¡3=
Ta
Size bp
9p
p16-W p16-M p16-U p16-M2 p16-U2
CAGAGGGTGGGGCGGACCGC TTATTAGAGGGTGGGGCGGATCGC TTATTAGAGGGTGGGGTGGATTGT TTATTAGAGGGTGGGGCGGATCGC TTATTAGAGGGTGGGGTGGATTGT
CGGGCCGCGGCCGTGG GACCCCGAACCGCGACCGTAA CAACCCCAAACCACAACCATAA CCACCTAAATCGACCTCCGACCG CCACCTAAATCAACCTCCAACCA
65 65 60 65 60
140 150 151 234 234
p16-W represents the nonmodified DNA primer. p16-M and p16-M2 represent the modified and methylated DNA primers. p16-U and p16-U2 represent the modified and unmethylated DNA primers. Differences between the modified methylated and unmethylated DNA primers are underlined.
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
rehydrated, and incubated for 10 minutes with 3% hydrogen peroxide in methanol in the dark, to block endogenous peroxidase activity. Sections were pretreated with ethylenediaminetetraacetic acid in Tris-buffer in a microwave oven for 30 minutes and blocked with 10% normal serum bovine in Tris-buffer. Subsequently, the sections were incubated with the anti p-16 mouse monoclonal antibodies (F-12; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) diluted 1:200 in Tris-buffer for 2 hours in the dark. Sections were then rinsed in Tris buffer (pH 7.4 –7.6) and finally incubated with Envision® (Dako Denmark A/S, Glostrup, Denmark) 1:2 in Tris buffer as a secondary antibody for 30 minutes in the dark at room temperature. After rinsing, staining was developed with 3,3=-diaminobenzidine tetrahydrochloride (DAB; Dako Denmark A/S), following the manufacturer’s instructions. The sections were dehydrated and coverslipped using Eukit (Sigma-Aldrich, St. Louis, MO) mounting medium. A previously known positive renal carcinoma that strongly expresses p16 protein was used as a positive control. Negative controls did not include the primary antibody, and it was replaced by 0.01 mol/l phosphatebuffered saline, and normal renal tissue was used as normal control. For scoring the p16 staining patterns, we used previously published criteria [14]. Specimens were considered immunohistochemically positive and, therefore, positive for p16 protein expression when cell nuclei showed more than 10% staining, regardless of staining of the cytoplasm. 2.4. Statistical analysis Qualitative variables are provided with their distribution frequencies. Quantitative variables were summarized as their mean, standard deviation, and range. Associations among qualitative variables were evaluated using the Pearson 2 test or Fisher exact test when 25% of expected associations were less than 5. For overall survival, the event was defined as deaths produced as a consequence of the tumor, with live patients or those dying of another cause excluded. Overall survival was calculated as the time elapsed from the date of surgery and the date of death or last follow-up. For disease-free survival, the event was defined as a diagnosis of a locoregional or distant recurrence in patients previously free of disease (i.e., in all those undergoing curative surgery). Disease-free survival was calculated as the time elapsed from the date of surgery to diagnosis of the first recurrence. The functions overall survival and disease-free survival were estimated by the KaplanMeier method and compared between groups using Breslow’s exact test. Data were fitted to a proportional risk regression model, and the assumption of proportionality of risks was confirmed. Adjusted relative risks and their significance were estimated. In each hypothesis contrast, the null hypothesis was rejected when the type I error was less
241
Table 2 Description of the population of 57 patients undergoing surgery for renal carcinoma Variable Sex Male Female Age (yrs) ⬎60 ⱕ60 Tumor size T1 and T2 T3 Lymph nodes N0 N1 Distance metastasis M0 M1 Stage I II III IV Grouped stage I and II III and IV Histologic type Clear cell Papillary cell Chromophobe cell Unclassified Bellini duct Grouped histology Papillary Nonpapillary Fuhrman grade I II III IV
No. (%) 39 (68.4) 18 (31.6) 31 (54.4) 26 (45.6) 35 (61.4) 22 (38.6) 54 (94.7) 3 (5.3) 56 (98.1) 1 (1.8) 32 (56.1) 2 (3.5) 22 (38.6) 1 (1.8) 34 (59.6) 23 (40.4) 38 (66.7) 5 (8.8) 8 (14.0) 2 (3.5) 4 (7.0) 5 (8.8) 52 (91.2) 13 (22.8) 31 (54.4) 12 (21.1) 1 (1.8)
than 0.05. All statistical analyses were performed using SSPS version 11.5 software (SPSS, Inc., Chicago, IL).
3. Results The mean age of our series of 57 patients diagnosed with renal carcinoma was 61.3 ⫾ 12.4 years, with a range of 36 – 83 years. A total of 39 patients (68.4%) were men, and 18 (31.6%) were women. Table 2 shows detailed information on the tumor characteristics. p16 methylation results were obtained for 48 patients because of a lack of specimens from the remaining 9 patients. The proportion of patients with p16 methylation was 22.9% (11 patients), and unmethylated p16 was 78.1% (37). The expression of p16 protein was determined in 34 of the 48 patients, and was positive in 16 (47.1%) and negative in 18 (52.9%). A lack of p16 expression was observed in 100%
242
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
Table 3 p16 methylation according to tumor characteristics in 48 patients undergoing surgery for renal carcinoma Variable
Sex Male Female Age (yrs) ⬎60 ⱕ60 Tumor size T1 and T2 T3 Lymph nodes N0 N1 Distance metastasis M0 M1 Stage I II III IV Grouped stage I and II III and IV Histologic type Clear cell Papillary cell Chromophobe cell Unclassified Bellini duct Grouped histology Papillary Nonpapillary Fuhrman grade I II III IV
No. p16 methylated (%)
No. p16 unmethylated (%)
3.1. Postoperative disease progression: Analysis of overall survival P value
1.00 8 (24.2) 3 (20.0)
25 (75.8) 12 (80.0)
2 (8.0) 9 (39.1)
23 (92.0) 14 (60.9)
7 (25.0) 4 (20.0)
21 (75.0) 16 (80.0)
11 (24.4) 0 (0)
34 (75.6) 3 (100)
11 (23.4) 0 (0)
36 (76.6) 1 (100)
5 (20.0) 2 (100) 4 (20.0) 0 (0)
20 (80.0) 0 (0) 16 (80.0) 1 (100)
7 (25.9) 4 (19.0)
20 (74.1) 17 (81.0)
6 (19.4) 0 (0) 4 (57.1) 0 (0) 1 (25)
25 (80.6) 4 (100) 3 (42.9) 2 (100) 3 (75)
0.01
0.70
1.00
1.00
0.08
0.10
0.50 4 (100) 33 (75) 0.60 4 (36.4) 5 (18.5) 2 (22.2) 0 (0)
Table 4 p16 protein expression according to tumor characteristics in 34 patients undergoing surgery for renal carcinoma Variable
0.70
0 (0) 11 (25)
The median follow-up time of our study was 76 months (6 years), with an interquartile range of 2–117 months. In our patient population, 6-year overall survival was 85%. All our survival analyses are referred to our median length of follow-up. Of the 48 patients, 7 died during follow-up, all as a consequence of the neoplasia. The overall survival of the patients with p16 gene methylation was 100% and of those without this gene alteration was 81% (P ⫽ 0.14). The relative risk of dying for those patients without methylation was undefined (P ⫽ 0.02) with respect to those with the alteration, because no deaths were recorded in the subset of patients with a methylated p16
7 (63.6) 22 (81.5) 7 (77.8) 1 (100)
Statistically significant: P ⬍ 0.05.
of the tumors showing methylation and in 36% of those without this genetic alteration. We also examined the relationship between the gene methylation status and p16 protein and the clinicopathologic variables described in Tables 3 and 4. A significant relationship was observed between patient age and the genetic aberration: 8% of patients older than 60 years had methylation, and 39.1% of those aged ⱕ60 years (P ⫽ 0.01) did not. Although the association with histologic type was not significant, it is noteworthy that all the papillary tumors expressed the p16 protein yet lacked promoter methylation. In 80% of the chromophobe carcinomas, no p16 protein expression was detected. Interestingly, 38.9% and 64.3% of the tumors of Fuhrman stages I and III, respectively, did not express the protein.
Sex Male Female Age (yrs) ⬎60 ⱕ60 Tumor size T1 and T2 T3 Ganglia affected N0 N1 Distance metastasis M0 M1 Stage I II III IV Grouped stage I and II III and IV Histologic type Clear cell Papillary cell Chromophobe cell Unclassified Bellini duct Grouped histology Papillary Nonpapillary Fuhrman grade I II III IV p16 methylation Methylation No methylation
No. p16 IHC (⫹) (%)
No. p16 IHC (⫺) (%)
11 (47.8) 5 (45.5)
12 (52.2) 6 (54.5)
9 (50.0) 7 (43.8)
9 (50.0) 9 (56.3)
11 (55.0) 5 (35.7)
9 (45.0) 9 (64.3)
1 (100) 15 (45.5)
0 (0) 18 (54.5)
16 (47.1) 0 (0)
18 (52.9) 0 (0)
11 (61.1) 0 (0) 5 (35.7)
7 (38.9) 2 (100) 9 (64.3)
11 (55.0) 5 (35.7)
9 (45.0) 9 (64.3)
12 (50.0) 2 (100) 1 (20.0) 1 (33.3)
12 (50.0) 0 (0) 4 (80.0) 2 (66.7)
2 (100) 14 (43.8)
0 (0) 18 (56.3)
4 (44.4) 9 (50.0) 3 (42.9)
57 (55.6) 9 (50.0) 4 (57.1)
0 (0) 16 (64.0)
9 (100) 9 (36.0)
P value 1.00
0.70
0.30
0.40
0.09
0.30
0.10
0.07
0.10
0.001
Statistically significant: P ⬍ 0.05.
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
gene promoter. When a tumor expressed p16, overall survival was 78%, and in the absence of expression, the rate was 93% (P ⫽ 0.17). The relative risk of dying was 2.30 times higher (95% confidence interval [CI] 0.50 –10.53) in the patients with the protein (P ⫽ 0.22), although the relationship was not statistically significant. We also examined overall survival in the patients stratified according to the classic factors used to predict the risk of death in patients with renal carcinoma (histologic type, tumor stage, and differentiation). Neither the methylation state nor protein expression served to establish groups with a different prognosis. In the multivariate analysis, we included clinicopathologic variables (age, sex, tumor stage, histologic type, and cell differentiation) and the molecular alterations examined (gene methylation and protein expression). Only methylation was found to have independent prognostic value: the absence of alteration confers an undefined risk of death (P ⫽ 0.03), independently of the other variables. With regard to expression of the p16 protein, the relative risk of dying was 8.47 times higher in patients who had the protein (95% CI 0.77–92.46) (P ⫽ 0.06). 3.2. Postoperative disease progression: Analysis of disease-free survival In the disease-free survival analysis, 1 patient with a distant metastasis preoperatively was excluded. Diseasefree survival for our study population was 79%. During follow-up, 10 patients had recurrence of their tumor. In patients with p16 gene methylation, disease-free survival was 90%, and in those without the genetic alteration disease-free survival, it was 75% (P ⫽ 0.30). The relative risk of recurrence in those without methylation was 2.78fold higher (95% CI 0.34 –22.22) than in patients with the aberration (P ⫽ 0.27). When the tumors expressed p16, disease-free survival was 73%, and if they lacked the expression of the protein, it was 80% (P ⫽ 0.40). The relative risk of tumor recurrence was 1.56 times greater (95% CI 0.35–7.02) in patients with the protein (P ⫽ 0.55). In the stratified disease-free survival analysis, neither gene methylation nor gene product expression served to establish groups of patients with a different prognosis. Neither did gene methylation nor protein expression had any prognostic value independent of the clinicopathologic variables.
4. Discussion Many genetic aberrations have been observed in renal carcinoma. Analysis of the loss of heterozygosity in several chromosomes of these tumors has indicated their polygenic origin (i.e., multiple genetic alterations are needed for the renal tumor to develop) [15]. The methylation of some genes plays an important role
243
in tumorigenesis as a gene inactivation mechanism. It has been shown that transcriptional gene silencing through methylation of the CpG islands is an alternative inactivation mechanism, which has been described in genes such as the Von Hippel-Lindau (VHL), retinoblastoma (Rb), p15, and p16 [16]. Some renal cell carcinomas show methylation of 5=CpG islands of the VHL gene (19%), and some retinoblastomas present de novo methylation in the Rb gene [17,18]. The incidence of gene methylation in different tumors ranges from 20% to 67%, and can be even higher in cell cultures of the same tumor type. For example, in 90% of colon cancer cells, the p16 gene promoter is methylated. Hypermethylation of the promoter of certain genes differs in prevalence, depending on the type of tumor analyzed, showing a tumor-specific and gene-specific pattern [19]. In our study, the incidence of methylation of the p16 promoter was 22.9% in renal tumors, and a lack of p16 protein expression was observed in 52.9% of these cases. Esteller et al. [19] observed p16 gene hypermethylation in a similar proportion (23%) of renal tumors to us. However, Kawada et al. [20] detected p16 hypermethylation in only 3.3% of the renal cancers they examined, and 5.5% lacked the p16 protein detected immunohistochemically. These investigators concluded that p16 gene inactivation is an infrequent event in primary renal carcinoma in the Japanese population [20]. We consider this to indicate that p16 methylation in renal tumors is also population specific, although further studies are needed to establish prevalences in the different populations. Here, we noted that hypermethylation of the p16 promoter led to transcriptional silencing, which is a common observation in renal cell carcinoma. Kawada et al. [20] also noted a close link between the immunohistochemical expression of p16 and hypermethylation of the gene promoter in renal tumors. This correlation has also been described in other tumors, such as gastrointestinal and hepatic [21–24]. In the present study, in no tumor with p16 methylation was the nuclear expression of the protein detected. Of the 18 tumors immunohistochemically negative for the gene product, gene methylation was noted in 9 (50%). In the remaining 50% of the patients in whom p16 methylation was absent, there must be some other genetic defect responsible for the protein not being encoded. The mechanism of inactivation of the p16 gene in this group of tumors remains unclear because homozygous deletions and point mutations are rare in renal carcinoma [25] and other tumor types [26]. Some investigators [21] suggest that this genetic silencing could be the result of a posttranscriptional modification in tumors lacking gene methylation. We detected no significant relationship between p16 gene methylation status and the expression of the protein and the stage or degree of cell differentiation of the tumors. Nevertheless, it is of interest that none of the papillary tumors showed p16 methylation, yet they all expressed the p16 protein. In this histologic type, we have not detected aberration of the p16 gene. Several investigators [27] argue
244
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245
that genetic analysis is an important tool for the diagnosis and prognosis of renal carcinoma, and have suggested different genetic patterns depending on the histologic tumor type. In our patients, p16 methylation was significantly correlated with age and was found to be most frequent in patients younger than 60 years. A link between gene methylation status and age has also been reported for other tumors [28]. With regard to the prognostic value of aberrant methylation of the p16 gene, in our study, patients with p16 hypermethylation had a 6-year overall survival of 100%, and those without hypermethylation had a rate of 81% (P ⫽ 0.14). Although not significant, this difference indicates a clinical tendency in that there were no deaths in the group of patients with the gene aberration. Moreover, patients with p16 promoter hypermethylation had a lower risk of dying than those without this aberration, regardless of the classic prognostic factors described for renal cancer included in the multivariate analysis. To our knowledge, this protective effect has not been previously described in renal carcinoma or any other type of tumor. We have sought an explanation for this in the interpretation of colorectal tumorigenesis, in which the successive genetic aberrations that accumulate at each stage have been widely explored. Results have led to the hypothesis that the colorectal tumor can occur via 2 different mechanisms: 1 of chromosome instability, whereby oncogenes and gene suppressors are consecutively altered; or via the instability of microsatellites (MSI) or mutant phenotype that occurs from the aberration of repair genes, especially hMLH1 and hMSH2 [29,30]. The MSI mechanism has been associated with a less aggressive component. In colorectal cancer, methylation of the p16 gene has been considered as part of the “methylator phenotype,” which includes the methylation of CpG islands described by Toyota et al. [30]. This phenotype defines a group of tumors characterized by the methylation of multiple loci, including the hMLH1 promoter. Several investigators [30,31] have reported that this phenotype is associated with MSI. This association can be explained by the theory that epigenetic changes, which include methylation of the promoters of certain genes such as p16 and hMLH1 blocking their transcription, contribute to a modification of the actions of the repair genes that condition the MSI phenotype [32,33]. It is generally accepted that patients who have colorectal tumors develop via the MSI route have a better prognosis [34,35]. Norrie et al. [21] observed that a lack of p16 expression was related to MSI in colorectal tumors and that this subset of tumors had different clinicopathologic features. Regardless of the mechanism whereby p16 gene aberrations contribute to renal tumorigenesis, the findings of our study indicate that methylation of the gene confers a better prognosis to patients with cancer of the kidney. An interesting next step in this investigation will be to ascertain the mutant phenotype in this patient population, to
check its association with the p16 aberration and its possible prognostic value.
5. Conclusions Our findings indicate that p16 gene aberrations are frequent events among the molecular modifications already described for renal carcinoma and that methylation of the gene promoter is 1 of the most common mechanisms of silencing the p16 gene. Our findings also point to a protective role of p16 gene hypermethylation against the risk of dying that these patients carry.
References [1] Liggett WH, Sidransky D. Role of the p16 tumor suppressor gene in cancer. J Clin Oncol 1998;16:1197–206. [2] Serrano M, Hannon GJ, Beach D. A new regulatory motif in cellcycle control causing specific inhibition of cyclin D/CDK4. Nature 1993;366:704 –7. [3] Nobori P, Miura K, Wu DJ, et al. Deletions of the cyclin-dependent kinase-4 inhibitor gene in multiple human cancers. Nature 1994;368: 753– 6. [4] Herman JG, Merlo A, Mao L, et al. Inactivation of the CDKN2/p16/ MTS1 gene is frequently associated with aberrant DNA methylation in all common human cancers. Cancer Res 1995;55:4525–30. [5] Bird A. The essentials of DNA methylation. Cell 1992;70:5– 8. [6] Merlo A, Herman JG, Mao L, et al. 5= CpG island methylation is associated with transcriptional silencing of the tumour suppressor p16/CDKN2/MTS-1 in human cancers. Nature Med 1995;1:686 –92. [7] Moch H, Presti JC, Sauter G, et al. Genetic aberrations detected by comparative genomic hybridization are associated with clinical outcome in renal cell cancer. Cancer Res 1996;56:27–30. [8] Schullerus D, Herbers J, Chudek J, et al. Loss of heterozygosity at chromosomes 8p, 9p, and 14p is associated with stage and grade of non-papillary renal cell carcinomas. J Pathol 1997;183:151–5. [9] Cairns P, Tokino K, Eby Y, et al. Localization of tumor suppressor loci on chromosome 9 in primary human renal cell carcinomas. Cancer Res 1995;55:224 –7. [10] Kinoshita H, Yamada H, Ogawa O, et al. Contribution of chromosome 9p21–22 deletion to the progression of human renal cell carcinoma. Jpn J Cancer Res 1995;86:795–9. [11] Störkel S, Eble JN, Adlakha K, et al. Classification of renal cell carcinoma: Workgroup No. 1. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer 1997;80:987–9. [12] Fuhrman SA, Lasky LC, Limas C. Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 1982; 6:655– 63. [13] Guinan P, Sobin LH, Algaba F, et al. TNM staging of renal cell carcinoma: Workgroup No. 3. Union International Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer 1997;80:992–3. [14] Gerardts J, Hruban RH, Schutte M, et al. Immunohistochemical p16INK4a analysis of archival tumors with deletion, hypermethylation, or mutation of CDKN2/MTS1 gene. A comparison of four commercial antibodies. Appl Immunohistochem Mol Morphol 2000; 8:71–9. [15] Thrash-Bingham CA, Salazar H, Freed J, et al. Genomic alterations and instabilities in renal cell carcinomas and their relationship to tumor pathology. Cancer Res 1995;55:6189 –95.
M. Vidaurreta et al. / Urologic Oncology: Seminars and Original Investigations 26 (2008) 239 –245 [16] Kashiwabara K, Oyama T, Sano T, et al. Correlation between methylation status of the p16/CDKN2 gene and the expression of p16 and Rb proteins in primary non-small cell lung cancers. Int J Cancer 1998;79:215–20. [17] Herman JG, Latif F, Weng Y, et al. Silencing of the VHL tumorsuppressor gene by DNA methylation in renal carcinoma. Proc Natl Acad Sci USA 1994;91:9700 – 4. [18] Sakai T, Toguchida J, Ohtani N, et al. Allele-specific hypermethylation of the retinoblastoma tumor-suppressor gene. Am J Hum Genet 1991;48:880 – 8. [19] Esteller M, Corn PG, Baylin SB, et al. A gene hypermethylation profile of human cancer. Cancer Res 2001;61:3225–9. [20] Kawada Y, Nakamura M, Ishida E, et al. Aberrations of the p14ARF and p16INK4␣ genes in renal cell carcinomas. Jpn J Cancer Res 2001;92:1293–9. [21] Norrie MWA, Hawkins NJ, Todd AV, et al. Inactivation of p16INK4a by CpG hypermethylation is not a frequent event in colorectal cancer. J Surg Oncol 2003;84:143–50. [22] Schneider-Stock R, Boltze C, Lasota J, et al. High prognostic value of p16INK4 alterations in gastrointestinal stromal tumors. J Clin Oncol 2003;21:1688 –97. [23] Qin Y, Liu J, Li B, et al. Association of low p16INK4a and p15INK4b mRNAs expression with their CpG islands methylation with human hepatocellular carcinogenesis. World J Gastroenterol 2004;10:1276 – 80. [24] Kaneto H, Sasaki S, Yamamoto H, et al. Detection of hypermethylation of the p16INK4A gene promoter in chronic hepatitis and cirrhosis associated with hepatitis B or C. Gut 2001;48:372–7. [25] Kinoshita H, Yamada H, Ogawa O, et al. Contribution of chromosome 9p21–22 deletion to the progression of human renal cell carcinoma. Jpn J Cancer Res 1995;86:795–9.
245
[26] Okamoto A, Demetrick DJ, Spillare EA, et al. Mutations and altered expression of p16INK4 in human cancer. Proc Natl Acad Sci USA 1994;91:11045–9. [27] Junker K, Weirich G, Amin MB, et al. Genetic subtyping of renal cell carcinoma by comparative genomic hybridization. Recent Results Cancer Res 2003;162:169 –75. [28] Wiencke JK, Zheng S, Lafuente A, et al. Aberrant methylation of p16INK4a in anatomic and gender-specific subtypes of sporadic colorectal cancer. Cancer Epidemiol Biomarkers Prev 1999;8: 501– 6. [29] Lengauer C, Kinzler K, Vogelstein B. Genetic instability in colorectal cancers. Nature 1997;386:623–7. [30] Toyota M, Ohe-Toyota M, Ahuja N, et al. Distinct genetic profiles in colorectal tumors with or without the CpG island methylator phenotype. Proc Natl Acad Sci USA 2000;97:710 –5. [31] Hawkins N, Norrie M, Cheong K, et al. CpG island methylation in sporadic colorectal cancers and its relationship to microsatellite instability. Gastroenterology 2002;122:1376 – 87. [32] Kane MF, Loda M, Gaida GM, et al. Methylation of hMLH1 promoter correlates with lack of expression of hMLH1 in sporadic colon tumors and mismatch repair-defective human tumor cell lines. Cancer Res 1997;57:808 –11. [33] Ward RL, Cheong K, Ku SL, et al. Adverse prognostic effect of methylation in colorectal cancer is reversed by microsatellite instability. J Clin Oncol 2003;21:3729 –36. [34] Gryfe R, Kim H, Hsieh ET, et al. Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer. N Engl J Med 2000;342:69 –77. [35] Wright CM, Dent OF, Barker M, et al. Prognostic significance of extensive microsatellite instability in sporadic clinicopathological stage C colorectal cancer. Br J Surg 2000;87:1197–202.