Deregulation of the TP53/P14arf Tumor Suppressor Pathway in Low-Grade Diffuse Astrocytomas and Its Influence on Clinical Course

4884 Vol. 9, 4884–4890, October 15, 2003 Clinical Cancer Research

Deregulation of the TP53/p14ARF Tumor Suppressor Pathway in Low-Grade Diffuse Astrocytomas and Its Influence on Clinical Course

Takao Watanabe,1 Yoichi Katayama, Conclusions: Alternative disruption of the TP53/ ARF Atsuo Yoshino, Chiaki Komine, and p14 pathway represents a frequent event in low-grade diffuse astrocytomas and correlates with an unfavorable Takakazu Yokoyama clinical course. However, its value is unlikely to include Department of Neurological Surgery, Nihon University School of prognostic utility that is independent of other conventional Medicine, Tokyo 173-8610, Japan prognostic factors.

ABSTRACT INTRODUCTION Purpose: The chromosome 9p21 region harbors three Low-grade diffuse astrocytomas (WHO grade II) are well- tumor suppressor genes, p14ARF, p15INK4b, and p16INK4a, all differentiated tumors that grow slowly but display an intrinsic of which can be targets for hypermethylation-associated tendency for diffuse infiltration of the neighboring brain struc- inactivation in low-grade gliomas. p16INK4a and p15INK4b tures. They almost invariably recur and often progress to higher are critically involved in the RB1 pathway, whereas p14ARF malignancy, i.e., anaplastic astrocytoma (WHO grade III) or acts as an upstream regulator of the TP53 pathway. The role glioblastoma [WHO grade IV (1, 2)]. Such malignant transfor- of each tumor suppressor pathway in low-grade diffuse mation is responsible for the majority of the patient mortality (3, astrocytomas and their relationships with clinical behavior 4). During the past decade, much effort has been directed toward remain to be elucidated. identifying genetic alterations associated with the pathogenesis Experimental Design: We assessed the alterations of and progression of diffuse astrocytomas. Mutation of the TP53 the RB1/CDK4/p16INK4a/p15INK4b and the TP53/MDM2/ tumor suppressor gene has been demonstrated as one of the most p14ARF pathways in 46 WHO grade II astrocytomas and frequent genetic alterations in low-grade diffuse astrocytomas, analyzed their impact on prognosis. although its precise prognostic value is still debated (5–7). Results: The TP53/MDM2/p14ARF pathway was altered The CDKN2A and CDKN2B genes, which cluster together on chromosome 9p21, code for the structurally highly homolo- in 32 of 46 cases (70%) by either TP53 mutation (25 cases) or INK4a INK4b p14ARF methylation (9 cases). The RB1/CDK4/p16INK4a/ gous tumor suppressor proteins p16 and p15 , respec- INK4b tively. Alternative splicing of the CDKN2A gene results in the p15 pathway was disrupted in 6 of 46 cases (13%) by ARF INK4a expression of p14 , which is encoded by the unique exon 1␤ either RB1 methylation (1 case), p16 methylation (3 INK4a INK4a INK4b INK4b and exons 2 and 3 for p16 (8–11). p16 and p15 cases), or p15 methylation or homozygous deletion (3 2 cases). Generally speaking, individual tumors thus tended to inhibit CDK4 - and CDK6-mediated phosphorylation of retino- blastoma protein (RB1) and thereby negatively regulate cell display alteration of only one component from both path- ARF ways. Any independently analyzed genetic alteration failed cycle progression at the G1 checkpoint (12, 13), whereas p14 to provide statistically prognostic information. The alternate increases TP53 stability by abrogating MDM2 inhibition and or simultaneous presence of TP53 mutation and p14ARF thus facilitates TP53-mediated cell cycle arrest and apoptosis methylation emerged as a univariate predictor of a shorter (8–11). It appears therefore that the 9p21 gene cluster is linked -to both the RB1 and TP53 tumor suppressor pathways. Inacti ؍ progression-free survival (P 0.0456) but was not statisti- ARF INK4b INK4a cally significant when age and extent of surgery were in- vation of the p14 , p15 , and p16 genes has fre- cluded in the analysis. quently been observed in a variety of human tumors, including malignant gliomas, and is largely due to homozygous deletion or to hypermethylation of CpG islands in the promoter region (14–18). In low-grade gliomas, these genes are occasionally subject to aberrant promoter methylation (16–20). No study, Received 12/30/02; revised 6/29/03; accepted 7/2/03. however, has yet systematically analyzed such hypermethyla- The costs of publication of this article were defrayed in part by the tion in a large series of low-grade diffuse astrocytomas and payment of page charges. This article must therefore be hereby marked assessed the contributions of the RB1 and TP53 pathways to advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. their clinical behavior. This work was supported by Grants-in-Aid from the Ministry of Edu- The aim of the present study was to investigate the role of cation, Science, Sports and Culture of Japan for the High-Tech Research Center (Nihon University) and for Science Research. 1 To whom requests for reprints should be addressed, at Department of Neurological Surgery, Nihon University School of Medicine, 30-1 Oyaguchi-kamimachi, Itabashi-ku, Tokyo 173-8610, Japan. Phone: 2 The abbreviations used are: CDK, cyclin-dependent kinase; PFS, 81-3-3972-8111, ext. 2481; Fax: 81-3-3554-0425; E-mail: takao@med. progression-free survival; OS, overall survival; HR, hazard ratio; CI, nihon-u.ac.jp. confidence interval.

the RB1 and TP53 tumor suppressor pathways in low-grade conformational polymorphism followed by direct sequencing as diffuse astrocytomas and to clarify whether or not genetic al- described previously (7, 18). terations in each pathway affect their clinical outcome. We Differential PCR for Homozygous Deletions of the examined the alterations of the RB1/CDK4/p16INK4a/p15INK4b p14ARF, p15INK4b, and p16INK4a Genes and for Amplification and the TP53/MDM2/p14ARF pathways in WHO grade II astro- of the CDK4 and MDM2 Genes. To assess homozygous cytomas and attempted to correlate these abnormalities with deletions of the p14ARF (exon 1␤), p15INK4b, and p16INK4a (exon prognosis. 1␣) genes, differential PCR was carried out as described previously (16–18). The ␤-actin sequence was used for p15INK4b and p16INK4a deletions, and the glyceraldehyde-3-phosphate dehy- MATERIALS AND METHODS drogenase sequence was used for p14ARF deletion, as a refer- Patients and Tumor Samples. Tumor samples were ob- ence. Samples presenting Ͻ20% of the control signal were tained from patients operated on at Nihon University School of considered as homozygous deletions (16–18). Medicine between 1981 and 1998. The sections from all patients CDK4 and MDM2 amplifications were detected by differ- were reviewed and classified according to the WHO criteria ential PCR as described previously, using the IFN-␥ (IFNG) and (21). Optic tract, basal ganglia, thalamus, brain stem, and cer- dopamine receptor (DR) sequences as a reference, respectively ebellar tumors; pilocytic astrocytomas; gemistocytic astrocyto- (18, 25, 26). Values of greater than 2.7 and 3.02 for the CDK4/ mas [Ͼ20% of all component tumor cells (21)]; pleomorphic IFNG ratio and MDM2/DR ratio were regarded as amplification xanthoastrocytomas; and mixed oligoastrocytomas were ex- of the CDK4 and MDM2 genes, respectively (18, 25, 26). cluded from the study. Patients who were less than 18 years of Statistical Analysis. The Kaplan-Meier method was age at the time of operation were also excluded. After the above used to calculate the PFS and OS. We designated the PFS as the exclusions, 64 patients with newly diagnosed fibrillary astrocy- time period from the first operation to the point when tumor tomas (WHO grade II) were retained. Among these cases, 46 regrowth or recurrence was confirmed by an imaging study. OS tumor samples were available for methylation analysis, and this was defined as the interval between the first operation and the constituted the study population. In 18 cases (39%), differential date of death or the most recent evaluation. The log-rank test PCR was not possible because of limitations in the amount of was used to assess the degree of significance of the differences DNA available for analysis. At the initial treatment, 9 of the in different subgroups. The continuous variables, i.e., age and patients were given surgery alone, 8 patients were given imme- Karnofsky performance score, were categorized on the basis of diate radiation therapy, and 23 patients were treated with human the median value. The Cox proportional hazards model was used fibroblast IFN without radiation therapy as reported previously to identify the multivariate predictors of survival. When a po- (4). Twenty-two patients underwent radical resection (defined tential prognostic factor was judged to be independent and by the absence of radiographic remnant disease). Alkylating appropriate for the model, the HR and 95% CI were calculated. agent-based chemotherapy was given to 20 patients at tumor The relationships between various parameters were analyzed ␹2 recurrence. Twelve patients who had tumors recurring after the statistically by the test, Fisher’s exact test, or Student’s t test Ϯ initial surgery only or after surgery plus human fibroblast IFN as appropriate. All values are expressed as the means SD. The Ͻ received radiation therapy in combination with chemotherapy. significance level chosen was P 0.05, and all tests were At the last follow-up (January 2002), 22 patients were alive two-sided. The statistics were analyzed with a personal com- puter running Stat View J-5.0 software (Abacus Concepts, without disease, 3 were alive with disease, and 21 were dead of Berkeley, CA). the disease. Among the 25 survivors, the median follow-up period was 7.6 years (range, 3.1–18.8 years). Genomic DNA was extracted from paraffin sections as described previously RESULTS (22). DNA was also extracted from a total of seven samples of Alterations of the TP53/MDM2/p14ARF Pathway. normal brain tissue adjacent to tumors. DNA obtained from the 46 newly diagnosed low-grade diffuse ARF INK4b Methylation-Specific PCR for the p14 , p15 , astrocytomas was subjected to a p14ARF promoter methylation INK4a p16 , and RB1 Genes. Promoter hypermethylation of the study using the methylation-specific PCR. Among all of the ARF INK4b INK4a p14 , p15 , p16 , and RB1 genes was determined by tumors studied, p14ARF promoter methylation was present in 9 methylation-specific PCR (23). Sodium bisulfite modification of the 46 samples (20%; Fig. 1). We also evaluated seven was performed with a CpGenome DNA Modification Kit (In- samples of normal brain tissue, but none showed p14ARF meth- tergen, Oxford, United Kingdom) as described previously ylation. Mutation of the TP53 gene was assessed in the same (16, 17). samples analyzed for p14ARF promoter methylation and demon- The methods used for methylation-specific PCR of the strated in 25 of the 46 tumors (54%) studied (Table 1). One p14ARF, p15INK4b, p16INK4a, and RB1 genes were as described deletion with frameshift and 26 missense mutations with amino previously (16–18, 24). The amplified products were electro- acid exchange were identified. Two tumors contained double phoresed on 3% agarose gels and visualized with ethidium missense mutations. The hot spots for mutations included bromide. CpGenome Universal Methylated DNA (Intergen) and codons 175 (n ϭ 5), 248 (n ϭ 3), and 273 (n ϭ 8). p14ARF normal blood DNA were included in each set of the PCR as homozygous deletion and MDM2 amplification were not de- methylated and unmethylated controls, respectively. tected in any of the 28 tumors analyzed. Screening for TP53 Mutations. Exons 5–8oftheTP53 The patterns of the TP53/p14ARF pathway gene abnormal- gene were screened for mutations based on PCR single-strand ities in individual tumors were then compared. p14ARF was

Table 1 TP53 mutations in low-grade diffuse astrocytomas Nucleotide Amino acid Case Exon Codon substitution substitution 2 6 211 ACT3GCT Thr3Ala 3 5 175 CGC3CAC Arg3His 4 8 273 CGT3TGT Arg3Cys 6 6 220 TAT3TGT Tyr3Cys 8 7 246 ATG3ATA Met3Ile 10 8 273 CGT3TGT Arg3Cys 12 5 135 TGC3TAC Cys3Tyr 14 5 175 CGC3CAC Arg3His 16 6 205 TAT3TGT Tyr3Cys 17 8 262 3-bp deletion Frameshift 19 7 248 CGG3CAG Arg3Gln 21 8 273 CGT3TGT Arg3Cys 22 5 162 ATC3AGC Ile3Ser Fig. 1 Methylation-specific PCR of CpG islands in low-grade diffuse 25 8 273 CGT3TGT Arg3Cys astrocytomas. The p14ARF gene is methylated in cases 12 and 20 and 8 282 CGG3TGG Arg3Trp unmethylated in cases 1 and 28. The p15INK4b gene is methylated in case 26 7 248 CGG3CAG Arg3Gln 28 and unmethylated in cases 1, 12, and 20. The p16INK4b gene is 27 5 175 CGC3CAC Arg3His methylated in case 28 and unmethylated in cases 1, 12, and 20. The RB1 30 8 273 CGT3TGT Arg3Cys gene is methylated in case 1 and unmethylated in cases 12, 20, and 28. 34 8 273 CGT3TGT Arg3Cys S, 25-bp molecular marker; U, PCR product amplified by unmethylated- 37 7 245 GGC3AGC Gly3Ser specific primers; M, PCR product amplified by methylated-specific 8 265 CTG3CCG Leu3Pro primers; NC, normal control; PC, positive control. 38 6 205 TAT3TGT Tyr3Cys 40 8 273 CGT3TGT Arg3Cys 41 5 175 CGC3CAC Arg3His 42 7 248 CGG3CAG Arg3Gln 43 8 273 CGT3TGT Arg3Cys 3 3 hypermethylated in 7 of the 21 tumors (33%) with wild-type 46 5 175 CGC CAC Arg His TP53, as opposed to 2 of the 25 tumors (8%) with a mutant TP53. These alterations thus tended to occur independently, although this trend did not reach statistical significance (P ϭ 0.0590, Fisher’s exact test). When all abnormalities of the two Survival Analysis. For the entire study population, the genes were combined, 32 of the 46 tumors (70%) displayed median PFS was 7.2 years with a 5-year PFS rate of 57%, and alterations of the TP53/p14ARF pathway. the median OS was 9.3 years with a 5-year OS rate of 84%. Alterations of the RB1/CDK4/p16INK4a/p15INK4b Path- The results for the univariate statistical analyses of cate- way. Promoter hypermethylation of RB1, p16INK4a, and gorical variables affecting the PFS and OS are summarized in ARF p15INK4b was assessed in all 46 tumors and detected in 1 (2%), Table 2. TP53 mutation and p14 methylation, when analyzed 3 (6%), and 2 samples (4%), respectively (Fig. 1). In the seven independently, tended to be negative predictors of the PFS, although they failed to achieve statistical significance. With samples of normal brain tissue, there was no evidence of meth- respect to hot spot codon TP53 mutations, the PFS of patients ylation of these genes. Analyses of p16INK4a and p15INK4b with tumors carrying codon 175 mutations was shorter than that homozygous deletions and CDK4 amplification were possible in of those with other mutations, but the difference did not reach 28 cases. One tumor (4%) showed homozygous deletion of the INK4a INK4b INK4b INK4a statistical significance. RB1, p16 ,orp15 methylation p15 gene. p16 homozygous deletions and CDK4 am- status, alone or in combination, was not a statistically significant plification were not detected in any of the 28 tumors. predictor of PFS. Alternate or simultaneous presence of TP53 These alterations were mutually exclusive, except for one mutation and p14ARF methylation emerged as a significant pre- INK4a INK4b case that demonstrated both p16 and p15 methylation; dictor of a shorter PFS (Fig. 2). Patient age and extent of surgery the numbers were too small for statistical analysis. When all of also influenced the PFS significantly. Patients with tumors lack- the abnormalities were combined, 6 of the 46 tumors (13%) ing both TP53 and p14ARF alterations had a mean age of 33.6 Ϯ INK4a INK4b displayed alterations of the RB1/p16 /p15 pathway. 11.3 years, as opposed to an age of 35.9 Ϯ 12.3 years in the Correlation of TP53 and RB1 Pathway Gene Abnor- patients with all other tumors (P ϭ 0.5431, Student’s t test). malities. When all of the genetic abnormalities described TP53/p14ARF pathway gene abnormalities were present in 18 of above were combined and compared, simultaneous disruption of the 24 patients (75%) undergoing palliative surgery, as com- the TP53 and RB1 pathways was found in only one case, in pared with 14 of the 22 patients (64%) undergoing radical which mutation of the TP53 gene and hypermethylation of the surgery (P ϭ 0.4028, ␹2 test). To determine whether or not p16INK4b gene were detected. This case relapsed at 13 months alternative alteration of the TP53 and p14ARF genes remained after surgery, and the patient died at 14 months after recurrence. significantly associated with the PFS after adjustment for the Thirty-seven of the 46 tumors (80%) had genetic alterations in effects of age and extent of surgery, we applied a Cox propor- either the TP53 or RB1 pathway genes. An inverse relationship tional hazards model. Such multivariate analysis demonstrated between abnormalities of the RB1 pathway genes and the TP53 that the alternate or simultaneous presence of TP53 mutation pathway genes was evident (P ϭ 0.0072, Fisher’s exact test). and p14ARF methylation had no statistically significant effect on

Table 2 Univariate analysis of the association of factors with the PFS and OS in a total of 46 patients with low-grade diffuse astrocytomas No. of Median PFS Log-rank Median OS Log-rank Factor patients (mo) P (mo) P Age (yrs) Ͻ35 21 126 135 Ն35 25 54 0.0130 85 0.0323 Sex Male 22 56 127 Female 24 126 0.9743 98 0.5770 KPSa Ͻ80 16 56 87 Ն80 30 126 0.2099 127 0.2356 Radical resection Yes 22 126 135 No 24 47 0.0020 85 0.0106 TP53 mutation Negative 21 126 135 Positive 25 54 0.1197 91 0.2644 Hot spot codon TP53 mutation Codon 175 mutation 5 27 56 Other mutations 20 56 0.1031 98 0.0633 p14ARF methylation Negative 37 126 127 Positive 9 58 0.0895 78 0.0504 TP53/p14ARF aberration Negative 14 — 135 Any aberration 32 56 0.0456 87 0.1178 RB1/p16INK4a/p15INK4b aberration Negative 40 88 118 Any aberration 6 39 0.1514 47 0.0773 a KPS, Karnofsky performance score.

ARF Fig. 2 Kaplan-Meier PFS curves for TP53 and p14ARF alterations in Fig. 3 Kaplan-Meier OS curves for TP53 and p14 alterations in low-grade diffuse astrocytomas. low-grade diffuse astrocytomas. the PFS (P ϭ 0.0856). Older age (HR, 2.360; 95% CI, 1.014– 5.491; P ϭ 0.0463) and palliative surgery (HR, 3.219; 95% CI, DISCUSSION 1.304–7.944; P ϭ 0.0112) each qualified as an independent The RB1/CDK4/p16INK4a/p15INK4b pathway plays a cru- predictor of a shorter PFS in this model. In terms of the OS, age cial regulatory role in cell cycle progression (13). Another and extent of surgery displayed a prognostic utility on univariate important tumor suppressor pathway, TP53/p14ARF/MDM2, is analysis. Patients with alternative alteration of the TP53 and involved in the regulation of cell proliferation, apoptosis, and p14ARF genes also exhibited a tendency toward a shorter OS DNA repair (27, 28). It has become apparent that many of the (Fig. 3), although this trend lacked significance. Applying a Cox genetic alterations detected in diffuse gliomas affect genes that proportional hazards model using age and extent of surgery, encode members of these two critical regulatory pathways (1, older age (HR, 2.496; 95% CI, 1.012–6.025; P ϭ 0.0470) and 2). The present study demonstrated that the TP53 pathway was palliative surgery (HR, 3.142; 95% CI, 1.212–8.141; P ϭ predominantly deregulated (32 of 46; 70%) over the RB1 path- 0.0184) were found to be significantly unfavorable prognostic way (6 of 46; 13%). We further provided evidence that alterna- factors for the OS. tive alteration of the TP53 and p14ARF genes was a univariate

predictor of a shortened PFS. These findings suggest that the may be the crucial target in a considerable fraction of diffuse TP53 pathway may play an important tumor suppressor role in astrocytomas with intact TP53. low-grade diffuse astrocytomas. An important function for p14ARF as well as TP53 in A mutation of the TP53 suppressor gene has been observed carcinogenesis is evident in animal models. Mice lacking ex- in approximately one-half of diffusely infiltrating astrocytomas. pression of p19ARF (the mouse homologue to human p14ARF), It is well known that the TP53 mutation occurs at an early stage through selective disruption of exon 1␤, frequently develop of malignant transformation (1, 2, 7). Some clinical studies have spontaneous tumors including sarcomas, lymphomas, and glio- ARF indicated that TP53 mutation represented a significant adverse mas (37). Inactivation of p14 has been demonstrated as a indicator of PFS in low-grade astrocytomas and oligoastrocyto- frequent event in a variety of human neoplasms, although its mas (5, 6), whereas others have failed to demonstrate such a prognostic usefulness was not consistently identified in colorec- correlation (7). In the study of Peraud et al. (6), the prognostic tal cancer, hepatocellular carcinoma, and soft tissue sarcoma impact of TP53 status was closely related to the influence of the (34, 38, 39). In a subset of glioblastomas, homozygous deletion ARF gemistocytic subtype. In the present study, which exclusively of p14 has been shown to be associated with a shorter ARF examined supratentorial fibrillary astrocytomas in adults, TP53 survival (40). In the present study, the presence of p14 mutation tended to be associated with a shorter PFS, but the hypermethylation was associated with a shorter PFS, although relationship failed to achieve statistical significance. In accord- the difference was not statistically significant. Also, TP53 mutation, analyzed independently, did not have any apparent effect ance with the studies on colorectal cancer, Peraud et al. (6) ARF observed an unfavorable prognostic influence of codon 175 on prognosis. When combining TP53 and p14 analysis, a significantly decreased PFS was identified. These observations mutations exclusively in nongemistocytic astrocytomas. We suggest that abnormalities of each gene may have an equivalent also found that mutations at this codon were associated with a impact on tumor aggressiveness. shorter PFS and OS, although the difference was not statistically Our current understanding of the molecular mechanisms significant, possibly due to the limited number of patients in- underlying growth control in cells is still far from complete. Cell vestigated. biological experiments have suggested that deregulation of the In the present study, homozygous deletions of the p14ARF, G -S transition control system may confer a growth advantage p15INK4b, and p16INK4a genes or amplifications of the CDK4 1 to the cells; however, in some types of cultured cells, loss of and MDM2 genes were either absent or very rare. Although we G1-S transition gene function induces TP53-dependent apopto- did not completely analyze these alterations in full samples, this sis (41–43). This indicates that the TP53 pathway may prevent observation is consistent with previous reports examining low- cells with abnormalities of G1-S transition-regulatory genes grade gliomas (16–18, 29). Such phenomena may thus contrib- from aberrant proliferation and further suggests that simultane- ute to a more malignant phenotype. In contrast, hypermethyla- ous disruption of these two pathways would be required to ARF INK4b INK4a tion of the p14 , p15 , and p16 genes was detected produce unregulated cell cycling and thereby predispose sub- in a considerable fraction of low-grade astrocytomas, although jects to a more aggressive biological behavior. This view is no methylation was detected in seven normal brain tissue sam- supported by the findings of several previous studies demon- ples. In colorectal carcinogenesis, CpG island methylation of strating that dual inactivation of the G1-S transition control several key genes is an early event, which is detectable in system and the TP53 pathway was absent in low-grade astro- nonneoplastic mucosa, and is associated with microsatellite cytomas and oligodendrogliomas but frequent in high-grade instability (30, 31). During astrocytoma progression in patients astrocytomas and anaplastic oligodendrogliomas (18, 29). Our ARF with multiple biopsies, hypermethylation of the p14 gene or finding that concurrent disruption of the RB1 and TP53 tumor the p16INK4a gene was already present in a subset of low-grade suppressor pathways was present in only one tumor, which diffuse astrocytomas and was followed by homozygous dele- exhibited a worse clinical behavior, also supports such an as- tions of p14ARF and p16INK4a in glioblastomas derived there- sumption. from (16). These data suggest that aberrant methylation of the Because the present study used a retrospective design with 9p21 genes represents an early genetic event in astrocytoma a relatively short follow-up and relatively small patient number, tumorigenesis and may enhance genetic instability, leading to it remains inherently difficult to assess the actual prognostic subsequent structural DNA changes. factors in these slow-growing tumors, eliminating the confound- p14ARF plays a major role in the TP53 tumor suppressor ing variable of treatment effect on prognosis. Nevertheless, age pathway by antagonizing MDM2-mediated degradation of TP53 and extent of resection stood out as powerful determinants of (8–11). Thus, theoretically, p14ARF loss of function would be both the PFS and OS, in agreement with the results of previous predicted to reduce the frequency of concomitant TP53 muta- studies (3, 44). Such correlations continued to be significant tions (32). In lymphoma (33), hepatocellular carcinoma (34), when these two variables were examined simultaneously in a and glioblastoma (35), an inverse correlation between TP53 and Cox proportional hazards model. In contrast, the value of alter- p14ARF alterations has been observed, whereas contradictory native alteration of the TP53 and p14ARF genes did not include results were obtained in lung cancer (36). Such conflicting a prognostic utility in multivariate analysis. This may be ac- findings might imply that different roles for p14ARF are opera- counted for on the basis that the TP53 pathway gene abnormal- tive in the pathogenesis of each cancer type. In the present ities tended to interrelate with age and extent of surgery, which study, p14ARF hypermethylation was implicated in 7 of 21 surpassed their prognostic importance. A relatively higher fre- tumors (33%) with wild-type TP53 and appeared to be an quency of TP53 pathway gene abnormalities in patients with independent event to TP53 mutation, suggesting that p14ARF persistent disease after surgical resection may be of particular

importance, implying that the presence of these aberrations 13. Sherr, C. J. Cancer cell cycles. Science (Wash. DC), 274: 1672– could be associated with invasive tumors (thus being inaccessi- 1677, 1996. ble to extensive surgery). Although these findings need to be 14. Merlo, A., Herman, J. G., Mao, L., Lee, D. J., Gabrielson, E., Ј confirmed in a larger cohort of patients with longer follow-up, Burger, P. C., Baylin, S. B., and Sidransky, D. 5 CpG island methylation is associated with transcriptional silencing of the tumor suppressor our data suggest that combined molecular analysis of these two p16/CDK2/MTS1 in human cancers. Nat. Med., 1: 686–692, 1995. genes is itself unlikely to add significant prognostic information 15. Herman, J. G., Jen, J., Merlo, A., and Baylin, S. B. Hypermeth- to other conventional prognostic factors and could not therefore ylation-associated inactivation indicates a tumor suppressor role for a be used routinely in the clinical setting as a prognostic marker. p15INK4B. Cancer Res., 56: 722–777, 1996. In conclusion, the TP53 pathway was frequently deregu- 16. Nakamura, M., Watanabe, T., Klangby, U., Asker, C., Wiman, K., ARF lated in low-grade diffuse astrocytomas. Alternative alteration Yonekawa, Y., Kleihues, P., and Ohgaki, H. p14 deletion and of the TP53 and p14ARF genes was associated univariately with methylation in genetic pathways to glioblastomas. Brain Pathol., 11: 159–168, 2001. a decreased PFS, but this association was not significant when 17. Watanabe, T., Nakamura, M., Yonekawa, Y., Kleihues, P., and age and extent of surgery were included in the analysis. Thus, an Ohgaki, H. Promoter hypermethylation and homozygous deletion of the important tumor suppressor role is suggested for the TP53 p14ARF and p16INK4a genes in oligodendrogliomas. Acta Neuropathol., pathway in low-grade diffuse astrocytomas, although its value is 101: 185–189, 2001. unlikely to include a prognostic utility that is independent of 18. Watanabe, T., Yokoo, H., Yokoo, M., Yonekawa, Y., Kleihues, P., other conventional prognostic factors. and Ohgaki, H. Concurrent inactivation of RB1 and TP53 pathways in anaplastic oligodendrogliomas. J. Neuropathol. Exp. Neurol., 60: 1181– 1189, 2001. REFERENCES 19. Dong, S. M., Pang, C. S., Poon, W. A., Hu, J., To, K. F., Chang, A. R., and Ng, H. K. Concurrent hypermethylation of multiple genes is 1. Kleihues, P., and Ohgaki, H. Genetics of glioma progression and the associated with grade of oligodendroglial tumors. J. Neuropathol. Exp. definition of primary and secondary glioblastoma. Brain Pathol., 7: Neurol., 60: 808–816, 2001. 1131–1136, 1997. 20. Wolter, M., Reifenberger, J., Blaschke, B., Ichimura, K., Schmidt, 2. Kleihues, P., and Ohgaki, H. Primary and secondary glioblastomas: E. E., Collins, V. P., and Reifenberger, G. Oligodendroglial tumors from concept to clinical diagnosis. Neuro-Oncology, 1: 44–51, 1999. frequently demonstrate hypermethylation of the CDKN2A (MTS1, 3. Piepmeier, J., Christopher, S., Spencer, D., Byrne, T., Kim, J., p16INK4a), p14ARF, and CDKN2B (MTS2, p15INK4b) tumor sup- Knisel, J. P., Lacy, J., Tsukerman, L., and Makuch, R. Variations in the pressor genes. J. Neuropathol. Exp. Neurol., 60: 1170–1180, 2001. natural history and survival of patients with supratentorial low-grade 21. Kleihues, P., and Cavenee, W. K. (eds.). Pathology and Genetics of astrocytomas. Neurosurgery, 38: 872–879, 1996. Tumours of the Nervous System. Lyon, France: IARC, 2000. 4. Watanabe, T., Katayama, Y., Yoshino, A., Komine, C., Yokoyama, 22. Bru¨stle, O., Ohgaki, H., Schmitt, H. P., Walter, G. F., Ostertag, H., T., and Fukushima, T. Treatment of low-grade diffuse astrocytomas by and Kleihues, P. Primitive neuroectodermal tumors after prophylactic surgery and human fibroblast interferon without radiation therapy. central nervous system irradiation in children: association with an acti- J. Neuro-Oncol., 61: 171–176, 2002. vated K-ras gene. Cancer (Phila.), 69: 2385–2392, 1992. 5. Ishii, N., Tada, M., Hamou, M. F., Janzer, R. C., Meagher-Villemure, 23. Herman, J. G., Graff, J. R., Myo¨ha¨nen, S., Nelkin, B. D., and K., Wiestler, O. D., de Tribolet, N., and Van Meier, E. G. Cells with Baylin, S. B. Methylation-specific PCR: a novel PCR assay for meth- TP53 mutations in low grade astrocytic tumors evolve clonally to ylation status of CpG islands. Proc. Natl. Acad. Sci. USA, 93: 9821– malignancy and are an unfavorable prognostic factor. Oncogene, 18: 9826, 1996. 5870–5878, 1999. 24. Nakamura, M., Yonekawa, Y., Kleihues, P., and Ohgaki, H. Pro- 6. Peraud, A., Kreth, F. W., Wiestler, O. D., Kleihues, P., and Reulen, moter hypermethylation of the RB1 gene in glioblastomas. Lab. Inves- H. J. Prognostic impact of TP53 mutations and P53 protein overexpres- tig., 81: 77–82, 2001. sion in supratentorial WHO grade II astrocytomas and oligoastrocyto- 25. Biernat, W., Tohma, Y., Yonekawa, Y., Kleihues, P., and Ohgaki, mas. Clin. Cancer Res., 8: 1117–1124, 2002. H. Alteration of cell cycle regulatory gene in primary (de novo) and 7. Watanabe, K., Sato, K., Biernat, W., Tachibana, O., von Ammon, K., secondary glioblastomas. Acta Neuropathol., 94: 303–309, 1997. Ogata, N., Yonekawa, Y., Kleihues, P., and Ohgaki, H. Incidence and 26. Biernat, W., Kleihues, P., Yonekawa, Y., and Ohgaki, H. Amplifi- timing of p53 mutations during astrocytoma progression in patients with cation and overexpression of MDM2 in primary (de novo) glioblasto- multiple biopsies. Clin. Cancer Res., 3: 523–530, 1997. mas. J. Neuropathol. Exp. Neurol., 56: 180–185, 1997. 8. Quelle, D. E., Zindy, F., Ashmun, R. A., and Sherr, C. J. Alternative 27. Bartek, J., and Lukas, J. Pathways governing G1/S transition and reading frames of the INK4a tumor suppressor gene encode two unre- their response to DNA damage. FEBS Lett., 490: 117–122, 2001. lated proteins capable of inducing cell cycle arrest. Cell, 83: 993–1000, 28. Taylor, W. R., and Stark, G. R. Regulation of the G2/M transition by 1995. p53. Oncogene, 20: 1803–1815, 2001. 9. Kamijo, T., Weber, J. D., Zambetti, G., Zindy, F., Roussel, M. F., and 29. Ichimura, K., Bolin, M. B., Goike, H. M., Schmidt, E. E., Moshref, Sherr, C. J. Functional and physical interactions of the ARF tumor A., and Collins, V. P. Deregulation of the p14ARF/MDM2/p53 pathway suppressor with p53 and Mdm2. Proc. Natl. Acad. Sci. USA, 95: is a prerequisite for human astrocytic gliomas with G1-S transition 8292–8297, 1998. control gene abnormalities. Cancer Res., 60: 417–424, 2000. 10. Pomerantz, J., Schreiber-Agus, N., Liegeois, N. J., Silverman, A., 30. Issa, J. P., Ahuja, N., Toyota, M., Bronner, M. P., and Brentnall, Alland, L., Chin, L., Potes, J., Chen, K., Orlow, I., Lee, H. W., Cordon- T. A. Accelerated age-related CpG island methylation in ulcerative Cardo, C., and DePinho, R. A. The INK4a tumor suppressor gene colitis. Cancer Res., 61: 3573–3577, 2001. Arf product, p19 , interacts with MDM2 and neutralizes MDM2’s inhibi- 31. Shen, L., Kondo, Y., Hamilton, S. R., Rashid, A., and Issa, J. P. p14 tion of p53. Cell, 92: 713–723, 1998. methylation in human colon cancer is associated with microsatellite 11. Zhang, Y., Xiong, Y., and Yarbrough, W. G. ARF promotes MDM2 instability and wild-type p53. Gastroenterology, 124: 626–633, 2003. degradation and stabilizes p53: ARF-INK4a locus deletion impairs both 32. Esteller, M., Tortola, S., Toyota, M., Capella, G., Peinado, M. A., the Rb and p53 tumor suppression pathways. Cell, 92: 725–734, 1998. Baylin, S. B., and Herman, J. G. Hypermethylation-associated inactiva- 12. Hannon, G. J., and Beach, D. p15INK4b is a potential effecter of tion of p14ARF is independent of p16INK4a methylation and p53 muta- TGF-␤-induced cell cycle arrest. Nature (Lond.), 371: 257–261, 1994. tional status. Cancer Res., 60: 129–133, 2000.

33. Pinyol, M., Hernandez, L., Martinez, A., Cobo, F., Hernandez, S., and p16 aberrations confer poor prognosis in human colorectal cancer. Bea, S., Lopez-Guillermo, A., Nayach, I., Palacin, A., Nadal, A., Fer- J. Clin. Oncol., 19: 299–304, 2001. nandez, P. L., Montserrat, E., Cardesa, A., and Campo, E. INK4a/ARF 39. Orlow, I., Drobnjak, M., Zhang, Z. F., Lewis, J., Woodruff, J. M., alterations in human non-Hodgkin’s lymphoma mainly occur in tumors Bernnan, M. F., and Cordon-Cardo, C. Alterations of INK4A and with wild-type p53 gene. Am. J. Pathol., 156: 1987–1996, 2000. INK4B genes in adult soft tissue sarcomas: effect on survival. J. Natl. 34. Tannapfel, A., Busse, C., Weinans, L., Benicke, M., Katalinic, A., Cancer Inst. (Bethesda), 91: 73–79, 1999. Gei␤ler, F., Hauss, J., and Wittekind, C. INK4a-ARF alterations and p53 mutations in hepatocellular carcinomas. Oncogene, 20: 7104–7109, 40. Labuhn, M., Jones, G., Speel, E. J., Maier, D., Zweifel, G., Gratzl, 2001. O., Van Meier, E. G., Hegi, M. E., and Merlo, A. Quantitative real-time PCR does not show selective targeting of p14ARF but concomitant 35. Fulci, G., Labuhn, M., Maier, D., Lachat, Y., Hausmann, O., Hegi, inactivation of both p16INK4a and p14ARF in 105 human primary glio- M. E., Janzer, R. C., Merlo, A., and Van Meir, E. G. p53 gene mutation mas. Oncogene, 20: 1103–1109, 2001. and ink4a-arf deletion appear to be two mutually exclusive events in human glioblastoma. Oncogene, 19: 3816–3822, 2000. 41. Debbas, M., and White, E. Wild-type p53 mediates apoptosis by E1A, which is inhibited by E1B. Genes & Dev., 7: 546–554, 1993. 36. Gazzeri, S., Della-Valle, V., Chaussade, L., Brambilla, C., Larsen, C. J., and Brambilla, E. The human p19ARF protein encoded by the ␤ 42. Morgenbesser, S. D., Williams, B. O., Jacks, T., and DePinho, R. A. transcript of the p16INK4a gene is frequently lost in small cell lung p53-dependent apoptosis produced by Rb-deficiency in the developing cancer. Cancer Res., 58: 3926–3931, 1998. mouse lens. Nature (Lond.), 371: 72–74, 1994. 37. Kamijo, T., Bodner, S., van de Kamp, E., Randle, D. H., and Sherr, 43. Wu, X., and Levine, A. J. p53 and E2F-1 cooperate to mediate C. J. Tumor spectrum in ARF-deficient mice. Cancer Res., 59: 2217– apoptosis. Proc. Natl. Acad. Sci. USA, 91: 3602–3606, 1994. 2222, 1999. 44. Keles, G. E., Lamborn, K. L., and Berger, M. S. Low-grade hem- 38. Esteller, M., Gonza´lez, S., Risques, R. A., Marcuello, E., Mangues, ispheric gliomas in adults: a critical review of extent of resection as a R., Germa, J. R., Herman, J. G., Capella, G., and Peinado, M. A. K-ras factor influencing outcome. J. Neurosurg., 95: 735–745, 2001.

Downloaded from clincancerres.aacrjournals.org on September 23, 2021. © 2003 American Association for Cancer Research. Deregulation of the TP53/p14ARF Tumor Suppressor Pathway in Low-Grade Diffuse Astrocytomas and Its Influence on Clinical Course

Takao Watanabe, Yoichi Katayama, Atsuo Yoshino, et al.

Clin Cancer Res 2003;9:4884-4890.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/9/13/4884

Cited articles This article cites 43 articles, 14 of which you can access for free at: http://clincancerres.aacrjournals.org/content/9/13/4884.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/9/13/4884.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/9/13/4884. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.