Clinical Cancer Research The Future of Cancer Research: Science and Patient Impact Infection and Cancer: Biology, Therapeutics, and Prevention
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Enokida, H.
Right arrow Articles by Dahiya, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Enokida, H.
Right arrow Articles by Dahiya, R.
Clinical Cancer Research Vol. 11, 6582-6588, September 15, 2005
© 2005 American Association for Cancer Research

Imaging, Diagnosis, Prognosis

Multigene Methylation Analysis for Detection and Staging of Prostate Cancer

Hideki Enokida1,3, Hiroaki Shiina2, Shinji Urakami1,2, Mikio Igawa2, Tatsuya Ogishima1, Long-Cheng Li1, Motoshi Kawahara3, Masayuki Nakagawa3, Christopher J. Kane1, Peter R. Carroll1 and Rajvir Dahiya1

Authors' Affiliations: 1 Department of Urology, Veterans Affairs Medical Center and University of California, San Francisco, California; 2 Department of Urology and Biochemistry, Shimane University, Izumo; and 3 Department of Urology, Kagoshima University, Kagoshima, Japan

Requests for reprints: Rajvir Dahiya, Urology Research Center, Department of Urology, Veterans Affairs Medical Center, University of California, San Francisco, CA 94121. Phone: 415-750-6964; Fax: 415-750-6639; E-mail: rdahiya{at}urol.ucsf.edu.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: Aberrant gene promoter methylation profiles have been well-studied in human prostate cancer. Therefore, we rationalize that multigene methylation analysis could be useful as a diagnostic biomarker. We hypothesize that a new method of multigene methylation analysis could be a good diagnostic and staging biomarker for prostate cancer.

Experimental Design: To test our hypothesis, prostate cancer samples (170) and benign prostatic hyperplasia samples (69) were examined by methylation-specific PCR for three genes: adenomatous polyposis coli (APC), glutathione S-transferase pi (GSTP1), and multidrug resistance 1 (MDR1). The methylation status of representative samples was confirmed by bisulfite DNA sequencing analysis. We further investigated whether methylation score (M score) can be used as a diagnostic and staging biomarker for prostate cancer. The M score of each sample was calculated as the sum of the corresponding log hazard ratio coefficients derived from multivariate logistic regression analysis of methylation status of various genes for benign prostatic hyperplasia and prostate cancer. The optimal sensitivity and specificity of the M score for diagnosis and for staging of prostate cancer was determined by receiver-operator characteristic (ROC) curve analysis. A pairwise comparison was employed to test for significance using the area under the ROC curve analysis. For each clinicopathologic finding, the association with prostate-specific antigen (PSA) failure-free probability was determined using Kaplan-Meier curves and a log-rank test was used to determine significance. The relationship between M score and clinicopathologic findings was analyzed by either the Mann-Whitney U test, Kruskal-Wallis test, or the Spearman rank correlation test.

Results: The frequency of positive methylation-specific PCR bands for APC, GSTP1, and MDR1 genes in prostate cancer samples was 64.1%, 54.0%, and 55.3%, respectively. In benign prostatic hyperplasia samples, it was 8.7%, 5.8%, and 11.6%, respectively. There was a significant correlation of M score with high pT category (P < 0.001), high Gleason sum (P < 0.001), high preoperative PSA (P = 0.027), and advanced pathologic features. For all patients, the M score had a sensitivity of 75.9% and a specificity of 84.1% as a diagnostic biomarker using a cutoff value of 1.0. In patients with low or borderline PSA levels (<10.0 ng/mL), the M score was significantly higher in prostate cancers than in benign prostatic hyperplasias (2.635 ± 0.200 and 0.357 ± 0.121, respectively). ROC curve analysis revealed that the M score had a sensitivity of 65.4% and a specificity of 94.2% when 1.0 was used as a cutoff value. For all patients, M score can distinguish organ-confined (≤pT2) from locally advanced cancer (≥pT3) with a sensitivity of 72.1% and a specificity of 67.8%. Moreover, considering patients with PSA levels of <10 ng/mL, the M score has a sensitivity of 67.1% and a specificity of 85.7%. The ROC curve analysis showed a significant difference between M score and PSA (P = 0.010).

Conclusions: This is the first report demonstrating that M score is a new method for multigene methylation analysis that can serve as a good diagnostic and staging biomarker for prostate cancer.

The discovery and widespread use of serum prostate-specific antigen (PSA) has had a tremendous impact on all aspects of management of prostate carcinoma (1–3). PSA is the most sensitive diagnostic biomarker for prostate cancer detection to date. However, its low specificity has led to unnecessary biopsy for verifying the presence of prostate cancer (2, 3). Aberrant methylation profiles of gene promoters have been well-investigated in prostate cancer (4–6). Recent studies have shown that the methylation status of glutathione S-transferase pi (GSTP1) or adenomatous polyposis coli (APC) could individually distinguish prostate cancer from benign prostatic hyperplasia (4–6). Some investigators have shown that GSTP1 hypermethylation can be easily detected in body fluids such as blood plasma, urine, or ejaculate from patients with prostate cancer (7–10). Based on prior studies, it is possible that multigene methylation profiles can provide better diagnostic or prognostic values for prostate cancer (4–12). However, such studies are lacking in the literature. In the present study, we investigated whether multigene methylation analysis can be a good diagnostic and staging biomarker for prostate cancer.

The ability to predict the outcome or pathologic stage of prostate cancer is a worthwhile goal which will enable physicians to make treatment recommendations for patients with prostate cancer. Nomograms, using a combination of three variables (PSA, biopsy Gleason score, and clinical stage) are currently distributed and used clinically (13, 14). However, the utility of multigene methylation analysis as a pretreatment staging biomarker has not been reported.

We have previously shown that the methylation status of GSTP1 or MDR1 is a good biomarker for detecting prostate cancer and correlates with clinicopathologic features (15, 16). We hypothesize that multigene methylation analysis can be a good diagnostic and staging biomarker prior to treatment. To this end, we did methylation analysis of the APC gene and combined the results with the GSTP1 and MDR1 data from the same 170 prostate cancer and 69 benign prostatic hyperplasia patients. We used the data to calculate a methylation score (M score) that is the sum of the log of the hazard ratios (HR) for each gene, analyzed by multivariate logistic regression analysis for pathology (benign prostatic hyperplasia versus prostate cancer). We also related the M score to clinical and pathologic outcome. Using receiver operator characteristic (ROC) curve analysis to determine the optimal cutoff value for the M score, we evaluated the sensitivity and specificity of M score as a staging biomarker compared with PSA or Gleason sum. In addition, we assessed PSA failure-free probability against the clinicopathologic features of prostate cancer.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Tissue samples. A total of 170 newly diagnosed prostate cancer tissues from radical prostatectomies and 69 pathologically proven benign prostatic hyperplasia samples from transurethral resection were obtained from Shimane University Hospital (Izumo, Japan) from 1997 through 2003. Our routine diagnostic strategy for prostate cancer included serum PSA level, transrectal ultrasonography, color Doppler ultrasonography, and magnetic resonance imaging, which enabled us to detect the localization of prostate cancer before radical prostatectomy (17). The patients' background and clinicopathologic characteristics are summarized in Table 1. Each tumor was graded and staged according to the Gleason grading system (18) and the tumor-node-metastasis staging system (19). None of these patients had received androgen deprivation therapy. We used serum PSA levels after radical prostatectomy as a surrogate end-point, with a level ≥0.2 ng/mL designated as PSA failure. Forty-six patients with prostate cancer were excluded from the PSA failure-free probability study because of adjuvant hormonal therapy immediately after radical prostatectomy. PSA failure-free probability was determined as the percentage of patients without PSA failure. Follow-up ranged from 0.7 to 91.4 months, with a median of 33.9 months. Written informed consent was obtained from all patients.


View this table:
[in this window]
[in a new window]
  		Table 1. Clinical characteristics of prostate cancer and benign prostate hypertrophy patients

 
Tissue preparations. All of the benign prostatic hyperplasia and prostate cancer samples were fixed in 10% buffered formalin (pH 7.0) and embedded in paraffin wax. For histologic evaluation, 5-µm-thick sections were used for H&E staining. All of the samples were microscopically dissected and analyzed for methylation (20). In benign prostatic hyperplasia samples, the presence of high-grade prostate intraepithelial neoplasia and cancer were ruled out by microscopic analysis.

Nucleic acid extraction. Genomic DNA from all prostate samples was extracted using a commercial kit (Qiagen, Valencia, CA), and precipitated with ethanol. The concentration of DNA was determined with a spectrophotometer, and its integrity was checked by gel electrophoresis.

Methylation analysis. Genomic DNA from all prostate samples (100 ng) was subjected to sodium bisulfite modification using a CpGenome DNA Modification Kit (Intergen Co., Purchase, NY). Based on the functional promoter sequence of APC (21), methylation-specific PCR (MSP) and unmethylation-specific PCR (USP) primers were designed using MethPrimer software (http://www.urogene.org/methprimer) developed in our laboratory (22). Primers used for MSP and USP analysis are as follows: universal primers, 5'-TAATTTTTTTGTTTGTTGGGGATT-3' (sense), 5'-ACTACACCAATACAACCACATATC-3' (antisense); MSP primers, 5'-TATTGCGGAGTGCGGGTC-3' (sense), 5'-TCGACGAACTCCCGACGA-3' (antisense); USP primers, 5'-GTGTTTTATTGTGGAGTGTGGGTT-3' (sense), 5'-CCAATCAACAAACTCCCAACAA-3' (antisense). An initial PCR product was created with universal primers, which have no CpG sites in either forward or reverse primers, followed by a second nested PCR with primers specific for MSP or USP. For semiquantitative analysis, a preliminary suitable number of PCR cycles for each MSP and USP were carried out in order to determine the linear range of the reaction. To ensure this, at least one initial PCR was done using 32 cycles each for MSP and USP. Then, a suitable PCR cycle was chosen for each sample. The annealing temperature and PCR cycles used for MSP and USP primers were 64°C and 32 cycles, respectively. The sequences of primers for GSTP1 and MDR1, as well as their PCR conditions, were described previously (15, 16). In each assay, the absence of DNA template served as negative controls. The obtained MSP and USP products were analyzed by electrophoresis in 3% agarose gels and stained with ethidium bromide. With ImageJ software (http://rsb.info.nih.gov/ij), relative methylation levels (%) were calculated (15, 16, 23) by using the area under the curve corresponding to each band (MSP and USP). For methylation analysis of APC, we used 5.3% methylation as a cutoff value, which was the average percentage of methylation in 69 benign prostatic hyperplasia samples. Using these criteria, MSP positivity for APC was defined as those prostate cancers with a percentage of methylation of >5.3%, and negative methylation was <5.3%. The criteria for MSP positivity of GSTP1 and MDR1 were described previously (15, 16).

Bisulfite DNA sequencing analysis. Bisulfite-modified DNA (1 µL) was amplified using a pair of universal primers in a total volume of 20 µL. Direct bisulfite DNA sequencing of the PCR products using either forward universal primer or reverse universal primer was done according to the manufacturer's instructions (Applied BioSystems, Foster City, CA).

Statistical analysis. Using a previously reported analytic technique (24, 25), we calculated the M score for each sample, defined as the sum of the corresponding log HR coefficients, which were derived from multivariate logistic regression analysis of each methylated gene in the benign prostatic hyperplasia and prostate cancer samples (Table 2). The optimal sensitivity and specificity of the M score for diagnosis of prostate cancer and for staging was determined by ROC curve analysis using MedCalc Software (Mariakerke, Belgium). A pairwise comparison was employed to test for significance using the area under the ROC curve (AUC) analysis. For each clinicopathologic finding, the association with PSA failure-free probability was determined using Kaplan-Meier curves and a log-rank test was used to determine significance. The relationship between M score and clinicopathologic findings was blindly analyzed by either the Mann-Whitney U test, Kruskal-Wallis test, or the Spearman rank correlation test using StatView software (SAS Institute Inc., Cary, NC). A P value of <0.05 was regarded as statistically significant.


View this table:
[in this window]
[in a new window]
  		Table 2. Multivariate logistic regression analysis of gene methylation in a series of prostate cancer and benign prostatic hyperplasias

 

   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Methylation status of the APC promoter in prostate clinical samples. Representative results of MSP and USP assays for APC in prostate cancers and benign prostatic hyperplasias are shown in Fig. 1A. MSP-positive bands were present in the majority of prostate cancers, and less so in the benign prostatic hyperplasia samples. USP-positive bands were present in all of prostate cancers and benign prostatic hyperplasias. The result of the methylation study was also confirmed by bisulfite DNA sequencing. Figure 1B shows the results of a typical bisulfite DNA sequencing in a prostate cancer sample. In sample "P" (corresponding to Fig. 1A, lane "P") with both MSP and USP bands, there was a "T" peak along with a "C" peak at the CpG sites, indicating partial methylation (Fig. 1B, top). In sample "U" (corresponding to Fig. 1A, lane "U"), where no MSP band was observed, the CpG sites were completely unmethylated (Fig. 1B, bottom).



View larger version (55K):
[in this window]
[in a new window]
  		Fig. 1. Methylation status of the APC promoter in clinical samples. A, MSP and USP bands of 24 samples for benign prostatic hyperplasia and prostate cancer are shown. P and U, samples depicting partial methylation and unmethylation, respectively. B, bisulfite DNA sequencing of partially methylated (top), unmethylated (bottom) samples corresponding to lanes P and U, respectively. In partially methylated samples, there was a "T" peak along with a "C" peak at the CpG sites. In unmethylated samples, every CpG site was unmethylated. PM and UM, partial methylation and unmethylation, respectively.

 
Evaluation of M score: multigene methylation analysis with APC, GSTP1, and MDR1 for distinguishing prostate cancer from benign prostatic hyperplasia. APC methylation analysis showed positive MSP bands in 109 of 170 (64.1%) prostate cancers and 6 of 69 (8.7%) benign prostatic hyperplasias (Fig. 2A). As we have reported previously, positive MSP bands for GSTP1 methylation analysis were solely found in 92 of 170 (54.0%) prostate cancers, and in 4 of 69 (5.8%) benign prostatic hyperplasias, whereas that for MDR1 methylation analysis were found in 94 of 170 (55.3%) prostate cancers, and in 8 of 69 (11.6%) benign prostatic hyperplasias (15, 16). There was a significant difference in the methylation status of each gene between the series of prostate cancer and benign prostatic hyperplasias (Fig. 2A). As shown in Table 2, multivariate logistic regression analysis revealed that APC, GSTP1, and MDR1 methylation was a significant dependent predictor of pathology (benign prostatic hyperplasia versus prostate cancer; P < 0.001, P = 0.005, and P = 0.020, respectively). The individual gene HRs for pathogenesis (prostate cancer versus benign prostatic hyperplasia) were different from one another. For instance, cases with APC methylation are 7.597 times more likely to have prostate cancer than cases with negative methylation, whereas the HR for MDR1 is 3.247. For all patients, the M score determined by the sum of log HR was significantly higher in prostate cancers than in benign prostatic hyperplasias (2.913 ± 0.158 and 0.357 ± 0.121, respectively; Fig. 2B, left). The optimal cutoff value of the M score for distinguishing prostate cancer from benign prostatic hyperplasia was determined using the ROC curve. The M score had a sensitivity of 75.9% and a specificity of 84.1% when 1.0 was used as a cutoff value (Fig. 2B, right). In patients with low or borderline PSA levels (<10.0 ng/mL), the M score was still significantly higher in prostate cancers than in benign prostatic hyperplasias (2.635 ± 0.200 and 0.357 ± 0.121, respectively; Fig. 2C, left). ROC curve analysis revealed that the M score had a sensitivity of 65.4% and a specificity of 94.2% when 1.0 was used as a cutoff value (Fig. 2C, right). All statistical values were age-adjusted because the mean ages were statistically different between benign prostatic hyperplasias and prostate cancers (Table 1).



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 2. Correlation of gene methylation frequency with pathology (benign prostatic hyperplasia versus prostate cancer) and methylation score (M score) as a diagnostic biomarker. A, the methylation frequency in prostate cancer samples was significantly higher than that in benign prostatic hyperplasia samples for APC, GSTP1, and MDR1 genes (64.1% versus 8.7%, 54.0% versus 5.8%, and 55.3% versus 11.6%, respectively). B, M score, determined as the sum of the corresponding log HR for pathology (benign prostatic hyperplasia versus prostate cancer; Table 2), was also significantly higher in prostate cancer samples than in benign prostatic hyperplasia samples (left). B, ROC curve analysis determined that a cutoff value of 1.0 had optimal sensitivity (75.9%) and specificity (84.1%) for M score to distinguish prostate cancer from benign prostatic hyperplasia (right). C, M score was also significantly higher in prostate cancer samples with low or borderline PSA levels (<10.0 ng/mL) than in benign prostatic hyperplasia samples (left); in patients with low PSA levels the M score had high sensitivity (65.4%) and specificity (94.2%) for prostate cancer detection when 1.0 was used as a cutoff value (right).

 
Correlation of M score with clinicopathologic findings. Among prostate cancers, the M score showed a significant stepwise increase with advancing pathologic stage (1.34 ± 0.26 in pT2a, 2.92 ± 0.24 in pT2b, 3.84 ± 0.26 in pT3a, 4.21 ± 0.62 in pT3b, and 4.98 ± 0 in pT4; P < 0.001; Fig. 3A). Similarly, the M score increased as the Gleason sum increased (2.20 ± 0.23, Gleason sum <7; 3.58 ± 0.25, Gleason sum = 7; and 3.819 ± 0.29, Gleason sum >7; P < 0.001; Fig. 3B). With regard to preoperative PSA levels, the M score was higher in PSA >10 ng/mL (3.35 ± 0.25) than in PSA ≤10 ng/mL (2.64 ± 0.20; P = 0.027; Fig. 3C). Moreover, the M score was higher in advancing pathologic features as follows: in capsular invasion (Cap) [positive (3.80 ± 0.23) versus negative (2.47 ± 0.19; P < 0.001)], in seminal vesicle involvement [positive (4.88 ± 0.11) versus negative (2.80 ± 0.16; P = 0.002)], in pelvic lymph node metastasis (pN) [positive (4.60 ± 0.38) versus negative (2.57 ± 0.19; P = 0.001)], in venous involvement (v) [positive (3.92 ± 0.22) versus negative (2.44 ± 0.19; P < 0.001)], in lymphatic vessel involvement (Ly) [positive (3.77 ± 0.28) versus negative (2.61 ± 0.18; P = 0.001)], and in perineural invasion (PNI) [positive (3.41 ± 0.18) versus negative (1.86 ± 0.28; P < 0.001)] (Fig. 3C). We also observed that the M score was age-related in the total group of benign prostatic hyperplasias (P < 0.001) but not in the total group of prostate cancers (P = 0.108).



View larger version (37K):
[in this window]
[in a new window]
  		Fig. 3. Correlation of M score with clinicopathologic features. Among the overall prostate cancer patients: A, higher stage of disease (1.34 ± 0.26 in pT2a, 2.92 ± 0.24 in pT2b, 3.84 ± 0.26 in pT3a, 4.21 ± 0.62 in pT3b, and 4.98 ± 0 in pT4). B, higher Gleason sum (2.20 ± 0.23, Gleason sum < 7; 3.58 ± 0.25, Gleason sum = 7; and 3.819 ± 0.29, Gleason sum >7). C, higher preoperative PSA levels (3.35 ± 0.25 in PSA >10 ng/mL, 2.64 ± 0.20 in PSA ≤10 ng/mL) correlated with an increased methylation score (M score). There was significant correlation between M score and worse clinicopathologic findings. Cap, capsular invasion; SV, seminal vesicle involvement; pN, lymph node invasion; ly, lymphatic vessel invasion; and PNI, perineural invasion; N and P, negative and positive, respectively.

 
Prognostic features. We analyzed PSA failure-free probability as disease-free survival. Of the clinicopathologic features considered, only Gleason sum was significantly associated with poor outcome in univariable analyses (P = 0.022; Fig. 4.).



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4. Kaplan-Meier PSA failure-free survival curves of prostate cancer patients after radical prostatectomy, grouped according to the evaluated variables: Gleason sum, preoperative PSA, and pT category. Follow-up ranged from 0.7 to 91.4 months, with a median of 33.9 months. {bullet}, censored data points.

 
Evaluation of M score as a predictive biomarker for preoperative staging. To test the ability of the M score as a staging biomarker to distinguish between organ-confined (≤pT2) and locally advanced cancer (≥pT3), the optimal cutoff value of M score, PSA, and Gleason sum was determined using ROC curve. As shown in Fig. 5A, for all patients with prostate cancer, the M score had a sensitivity of 72.1% and a specificity of 67.8% when 4.0 is employed as a cutoff value. PSA had 69.4% and 69.5%, respectively, with a cutoff value of 9.0 ng/mL; Gleason score had 60.0% and 65.5%, respectively, with a cutoff value of 7. The corresponding AUC for each was 0.721, 0.724, and 0.687, respectively. Pairwise analysis for AUC showed no statistical difference among these three markers. Looking at the patients with PSA levels of 10 ng/mL or less, the M score had a sensitivity of 67.1% and a specificity of 85.7% with 3.3 as a cutoff value. PSA had 75.6% and 38.1%, respectively, with a cutoff value of 7.7 ng/mL; Gleason score had 66.7% and 57.1%, respectively, with a cutoff value of 7. The corresponding AUC for each was 0.780, 0.550, and 0.663, respectively. There was a significant difference between M score and PSA (P = 0.010; Fig. 5B).



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 5. Optimum sensitivity and specificity of M score, PSA, and Gleason sum for predicting pathologic stage. A, for all patients, ROC curve analysis was used to determine a cutoff value that had the optimum sensitivity and specificity for each variable to distinguish organ-confined (≤pT2) from locally advanced cancer (≥pT3). Pairwise comparison test showed no significant difference among M score, PSA, and Gleason sum. B, in patients with PSA levels of 10 ng/mL or less, there was a significant difference between M score and PSA. AUC, area under ROC curve; CI, confidence interval.

 

   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
We have reported previously that the methylation status of the GSTP1 or MDR1 gene promoter correlate with clinicopathologic features (15, 16). In this study, we found that the M score for APC, GSTP1, and MDR-1 genes can be used as a diagnostic biomarker for prostate cancer. This is the first study to integrate the methylation status of multigenes using the M score, which is the sum of the log HR analyzed by multivariate logistic regression analysis for pathology (benign prostatic hyperplasia versus prostate cancer). This analysis provides automatically adjusted statistical data (24), with each HR directly related to gene methylation in prostate cancer samples as compared with benign prostatic hyperplasia (methylation-negative) samples (Table 2). By adding the log HR of each gene in a multigene analysis, it is therefore possible to predict the risk of prostate cancer in individual patients. Similarly, Ray et al. employed multivariate Cox proportional hazards models for their multigene methylation analysis in medulloblastoma, and used the sum of the log HR as a risk score for each patient (25).

PSA is the most sensitive diagnostic biomarker for prostate cancer detection thus far. However, its low specificity has forced unnecessary biopsy of patients in order to exclude prostate cancer. Using various kinds of PSA analysis such as free-PSA, complexed-PSA, or total-PSA, and its combinations—% free-PSA or % complexed-PSA, many investigators have struggled to find better methods for prostate cancer detection (2, 3). However, the specificity of these tests is ~60% at best with 80% sensitivity (2, 3). To make matters worse, in patients with low or borderline PSA levels, any PSA analysis is a poor diagnostic tool for prostate cancer detection because of much lower specificity (2, 3). In the current study, there was significant difference in M score between benign prostatic hyperplasia and prostate cancer (Fig. 2B, left). For prostate cancer detection, using a cutoff value of 1.0, the M score had a 75.9% sensitivity and 84.1% specificity, which is much higher compared with that reported with PSA (Fig. 2B, right). Moreover, in patients with low or borderline PSA levels (<10.0 ng/mL), the M score had high sensitivity of 65.4% and specificity of 94.2% for prostate cancer detection when 1.0 was used as a cutoff value (Fig. 2C, right). Thus, the M score can be a very useful and improved diagnostic biomarker for prostate cancer detection, even in patients with low or borderline PSA levels. Several investigators have already shown that GSTP1 hypermethylation can be readily detected in bodily fluids such as blood plasma, urine, or ejaculate from patients with prostate cancer (7–10). Taken together, these findings indicate that the M score is applicable for use as a diagnostic biomarker. Therefore, by using both the M score and PSA for prostate cancer screening, it may be possible to reduce the number of unnecessary biopsies, however, additional study is warranted to verify this hypothesis.

Predicting the probability of the final pathologic stage is a worthwhile goal so that physicians can make appropriate treatment recommendations for patients with prostate cancer. Nomograms using a combination of three variables (serum PSA, biopsy Gleason score and clinical stage) are already distributed and used clinically (13, 14) and have been verified in two studies (26, 27). Using ROC curve analysis, they analyzed staging probability based on the nomograms; however, the AUC reported in the two studies differ (0.787 versus 0.684). One reason for the discrepancy may be related to the fact that the clinical (tumor-node-metastasis) stage was subjectively determined by different individuals, thereby introducing a bias in the results. It would be best if pathologic staging were predicted without subjective variables. Interestingly, in this study, the M score showed significant correlation with worse clinicopathologic features such as higher pT and pN categories, higher Gleason sum, capsular extension, involvement of seminal vesicles, veins and lymphatic vessels, and higher preoperative PSA values (Fig. 3). Maruyama et al. also showed that there was significant correlation between their method for multigene methylation analysis and clinicopathologic features (12). However, in their study, the correlation with pathologic stage was less significant (P = 0.04) compared with our results (P < 0.001; Fig. 3) and no cutoff value was used in their analysis. Using ROC curve analysis, we were able to determine a cutoff value for the M score as a staging biomarker, which enabled us to distinguish organ-confined prostate cancer (≤pT2) from locally advanced prostate cancer (≥pT3). As shown in Fig. 5A, the M score has the sensitivity and specificity to serve as a good predictive staging biomarker like PSA or Gleason sum. Moreover, in patients with PSA levels of 10 ng/mL or less (Fig. 5B), the M score showed a sensitivity of 67.1% and a high specificity of 85.7% compared with corresponding PSA values of 38.1%. ROC curve analysis showed a significant difference between M score and PSA in this category. Our data also indicates that the M score can be a useful biomarker not only in distinguishing prostate cancer from benign prostatic hyperplasia, but also for predicting the final pathologic stage before medical treatment. Among patients with PSA levels of 10 ng/mL or less, the M score could predict the final pathologic stage more precisely than other biomarkers (Fig. 5B). However, a prospective study will be necessary to confirm this idea. We also analyzed PSA failure-free probability as disease-free survival. Of the clinicopathologic features considered, only Gleason sum was significantly associated with poor outcome (Fig. 4). Currently, there are no other reports demonstrating a significant correlation between multigene methylation analysis and PSA failure-free probability.

In conclusion, this is the first study to integrate the methylation status of multigenes using the M score which reflects the comprehensive methylation status of prostate tissues and is useful as a biomarker for detection and staging of prostate cancer. To elucidate the practical effect of M score in predicting prostate cancer outcomes, it will be necessary to include more genes from prostate tissue biopsy and body fluid samples in the future.


   Footnotes
 
Grant support: NIH grants RO1CA101844, RO1AG21418, and T32DK07790, VA Merit Review, and REAP grants.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 3/23/05; revised 5/17/05; accepted 6/ 8/05.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Hernandez J, Thompson IM. Prostate-specific antigen: a review of the validation of the most commonly used cancer biomarker. Cancer 2004;101:894–904.[CrossRef][Medline]
  2. Partin AW, Brawer MK, Bartsch G, et al. Complexed prostate specific antigen improves specificity for prostate cancer detection: results of a prospective multicenter clinical trial. J Urol 2003;170:1787–91.[CrossRef][Medline]
  3. Okihara K, Cheli CD, Partin AW, et al. Comparative analysis of complexed prostate specific antigen, free prostate specific antigen and their ratio in detecting prostate cancer. J Urol 2002;167:2017–24.[CrossRef][Medline]
  4. Li LC, Carroll PR, Dahiya R. Epigenetic changes in prostate cancer: implication for diagnosis and treatment. J Natl Cancer Inst 2005;97:103–15.[Abstract/Free Full Text]
  5. Yegnasubramanian S, Kowalski J, Gonzalgo ML, et al. Hypermethylation of CpG islands in primary and metastatic human prostate cancer. Cancer Res 2004;64:1975–86.[Abstract/Free Full Text]
  6. Nakayama M, Gonzalgo ML, Yegnasubramanian S, Lin X, De Marzo AM, Nelson WG. GSTP1 CpG island hypermethylation as a molecular biomarker for prostate cancer. J Cell Biochem 2004;91:540–52.[CrossRef][Medline]
  7. Gonzalgo ML, Nakayama M, Lee SM, De Marzo AM, Nelson WG. Detection of GSTP1 methylation in prostatic secretions using combinatorial MSP analysis. Urology 2004;63:414–8.[CrossRef][Medline]
  8. Gonzalgo ML, Pavlovich CP, Lee SM, Nelson WG. Prostate cancer detection by GSTP1 methylation analysis of postbiopsy urine specimens. Clin Cancer Res 2003;9:2673–7.[Abstract/Free Full Text]
  9. Cairns P, Esteller M, Herman JG, et al. Molecular detection of prostate cancer in urine by GSTP1 hypermethylation. Clin Cancer Res 2001;7:2727–30.[Abstract/Free Full Text]
  10. Goessl C, Krause H, Muller M, et al. Fluorescent methylation-specific polymerase chain reaction for DNA-based detection of prostate cancer in bodily fluids. Cancer Res 2000;60:5941–5.[Abstract/Free Full Text]
  11. Tokumaru Y, Harden SV, Sun DI, Yamashita K, Epstein JI, Sidransky D. Optimal use of a panel of methylation markers with GSTP1 hypermethylation in the diagnosis of prostate adenocarcinoma. Clin Cancer Res 2004;10:5518–22.[Abstract/Free Full Text]
  12. Maruyama R, Toyooka S, Toyooka KO, et al. Aberrant promoter methylation profile of prostate cancers and its relationship to clinicopathological features. Clin Cancer Res 2002;8:514–9.[Abstract/Free Full Text]
  13. Partin AW, Mangold LA, Lamm DM, Walsh PC, Epstein JI, Pearson JD. Contemporary update of prostate cancer staging nomograms (Partin tables) for the new millennium. Urology 2001;58:843–8.[CrossRef][Medline]
  14. Khan MA, Partin AW. Partin tables: past and present. BJU Int 2003;92:7–11.
  15. Enokida H, Shiina H, Urakami S, et al. Ethnic group-related differences in CpG hypermethylation of the GSTP1 gene promoter among African-American, Caucasian and Asian patients with prostate cancer. Int J Cancer 2005;116:174–81.[CrossRef][Medline]
  16. Enokida H, Shiina H, Igawa M, et al. CpG hypermethylation of MDR1 gene contributes to the pathogenesis and progression of human prostate cancer. Cancer Res 2004;64:5956–62.[Abstract/Free Full Text]
  17. Shigeno K, Igawa M, Shiina H, Wada H, Yoneda T. The role of colour Doppler, ultrasonography in detecting prostate cancer. BJU Int 2000;86:229–33.[CrossRef][Medline]
  18. Gleason DF, Mellinger GT. Veterans Administration Cooperative Urological Research group. Prediction of prognosis for prostatic adenocarcinoma by combined histologic grading and clinical staging. J Urol 1974;111:58–64.[Medline]
  19. Hermanek P, Hutter RVP, Sobin LH, et al. Prostate. Illustrated guide to the TNM/pTNM classification of malignant tumors. 4th ed. Heidelberg: Springer-Verlag; 1997. p. 272–80.
  20. Dahiya R, Lee C, McCarville J, Hu W, Kaur G, Deng G. High frequency of genetic instability of microsatellites in human prostatic adenocarcinoma. Int J Cancer 1997;72:762–7.[CrossRef][Medline]
  21. Jaiswal AS, Narayan S. Upstream stimulating factor-1 (USF1) and USF2 bind to and activate the promoter of the adenomatous polyposis coli (APC) tumor suppressor gene. J Cell Biochem 2001;81:262–77.[CrossRef][Medline]
  22. Li LC, Dahiya R. MethPrimer: designing primers for methylation PCRs. Bioinformatics 2002;18:1427–31.[Abstract/Free Full Text]
  23. Umbricht CB, Evron E, Gabrielson E, Ferguson A, Marks J, Sukumar S. Hypermethylation of 14-3-3 sigma (Stratifin) is an early event in breast cancer. Oncogene 2001;20:3348–53.[CrossRef][Medline]
  24. Hata K, Akiba S, Hata T, Miyazaki K. A multivariate logistic regression analysis in predicting malignancy for patients with ovarian tumors. Gynecol Oncol 1998;68:256–62.[Medline]
  25. Ray A, Ho M, Ma J, et al. A clinicobiological model predicting survival in medulloblastoma. Clin Cancer Res 2004;10:7613–20.[Abstract/Free Full Text]
  26. Augustin H, Eggert T, Wenske S, et al. Comparison of accuracy between the Partin tables of 1997 and 2001 to predict final pathological stage in clinically localized prostate cancer. J Urol 2004;171:177–81.[CrossRef][Medline]
  27. Penson DF, Grossfeld GD, Li YP, Henning JM, Lubeck DP, Carroll PR. How well does the Partin nomogram predict pathological stage after radical prostatectomy in a community based population? Results of the cancer of the prostate strategic urological research endeavor. J Urol 2002;167:1653–8.[Medline]



This article has been cited by other articles:


Home page
 Clin. Cancer Res.Home page
M. Roupret, V. Hupertan, D. R. Yates, J. W.F. Catto, I. Rehman, M. Meuth, S. Ricci, R. Lacave, G. Cancel-Tassin, A. de la Taille, et al.
Molecular Detection of Localized Prostate Cancer Using Quantitative Methylation-Specific PCR on Urinary Cells Obtained Following Prostate Massage
Clin. Cancer Res., March 15, 2007; 13(6): 1720 - 1725.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
S. Urakami, H. Shiina, H. Enokida, H. Hirata, K. Kawamoto, T. Kawakami, N. Kikuno, Y. Tanaka, S. Majid, M. Nakagawa, et al.
Wnt Antagonist Family Genes as Biomarkers for Diagnosis, Staging, and Prognosis of Renal Cell Carcinoma Using Tumor and Serum DNA
Clin. Cancer Res., December 1, 2006; 12(23): 6989 - 6997.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
S. Urakami, H. Shiina, H. Enokida, T. Kawakami, K. Kawamoto, H. Hirata, Y. Tanaka, N. Kikuno, M. Nakagawa, M. Igawa, et al.
Combination analysis of hypermethylated Wnt-antagonist family genes as a novel epigenetic biomarker panel for bladder cancer detection.
Clin. Cancer Res., April 1, 2006; 12(7 Pt 1): 2109 - 2116.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Enokida, H.
Right arrow Articles by Dahiya, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Enokida, H.
Right arrow Articles by Dahiya, R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online