Friday, August 1, 2014

Prostate Cancer, Methylation, and Prognostics

There seems to be a continuous flow of genes, miRNAs, epigenetic factors including methylation, SNPs and the like all both diagnostic and prognostic for various cancers. A decade ago one looked for a gene, some gene that somehow got broken, changed, deleted, or the like. The paradigm was the Philadelphia chromosome of a cut and paste example. With the understanding we now have of methylation we see the same occur here, and methylation can be acquired and/or genetically inherited (see imprinting examples). However methylation is still somewhat poorly understood; what causes it, why does it work positively in some cases and negatively in others?

Methylation is but one of the many facets of what we now see as causes of Cancer. We depict a short summary below.


We examine the work of Wojno et al which has received recent interest. They examine the impacts of methylation upon 3 genes and see their presence as prognostic of potential aggressive prostate cancer. Specifically they conclude:

The diagnosis of prostate cancer is dependent on histologic confirmation in biopsy core tissues. The biopsy procedure is invasive, puts the patient at risk for complications, and is subject to significant sampling errors.

An epigenetic test that uses methylation-specific polymerase chain reaction to determine the epigenetic status of the prostate cancer–associated genes GSTP1, APC, and RASSF1 has been clinically validated and is used in clinical practice to increase the negative predictive value in men with no history of prostate cancer compared with standard histopathology. Such information can help to avoid unnecessary repeat biopsies.

The repeat biopsy rate may provide preliminary clinical utility evidence in relation to this assay’s potential impact on the number of unnecessary repeat prostate biopsies performed in US urology practices.

DNA methylation normally can result in the silencing of genes by interrupting the normal process of promoters. CpG islands are often hypermethylated and thus the gene which may regulate cell proliferation is silenced. This may result in uncontrolled cell growth. For example genes controlling MYC are not produced and MYC may then result in excess cell cycle proliferation. Methylation is hypermethylated in the regions of intergenic regions and in repetitive elements and this hypermethylation silences these regions and facilitates normal cell DNA transcription of the gene. Disruption of DNA, namely hypomethylation, in the intergenic and repetitive regions may result in possible loss of imprinting. This hypomethylation is also related to the production of lncRNAs which may in turn interfere with normal gene transcription.

Decitabine is a DNMT inhibitor. Namely, it inhibits the DNA methyltrasferases that facilitate methylation (such as DNMT3 which are de novo and DNMT1 which is maintenance). Decitabine thus has then tendency on the specific hematologic cell lines in MDS to remove methylations which have caused the aberrant cell line proliferations and allow for the return of homeostasis. MDS is a quasi-malignant condition originating in the bone marrow which may in many cases result in Acute Myelogenous Leukemia. With the use of decitabine or a similar DNMTI azacitidine, demethylation of these rapidly reproducing cells may be achieved and possible a normal state of homeostasis achieved.

The use of pharmaceuticals that alter the methylation patterns of DNA can have lasting effects because those patterns may last through subsequent mitotic changes. On the one hand that may be beneficial as is the case with MDS but such broad demethylation may also alter other segments of the DNA altering essential control elements and pathways. In cell development there are two sensitive periods; germ cell development and early embryonic development. It is during these periods that methylation is cleared and reset and that a drug-like a DNMTI would pose a serious risk to the proper resetting of the marks and could result in substantial DNA expression damage.

In summary we will examine the three gene methylation proposition with this test. We summarize this below:


Let us examine what the study presents in a bit more detail. Basically it does the following:

1. It examines three gene products; GSTP1, APC, and RASSF1

2. It determines if the genes are methylated so that the gene products are suppressed.

3. If that is the case after a first biopsy which is deemed normal, then a second biopsy is mandated due to the high incidence of a positive result on the second biopsy for PCa.

Specifically from the paper by Partin et al on the same topic the authors’ state:

The DOCUMENT multicenter trial in the United States validated the performance of an epigenetic test as an independent predictor of prostate cancer risk to guide decision making for repeat biopsy. Confirming an increased negative predictive value could help avoid unnecessary repeat biopsies. We evaluated the archived, cancer negative prostate biopsy core tissue samples of 350 subjects from a total of 5 urological centers in the United States. All subjects underwent repeat biopsy within 24 months with a negative (controls) or positive (cases) histopathological result. Centralized blinded pathology evaluation of the 2 biopsy series was performed in all available subjects from each site.

Biopsies were epigenetically profiled for GSTP1, APC and RASSF1 relative to the ACTB reference gene using quantitative methylation specific polymerase chain reaction. Predetermined analytical marker cutoffs were used to determine assay performance. Multivariate logistic regression was used to evaluate all risk factors.

The epigenetic assay resulted in a negative predictive value of 88% (95% CI 85-91). In multivariate models correcting for age, prostate specific antigen, digital rectal examination, first biopsy histopathological characteristics and race the test proved to be the most significant independent predictor of patient outcome.

The DOCUMENT study validated that the epigenetic assay was a significant, independent predictor of prostate cancer detection in a repeat biopsy collected an average of 13 months after an initial negative result. Due to its 88% negative predictive value adding this epigenetic assay to other known risk factors may help decrease unnecessary repeat prostate biopsies.

Recall that the negative predictive value or NPV is defined as:

 

Thus an NPV of 88% for the sample size used implies that it is fairly high in predicting True Negatives a priori but may still miss a percentage. There of course is the issue of the pathologist missing the PCa as well. This could be done by a sampling deficiency or confusion on a reading. It is not clear if for example a HGPIN is read.

Thus focusing on methylated genes, specifically just 3 of them, GSTP1, APC and RASSF1, they were able in a small sample to ascertain that there would be no need of a second biopsy if they were found to be unmethylated in the first.

Recall the effects of methylation as we show below:


From an article in Medical Express[1] as well as from an article in Eureka[2] as well as from an article in Science Codex[3] we have the following summary:

More than one million prostate biopsies are performed each year in the U.S. alone, including many repeat biopsies for fear of cancer missed. Therefore there is a need to develop diagnostic tests that will help avoid unnecessary repeat biopsies. Two independent trials have now validated the performance of an epigenetic test that could provide physicians with a better tool to help eliminate unnecessary repeat prostate biopsies, report investigators in The Journal of Urology.

In the previously reported independent MATLOC (Methylation Analysis To Locate Occult Cancer) trial, a multiplex epigenetic assay (ConfirmMDx for Prostate Cancer) profiling the APC, GSTP1 and RASSF1 genes demonstrated a negative predictive value of 90%. GSTP1 methylation is a specific biomarker for (prostate) cancer and this gene is methylated in up to 90% of prostate cancer cases. Additionally, APC and RASSF1 are important field effect markers and increase the diagnostic sensitivity of the assay.

A second multicenter study, DOCUMENT (Detection Of Cancer Using Methylated Events in Negative Tissue), has validated the performance of the epigenetic assay used in the MATLOC trial as an independent predictor of prostate cancer risk to guide decision making for repeat biopsy. In the DOCUMENT study patients with a negative biopsy were evaluated to identify those at low risk for harboring cancer missed, through biopsy sampling error, who could forego an unnecessary repeat biopsy. The validation study resulted in a negative predictive value of 88%.

"This epigenetic assay is a significant, independent predictor and has been shown to be the most valuable diagnostic aid of all evaluated risk factors in two independent trials," comments Alan W. Partin, MD, PhD, of the James Buchanan Brady Urological Institute, The Johns Hopkins University School of Medicine, Baltimore, Maryland. "Negative findings of this assay could be used to reduce concern over unsampled cancer and effectively avoid unnecessary repeat biopsies."

A total of 350 patients were enrolled in the DOCUMENT trial from five geographically dispersed medical centers: Cleveland Clinic, Eastern Virginia Medical School, Lahey Hospital & Medical Center, Johns Hopkins University, and University of California Los Angeles. Patients were grouped into those with two consecutive negative biopsies (controls) and those with a negative biopsy followed by a positive biopsy within 24 months. The initial archived, negative for cancer, prostate biopsy core tissue samples were evaluated. All of the men underwent a repeat biopsy on average one year after the initial biopsy.

Only biopsies with a minimum of eight cores per biopsy, collected no earlier than 2007, were included in the study, while initial biopsies with atypical cells suspicious for cancer, i.e. atypical small acinar proliferation by the sites' pathologists, were excluded, since this would have triggered a repeat biopsy based upon histopathology alone.

After correcting for age, prostate specific antigen (PSA), digital rectal exam, histopathological characteristics of the first biopsy, and race, this epigenetic test proved to be the most significant, independent, and strongest predictor of patient outcome with an odds ratio of 2.69 as well as the most valuable diagnostic aid of all evaluated risk factors. The slightly decreased sensitivity of the DOCUMENT trial compared to the MATLOC trial is most likely associated with a higher PSA screening prevalence in the DOCUMENT cohort.

It is important to note the following:

1. The genes selected have been studied for over two decades and especially as regards to their hypermethylation status.

2. The test is an early prognostic test which when combined with prostate biopsy data, especially a benign reading on the prostate biopsy.

3. The test has reasonable statistics given its small sample size but as we shall see there is substantial variability in these tests.

Methylation is but one of the very many changes we see in cancers. Whether they are cause or effect is yet to be determined. We know, for example, that methylation is causal in MDS, myelo-dysplasia syndrome, a precursor often of ALL. However, the cause of this methylation is still problematic. Drugs like DNMT inhibitors, azacitidine and decitabine are methylation inhibitors that seem to work for a while in this disorder. However what they do to other cells is uncertain.

In this study which we examine, there is a multiplicity of questions.

1. What causes the methylation?

There is still a lack of clear process as to how methylation occurs. As we will note shortly there are hypothesis that it is a result from inflammatory states and others that there may be secondary effects of autoimmune conditions. The elements of the process are understood to some degree but no true full causal chain is established.

2. Why are these gene sites methylated?

Why specific CpG islands at specific genes are methylated is still unknown. Are these initial locations or are the related to other as yet to be determined sites. Are certain CpG site more susceptible? Is there some histone issue wherein the histones are demethylated opening the DNA and thus allowing CpG methylation?

3. Why not try the DNMT inhibitors if these methylations are causal for PCa?

DNMT inhibitors are used for MDS with some short term success. Can they be used here as a step towards reversal? The problem is that we do not know what DNMTI sequellae are. The unintended consequences could be significant. Can we develop cell specific DNMTIs or are they even too potent?

4. Are many cancers caused by methylation or demethylation? If so which ones and why?

We have examined several different cancers and there is a large putative collection of genes, translocations, miRNAs, SNPs, methylations, and the like. Almost weekly and perhaps soon daily we will be seeing alleged markers for every element of a cancer development. Two decades ago we saw just gene issues. Today we have a problem of ascertaining the chicken or the egg. This is both a challenge and an opportunity. The problem, however, is that each time one of these has been developed, we see a Press release frenzy as it be a sine qua non.

5. Are the observations made by the researchers causal or coincidental?

This is always a critical question. Is this specific methylation of genes the cause of ultimate PCa or the coincidental effect of some other factor?

6. Are the methylation observations drivers or therapeutic approaches?

This question is a follow on to our question regarding DNMTIs.

7. Should we be examining more cancers from an epigenetic point of view rather than a genetic mutation perspective?

We have examined a few cancers for methylation as drivers for malignancy and metastatic behavior.

8. If as we have seen anecdotally, patients determined to have widespread HGPIN on a first 24 core biopsy and then none on a second 24 core biopsy, is this possibly a demethylation result, a stem cell result, or some other factor.

We have examined HGPIN in previous analyses. Although not considered a true PCa it has been considered as a natural pre-cursor. Namely many assume that HGPIN will always turn into PCa. However we have anecdotally seen patients where the HGPIN actually regresses, to the level of being unidentifiable at on a high density core biopsy. Thus we have asked if the first biopsy actually precipitated the remission, if so how and why, or was there some other factor. Unfortunately there is inadequate data for this study.

9. What of the stem cell factor in PCa?

PCa stem cells are always a concern. We have discussed them at length and basically research continues into these cells in PCa. Yet it does beg the question; what cells need be methylated and are certain cells more likely than others? Furthermore we can ask; how do we segregate prostate cells so as to ascertain the severity of methylation?

10. PTEN mutations have been proposed as a major causal effect of PCa. How does this relate to epigenetic factors?

We have shown that PTEN is controlled in the process of p53 and in turn the MDM2 and its RAS precursor control. Thus the focus on RAS derivatives is consistent with the PTEN argument.

11. Are epigenetic factors the results of inflammatory states? Namely in a patient with for example Type 2 Diabetes and elevated inflammatory status, does this become an initiating factor for methylation?

Hartnett and Egan have recently written:

Recently, epigenetic alterations, in particular alterations in DNA methylation, have been observed during inflammation and inflammation-associated carcinogenesis. The mediators of this, the significance of these changes in DNA methylation and the effect this has on gene expression and the malignant transformation of the epithelial cells during IBD and CAC are discussed in this review. The recent advances in technologies to study genomewide DNA methylation and the therapeutic potential of understanding these molecular mechanisms are also highlighted.

Dayeh et al focus on these types of methylation induced by Type 2 Diabetes in a recent paper as well.

Also it is worth examining the summary by Kundu and Surh on inflammation and cancer. They state:

DNA methylation, the covalent addition of a methyl group to the 5-position of cytosine base in the DNA, represents a critical epigenetic control of gene expression. The perturbation of methylation patterns as either aberrant loss of cytosine methylation in transforming genes or inappropriate gain of cytosine methylation in tumor suppressor genes has been involved in various human malignancies.

The most predominant aberrant DNA methylation is hypermethylation that typically occurs at the CpG islands located in the promoter regions of tumor suppressor genes. Promoter hypermethylation of tumor suppressor genes, such as p16, von-Hippel Lindau (VHL), adenomatous polyposis coli (APC), breast cancer susceptibility gene (BRCA1), retinoblastoma (Rb), E-cadherin (CDH1), p73, p53, and p57, results in transcriptional silencing . By analyzing the methylation status of 11 candidate cancer-related genes in cutaneous squamous cell carcinomas, Murao et al. have demonstrated that the promoter hypermethylation of CDH1, p16, Rb1 and p14 results in the loss of respective protein production.

Therefore, the epigenetic silencing of tumor suppressor genes by promoter CpG island hypermethylation perturbs cell cycle control, apoptosis, DNA repair and cell adhesion, and is recognized as an important mechanism in the tumorigenesis. However, global hypomethylation of certain genes, e.g., insulin-like growth factor-2 (IGF-2), can also result in genomic instability, accelerating malignant transformation.

As Donkena et al state:

“Oxidative stress” is the state of a cell, which is characterized by excess production of reactive oxygen species (ROS) and/or a reduction in antioxidant defenses responsible for metabolism. ROS are formed as a natural byproduct of the normal metabolism of oxygen. Under normal circumstances, the cell is able to maintain an adequate homeostasis between the formation of ROS and its removal through enzymatic pathways or via antioxidants. If, however, this balance is disturbed, then oxidative stress occurs.

This generates an imbalance of production/removal of ROS, which is either directly or indirectly involved in initiation, promotion, and progression phases of carcinogenesis. Oxygen radicals may cause damage to DNA and chromosomes, induce epigenetic alterations, interact with oncogenes or tumor suppressor genes, and impart changes in immunological mechanisms.

The extent of ROS induced oxidative damage can be exacerbated by a decreased efficiency of antioxidant defense mechanisms. …. Hypermethylation of a combination of genes including APC, RASSF1A, PTGS2, PDLIM4, and MDR1 could distinguish cancer from benign prostate tissues with sensitivities of 97.3%–100% and specificities of 92%– 100%. The increase in methylation of these genes with cancer progression indicates that they could be used for biomarkers for both diagnosis and risk assessment  

Furthermore, we showed significant differences in the frequency of methylation at individual CpG sites of PITX2, PDLIM4, KCNMA1, GSTP1, FLNC, EFS, and ECRG4 in recurrent and nonrecurrent subtypes of prostate tumors. Indeed, hypermethylation of a CpG island in PITX2 portended an increased risk of prostate cancer recurrence and was a predictor of distant disease recurrence in tamoxifen-treated, node-negative breast cancer patients.

12. HGPIN has always been an issue of concern. In some cases it appears to revert to fully benign states and in others it progresses to PCa. The question is: How useful would such a test be in HGPIN conditions?

The concern always is the HGPIN cells are confined to the glandular portions of luminal and basal cells. Is it necessary to get HGPIN cells to test or can the test be performed on other cells? Specifically how large a sample is necessary to get adequate sensitivity and specificity?


1.     Ahmed, H, Promoter Methylation in Prostate Cancer and its Application for the Early Detection of Prostate Cancer Using Serum and Urine Samples, Biomarkers in Cancer 2010:2.
2.     Armstrong, L., Epigenetics, Garland (New York) 2013.
3.     Brooks, J., et al, CG Island Methylation Changes Near the GSTP1 Gene in Prostatic Intraepithelial Neoplasia, Cancer Epidemiology, Biomarkers & Prevention , Vol. 7, 531-536, June 1998
4.     Chesire, D., W. Isaacs, β-Catenin signaling in prostate cancer: an early perspective, Endocrine-Related Cancer (2003) 10 537–560.
5.     Dayeh, T., et al, Genome-Wide DNA Methylation Analysis of Human Pancreatic Islets from Type 2 Diabetic and Non-Diabetic Donors Identifies Candidate Genes That Influence Insulin Secretion, March 2014, Volume 10, Issue 3.
6.     Donkena, K., et al, Oxidative Stress and DNA Methylation in Prostate Cancer, Hindawi Publishing Corporation, Obstetrics and Gynecology International, Volume 2010.
7.     Francis, J., et al, b-Catenin Is Required for Prostate Development and Cooperates with Pten Loss to Drive Invasive Carcinoma, PLOS Genetics, www.plosgenetics.org ,  1 January 2013,| Volume 9, Issue 1, e1003180.
8.     Gonzalgo, M., et al, Prostate Cancer Detection by GSTP1 Methylation Analysis of Post biopsy Urine Specimens, Clin Cancer Res 2003; 9,  2673-2677.
9.     Gordon, M., et al, RASSF1 Polymorphisms in Cancer, Molecular Biology International, Volume 2012, Article ID 365213, 12 pages
10.  Hanna, C., G. Kelsey, The specification of imprints in mammals, Heredity (2014) 113, 176–183.
11.  Hartnett, L., L. Egan, Inflammation, DNA methylation and colitis-associated cancer, Carcinogenesis vol.0 no.0 pp.1–9, 2012.
12.  He, T., et al, Identification of c-MYC as a Target of the APC Pathway, Science, 281, 1509 (1998).
13.  Hotchkiss, R., L. Moldawer, Parallels between Cancer and Infectious Disease, NEJM, 371;4, July 24, 2014.
14.  Jaiswal, J., S. Naryan, p53-dependent Transcriptional Regulation of the APC Promoter in Colon Cancer Cells Treated with DNA Alkylating Agents, THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 21, Issue of May 25, pp. 18193–18199, 2001
15.  Kundu, J., Y. Surh, Inflammation: Gearing the journey to cancer, Mutation Research 659 (2008) 15–30.
16.  Laborde, E., Glutathione transferases as mediators of signaling pathways involved in cell proliferation and cell death, Cell Death and Differentiation (2010) 17, 1373–1380.
17.  Mancini et al, MDM4 (MDMX) localizes at the mitochondria and facilitates the p53-mediated intrinsic-apoptotic pathway, The EMBO Journal (2009) 28, 1926–1939.
18.  Marks, F., et al, Cellular Signal Processing, Garland (New York) 2008.
19.  Meirs, I., Glutathione S-Transferase pi (GSTP1), Atlas Genet Cytogenet Oncol Haematol. 2010; 14(12).
20.  Morin, P., A. Weeraratna, The APC Tumor Suppressor Pathway, El Deiry (Editor), Tumor Suppressor Genes, Humana (New York) 2003.
21.  Muller et al, p53 and its mutants in tumor cell migration and invasion, J. Cell Biol. Vol. 192 No. 2 209–218.
22.  Partin, A., et al, Clinical Validation of an Epigenetic Assay to Predict Negative Histopathological Results in Repeat Prostate Biopsies, J Urol. 2014 Apr 18.
23.  Patten, M. et al, The evolution of genomic imprinting: theories, predictions and empirical tests, Heredity (2014) 113, 119–128.
24.  Rabiau, N., et al, Methylation Analysis of BRCA1, RASSF1, GSTP1 and EPHB2 Promoters in Prostate Biopsies According to Different Degrees of Malignancy, in vivo 23: 387-392 (2009)
25.  Richter, A., et al,  The RASSF proteins in cancer; from epigenetic silencing to functional characterization, Biochimica et Biophysica Acta xxx (2009) xxx–xxx.
26.  Wan, X., et al, Activation of β-Catenin Signaling in Androgen Receptor–Negative Prostate Cancer Cells, Clin Cancer Res, 2012.
27.  Wang, T., et al, Glutathione S-transferase P1–1 (GSTP1–1) Inhibits c-Jun N-terminal Kinase (JNK1) Signaling through Interaction with the C Terminus, THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 24, Issue of June 15, pp. 20999–21003, 2001.
28.  Wantanabe, M., et al, APC Gene Mutations in Human Prostate Cancer, (Jpn J Clin Oncol 26: 77-81, 1996.
29.  Wojno et al, Reduced Rate of Repeated Prostate Biopsies Observed in ConfirmMDx Clinical Utility Field Study, Am Health Drug Benefits. 2014;7(3):129-134, www.AHDBonline.com