| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Authors' Affiliations: 1 Breast Cancer Program, Department of Oncology, Johns Hopkins University School of Medicine and 2 Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
Requests for reprints: Kala Visvanathan, Johns Hopkins Bloomberg School of Public Health, E 6142, 615 North Wolfe Street, Baltimore, MD 21205. Phone: 410-614-1112; Fax: 410-614-2632; E-mail: kvisvana{at}jhsph.edu.
The term "epigenetics" was first coined in the 1940s to describe the interaction between genes and environment in the development of specific phenotypical traits (1). Within a decade, DNA methylation was identified as the first epigenetic marker (2). Sixty years later, there is compelling evidence for the role of aberrant epigenetic gene regulation, including DNA methylation in tumorigenesis. Currently, epigenetics refers to heritable changes in gene expression that occur without changes in nucleotide sequences. These changes may involve chemical modifications of the DNA itself, such as DNA methylation or modifications of proteins that are closely associated with DNA, such as the histones that bind and compact DNA into chromatin packages (3). DNA methylation occurs at cytosine bases located 5' to a guanosine, so-called CpG dinucleotides (3). Short regions of CpG dinucleotides known as CpG islands are found in the proximal promoter region of over half of human genes. Methylation of these gene promoters is generally not detected in normal tissues but hypermethylation of CpG islands resulting in a loss of gene function is a common feature of almost every type of cancer. Indeed, it has been postulated that CpG island methylation may serve as the "second hit" of the Knudson two-hit hypothesis to inactivate the normal allele of a tumor suppressor gene in certain hereditary cancer syndromes, such as the familial breast and ovarian cancer syndrome linked to BRCA1 mutations (3). Epigenetic silencing of genes, such as BRCA1, has also been observed in sporadic tumors showing the importance of both genetic and epigenetic phenomena in tumorigenesis (4). The observation that DNA methylation of the same gene occurs in both premalignant lesions, such as atypical hyperplasia of the breast (5) and carcinoma, suggests that it might serve as a potential marker for early detection or risk assessment. Prognosis (6, 7) and response to treatment (8) could also vary based on the epigenetic characteristics of a specific tumor. Lastly, unlike germ-line mutations, epigenetic modifications, such as DNA methylation or histone acetylation, can potentially be reversed, making them appealing preventive and therapeutic targets (3). The challenge scientists and clinicians are facing now is whether identification of epigenetic changes, such as DNA methylation in breast tissue, can truly inform the early detection, prevention, and treatment of cancer.
In this issue of Clinical Cancer Research, Yan et al. (9) tackle aspects of this thorny question through a detailed analysis of the methylation patterns in the RASSF1A tumor suppressor gene promoter in human breast tissues. This promoter is among the most frequently methylated sites in human cancer and its methylation leads to silencing of RASSF1A expression (10). Using microdissected human breast tissues and well-validated methylation-specific oligonucleotide microarrays, a 4-kb region surrounding the transcription start site of the RASSF1A gene was mapped in 23 invasive breast cancers and 17 normal breast tissues from healthy women. These studies showed that the outer flanking regions of the RASSF1A gene CpG island are heavily methylated in both malignant and normal tissues, whereas differential methylation of the core of the CpG island is observed—high-level methylation (40-90%) in 75% of the cancers versus low-level methylation (20%) in the normal control samples. These findings are consistent with results from other investigators in studies of breast tissues and fluids and suggest that RASSF1A promoter hypermethylation may play an important role in breast tumor development (11–17). A logical question is whether such RASSF1A methylation might be a hallmark of a field defect in breast tissue, particularly in the affected breast. Thus, Yan et al. (9) also compared RASSF1A promoter core methylation in primary breast cancers, with histologically normal adjacent tissue and with normal breast tissues from healthy women. The mean percentage methylation was highest in tumor tissues; it was also significantly increased in histologically normal tissues adjacent to tumor compared with normal tissues from unrelated healthy individuals. However, there was significant overlap in the percentage methylation among tumors and adjacent tissue, complicating the interpretation of methylation levels at either end of the spectrum. A careful analysis of RASSF1A methylation in normal tissues obtained at progressively greater distance from the primary tumor suggested a gradient in some but not all breasts such that the extent of methylation was greater in the tissue within a 1-cm circumference of the tumor compared with tissue obtained from 2 to 4 cm. Provocatively RASSF1A methylation was also seen in normal tissue obtained from the contralateral breast in two women with RASSF1A core promoter methylation in the cancer.
The interplay between genetic and epigenetic factors in carcinogenesis is also reinforced in this article by the finding that loss of heterozygosity in the chromosome 3p region that encompasses the RASSF1A gene is a common feature of breast cancer and that similar changes may be found in adjacent normal tissues. It seems that these epigenetic and genetic changes may be observed separately or in concert with each other. Finally, findings of this type are not restricted to the RASSF1A gene. Indeed, in this study, a global screen of paired invasive ductal carcinomas with adjacent normal tissues and additional breast samples obtained from healthy individuals identified several loci that are preferentially methylated in tumor versus normal tissue; a careful analysis of three such genes (CYP26A1, KCNAB1, and SNCA) suggests that their methylation in the primary tumors can be associated with methylation in the adjacent normal tissues.
The strengths of this study lie in the promise of epigenetic markers in all phases of breast carcinogenesis, the importance of the questions posed, the focus on microdissected tissues, and the use of sophisticated molecular techniques to identify, evaluate, and confirm the presence of these markers. It confirms the observation by Fackler et al. (18) that cumulative methylation of a panel of genes (RASSF1A, TWIST, HIN1, and Cyclin D2) is seen more commonly in primary invasive breast cancers than reduction mammoplasty specimens from well women and that normal tissue adjacent to primary breast cancer frequently harbors the same methylation profile as the cancer. Whether this latter finding reflects an inadequate surgical margin of a poorly circumscribed malignancy or early premalignant changes needs further evaluation. A methodologic limitation of methylation studies to date has been the considerable difficulty in obtaining sufficient tissue that is representative of the entire organ being evaluated. Unfortunately, studies to detect DNA methylation in the blood are in their infancy in breast cancer; therefore, such tests are not yet a suitable surrogate marker for the tissue. Although DNA methylation has been detected in breast fluid (19), the routine use of nipple aspirate or ductal lavage has been limited by the inconsistency of breast fluid production. (20). Random fine-needle aspiration of the breast may be an option (20). Ultimately though, a robust approach to obtain relevant samples with minimal discomfort and cost needs to be identified before large clinical studies can be undertaken.
If a panel of methylation markers is confirmed to be associated with the development of breast cancer, the next challenge is how that knowledge might better inform clinical practice. For example, methylation markers may help to identify women at high risk for disease by molecular criteria that might supplement or even replace traditional clinical and epidemiologic information about risk. Lewis et al. (21) reported a higher frequency of methylation of both RASSF1A and APC genes in unaffected women at high risk for breast cancer compared with those at low or intermediate risk based on the Gail model. If such changes do indeed occur earlier than abnormal histologic findings and are associated with subsequent development of breast cancer, then methylation markers in breast samples could potentially identify women at increased risk for breast cancer who might be candidates for targeted screening and prevention strategies and also serve as intermediate biomarkers in chemoprevention trials. Alternatively, for women diagnosed with breast cancer, knowledge about the methylation status of the tumor tissue may contribute to a better estimation of prognosis and the prediction of need for and/or response to specific therapies. In addition from their epigenetic mapping studies, Yan et al. (9) suggest that the identification of methylated markers in apparently normal tissue adjacent to tumor may enable surgical margins and radiation portals to be better defined, whereas absence of aberrant markers might minimize the need for further surgery or radiotherapy. Such a test might be especially useful in the management of ductal carcinoma in situ, an increasingly common scenario where the majority of women currently undergo excision, radiotherapy, and/or tamoxifen therapy to prevent future complications in a minority of survivors.
Although the study of Yan et al. (9) focused largely on RASSF1A methylation, a higher sensitivity and specificity for detecting breast tumors has been attained with the measurement of multiple markers, reflecting the heterogeneity of breast cancer. High sensitivity and specificity, an important feature of any test, is particularly important to minimize false-positive results when evaluating healthy women where the presence of methylation is rare. Early efforts to establish a quantitative multiplex PCR-based assay for methylated genes for use in breast cancer diagnosis and risk assessment have been reported by Fackler et al. (14). This assay was found to be twice as sensitive as cytopathology for detection of methylated markers in ductal lavage fluids collected from women undergoing mastectomy for breast cancer. Interestingly, a panel of nine genes was not more useful than a panel of five genes, suggesting that a panel of a limited number of carefully selected genes may be sufficient.
In summary, promoter hypermethylation of specific tumor suppressor genes, including RASSF1A, has been observed in breast cancers as well as the adjacent normal tissue and in some instances in the contralateral unaffected breast. The significance of these findings in the context of breast cancer risk, prevention, and treatment remains to be elucidated. As we attempt to extend these findings in the clinical trial arena, we should recall the counsel of William L. McGuire (22) that assays for breast cancer risk assessment, prognosis, and response prediction must first be biologically plausible; such is the case with methylation markers. The imperative to optimize assays and validate panels of DNA methylation markers in well-designed clinical studies in both high risk and affected populations is now upon us.
 |
Footnotes |
|---|
Received 8/10/06; accepted 8/30/06.
|
References |
|---|
 Top References |
|---|
(stratifin) is an early event in breast cancer. Oncogene 2001;20:3348–53.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
