Epigenetic Biomarkers

Pancreatic cancer is very "sneaky" due to the late onset of symptoms that prevent an early prognosis and surgical removal of the tumor. The cancer cells make use of the hallmark of evading growth suppressors; tumor suppressor genes are hyper-methylated, thus they are silenced. However, we can make use of the hyper-methylation of the cancer cells to our advantage: DNA methylation patterns can be detected through a methylation-specific PCR (MSP) analysis (1).

Introduction
There are many classical clinical ways for the scanning of pancreatic tumors: ultrasound, computed tomography scanning, MRI, endoscopic ultrasound...etc. However, some clinical scanning are invasive. With any imaging technique, there are limitations. With the intake of certain bio-markers, the first few screenings may show accurate scanning. Over time, the body may adapt to the presence of these bio-markers, causing either a false positive or negative of the test result. How does one determine whether if the chemotherapy is working if the results are misleading? There are other tests that analyze the samples from the patients such as cytopathological assessment to accompany the scanning results. For cytopathological assessment however, the cells must be intact to provide an accurate reading. Since we are concentrating on the cancer of the pancreas, there are pancreatic proteolytic enzymes in the pancreatic juice. The chances of obtaining intact cells for an accurate reading are not reassuring (1). So rather than introducing something new to the body and the cancer cells, we can make use of what is already present: the methylation of the CpG islands.

Why DNA methylation?

The article by Fukushige and Horii (1) offers 4 reasons:

1. Incidences of aberrant DNA methylation of specific CpG islands are higher than those of genetic defects.
2. The aberrant DNA methylation seen in cancer cells can be sensitively detected, even when it is embeded in substantial amounts of contaminating normal DNA.
3. Detection of aberrant DNA methylation is technically simple; it can be detected using MSP.
4. Aberrant DNA methylation seems to occur in early-stage tumors, causing loss- and/or gain-of-function of key processes and signalling properties.

 Cancer is the result of minimum of 14 significant mutations; the perfect combination of certain mutations as it was put in class. CpG island promoters are usually unmethylated if the gene was expressed. For the first reason, the article does not offer a full explanation. As for reasons 2 and 3, the article also fails to describe the mechanics of MSP and how DNA methylation is detected. But those aside, reason 4 seems promising in the prognosis of pancreatic cancer. If aberrant DNA methylation can be detected in the early-stages, then early detection leads to a higher chance of surgical removal of tumor and can be used as a predictor of when the tumor may metastasize.

Purpose
The detection of aberrant DNA methylation in the CpG islands is one of the most common molecular lesion of the cancer cell. The hyper-methylation of CpG islands promoters are associated with the over expression of DNMT1 (DNA methylatransferase). The article however does not go into detail of how the normal CpG islands are hyper-methylated by DNMT1, only associating the idea that "DNA methylation alterations participate in multistage pancreatic carinogenesis" (2). Utilizing that idea, the researchers performed an experiment by performing a bacterial artifical chromosome array-based methylated CpG island amplification on tissue samples of normal pancreas, noncancerous pancreatic tissue from patients with ductal adenocarcinomas and cancerous tissue. To be able to differentiate between the tissue samples through the amplification of the CpG islands, not only the epigenetic biomarkers are able to aid in prognosis but in determining of the stages of the tumor.

Methods
15 tissue samples were obtained from patients without ductal adenocarcinomas, but the patients have a history of other tumors for the control of "normal tissue of normal pancreas". 32 tissue samples of nearby regions of the dutcal adenocarcinoma tumor but without the presence of contaminating cancer cells were obtained from patients for the noncancerous tissue group. 91 tissue samples were obtained from ductal adenocarcinomas patients for the cancerous tissues, The researchers extracted DNA from the tissues and methylation of CpG islands were analyzed using a custom-made array with 4361 bacterial artificial chromosome clones that can be used to detect cytosine methylation on 23 chormosomes and the sex chormosomes. The arrays were scanned and the signal rations were normalized for each group (2). 

Results
 
The three groups: C as in normal pancreatic control, N as in noncancerous tissue and T as in cancerous tissue were classified by the signal ratio of the bacterial artificial chromosomes and the methylation of DNA from samples. From the data, if there are more signal ratios, then there are more methylations of DNA of the amplified CpG islands promoters. A comparison of the normal control and the cancerous tissue sample shows a significant increase of hypermethylation in the cancerous tissue. If one was to look at the progression from normal to noncancerous, the hypermethylation increases in CpG island promoters (2).

Discussion

• The data indicates different stages of tumor progression from normal to noncancerous to cancerous tissue by increased hypermethylation of CpG island promoters. DNA methylation can be used as a epigenetic biomarker to not only aid in the diagnosis of pancreatic cancer but in prognosis of the cancer as well.

• As we learned in class today, epigenetic changes does not involve a base pair change (from A to G) but a unmethylated C in to a methylated C in CpG islands. A methyl group attaching to C (causing a covalent modification of DNA) will result in a silenced gene. With the added information of the low occurrence of CpG islands in the genome due to genomic defect, hypermethylation of these promoters can be used as indicators of cancer. 

• As mentioned in the article by Gotoh et al. (2011), pancreatic cancers are associated with the overexpression of DNA methyltransferase which result in hypermethylation of CpG island promoters. With hypermethylation, many tumor suppressor genes are silenced. If one can narrow down to the specific cause of the overexpression of DNA methyltransferase, one may be able to interfere with evading of growth suppressors.

• I was a bit unsure of the control used for the experiment. The control was "normal" pancreatic tissue, yet the samples that were used were obtained from patients with other tumors. If other tumors were already present, can we assume that the tissue has cancerous cells, thus it cannot be classified as normal tissue? Or is it the fact that we are concentrating only on ductal adenocarcinomas methylation patterns that other tumors do not interfere with the array? The array was done on all 24 chromosomal pairs. It would be nice if we could narrow down to specific chromosomes so a better comparison. The data generated by the experiment shows a distinct difference, but it bothers me that we can only use the epigenetic biomarkers to quantify the presence of pancreatic tumor. 

The last thing I want to comment on is the sensitivity of the test. I understand that the incidences of abnormal hypermethylation of CpG island promoters are rare in normal tissue, but if one was to compare normal tissue to noncancerous tissue, how will we account for some genetic differences for the methylation patterns? Both C and N tissue sample signal ratios look similar. We need a border line between having cancerous tissue and noncancerous tissue.  Otherwise, the utilization of DNA methylation as epigenetic biomarkers seems promising.


Works Cited

1) Fuskushige, S., Horii, A. Road to Early Detection of Pancreatic Cancer: Attempts to Utilize Epigenetic Biomarkers, Cancer Letters, (2012)

2) Gotoh, M. et al. Diagnosis and Prognostication of Ductal Adenocarcinomas of the Pancreas Based on Genome-Wide DNA Methylation Profiling by Bacterail Artifical Chromosome Array-Based Methylated CpG Island Amplification, Journal of Biomedicine and Biotecnology, (2011)

Images from
a) http://en.wikipedia.org/wiki/DNA_methylation
b) http://www.cancer.gov/cancertopics/understandingcancer/cancergenomics/page33
c) Reference 2