4.4. Polymorphism: From Gene Mapping to Pharmacogenomics

pp. 42-44 in Bioethics and the Impact of Human Genome Research in the 21st Century

Author: Mark K. Skolnick (Chief Scientific Officer, Myriad Genetics, Inc., USA)

Editors: Norio Fujiki, Masakatu Sudo, and Darryl R. J. Macer
Eubios Ethics Institute

Copyright 2001, Eubios Ethics Institute All commercial rights reserved. This publication may be reproduced for limited educational or academic use, however please enquire with the author.
In this presentation, I will trace the history of the discovery and use of DNA polymorphisms, from their origin at the root of the genome project, through their use in the mapping and cloning of diseases, using BRCA1 and BRCA2 as examples. Causal DNA polymorphisms are revealed in high-risk individuals through genetic testing which brings to the forefront both technological and ethical issues. I will demonstrate how, through well-documented clinical utility, genetic testing for BRCA1 and BRCA2 has proven to be a prime example of a valuable pharmacogenomic application of DNA polymorphisms. Systematic discovery of such causal variants should yield a new gene based medicine.

Systematic analysis of DNA polymorphisms has evolved as a major aspect of the field of genomic analysis, which is the study of genomes rather than genes, resulting in an industrial scale approach to biology and it's applications.

Genomics has evolved from a number of advances, but in particular, genomic science could not have arrived without the advent of recombinant DNA (known as gene cloning), DNA sequencing, and the polymerase chain reaction (PCR). Collectively, these lead to the completion of the first draft of the Human Genome this year. We now know that the Human Genome is 3.12 billion DNA base pairs in length, although there is still debate how this sequence is decoded into genes and whether the genome will eventually be seen as small as 40,000 genes or as many as 150,000 genes. To date there are only about 5,000 genes which are very well characterized.

With the conceptual breakthroughs, there have been accompanying technical breakthroughs. Automated DNA sequencing has made it possible to evolve from the 1,000s of bases which a researcher could produce in a day, to over 100,000 bases a day per sequencer. Capillary sequencers have increased the number of bases a day to up to 500,000 per sequencer and this number will increase another order of magnitude in the next few years. We can only guess at the capacity of DNA chips, but with a DNA chip reader, they will certainly exceed 50,000,000 bases a day in the near future. The value obtained by high throughput sequencing comes from the use of oligonucleotide synthesizers, which can make 48 sequencing primers a day, and high-throughput thermal cycles which can produce thousands of PCR reactions an hour.

All genomic analysis requires two elements added to the molecular tools described above. Advances in both bioinformatics and automation lie at the core of past and future progress. Just as the Internet has enabled the sharing of knowledge, robotics is at the core of today's high-throughput systems. Microfluidics and the next generation of computers, molecular analysis technologies, bioinformatics, and telecommunication breakthroughs will be at the core of the ultra high-throughput, low cost analyses which are on the horizon.

Before 1980, DNA sequence variation was revealed indirectly through isoenzymes and blood and tissue antigens. In 1980 the concept of using restriction enzymes to reveal restriction fragment length polymorphisms, or RFLPs, led to the development of genetic maps. This pre-PCR technique used southern blots and was clumsy and slow. But it could reveal both single nucleotide polymorphisms, called "SNPs", or insertions of blocks of DNA. The analysis of a variable number of inserted blocks of DNA started with minisatellite fingerprints and progressed to analysis of a variable number of tandem repeats, called "VNTRs". With the advent of PCR these techniques evolved into the analysis of microsatellite repeats, especially variable length runs of the DNA bases "C" and "A", known as CA repeats which are very frequent in the human genome. This analysis generalized to all short tandem repeats, known as STRs. We are now entering the SNP era. Single Nucleotide polymorphisms or SNPs can now be seen via PCR rather than as RFLPs and hundreds of thousands of SNPs are now known.

At Myriad we developed a very useful tool for mapping common disease genes. We created an integrated map of nearly 7,000 short tandem repeat markers and selected 630 of them for our automated genotyping analysis. The selected markers have a high degree of variability, leading to 74% of the individuals tested being heterozygotes, that is having two different underlying DNA sequences at the marker. This tool gave us an average spacing of 5.38 centiMorgans. An early version of this tool was the initial step used to find genes using a technique called positional cloning. Positional cloning analyzes the pattern of diseases in families and relating these patterns to patterns of DNA polymorphisms. Utah has a special role in providing large multigenerational families which greatly accelerate the process. Because Mormons keep extensive publicly available genealogical records, we were able to construct a computerized genealogy of the Utah descendants of the Mormon pioneers which was then linked to a statewide tumor registry with cases from 1966 to present. In our search for BRCA1, we were able to extract large breast cancer families from this resource for genetic analysis. Early polygamy and large families, led to enormous 7 generation descent groups with dozens of cases of breast and ovarian cancer. Furthermore, this population was eager to participate in our studies.

Once a gene is localized to a specific chromosome, detailed analysis of more polymorphisms leads to the narrowing of the region which contains the gene. The region defined by the polymorphisms is sequenced, candidate genes in the region are identified, and these are resequenced in representative members of each family until one gene with abnormal sequences which would destroy the gene product is identified. This process yielded BRCA1, the first common cancer gene to be identified by positional cloning. The mapping and isolation of BRCA1 was a much publicized endeavor, as this was the first race for an important cancer predisposing gene. Mary Claire King first found that DNA polymorphisms some 30 million bases away from BRCA1 showed a pattern of genetic linkage in 1990. Many groups contributed to the localization of the gene over the next few years, but we were all confused by some data which pointed to the wrong region. Finally in 1993 we defined the proper region and the gene was cloned the following year.

Although the average gene is about 1,000 to 1,500 DNA bases long, BRCA1 was 5,592 bases long, greatly increasing the difficulty in developing a genetic test for breast and ovarian cancer predisposition.

Just as BRCA1 was cloned, we mapped BRCA2, in collaboration with Prof. Stratton's group. A year later, his group found a fragment of BRCA2, and shortly thereafter we independently found the entire gene. This gene posed even greater problems for genetic testing, being approximately twice the size of BRCA1.

We solved the problems presented to us by creating new automation, miniaturizing reaction volumes, implementing rigorous reagent quality control, and creating a complex system of sample tracking and result driven reporting. We were able to create the first, and to date only, high-throughput genetic testing laboratory. In this application of genomics, DNA polymorphisms are not used for gene discovery, but rather to distinguish carriers of normally functioning genes from carriers of genes with abnormal sequences. This process begins with the collection of a blood sample in a bar coded tube. DNA is isolated, 82 PCR reactions are created and sequenced in forward and reverse, and the sequences are compared with the wild type to look for abnormal sequences. If an abnormal sequence is found, that PCR reaction is repeated and the analysis is repeated to confirm the result. We have never known this highly automated system to mix up a sample or give a false positive result.

The high survival rate of women diagnosed with early-stage breast cancer warrants heightened surveillance for women who carry mutations in BRCA1 and BRCA2, commencing at an early age in recognition of the earlier age of onset of hereditary breast cancer. The "Cancer Genetics Studies Consortium" of the National Human Genome Research Institute has specifically recommended that women with mutations in BRCA1 and BRCA2 undergo annual or semiannual clinician breast examinations as well as annual mammography, to commence before age 35. Suspicious lesions in young women known to carry mutations should be evaluated promptly. Women with mutations in BRCA1 and BRCA2 might also benefit from investigational imaging methods not generally available for routine screening.

Because oral contraceptives reduce the risk of ovarian carcinoma among women in the general population by 50%, it has been suggested that they may be effective in reducing cancer risk for carriers of mutations in BRCA1 and BRCA2. Data specific to genetically susceptible women are not yet available but a significant association between oral contraceptive use and absence of ovarian cancer has been observed among women in the Gilda Radner hereditary ovarian cancer registry. In this context, it should be noted that the preponderance of evidence so far suggests that oral contraceptives do not increase risk of breast cancer.

An NIH consensus development panel on ovarian cancer concluded that "the risk of ovarian cancer from families with hereditary ovarian cancer syndromes is sufficiently high to recommend prophylactic oophorectomy in these women at age 35 years of age or after childbearing is completed." The most important concern regarding prophylactic oophorectomy is the possibility of subsequent peritoneal carcinomatosis, which has been documented in 2% to 11% of women with hereditary risk who have undergone this procedure. Most studies of this phenomenon were conducted before direct genetic testing for BRCA1 and BRCA2 was available, and consequently there are few data regarding the risks of peritoneal carcinoma following prophylactic oophorectomy for carriers of mutations in BRCA1 and BRCA2. This remains an area of ongoing investigation.

In conjunction with increased surveillance, it may be possible to reduce breast cancer risk by "selective estrogen receptor modulators" (SERMs). Tamoxifen, for instance, was recently demonstrated to reduce the risk of breast cancer by 45% in women with an elevated risk of the disease. Significantly for young women with hereditary risk, side-effects of tamoxifen (such as endometrial carcinoma and thrombo-embolic events) were not observed in any of the participants under age 50. The family history criteria used for defining high risk in women under age 50 in this study is very likely selected for women with mutations in BRCA1 and BRCA2, and these genes are being analyzed in a subgroup of the study participants.

Prophylactic simple mastectomy is chosen by only a minority of women with cancer-predisposing mutations. This procedure does not eliminate the possibility of breast cancer but has been shown in high-risk women to reduce the risk by more than 90%. Prophylactic mastectomy may be chosen by women whose mammographic assessment is compromised (for example, by extensive fibrocystic change), or by women whose experience and perception of breast cancer has been influenced by relatives or close friends who suffered or died from early-stage disease.

Increasing numbers of patients are requesting and obtaining coverage for the cost of genetic testing from their insurance carrier or managed care organization. In the United States, virtually all insurance companies and many managed care organizations are expanding coverage to include genetic testing services and are developing guidelines to determine eligibility for counseling and testing. As testing for hereditary cancer risk has become more widespread, it has become apparent that "the perceptions of discrimination far exceed the reality." Although testing for hereditary breast-ovarian cancer syndrome has been available since 1996, instances of health insurance discrimination in healthy individuals based on these tests appear to be exceedingly rare. In addition, the Health Insurance Portability and Accountability Act of 1996 makes it illegal for group health plans to call genetic information a "pre-existing condition" so that they can use it to deny or limit coverage. Many states have also passed laws that prevent genetic discrimination. For healthy women, there is little evidence to suggest that the identification of a hereditary cancer syndrome will jeopardize their access to health insurance. For patients already diagnosed with cancer, the additional risk to insurability posed by genetic testing would seem to be small indeed.

In conclusion, we now see before us an expanding field of application of DNA polymorphisms to medicine called pharmacogenomics. In the next decade we will discover many polymorphisms in genes which predispose individuals to different diseases or which lead individuals to metabolize drugs differently. This will lead to a new classification of diseases by genotype rather than phenotype. As we know the specific causality of diseases, we can design drugs more effectively both by using the indicated genes as drug development targets and by prescribing drugs specifically for subclasses of diseases. We now think of genetic testing as appropriate for rare high-risk individuals, but in the future, we will all know our genetic profiles and be able to specifically prevent diseases from occurring and treat them when they do occur in a highly effective fashion. The era of gene-based medicine is around the corner, and analysis of DNA polymorphisms both for drug discovery and for drug prescription is at the heart of this new era. As with any new technology, there are benefits and challenges, and it is only by workshops such as this, which analyze and confront those challenges, that this new era can properly unfold.

Please send comments to Email < asianbioethics@yahoo.co.nz >.

To contents page
To Eubios book list
To Eubios Ethics Institute home page