| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Chemistry Forum |
Department of Pathology and Laboratory Medicine, University of Pennsylvania Medical Center, 3400 Spruce St., Philadelphia, PA 19104-4283.
a Author for correspondence. Fax 215-349-5090; e-mail donaldyo{at}mail.med.upenn.edu
 |
Introduction |
|---|
 Top Introduction TextReferences |
|---|
The opening presentation by Bradley W. Popovich, PhD, Director, DNA Diagnostic Laboratory, and Director, Clinical Genetic Services Laboratory, Oregon Health Sciences University, set the background for the conference by describing the evolution of DNA diagnostic testing. DNA tests generally are extremely accurate, minimally invasive, and cost-effective. Many hundreds of genetic conditions are already diagnosed through DNA testing. The number and variety of genetic tests will expand rapidly as an outgrowth of the Human Genome Project.
|
Text |
|---|
|
TopIntroductionTextReferences |
|---|
Three methods predominate genetic testing: Southern blot analysis, PCR, and DNA sequencing. Southern blot analysis has been used since the early 1980s, but it is extremely labor-intensive and the turnaround time is ~1 to 2 weeks. The cost is between $250 and $500 per test. PCR has been in use clinically since ~1990. It has a turnaround time of 1 to 5 days at a cost of between $100 and $300 per test. DNA sequencing, which often combines PCR with other types of testing, has been used clinically since ~1995. The turnaround time is 1 to 4 weeks, depending on the test. DNA sequencing currently is very expensive, and its analytical sensitivity is unknown for many genetic targets.
The final phase of genetic testing is genetic counseling of patients and, potentially, their family members. Two parameters are important for the proper interpretation of the relevance of genetic test results: the positive predictive value of the test, the probability that an individual with a positive test result will have the phenotype; and the negative predictive value, the probability that an individual with a negative test result will not have the phenotype. Both positive and negative genetic test results impact an extended family. There is considerable debate about the duty to inform the extended family of genetic disposition. A major issue is balancing the confidentiality of genetic information for the patient and the rights of family members to know their genetic risk. An additional ethical issue for the laboratory to consider is whether there is a need to inform and counsel patients and family members about the availability of improved genetic tests.
role of the human genome project
The Human Genome Project is making tremendous advances in our
understanding of the human genome sequence. Identification of
individual genes, however, will be more laborious. The effort
eventually will determine every gene, but not every disease. The
project will be able to identify genes and mutations or polymorphisms
that might be useful for prognosis and patient management, but not
necessarily for prediction and prevention of diseases. It should be
emphasized that gene therapy will lag behind gene diagnostics and will
not be available for many diseases.
There are likely to be several areas for future application of genetic disease testing. Among these are cardiovascular diseases and cancer. There is likely to be further development of multiplex genetic testing, including multiple mutations in a single gene, multiple genes causing single disease phenotypes, and multiple genes causing multiple phenotypes screened simultaneously.
consequences of genetic testing
Changes in the medical testing paradigm are likely to occur as a
result of genetic testing. Genetic testing for a given disease
generally is inexpensive, very accurate, and minimally invasive.
Referral patterns may change when nonspecialist physicians and others
can make difficult diagnoses and management decisions for patients
through available online databases. Nevertheless, genetic test results
should not allow employment or insurance discrimination or lead to the
persecution of patients. Comprehensive Federal protective legislation
that will protect the patient and at the same time will not limit
legitimate biomedical research is needed.
Genetic test results should be shared with first- and second-degree relatives. This sharing of medical information would be facilitated by a single payer healthcare system, as in Canada, and a national medical database. The evolution of genetic testing will be hastened by addressing the patenting of the human genome. Currently, patient access to genetic testing is restricted, which increases the cost of testing, affects the overall quality of genetic testing, and also impacts biomedical research.
Barbara Fuller, JD, Senior Policy Analyst, National Human Genome Research Institute, discussed "Public Policy Issues and Genetic Information". She addressed the ethical, legal, and social issues of genetic testing, most specifically, the informed consent for genetic testing as well as discrimination in health insurance and employment, privacy issues, and confidentiality. In addition, she discussed familial rights and responsibilities. The public is worried about genetic testing: In one survey, 86% said they were very concerned or somewhat concerned that insurance companies or their employers would use genetic information against them. When the Human Genome Project was started, money was set aside for exploration of the ethical, legal, and social impact of knowing the complete human genome sequence, not just for the research to obtain the sequence.
informed consent
Informed consent for genetic testing is required because of the
potential for consequences other than medical or physical risks. For
example, the use of BRCA1 testing to screen for breast
cancer is straightforward, but the ramifications may be complex.
Increased anxiety and a change in self-image may be generated. If a
woman is told that she tests positive for a BRCA1 mutation,
how will her privacy and issues of confidentiality be affected? Altered
family relationships may follow identification of some members as
positive and others as negative. Social and group stigmas may ensue.
Insurance and employment discrimination is a real fear. But what are
the realities? The value of the predictive information is unknown. The
impact of predictive testing on behavior is largely unknown. The
biggest current problem is the difficulty in quantifying and qualifying
social risks. The problem will grow in the future with the exponential
development of new tests.
state legislation
Approximately 30 states have enacted legislation concerning
genetic testing. Numerous bills are pending before Congress to prohibit
health insurance discrimination. The definition of genetic information
has been an important issue in State legislation. There are many issues
to be considered, including privacy concerns and health insurance and
workplace discrimination. Privacy affects research records and medical
records. Prohibitions on discrimination, as well prohibitions on access
to genetic information, are necessary. Medical records and research
records should be viewed as separate entities. Research protocols
require a higher standard of protection because the research subjects
should not be informed of their own test results. If the information is
included in an individual's medical record, it is afforded the usual
protection of medical records. State laws vary greatly in the degree of
protection of medical records. Only some states have a privacy
component in their genetic discrimination laws that prevent health
insurers from gaining access to medical record information and using
it.
federal legislation
Genetic information is now being used by the courts. Many Federal
laws and regulations are in place to protect patient rights. They
include certificates of confidentiality, designed specifically to
protect individuals participating in research, so that researchers do
not have to release research results if subpoenaed by a court.
The testing of children should be prevented unless a medical intervention is available, or unless there is a childhood onset of the disease. If it is an adult-onset disease, there is no intervention possible.
Physicians should be aware of liability issues surrounding the standard of care. Doctors have a duty to disclose the availability of tests, as well as their benefits and risks. Currently, the most visible issue is the duty to warn the patient of genetic transferability, the liability to blood relatives, and the duty to keep information confidential.
Joel White, Legislative Assistant to US House of Representatives member Christopher Shays (Republican) of Connecticut, discussed pending Federal legislation. This includes: (a) HR 3900, the Consumer Health and Research Technology (CHART)1 Protection Act; (b) HR 4250, the Patient Protection Act; and (c) HR 306, a Genetic Testing bill. There are many competing and complex issues that must be addressed. The CHART bill was introduced because there is no comprehensive Federal law governing privacy interests. There is a patchwork of State laws, which are becoming increasingly more and more difficult for laboratories and others to follow as business crosses state lines. There are several Federal acts that cover specific aspects of an individual's identifiable health information, e.g., the Federal Privacy Act, the Social Security Act, and the Americans with Disabilities Act, but there is no comprehensive Act encapsulating all of these issues.
federal legislation
About 20 years ago, the Federal government released its first
report on medical records and the need to enact Federal privacy
legislation, but to date no legislation has become law. HR 3900, the
CHART Protection Act, would delineate both the appropriate and
inappropriate uses of healthcare information and establish strong civil
and criminal penalties for misuse or disclosure of an individual's
identifiable health information. The bill is not concerned with
anonymized information. It is designed to not inhibit research, and it
differs from other bills that require separate authorization for use of
an individual's health information. HR 3900 provides single-tiered
authorization for the release of information and establishes
administrative safeguards and procedures. The issue of preemption of
State laws is addressed in the bill. Because of the frequent
involvement of many states in the management of an individual's
privacy issues, the bill is intended to broadly and generally
preempt State law in this area and to establish Federal legislation as
the standard of protection.
The US House of Representatives has addressed managed care reform. A bill passed on July 25, 1998, included a section on medical record confidentiality. The bill was never taken up in the Senate, but may be addressed again in 1999. The Genetic Information Nondiscrimination and Health Insurance Act was introduced in 1997, but it applied only to genetic information. HR 3900 protects the whole medical record. There are some major opportunities and some real dangers as a result of the Human Genome Project. There is a need to legislatively balance the needs of business and society on the one hand, and the individual's expectations of confidentiality on the other. Yet, in spite of the importance of this issue, only a few staff members are researching confidentiality issues in either the Senate or House.
Anne Phelps, Health Policy Advisor for the Senate Labor and Human Resources Subcommittee on Public Health and Safety, discussed the general issues of genetic testing and health insurance coverage. She reviewed the history of the Federal legislation, the scope of the pending bills, and their prospects for passage. The 104th Congress passed the Health Insurance Portability and Accountability Act of 1996, which speaks to relevant genetic issues. The Act for the first time addresses the voluntary private health insurance market and deals primarily with discrimination in terms of access to coverage of health insurance and its portability. The Act basically prohibits discrimination against individuals in a group setting, particularly those in an employer-based health insurance system, with respect to health information or health status, including medical conditions, family history, and disabilities. Genetic information is also addressed. The Health Insurance Portability and Accountability Act states that genetic information should not be considered a preexisting condition. Individuals cannot be denied health insurance coverage on the basis of a genetic predisposition or genetic test result. These provisions primarily apply to the group insurance market. The legislation states that an individual cannot be discriminated against for eligibility of coverage, or if the individual is a member of a group, he or she cannot be singled out from that group and be charged a higher premium than everyone else in the group. However, the possibility of charging the whole group a higher premium based on one or two individuals is not addressed. There are no limitations on what rates can be set. Although an individual might have access to insurance coverage in the individual market, he or she might be charged an exorbitant rate, which may be based on any type of health information, including genetic information.
state laws
Currently, there are ~30 different State laws dealing with
genetic information and health insurance. These are quite varied. Some
prohibit requiring an individual to take a genetic test, and some
prohibit altering the condition, coverage, or benefits based on genetic
testing. Some laws also prohibit the use of genetic testing information
for the determination of rates. Each of these State laws has at its
core the definition of genetic information. Some definitions have
focused solely on the tests themselves, others have used a broader
definition that includes routine blood tests, family history, physical
examination, or medical records. Currently, few health insurers ask
for, or use, genetic test results to determine coverage or rates.
pending federal legislation
In the 105th Congress, ~10 bills were introduced to address
discrimination and privacy issues relating to the use and disclosure of
genetic and medical information. Some of the bills prohibit
discrimination in health insurance coverage and/or employment on the
basis of genetic information, others define confidentiality
requirements for the use of the actual physical DNA sample, and still
others address insurance and employment practices. Some of the bills
focus on the use of genetic information for research purposes.
Three major Senate proposals dealing with genetics specifically are (a) S-422, introduced by Senator Pete Domenici (Republican, New Mexico), which attempts to cover all issues, including health insurance discrimination, employment discrimination, and the use, storage, collection, and analysis of DNA or tissue samples; (b) S-89, introduced by Senator Olympia Snowe (Republican, Maine), the companion to HR 306, introduced by Representative Louise Slaughter (Democrat, New York), which concerns health insurance discrimination; and (c) S-1045, introduced by Senator Tom Daschle (Democrat, South Dakota), which concerns primarily the issue of employment discrimination. All of these bills were referred to the Senate Labor and Human Resources Committee, which held a hearing on genetic information and health insurance discrimination earlier in 1998. S-89, Senator Snowe's bill, effectively outlines some of the key principles involved in genetics and health insurance discrimination. The Genetic Information and Services bill, S-2330, became the Patient's Bill of Rights. It was introduced because patients throughout the country have a real fear of genetic discrimination, especially with regard to health insurance. The two main issues in front of Congress in 1999 are medical records privacy and the discriminatory use of genetic information by health insurers.
Judy Yost, Director, Division of Outcomes and Improvement, Health Care Financing Administration, discussed how the Clinical Laboratory Improvement Act of 1988 (CLIA) impinges on genetic testing. CLIA is still being challenged by some organized medical groups who want to reduce CLIA standards because they believe these are too burdensome; however, genetics groups are petitioning to make CLIA more stringent. In 1992, the Institute of Medicine published a report indicating that the CLIA standards were not stringent enough to oversee genetic testing. Subsequently, a NIH/Department of Energy (NIH-DOE) task force was convened to review all facets of genetic testing. It produced a final report in September 1997, containing ~14 recommendations, some of which address the quality of genetic testing.
clia and genetic testing
The NIH-DOE Task Force recommendations for genetic testing include
making the CLIA requirements more stringent for proficiency testing,
quality controls, personnel qualifications, and record keeping. The
Institute of Medicine also recommends the creation of a laboratory
specialty for genetic testing, with accreditation by a private
organization.
Proficiency testing was an area of major concern of the NIH-DOE Task Force. At present, CLIA does not require proficiency testing for genetic tests. Very few proficiency panels are now available for genetic testing. Therefore, many genetics laboratories do voluntary interlaboratory comparisons as an interim step. Several proficiency testing providers do have proficiency panels for genetic testing, although they are not currently approved under CLIA. If proficiency testing is not available, a laboratory is obligated twice a year to demonstrate the accuracy of the tests that they are performing.
The record-keeping portion of CLIA, called patient test management, requires the laboratory to use and maintain a system ensuring proper patient preparation, specimen collection and identification, preservation, transportation, and processing as well as accurate reporting of results. There are also record retention requirements, averaging ~2 years for most types of tests.
CLIA requires that a reference range be included with other interpretive information for the test result. In addition, laboratories must refer genetic tests only to laboratories that are CLIA certified for genetic testing. Confidentiality must be maintained. The quality provisions of CLIA apply to all tests, including those tests that are developed in-house, not just those that are cleared by the Food and Drug Administration for use in laboratories under the Premarket Approvall or 510(k) process.
CLIA has special quality-control requirements for cytogenetics. However, at present they are not very complete and are also out of date. Most genetic tests are highly complex. Therefore, the laboratory director needs to be an MD or a PhD with appropriate experience. For cytogenetics, CLIA currently requires the technical supervisor to be an MD or a DO with 4 years of training or experience in genetics, with 2 of these years in clinical cytogenetics, or a doctoral degree with 4 years training or experience in clinical genetics. The CLIA regulations do not specify qualifications for the technical supervisor of molecular genetics or biochemical genetics.
The technical supervisor for a genetics laboratory is responsible for all technical and scientific oversight of the laboratory, including test method selection, test verification, proficiency testing, program selection, quality-control procedures, troubleshooting with equipment or test systems, problem solving, and remedial actions as well as employee training. All aspects of CLIA apply to genetic testing. Some of the requirements clearly need to be updated or expanded. The recommendations that have been made by the CLIAC subcommittee are currently under review by both the CDC and Health Care Financing Administration before incorporation into final CLIA regulations. In particular, proficiency testing and personnel requirements must be addressed, including the role of the genetic counselor. The other areas of genetic testing that are only indirectly related to CLIA, but which are very important to genetic testing, are interpretation of results, informed consent, and counseling.
Brian Ward, PhD, Vice President of Operations, Myriad Genetics Laboratories, Salt Lake City, UT, discussed the economic aspects of genetic testing. There are primarily three types of genetic tests: (a) Research tests are associated with the discovery of the test and proof of the concept of the analysis. Initially these tests are of unproven diagnostic value. (b) Investigational tests become such when they are first applied to patients. Their performance is characterized. Their analytical sensitivity and specificity are established. When the tests are of demonstrable value, their results can be reported. Investigational tests may be reimbursed. (c) The generally accepted test is the final category. It uses an established protocol with equivalency between multiple laboratories, and often it becomes part of a standard of care or standard of medical practice.
profitability of genetic testing
Genetic testing should be profitable. This does not mean that it
is required to generate a positive cash flow for the laboratory.
Genetic testing should be associated with a positive currency flow. The
currency may differ in the academic research setting, in the
biotechnology setting, and in the hospital or corporate laboratory
setting. In the academic laboratory, the currency is funding for
research. The currency in the biotechnology laboratory is intellectual
property and proprietary positions, or patent positions. However, in
the biotechnology diagnostic laboratory, concerns for patient care
should supersede those for profitability. Some of the rewards and
challenges are similar for biotechnology companies and successful
academic laboratories.
In dollar terms, the gross margins in clinical laboratories are quite small. In the clinical diagnostic laboratory, the percentages of gross margin for the typical genetic testing laboratory are somewhere between 30% and 50%. This compares favorably with 7–8% for a supermarket, but is much less than the 300-1000% for some orphan drugs made by pharmaceutical companies. Genetic tests, in general, are high cost and low volume, very different from the typical clinical laboratory tests. The profitability of a clinical laboratory test, regardless of type, is influenced by the rate of reimbursement.
Established clinical laboratories have less freedom than biotechnology laboratories in that they can provide only generally accepted tests. These are more frequently high-volume/low-cost tests. Genetic testing in the US typically is performed in esoteric laboratories or in special genetic subsections. The transition from an investigational to a generally accepted test, i.e., one that is widely available and recognized as standard of care by the medical community, is protracted.
paying for genetic tests
Patients often must pay for genetic tests directly. Both
commercial and self-funded insurance companies may reimburse for
genetic tests. Health maintenance organizations (HMOs) and government
agencies may also pay for tests, albeit at a low rate. The appropriate
charge for a test may be determined using top-down or bottom-up
pricing. In the top-down pricing approach, a price is decided and
supported by a defensible and consistent coding strategy, using CPT
codes to achieve that price. In the bottom-up strategy, the most
appropriate CPT code to apply to the assay is considered, and then the
typical allowable reimbursement is reviewed to arrive at a charge. Both
of these strategies are defensible.
New CPT codes for genetic testing became available in January 1999. To maximize reimbursement for genetic testing, efforts should be made to seek reimbursement on a case-by-case basis at the time of test introduction. This may have greater success with a patient advocacy program behind it, which may take the issue directly to an insurance company. Preauthorization or predetermination of benefits should be sought. Letters from physicians supporting the medical necessity of the test are useful in persuading insurance companies to pay. All these efforts should be supported by scientific publications, letters of medical necessity, and requests from referring physicians and patients. An appropriate ICD-9 code is solicited from requesting physicians.
obtaining reimbursement from third-party payers
Personal contact should be made with the claims staff of the
insurance company or the HMO covering the patient being tested.
Contacting the medical director is useful, but perseverance is
essential. Low reimbursements or denials may be appealed. During this
process, it is important to emphasize the complexity of the test, what
is needed for a positive or negative diagnosis, the difficulty and
accuracy required in making a diagnosis, and the scientific validity of
the test.
Education at all levels is important. It is useful to develop a standardized reimbursement strategy, such as the patient paying a 20% co-payment up front, followed by a reimbursement for the usual and customary charges by the insurance company, with the balance billed to the patient. This strategy should be followed by the development of testing guidelines for both HMOs and major insurance companies. Patient consent should be obtained before contacting insurance companies or other insurers. Test results should not be given to the insurance companies unless written authorization is received from the patient to release the results. Billing records and laboratory records are maintained separately.
The decision to introduce a new genetic test should involve not only consideration of the fiscal, academic, or other pertinent benefits, but also whether the test will help in new medical paradigms. Many different modeling tools are available to evaluate the value of a new genetic test, including modeling of relative risk vs economic benefit, or predictive outcomes testing with risk-benefit analysis for patient care.
Eric Green, MD, PhD, Acting Chief, National Human Genome Research Institute, Genome Technology Branch, National Institutes of Health, reviewed the state of the Human Genome Project. The Project is centered on understanding the structure and function of the DNA molecule. The Human Genome Project is a coordinated international effort to study the genomes of humans and several well-studied organisms. The project is in the 8th of a projected 15 years, with initial goals of mapping completely and then sequencing the entire genomes of a very well circumscribed group of organisms. The project occurs in two major phases, a mapping or organizational phase, and a reading phase of the DNA sequence—the actual sequencing phase.
the human genome
The haploid human genome has ~3 billion bases or 3000 megabases,
containing an estimated 50 000 to 100 000 genes. Human chromosomes
range in size from ~50 to ~260 megabases, with an average size of
~130 megabases, much larger than in the studied microorganisms. The
ultimate goal of the Human Genome Project is to identify the precise
gene sequences of these organisms. Over the past few years, the yeast
and the Escherichia coli genomes were completely deduced and
are now fully available as Internet databases. The nematode genome will
soon be completely sequenced. This will represent the first complete
sequence of a multicellular organism. Sometime around the year 2000,
the fruit fly genome will be sequenced.
mapping the genome
To sequence a genome, geneticists first organize the DNA, map it,
and then sequence it. There are three different types of maps. The
cytogenetic map represents the appearance of chromosomes when properly
stained and examined microscopically. The second type of genetic map
defines the location of polymorphic genetic markers along a chromosome.
These maps show the presence of differences in DNA sequences, which can
range from an identifiable phenotype, such as a disease, to, more
commonly, innocent sequence differences, known as polymorphisms. The
third type of map is the physical map.
During meiosis, there is an exchange of genetic information by a process known as meiotic recombination or "crossing over", which leads to the formation of two new hybrid chromosomes that were not represented in either of the parents. Genetic mapping simply involves the process of measuring the frequency of such cross-over events. It involves monitoring the inheritance of specific genetic markers from one generation to the next. The Human Genome Project involves making high-resolution maps. Polymorphic stretches of DNA that consist of reiterated di-, tri-, and tetranucleotide repeats are sprinkled throughout the genomes of humans and other organisms. These polymorphic stretches are small enough that the blocks of DNA can be amplified by PCR.
The most common type of dinucleotide repeat in the human genome is the CA repeat. There are so many repeats that when they get to be a certain size they actually differ between the different copies of a chromosome. The two alleles inherited from the mother and father will be different in size. Because they are of the order of a few hundred bases, they can be amplified using PCR.
The real revolution brought on by the Human Genome Project was the construction of genetic maps for all human chromosomes, providing a very high-resolution set of markers, most typically di-, tri-, and tetranucleotide repeat markers. A goal of the Human Genome Project was to make such high-resolution microsatellite PCR-typeable genetic maps. This has been accomplished for all of the human chromosomes. The initial goal was to have 1500 of these microsatellite markers evenly distributed across the human genome. This goal was actually accomplished in 1993. About 10 times as many markers have been identified as were originally envisioned as necessary.
Physical maps vary with the landmarks that are used to construct them. The most conventional type of map is the clone-based landmark map. Physical mapping by clone-based methods is quite analogous to assembling a jigsaw puzzle. In the case of DNA cloning and mapping, the pieces of the puzzle come from the starting DNA, which is broken by some method, but is then reassembled by recombinant DNA cloning. It is easier to reassemble larger fragments for molecular cloning and mapping than smaller fragments. In 1987, the yeast artificial chromosomes (YAC) was developed, which dramatically changed the way DNA is mapped. YACs enable the cloned DNA to propagate, not as an extra chromosomal element, as with bacterial cloning, but as a linear artificial chromosome. Because it is being propagated as an artificial chromosome, the DNA insert can actually be quite large, on the order of 100 000 to >1 million base pairs of DNA.
yeast artificial chromosomes
YACs rapidly became the fundamental material used to assemble, at
least, the first generation of clone-based maps. PCR had a major role
in detecting landmarks and led to the concept of the sequenced tag site
(STS). A STS is a DNA sequence of ~60 to 1000 base pairs. From such
DNA sequences, two oligonucleotide primers can be designed and used in
a PCR assay that will specifically detect this STS.
A given stretch of DNA can be thought of as consisting of a series of STSs. Each one is indicated by a unique geometric shape because it is associated with a unique PCR assay and specific PCR primers. Thus maps of human chromosomes could be assembled, using YAC clones as the pieces of DNA and STSs as the landmarks on which the maps would be based. A YAC library, which is all of the pieces of the puzzle, can be thought of as consisting of a series of clones of different compositions of STSs. When they exist in the library, it is not known which clones contain which STSs. It is possible to make PCR assays systematically to define different STSs. In this way it is possible to determine that certain clones contain multiple STSs.
Using this procedure, researchers can essentially determine the STS content of clones. The STSs are unique; therefore, if two clones contain the same STS, they must, by definition, overlap. From the STSs, it is possible to assemble what is known as a contig. A contig is an overlapping series of clones that together contain a contiguous segment of the starting DNA. A contig provides access to the DNA and some information about the order of these landmarks.
This general strategy of YAC-based STS content mapping rapidly became the dominant approach that was used by the Human Genome Project for constructing physical maps of human chromosomes. The goal, set several years ago, was to use strategies such as this to map and order STSs across the human genome. A STS occurs, on average, every 100 kb along all human chromosomes. The goal has been reached for some chromosomes, and it is very close for the other remaining chromosomes. This means 30 000 PCR assays had to be developed and optimized.
timetable for the human genome project
The construction of genetic maps and then physical maps
represented the early mapping phase of the Human Genome Project. The
initial goals for genetic mapping were reached in ~1994, and some
time around 1996 or slightly thereafter the goals for physical maps
were essentially reached. The year 1998 marked a very important
transition point because the second major phase of the Human Genome
Project, the sequencing phase, began.
Initially, the plan was to complete the human genome sequence by 2005. But this date has now been moved up to 2003 to correlate with the 50th anniversary of the discovery of the Watson-Crick model of the double helix. Sequencing the human genome will be the dominant effort in the Human Genome Project over the next 3 or 4 years. Three billion bases must be sequenced. The gold standard for DNA sequencing remains the classic technique of dideoxy chain-termination sequencing. This approach does not work for the kind of industrial process that is needed to sequence the human genome. A more automated fluorescent-based technology is now used. Instead of being tagged with radioisotopes, the DNA is tagged with four fluorescent dyes: one that corresponds to A, one that corresponds to C, one to G, and one to T. Instruments have been designed with lasers that are able to detect these different colored dyes as they are being electrophoresed on an acrylamide gel. Because of the four different fluorescent dyes, all four sequencing reactions (A, C, G, and T) can be placed in one lane of a gel and the sequence is readily apparent.
A new technology, the bacterial artificial chromosome (BAC), will supplement YACs. BACs are smaller than YACs, but bigger than cosmids and conventional plasmids. With BAC contigs instead of YAC contigs, each BAC is individually subjected to "shotgun sequencing", which is now the predominant method for sequencing the human genome. Shotgun sequencing is relatively straightforward. It consists of two major phases: a random shotgun phase and a directed finishing phase. In the random shotgun phase, the DNA of each individual BAC is purified and then broken at random into ~2-kb fragments. Random subclones are then sequenced. This is repeated many times. Every nucleotide on a BAC is statistically sampled about six to eight times. A computer lines up the sequences. When two sequences are the same, it overlays them and starts to assemble a sequence contig similar to a clone contig. In this way, virtually all of the sequence is assembled. The second phase is known as directed finishing, which is a polishing stage. Small bits of missing sequence are retrieved to complete the contig.
With the Human Genome Project, the error rate is no greater than one error every 10 000 bases. All the data produced in the Human Genome Project are put onto the World Wide Web every night. The sequencing centers themselves have no more advanced access to their own data than anyone else. Eight to 10 centers around the world are being set up to build large production teams for shotgun sequencing. Each group probably already has between 10 and 100 ABI-type instruments or similar sequencing instruments, and over the next year will scale up even more. The sequencing phase of the Human Genome Project has just begun. Yet, as of October 1998, only ~6.5% of the human genome had been finished. Another 10–15% is pre-finished, i.e., a large portion of the sequence data is available, but it is not quite polished and perfectly finished.
gene and disease associations
Positional cloning is important in developing associations of
diseases with particular genes. The process usually starts with the
study of an individual family or group of families in which a genetic
disease is clearly being passed from one generation to the next, and
then analyzing that family with many genetic markers. The objective is
to identify a genetic marker that co-inherits with the disease and to
demonstrate a strict correlation between the inheritance of the disease
and the inheritance of a genetic marker. Usually, when this is done
successfully, it is possible to limit the region of the genome that
contains that gene to something relatively small, especially with
improved genetic maps, that can then be analyzed by cloning the DNA,
identifying genes, and eventually identifying mutations.
The Human Genome Project has helped this process by providing thorough genetic maps; as a result, the likelihood of identifying a correlation between even a rare genetic disease and a specific genetic marker has been markedly enhanced. With an overlapping set of clones, identification of genes is increasingly straightforward. With the improved sequencing that is now being done, most of the sequence information about every region of the genome will be readily available. Mutation detection technology is improving steadily, enabling these specific mutations to be associated with specific diseases.
There has been a remarkable correlation between the products of the Human Genome Project and the identification, by the human genetics and molecular genetics community, of an increased number of disease genes by positional cloning strategies. The next challenge is positional cloning of relatively simple genetic bonds. This is much more profound from the point of view of clinically relevant disorders, i.e., polygenic diseases. These diseases, e.g., diabetes, cardiovascular disease, and cancer, are much more common. The involvement of multiple genes has subtle influences on the morbidity and mortality of patients.
Richard Press, MD, PhD, Director of Molecular Pathology, Oregon Health Sciences University, discussed the points to consider when introducing a new test. The decision should be based on several criteria, one of which is that the disease condition should be common. It could be less common if a specific local need exists. The test should improve patient care and also yield a positive financial return. Molecular diagnostic tests for infectious diseases fulfill these criteria. Molecular diagnostic tests for cancer have been slow to reach clinical fruition, primarily because they have not been shown to improve patient care. Exceptions are the tests for hematopoietic neoplasms, gene rearrangement tests, and tests for specific chromosomal translocation break points, which have been immensely useful in routine patient care.
acceptance of genetic tests
Some tests for classical inherited diseases—the Mendelian single
gene diseases—have not gained clinical acceptance because the diseases
typically are untreatable. Likewise, tests for neurogenetic disorders
have not been readily accepted because few of these diseases are
treatable at present. Predispositional diseases, typified by vascular
syndromes, meet the ideal criteria because they are common and
generally treatable. There are many of these diseases with data that
indicate that these tests can improve patient care. However, molecular
diagnostic tests generally need to have improved analytical
performance, increased sensitivity and specificity, and positive and
negative predictive values. They have the potential to replace more
poorly performing tests and, therefore, to improve the ultimate
economic picture, in direct test costs, in long-term management of
disease, and ultimately in public health costs. Clearly, molecular
diagnostic tests need to demonstrate better medical outcomes.
Molecular tests are moving in the direction of predispositional disease testing. Good examples of molecular tests that improve clinical performance are tests for microbiology or virology. There are many examples in the virology arena, e.g., cytomegalovirus infection and transplant patients. Molecular HIV viral load testing has also led to better medical outcomes. There are numerous examples of how these tests are useful and ultimately save healthcare dollars. Screening for hemochromatosis has been shown to be cost-effective to society, primarily by preventing expensive chronic disease.
Mark Sobel, MD, PhD, National Cancer Institute, National Institutes of Health, discussed ethical issues arising from molecular tests. He stated that the public consciousness toward genetic testing has been raised, with much of the public's focus on the abuse, or the perception of abuse, of molecular testing and the need to protect the public.
historical background of ethical issues
In the mid-1970s, at the behest of Congress, a group was convened
to discuss the implications of biomedical research. Before World War
II, there were no Federal regulations to protect the public from abuses
of medical research. The Nuremberg trials highlighted the abuses
perpetrated by the Nazis and led to the Nuremberg code and the idea of
informed consent for procedures. This was followed by the Declarations
of Helsinki.
After World War II, the US government appointed a National Commission, which developed a report entitled "Ethical Principles and Guidelines for the Protection of Human Subjects of Research". This report established three ethical principles related to research and clinical diagnostics. These are respect for persons (otherwise called personal autonomy), beneficence, and justice. The first principle, respect for persons, means that individuals have the right to decide what should be done with their bodies and, by extension, what may be done with specimens removed from their bodies. Beneficence means that if research is to be done, it must be for a good purpose. Justice, the third principle, implies that research should not be done on underprivileged people, prisoners, or others who would not necessarily benefit from the outcome of the research.
institutional review boards
One of the recommendations of the government report was that
institutional review boards (IRBs) be formed to judge and approve
research protocols that involve human subjects. In essence, the IRBs
are managed or controlled by the Office for Protection from Research
Risks (OPRR) of the Federal government. The relevant Federal Code of
Regulations (Protection of Human Subjects, Title 45, Part 46),
abbreviated as 45CFR46, was published in June 1991.
It is sometimes difficult to differentiate between tests for research and clinical care. If an approved test is ordered by a clinician for the clinical care of a patient and is to be performed in a CLIA-certified laboratory, it can be construed as a diagnostic test. However, if the test is not done in such a laboratory, the use of information from the test for the clinical management of a patient is prohibited. If a test is reimbursed by a third party, it is generally acceptable to use it diagnostically.
It is widely believed that it is appropriate to use previously analyzed specimens as positive or negative controls or to standardize equipment. However, some questions have now been raised as to whether it is appropriate to do this without the knowledge and permission of the individuals who gave the specimens. There is increasing sensitivity to this issue, and it will be reconsidered when new informed consent guidelines are developed.
ethics and identification
There are four categories of specimens for which ethical issues
arise. The first involves the anonymous specimen, in which the name of
the individual who provides a specimen is never associated with the
sample. The specimen receives only a random code. A more common
situation arises when an aliquot of a blood, urine, or tissue specimen
that has been collected for clinical analysis, and therefore has been
coded with an individual's name and medical record number, is taken
for testing as part of a particular study. Initially, demographic
information might be associated with the specimen, but then the
identification of the individual is stripped irretrievably from the
sample. A new, random alphanumeric code is provided, and the key that
links the new alphanumeric random code to the original identification
is completely and totally destroyed. This is called an anonymized
specimen.
The third—and most controversial—category, is the "linked specimen". This category arises when someone has a key to the code, which means that it is theoretically possible for the code to be broken and the identity of a sample to be determined. Many ethicists, and OPRR, believe that this situation is no different from an identified sample because the specimen can be traced back to its donor. Essentially, this interpretation expands the number of tests that are subject to regulations.
definition of a human subject
OPRR is concerned with the definition of a human subject. Its
first concern is whether an intervention or an interaction with a
living person occurs because of this research. Under this
interpretation, specimens from autopsies or archival material from
deceased patients are not subject to these regulations because a
non-living person is not a human subject. Many ethical and regulatory
groups are perturbed by this exemption because they believe that
genetic information affects not only the person from whom a sample was
obtained, but also the individual's relatives, either current or
future. They argue that a living subject exists in the form of a family
member. There has been no immediate move to change the definition, but
it is being discussed actively. If a patient is alive, a human subject
is involved, and a laboratory must follow the OPRR regulations.
If samples are anonymized, the regulations do not apply. However, if there is a key to the code so that someone could potentially correlate the specimen with a particular individual, a human subject is involved according to the current interpretation of the regulations. Some exemptions are provided for in the regulations, such as use of retrospective information. A second concern is associated with specimens in the public domain, as with existing cell lines in various national repositories. OPRR's opinion is that because these are publicly available specimens, the testing is permissible.
The ultimate question is: "Will the information be recorded in such a way that it cannot be linked to the subject?" The level of sensitivity needs to be increased when these issues are considered. Exemptions are possible when the samples are anonymized or if they are from a subject who is no longer living and the identities will not be published. Although these regulations apply only to Federally funded research, 99% of the medical institutions in the US have voluntarily set up a plan with the OPRR. Within Congress, there is a movement to broaden the impact of the Federal regulations to make IRB approval a requirement for any type of study.
essentials for human studies
The first criterion for any study is that it poses, at most, a
minimal risk to the subject. The question of whether the risk in any
genetic test is minimal is an issue that is now being widely debated.
IRBs are now being asked to assess whether the personal rights—the
personal autonomy of the subjects—are being considered and protected.
Another issue is impracticability. This arises when a new study requires informed consent from anyone who gave a sample that is included in the study. This could substantially increase the cost of a study. However, cost is generally not an acceptable reason for impracticability. An acceptable reason might be that the time involved in getting such permission for samples that were collected from different places over varying periods of time would delay the study so long that there would be a loss of benefit to the public. Some IRBs are very strict and others are more liberal in their interpretation of impracticality. Another issue that has arisen but rarely affects the clinical diagnostician is the requirement to inform the subjects after completion of a study. This issue is most applicable to behavioral research, where the purpose of the study would be obviated if subjects knew in advance exactly what was going to be done.
The ethical issues of greatest concern are privacy, confidentiality, and security. Privacy is respect for personal autonomy and freedom from uninvited intrusions. Although this interpretation was originally defined for active, therapeutic regimens, it also applies to tissue taken from a human because an intrusion occurs on a part of what was once his or her body. Confidentiality means that information will not only be held in confidence, but it also will not be disclosed without the consent of the person involved in that study. Concern with confidentiality has been heightened with the widespread availability of computerization and the Internet. The third concept is of security. An extension of this is compartmentalization, by which patient identity, clinical information, and research information are maintained separately.
definition of genetic tests
The Task Force on Genetic Testing describes a genetic test as the
analysis of human DNA, RNA, chromosomes, proteins, or metabolites to
detect heritable disease-related genotypes, mutations, phenotypes, or
karyotypes for clinical purposes. Such purposes include predicting risk
of disease, identifying carriers, establishing prenatal and clinical
diagnosis or prognosis, monitoring, and screening, both prenatally and
in newborns; however, the definition excludes tests conducted purely
for research with the caveat that research results are kept separate
from the medical record.
Senator Pete Domenici (Republican, New Mexico) has proposed to Congress a bill that he feels would protect genetic privacy. The proposed bill includes a broader definition of genetic information as information gathered from a human DNA sample relating to molecular genotype, information from mutation analysis, or information about the nucleotide sequence of a gene. In his proposed bill, Senator Domenici defines a DNA sample as a human tissue sample from which DNA is to be extracted or DNA that is actually extracted from such a sample, not including a tissue sample that is taken as a biopsy or an autopsy unless it is for a DNA test, in which case the exemption no longer applies.
Several regulatory panels feel that the blanket or general consent signed by an individual on admission to a hospital is really assent rather than consent. This is leading to so-called "layered" consent, where separate consents are given for different procedures. There are many unanswered questions related to the responsibility of the laboratory and the clinical care worker to the patient. Should the patient be told the test result? What is the responsibility to the family for tests that are sensitive or that imply inherited conditions or predispositions? If the studies are for an inheritable disorder, should a genetic counselor be involved in explaining those results? Does the test fall into the category of a research test or a clinical test? Is it appropriate to do BRCA1 testing? The current guidelines in the Human Genome Institute are that the test should be done in a research protocol environment because the implications of a negative result are not fully understood. The appropriateness of making clinical care decisions based on positive test results is not completely clear at present.
future considerations
From the ethical standpoint, the future will include more
restrictions on research because of new guidelines that will strengthen
the need to obtain informed consent and IRB approval for many
activities that currently are performed without such approval.
Donald J. Pochopien, PhD, JD, from McAndrews, Held & Malloy, Ltd., reviewed issues related to laboratory liability. The first concern is negligence. Lawyers look for individuals who have breached their duties, which in turn provides the potential for negligence. Duties can be as simple as exercising typical care in a laboratory setting. Lawyers look next for causation of injury to someone, either by the release of information or through acts outside the realm of the duty that cause an injury. Duty can arise through a State statute or through a moral obligation. It can arise by custom in an industry, e.g., a clinical laboratory, or it can be set by one's own standard of care in situations where written laboratory policies are not followed and injury results.
the concept of duty
The medical industry has duties. A laboratory can be sued on the
basis of something that it did that should not have been done, or
something the laboratory did not do when it should have. Duty may arise
from a person's rights. When individuals release or waive a particular
right, it does not follow that all of their rights have been
waived. Rights arise by statute, by moral principles, or by the
Constitution. The right to privacy is one of an individual's
constitutional rights, although it does not appear in the Constitution.
The public disclosure of private facts about an individual without his
or her consent is a violation of that individual's rights. Privacy
rights impact the laboratory. There are specific AIDS rights and
genetic testing rights. To protect itself, a laboratory should develop
and implement written policies.
Those in charge of laboratories should know about State statutes and develop appropriate pertinent policies. Such statutes generally are a state-by-state issue, although society will benefit when privacy with regard to the release of genetic information becomes a Federal issue. The Federal government currently can act on the release of genetic information under clauses that affect interstate commerce. The transportation of samples between states is interstate commerce.
patenting of genes
The patenting of genes is an important issue. There are several
types of patents. Inventions that are patentable include any new and
useful process, machine, manufacture, or composition of matter, and any
new and useful improvement thereof. An improvement to some thing that
is patented or exists is patentable provided it meets all of the other
statutory requirements. The essential criteria are novelty, usefulness,
and non-obviousness.
If a gene was known previously, but it is newly sequenced, it meets the criterion of newness. If the same gene is identified in several species, it is hardly non-obvious. For patent law, the definition of a gene covers the structure having all its naturally occurring sequences of promoters, enhancers, the coding region, and any other regulatory elements. Thus, it includes the entire coding sequence and whatever else makes it work. The claim is basically for a DNA sequence.
Demonstration of usefulness requires some type of qualitative or quantitative assay as a minimal requirement. With regard to non-obviousness, the nucleotide sequence would be non-obvious in the absence of any analogous sequences elsewhere. An example of a claim for a gene might speak of a gene comprising a certain sequence. That sequence would then be submitted on disks to the patent office, which would search all its databanks for sequences that are similar or identical. For a gene to be patentable, it must be isolated and purified. The sequence of nucleotides may be patentable as a gene if all the enhancers, promoters, and regulatory elements are identified. The most common claim is for a cDNA, which encodes a particular protein. It is also possible to claim antibodies to a protein.
obtaining a patent and challenging a patent
When an individual attempts to prosecute a patent before the
patent office to obtain the patent, the burden of proof requires only a
51% certainty, the so-called preponderance of evidence. However, if
someone wants to attack the patent, the patent holder must prove by
clear and convincing evidence its validity with an ~75% certainty.
If an individual tries to invalidate the patent, he or she must show
that the subject of the patent was not new or was not obvious as of the
date of the patent. It is possible to patent new uses of an old
product. New uses of genes will be patented in time.
Under statute, a patent is considered infringed upon when someone without authority makes, uses, offers to sell, or sells any patented invention within the US, or imports into the US any patented invention during the term of the patent. The remedy for infringement is a civil action. This requires the individual holding a patent to take the affirmative step of enforcing the patent. A patent right involves the exclusive right to make, use, and sell the invention, and to prevent others from doing this. For the patent holder to enforce this right, he or she must go to Federal court because patents are conferred by the Federal government.
All patents are presumed to be valid. Each claim is viewed independently of the others, and the burden of establishing invalidity is on the party who challenges the patent. The defense to the challenge of a patent is that the defendant is not infringing on it or is not doing what is charged, or that the patent is invalid. Other defenses might include that the patent is not enabling or that the patent holders failed to disclose the best mode of practicing the invention at the time.
court responses to infringements of patents
A patent holder has several remedies if a patent is challenged. A
court may grant an injunction, but this is not mandatory. The courts
may look, especially with regard to diagnostic testing, at whether
there may be an underlying policy consideration, although this has not
really been tested. A court will look at what happens with regard to an
individual patient's rights. The patient's rights probably will not
be hurt, but if a person holds a patent and is able to supply all the
needs of the industry, it is very likely that a court will shut down
the operations of an individual who infringes on a patent.
Upon finding for the claimant, a court is required to award the claimant damages that are adequate to compensate for the infringement. The minimum is a reasonable royalty, although it could be much more. The court makes an award based on its assessment of how willful the infringement was, how much profit was made from it, and whether the test was a good assay. The court considers whether the test was performed properly. If not, and the assay was given a bad name, or if the price charged had eroded the market, the court may decide that the reputation of the patent has been ruined. Damages are set as if the market had not been eroded during the patent term. Courts can also award, exceptionally, attorney's fees. If an individual has deliberately continued to infringe on the patent after receipt of a letter from a lawyer for the patent holder, the damages are likely to be greater. In such a situation, the damages are trebled and attorney's fees are awarded.
There is an exception with regard to infringement when a medical practitioner performs a medical activity that constitutes infringement. The provisions applicable to damages, attorney's fees, and injunctions are not applied against the medical practitioner or against his or her related healthcare entity with respect to such medical activity. Here, medical activity is defined as the performance of a medical or surgical procedure on the body. Basically, it is not possible to obtain patents on surgical procedures. Biotechnology companies also sought a similar exception with regard to practicing on the body because they were looking at an inveigled gene therapy, which basically is a medication. Thus, if biotechnology companies were to treat the body by inveigled gene therapy, they certainly would not want to have patent protection for it. Accordingly, such treatments do not fall under the exception.
Summary
The field of genetic testing has evolved rapidly, with a host of
challenges that are new. The American Association for Clinical
Chemistry's Eighth Clinical Chemistry Forum addressed some of the
current issues in genetic testing. The exciting meeting highlighted the
dynamic state of genetic testing, with the unraveling of the human
genome assuredly adding a broad spectrum of new diagnostic tests.
However, the issue of how to manage the often conflicting demands of
confidentiality of patient information on the one hand and a patient's
need to know genetic findings on the other has not been satisfactorily
resolved. Genetic test results have a potential impact beyond those
involved in managing a single patient. Genetic tests detect
predisposition to disease, which may or may not lead to actual disease,
in effect making an entire family the patient. The tests may detect
presymptomatic diseases, raising new issues of patient management.
Congress is making efforts to address the confidentiality of genetic
information. In some proposed bills, only genetic test results are
considered, whereas in others, all medical information is considered.
Several states are ahead of Congress in addressing the issues, and
unless Congress moves rapidly and acts decisively, the result could be
a plethora of different laws with different requirements. A further
challenge with genetic tests is to ensure proper reimbursement. Such
tests typically are new and initially will not be seen by third-party
payers as having have demonstrated clinical value; however, many of
these tests will be needed clinically.
|
Footnotes |
|---|
|
References |
|---|
|
TopIntroductionTextReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
