NHGRI logo

Promoting Safe and Effective Genetic Testing in the United States

Chapter 3

Ensuring the Quality of Laboratories Performing Genetic Tests

Over 500 clinical laboratories in the United States perform chromosomal, biochemical, and/or DNA-based tests for genetic diseases (see Appendix 3). These laboratories must comply with regulations under the Clinical Laboratory Improvement Amendments of 1988 (CLIA), which include biennial inspection, some proficiency testing, and requirements of the specialty in which the laboratory is certified. Although clinical cytogenetics is a specialty under CLIA, there is no broader genetics specialty and, consequently, no special requirements for laboratories performing DNA-based and other types of genetic tests. No proficiency testing programs in genetics or cytogenetics are required under CLIA. New York State requires any laboratory performing tests on New York residents (even if those laboratories are outside of New York) to participate in its quality assurance programs in DNA-based and biochemical genetics. These programs involve onsite inspection but not formal proficiency testing.a A number of organizations have voluntary programs for quality control of genetic tests; they are described later in this chapter. In a survey conducted for the Task Force in early 1995, 11 percent of biotechnology companies that provide genetic tests and 16 percent of nonprofit (primarily university-based) molecular (DNA) labs reported that they neither participated in a formal proficiency testing program nor shared samples informally for quality control (see Appendix 3). According to the survey, about 15 percent of laboratories performing clinical DNA-based tests were not registered under CLIA (see Chapter 5 for further discussion of this problem).

Although the vast majority of laboratories providing genetic tests perform adequately, the Task Force has two concerns. First, even though most laboratories voluntarily participate in quality programs addressed specifically to genetic tests, they are not required to do so. Consequently, providers and consumers have no assurance that every laboratory performs adequately. Occasionally errors are made. Second, that current requirements under CLIA, with which clinical laboratories must comply, are inadequate to ensure the overall quality of genetic testing because they are not specifically designed for genetic tests and because they do not give sufficient emphasis to pre- and post-analytic phases of testing. Voluntary programs are also lacking on this second point.

In this chapter, we first describe the principles that laboratories should follow in adding new genetic tests to their repertoire. We then consider CLIA's framework for laboratory quality and, in view of gaps in CLIA in the area of genetics, other programs for assessing and improving test performance. We then indicate our concerns about ensuring the quality of the pre- and post-analytic phases of predictive genetic testing. We conclude with brief consideration of the need for a central repository of materials for genetic testing, direct marketing, and international standardization of quality assurance methods.

PRINCIPLES FOR LABORATORIES ADOPTING NEW GENETIC TESTS

No clinical laboratory should offer a genetic test whose clinical validity has not been established, unless it is collecting data on clinical validity under either an IRB-approved protocol or conditional premarket approval agreement with FDA (one of the options presented in Chapter 2). The service laboratory should justify and document the basis of decisions to put new tests into service. In accord with the recommendations in Chapter 2, a clinical laboratory that develops a genetic test would have to submit its data on analytical and clinical validity to external review before offering the test for clinical practice.b If the test has been developed elsewhere, clinical laboratories should carefully review evidence for test validity. If external review by professional societies has led to the publication of indications and guidelines for use, laboratories should adhere to them. Regardless of where the test to be adopted was developed, clinical laboratory directors are responsible for ensuring the analytic validity of each genetic test their laboratory intends to offer before they make the test available for use in clinical practice (outside of an investigative protocol). (Methods for assessing analytical validity are summarized in Chapter 2.)

Before routinely offering genetic tests that have been clinically validated, a laboratory must conduct a pilot phase in which it verifies that all steps in the testing process are operating appropriately. In establishing the pilot phase, the laboratory should define endpoints, such as number of tests to be performed,c and the procedures to be used to review the findings, including the organizational body that will review them. If the outcome of this review reveals that the laboratory is not as competent as other laboratories in performing the test, or the test does not detect as many people with the genetic alteration as anticipated, the laboratory should not proceed to report patient-specific results without attempting to rectify the problems. If demand is not sufficiently high to be able to maintain a high level of quality, the laboratory should institute special procedures to ensure quality.

During the pilot phase, confidence in the analytic validity of the test can be gained by splitting specimens with another laboratory.d This phase can be used to detect and correct problems in test requisitions, specimen transport, data analysis and transcription, reporting of results, and user satisfaction. It can also be used to establish that laboratory staff are capable of deciding whether each requisition for the test meets established criteria, and the staff is capable of performing the tests and interpreting the results correctly. The pilot phase should employ laboratory practices as similar as possible to those planned when the test becomes routinely available.

CLINICAL LABORATORY IMPROVEMENT AMENDMENTS OF 1988 (CLIA)

A statutory framework for ensuring laboratory quality was laid down by Congress in the Clinical Laboratory Improvement Act of 1967 and greatly expanded in the Clinical Laboratory Improvement Amendments of 1988 (CLIA). Any laboratory performing "examination of materials derived from the human body for the purpose of providing information for the diagnosis, prevention, or treatment of any disease or impairment of, or the assessment of the health of, human beings" must comply with CLIA.2 Implementation of CLIA is the responsibility of the Health Care Financing Administration (HCFA) and the Centers for Disease Control and Prevention (CDC). Under CLIA, these Federal agencies have developed requirements for laboratory quality assurance and control, personnel, patient test-management and, if a proficiency program is not available, interlaboratory comparison of assays. The stringency of these requirements depends on the complexity level and specialty to which tests are assigned. Despite these basic provisions, the Task Force has serious concerns as to whether CLIA adequately assures the quality of genetic tests in clinical use.

Complexity Ratings

CDC assigns a complexity level to a test according to predetermined criteria. Simple tests are categorized "waived." The remainder are assigned ratings of either "moderate" or "high" complexity. Laboratories performing high complexity tests have more stringent personnel and quality control requirements.

Over 17,000 clinical laboratory tests have been assigned a complexity level.e Any test for which CDC has not determined test complexity is considered to be high complexity by default.f Any home-brew method, or change in procedure that can affect laboratory performance (sensitivity, specificity, accuracy, precision) falls under the high complexity category until it is rated differently by CDC. Under the rating scheme, a genetic test that can be used predictively might receive a rating of moderate complexity despite the importance of ensuring that the provider and the patient understand the uncertainty of the prediction and the implications for decision-making. Both the creatine phosphokinase test, which can be used as a screening test for Duchenne muscular dystrophy, and alpha-fetoprotein (AFP), which is used as a predictive prenatal test for neural tube defects and Down syndrome, are rated as moderate complexity. Despite multiple uses, a test method gets only one rating based on the seven criteria (see box entitled Complexity Determinations under CLIA) that reflect the complexity of performing the test. All cytogenetic tests are rated high complexity, which seems appropriate. The Task Force recommends that tests that can be used for purposes of predicting future disease be given a rating of high complexity.

CLIA Specialties

Laboratories performing tests of moderate or high complexity must also conform to the

requirements of the specialties to which tests are assigned. Laboratories can perform tests only in specialties for which they are certified. Although there is a cytogenetics specialty, there is no genetics specialty.g The specialty categories under CLIA are based on traditional laboratory practice; each specialty tends to involve somewhat similar technologies, although this is not the case in all instances. Each analyte is assigned to only one specialty.

Establishing a specialty for genetics presents a number of problems. For example, specialty designations are administratively linked to Medicare payment specialty designations, and any changes in specialty designations must take this into account. In addition, genetic tests use a wide variety of technologies, some of which are used in other (non-genetic) types of tests. For instance, DNA is the analyte in some tests for predicting genetic susceptibility and also in some tests for infectious agents. Sometimes the same test is used for purposes of genetic prediction (in healthy individuals), genetic diagnosis (in individuals with symptoms), and non-genetic diagnosis or prognosis. For instance, the creatine phosphokinase assay can be used to screen for carriers of muscular dystrophy and affected infants, but it is also used in the diagnosis of myocardial infarction. Despite these problems a genetics specialty is needed.

The Task Force welcomes the intention of CDC to create a genetics subcommittee of the Clinical Laboratory Improvement Advisory Committee (CLIAC), which advises on policies under CLIA. The Task Force urges this subcommittee to consider the creation of a specialty of genetics that would encompass all predictive genetic tests that satisfy criteria for stringent scrutiny. If a specialty of genetics is not feasible, the subcommittee should consider a specialty or subspecialty of molecular genetics for DNA/RNA-based tests. In the latter case, it must then address how to ensure the quality of laboratories performing nonDNA/RNA genetic tests. Although DNA-based tests will comprise the largest proportion of predictive tests, for disorders with great allelic diversity, gene product tests might have greater sensitivity than DNA-based tests, at least until technologies that can detect a large proportion of all possible mutations become applicable to clinical testing. The subcommittee should also consider assigning tests that have widely different uses to more than one specialty. This will facilitate assigning separate billing and reimbursement codes for each use of a genetic test when the uses are vastly different.

LABORATORY PERSONNEL

Personnel requirements under CLIA, particularly at the level of laboratory director, depend on the specialty and complexity categories to which tests or analytes are assigned. Without a genetics specialty, genetic tests fall into other specialties for which requiring special training in genetics would be superfluous for many of the other tests in those specialties.

Laboratory Director

Under CLIA, a laboratory director must possess a current license as a laboratory director in the state in which the laboratory is located and be either (1) a pathologist; (2) a physician licensed to practice medicine or osteopathy; or (3) a board-certified doctoral scientist (Ph.D.).h The Task Force recommends that for laboratories performing high complexity tests in the proposed specialty of molecular genetics, as well as in biochemical genetics and cytogenetics, personnel serving as directors or technical supervisors must have formal training in human and medical genetics, as documented by holding certification from an organization that assesses knowledge of human and medical genetics as part of its certification process, such as the American Board of Medical Genetics.

Testing Personnel

CLIA imposes minimal academic qualification requirements for testing personnel.i This is reasonable in the area of genetics because most current medical technology training programs include little, if any, exposure to genetics or molecular biology. Several formal training programs for cytogenetics technical staff are available, but there are very few certificate- or diploma-track genetics training programs for technicians or technologists in the U.S. Consequently, most technicians in molecular genetic testing laboratories are trained on the job. Broad backgrounds in genetics are unlikely, as is a familiarity with specialized methodologies involved in molecular genetics testing. Training programs for laboratory technicians/technologists need more human and medical genetics content than are currently available in the U.S.

The College of American Pathologists (CAP) specifies a B.S. degree or equivalent in the biological sciences for technologists engaged in genetic testing. Neither CAP nor the American College of Medical Genetics (ACMG) requires personnel to be licensed medical technologists but some States require it. Many States offer a special licensure for cytogenetics technicians; this is a desirable attribute where available. California is presently trying to develop a similar licensing mechanism for molecular genetics technicians. The National Certification Agency is working with the Association of Genetic Technologists to develop certification in genetics. Licensing of technologists performing genetic tests can then be linked to certification. Most clinical molecular genetics laboratories employ technicians with a molecular biology research background.

Biochemical genetic techniques resemble those used in other, more routine, areas of clinical chemistry. For this area, therefore, Federal, State, and professional requirements for clinical chemistry laboratory personnel are sufficient, as long as the technologists work under a director who is a certified biochemical geneticist.

MONITORING LABORATORY PERFORMANCE

Because laboratories provide services to providers and patients in many States it is clearly more desirable to have a rigorous Federal standard for certification or accreditation than fifty different State standards. Moreover, interstate genetic testing is unavoidable when only one or a few laboratories in the country provide tests. A national accreditation program for laboratories performing genetic tests, which includes proficiency testing and on-site inspection, is needed to promote standardization across the country. Such an accreditation program can occur more readily if a genetics specialty were established under CLIA. Until such time as a genetics specialty is established under CLIA, laboratories performing DNA/RNA-based tests for predictive purposes should choose to voluntarily participate in the CAP molecular pathology program including the CAP/ACMG molecular genetics proficiency testing program. Laboratories performing genetic tests on analytes not covered in the CAP/ACMG program, such as Tay-Sachs carrier screening and newborn screening, should participate in the available proficiency programs.

Proficiency Testing

Proficiency testing (PT) is mandated by CLIA to externally evaluate the quality of a laboratory's performance. For PT, a laboratory is provided with specimens whose composition of an analyte is known to the supplier but not to the recipient laboratories. They are expected to analyze the specimen the same way they would a patient's specimen. Each laboratory performing moderate or high complexity tests is required to enroll in an approved PT program for all specialties/subspecialties, analytes, or tests for which the laboratory is certified and for which a PT program has been recognized by HCFA. Any laboratory that fails a proficiency test must take corrective action.j HCFA takes an educational approach to PT and works with the laboratories that have problems to help improve performance. Sanctions can be applied to those laboratories repeatedly unable to perform satisfactorily. These include suspension of the CLIA certificate to perform that test or specialty. If its certificate is suspended, the laboratory is not eligible for Medicare/Medicaid reimbursement, since such reimbursement requires a CLIA license with no restrictions.

So far, the Department of Health and Human Services has approved 19 PT programs under CLIA. It has not approved proficiency testing programs for genetic tests because such tests do not measure regulated analytes for PT purposes as currently listed in the regulations. New York State and a few regions have cytogenetics PT programs. CAP and ACMG jointly administer PT in cytogenetics, fluorescent in situ hybridization, biochemical genetics, and molecular genetics. In collaboration with the Foundation for Blood Research (FBR), CAP has a PT program for prenatal screening of neural tube defects and Down syndrome. CDC has a PT program for newborn screening tests, including hemoglobinopathies. PT is also available for laboratories worldwide performing Tay-Sachs screening. Responding to a survey conducted for the Task Force in July 1997, CAP, FBR, CDC, and the International Tay-Sachs program reported that most laboratories known to them were participating in their respective programs.k

Although genetic tests do not appear on the list of regulated analytes for PT purposes under CLIA, laboratories must establish the accuracy and reliability of a test by methods of their own choosing. This can include participation in one of the voluntary PT programs. As the PT programs mentioned above are not approved by CLIA, no laboratory is obliged to use them and can establish accuracy and reliability by another method, although it must make the data available for onsite inspection under CLIA (see below). If they do participate and do not perform adequately, laboratories will usually improve performance. If, however, they continue to fail to meet PT criteria, they are not obliged to stop testing as participation is voluntary. A few laboratories participating in the PT programs recently surveyed do not always correctly analyze all PT specimens. According to the Tay-Sachs program, one or two per year do not improve and usually stop testing.

Information collected in conjunction with PT sometimes reveals outliers among laboratories. For instance, a survey conducted by the FBR/CAP prenatal screening PT program found a few laboratories that did not follow established criteria in accepting specimens.

Participation in well-established proficiency testing programs for genetic tests must be required under CLIA once a genetics specialty is established. When no relevant proficiency testing programs exist, laboratories must, whenever possible, participate in inter-laboratory comparison programs and help develop them if none exist in their particular area of testing.

Proficiency testing programs should be broadly based since the number of genetic disorders is very large and the analytical approaches to testing are numerous. It is unlikely that proficiency challenges will ever be constructed for every rare disease or every rare mutation in common diseases for which a given laboratory might test. Because of the similarity of techniques used in biochemical genetics, proficiency in these techniques applied to one or a few analytes is a reasonably good indicator of proficiency in other uses of the technique. CAP/ACMG is expanding the PT offering in molecular genetics to a greater number of disorders in order to get more complete demonstrations of proficiency.

Onsite Inspection

All CLIA-certified laboratories are routinely inspected on a two-year survey cyclel by one of three types of organizations: (1) HCFA regional offices and State agencies; (2) private non-profit organizations that have applied for and received deemed status because they provide reasonable assurance that the laboratories they accredit, which enables the laboratory to obtain a CLIA certificate, meet the conditions required by Federal law and regulation;m (3) State-exempt licensure programs. States that have programs that license laboratories and provide HCFA with reasonable assurance that their criteria are equivalent to or more stringent than those specified under CLIA can apply for exempt status. So far New York, Oregon, and Washington (state) have exempt status. California, Florida, and Georgia are under review (as of July 1997). Regardless of the organization under whose auspices inspections are conducted, the surveyors are laboratory professionals who are trained to determine compliance with CLIA regulations (or a program that is determined to be equal to or more stringent than CLIA). Even though genetics is not a specialty, surveyors are expected to examine the quality of genetic tests. This should include inspection of the records of how the laboratory performed on genetic PT programs in which it participated voluntarily. It is not clear, however, that all CLIA surveyors currently are sufficiently knowledgeable to assess the performance of molecular genetics laboratories.n

CAP has deemed status to conduct inspections in several specialties, but since genetics is not a specialty under CLIA, the CAP program does not have deemed status in genetics. In the CAP genetics program, laboratories who voluntarily (and for a fee) participate in the program are inspected. The surveyors use a checklist covering all aspects of quality assurance and quality control, from specimen accessioning to final sign-out. Compliance with some items on the checklist is optional; for others, compliance is mandatory.o Following inspection, the laboratory receives a written report and is expected to respond to CAP in writing regarding correction of any deficiencies in the mandatory categories. In areas in which it does not have deemed status, such as genetics, CAP has no authority to grant accreditation for CLIA purposes.

Making Laboratory Performance Assessments Public

HCFA annually publishes a list ("Laboratory Registry") that identifies all poor performance laboratories, the reason enforcement actions were taken and type of enforcement, and the name of the laboratory director. The Registry is available to the public upon request, and will soon be accessible on the Internet. Survey findings are also available through the Freedom of Information Act, once the laboratory has the opportunity to respond with its Plan of Action. CAP reports PT results for regulated analytes (i.e., those for which CLIA requires PT) to HCFA. It does not report PT results directly to the public because it maintains that PT alone is insufficient to demonstrate laboratory quality. CAP does make accreditation status available through its toll-free hotline (1-800-LAB-5678), and a CAP-published list of accredited laboratories.p As CAP is not deemed to accredit in areas of genetics, it does not make the results of its assessments of genetic test performance public.

The Task Force recommends that CAP/ACMG periodically publish, and make available to the public, a list of laboratories performing genetic tests satisfactorily under its voluntary program. Other PT programs should also publish the names of laboratories performing satisfactorily if they do not already do so. Until then, publication of results in voluntary proficiency and other quality assurance programs enable providers and consumers to select approved laboratories and also serve as an incentive for laboratories to participate in the CAP/ACMG quality assessment program. The information on laboratories performing satisfactorily should be readily accessible to consumers and providers.

Publishing the names of laboratories performing satisfactorily would advise users that labs not appearing on the list have either not submitted to external review or have not performed adequately. Directories of laboratories providing genetic tests (e.g.. HELIX - see Chapter 5) should also publish information on listed laboratories 19 satisfactory participation in PT and other quality control programs specific for genetic tests. The Association for Molecular Pathology publishes information on the quality of laboratories, and the National Organization for Rare Diseases and the Alliance of Genetic Support Groups make it publicly available. Managed care organizations and other third-party payers should limit reimbursement for genetic tests to the laboratories on published lists of those satisfactorily performing genetic tests. Implementation of this recommendation is especially important as more managed care organizations move to restrict access to laboratory services for their members to a single laboratory with whom each organization contracts. Such a laboratory might not have participated or performed satisfactorily in a quality control program.

A CENTRAL REPOSITORY OF CELL LINES AND DNA

Making cell lines or DNA containing disease-related mutations available to many laboratories would be useful in the validation of new tests, calibration, standardization, and quality control. To accomplish this, appropriate specimens from patients, carriers, and controls should be available through a centralized repository in order to facilitate their availability to aid in analytical validation, improving quality, and other needs. Resources such as the National Institute of General Medical Sciences' Human Genetic Mutant Cell Repository (housed at the Coriell Institute for Medical Research) and the American Type Culture Collection should be utilized. It should be impossible to trace samples in a repository to the individuals from whom they were obtained. The samples should not be used for any purpose from which a profit could be derived, such as the sale of unusual probes. A central repository of analytes for standardizing biochemical and other types of tests, including those used for screening, is also needed. Some mechanism for ensuring the composition and concentration of these standards, such as FDA review, is needed.

THE IMPORTANCE OF THE PRE- AND POST-ANALYTIC PHASES OF TESTING

In the pre-analytic phase, laboratories sometimes give information about the test to providers and consumers. Informed consent can be obtained, and data are requested from those to be tested. In the post-analytic phase, test results are given to the provider and patient, often with an interpretation. Genetic counseling services can be provided or arranged by laboratories, but are the responsibility of the referring provider.

Pre-analytic Phase

The Task Force is concerned about the quality of information made available to providers and consumers who are considering testing. Some materials have serious omissions that impair the ability of providers and consumers to make informed decisions about testing. In a comparison of four different brochures made available by organizations offering testing for genetic susceptibility to breast cancer, the Task Force found striking discrepancies. Physicians or consumers reading one brochure might, as a result, make a different decision than if they read another organization's brochure. It is the responsibility of health care providers, not the clinical laboratory, to provide information to the individual offered or considering testing, but material made available by laboratories is often used. The completeness and accuracy of this material is, therefore, extremely important.

Obtaining informed consent helps ensure that the person voluntarily agrees to testing and has some understanding of the reasons for testing. Informed consent is appropriate for predictive genetic tests, particularly those for which stringent scrutiny is needed. The Task Force is of the opinion that laboratories should obtain documentation of informed consent when appropriate and should not perform an analysis if documentation is lacking. The most rigorous documentation is for the laboratory to be sent a signed copy of the patient's consent. It is less rigorous to ask the ordering physician to check a box on the laboratory requisition indicating that consent has been obtained.

Because of the complexities of assessment and interpretation, requisitions for many genetic tests require more intake information than those for virtually any other clinical laboratory procedure. In addition to routine information, genetic test requests often must include the reason for requesting the test, any relevant clinical or laboratory information, the person's age and ethnicity, and notation of family history of the disorder in question (along with a full pedigree for tests involving linkage analysis). If information that is critical to the performance or the interpretation of the test cannot be obtained, or if the information that is provided suggests that the patient is not an appropriate candidate for testing, the physician must be contacted. There is consensus, for instance, that minor children should not be tested for adult-onset disease for which no diagnostic or therapeutic interventions are needed before adulthood (see Chapter 1). Yet some laboratories report testing children (see Chapter 4 and Appendix 3). Most authorities agree that healthy women without a family history of breast cancer should not be tested for inherited susceptibility mutations for breast cancer except under investigative protocols to gather data on the penetrance of these mutations, and that women with a family history of the disease should only be tested if an inherited susceptibility mutation is found in an affected relative.5-7 Consequently, laboratories must ascertain the presence of a family history before accepting a specimen. At least one laboratory is offering testing to Ashkenazi Jewish women without a family history.8 In general, laboratory personnel must be competent to recognize what information is needed and what the criteria are for accepting specimens. When in doubt, they must communicate with the ordering provider.

Post-analytic Phase

Increasingly, genetic tests will be requested by providers without much or any training in genetics. (Recommendations on ensuring provider competence appear in Chapter 4.) Accurate and comprehensible interpretation of genetic test results by the clinical laboratories is critical to ensure that the provider understands the implications and can explain them to the persons who were tested. Genetic test results must be written by the laboratory in a form that is understandable to the non-geneticist health care provider. The quality of laboratories' written interpretations of genetic test results should be included in the overall assessment of laboratories providing genetic tests.

Some laboratories also make genetic counselors available to discuss results with physicians. If testing of other relatives is an option, a potential conflict of interest arises as the counselor might want to promote additional business.

Ensuring the Quality of Pre- and Post-analytic Phases

One way of improving laboratory performance is to have more rigorous standards with which laboratories must comply. The Task Force is of the opinion that not enough emphasis is placed on the pre- and post-analytic phases in CAP's molecular pathology and special chemistry programs. The Task Force recommends that CAP and ACMG seek advice and input from consumer groups such as the Alliance of Genetic Support Groups, as well as from the National Society of Genetic Counselors (NSGC), on educational, psychological, and counseling issues in pre- and post-analytic components of genetic testing that are of direct concern to consumers.

Under CLIA, the rating system used to establish the complexity of tests does not give sufficient weight to these phases (see box entitled Complexity Determinations under CLIA). CDC should consider how the pre- and post-analytic phases of predictive genetic testing can be given greater weight in CLIA standards and regulations.

DIRECT MARKETING OF GENETIC TESTS TO THE PUBLIC

Many clinical laboratories advertise the availability of tests directly to the public (see Appendix 4). Great care must be taken that information on genetic tests presented directly to the public is accurate and includes risks and limitations, as well as benefits. The informational material should be sensitive to the knowledge level of the general public. In addition to describing the benefits and risks of the genetic test(s), including discrimination issues and the potential emotional impact on individuals and family members, the material should describe those for whom testing is appropriate (e.g., couples planning to have children for carrier tests, and individuals with a family history of a late-onset disorder for which genetic predispositions can be detected), and should emphasize that all genetic testing is voluntary, often requiring informed consent. Consumers should discuss testing options with a health care provider competent in genetics prior to having specimens collected for analysis.

The Task Force is concerned that no mechanism exists for the review of the accuracy of informational material on genetic tests made available either to providers or consumers, except for the labeling materials on kits that must be reviewed by FDA in premarket applications. As already noted, most genetic tests are marketed as services, not kits. Although complaints concerning inaccurate information can be made to FDA, the Federal Trade Commission, the Consumer Product Safety Commission, or the consumer protection divisions in the offices of most States 19 Attorneys General, harm could be done from exaggerated claims before complaints are filed or acted on. The external review of tests before they enter clinical use (see Chapter 2) should include examination of proposed informational material.

In accord with laws in most States, clinical laboratories in the U.S. require that specimens for the vast majority of tests come from a physician or are reported to a physician. A few laboratories accept specimens for predictive genetic testing directly from consumers without the intervention of their own physician. In such cases, a physician affiliated with the testing laboratory, who is a specialist but may be previously unknown to the patient, can order the test. As DNA can be isolated and amplified from cells in saliva or scraped from the buccal mucosa, it is possible for lay people to collect their own specimens. FDA has the authority to regulate this practice if the laboratory supplies or requires use of a specially designated collection device or container to send specimens from the person's home to the laboratory. The Task Force discourages advertising or marketing of predictive genetic tests to the public.

INTERNATIONAL HARMONIZATION

At present, no mechanism exists to create international standards of laboratory quality and proficiency for genetic tests. Current United States regulations require any foreign laboratories performing clinical laboratory tests on U.S. residents to hold a CLIA certificate even if their nation 19s laboratory standards are more stringent than those of CLIA. The Task Force recommends that efforts should be made to harmonize international laboratory standards to ensure the highest possible laboratory quality for genetic tests. A proposed European Union Directive on "In Vitro Diagnostic Medical Devices," with which FDA is cooperating, will harmonize the situation for assessing medical devices including genetic test kits and reagents. This Directive, however, does not extend to tests provided as services, similar to the situation in the U.S.

REFERENCES

  1. Kaback M, Lim-Steele J, Dabholkar D, Brown D, Levy N, Zeiger K: Tay-Sachs disease--carrier screening, prenatal diagnosis, and the molecular era. An international perspective, 1970 to 1993. The International TSD Data Collection Network. JAMA 1993;207:2307-2315.

  2. Bogdanich W: False negative. Medical labs, trusted as largely error-free, are far from infallible. Wall Street Journal Feb. 2,1987:1.

  3. Public Law 100-578: Clinical Laboratory Improvement Amendments of 1988. 1988;42 USC 263a.

  4. Palomaki GE, Knight GJ, McCarthy JE, Haddow JE, Donhowe JM: Maternal serum screening for Down syndrome in the United States: A 1995 survey. American Journal of Obstetrics and Gynecology 1997;176:1046-1051.

  5. Burke W, Kahn MJE, Garber JE, Collins FS: "First Do No Harm" applies to cancer susceptibility testing too. Cancer Journal from Scientific American 1996;2:250-252.

  6. Weber B: Breast cancer susceptibility genes: Current challenges and future promises. Annals of Internal Medicine 1996;124:1088-1090.

  7. Blue Cross and Blue Shield Association Technology Evaluation Center (TEC): Executive Summary of TEC Assessment on Genetic Testing for inherited BRCA1 or BRCA2 mutations. 1997.

  8. Schulman JD, Stern HJ: Genetic predisposition testing for breast cancer. Cancer Journal from Scientific American 1997;2:244-249.

Top of page

Last updated: April 01, 2006