CDC Releases Recommendations for Molecular Testing

Jacquelyn K. Beals, PhD

June 15, 2009

June 15, 2009 ( UPDATED June 19, 2009 ) — The Centers for Disease Control and Prevention (CDC) has released recommendations for best practices in genetic testing for heritable diseases and conditions. Published June 12 in the Morbidity and Mortality Weekly Report (MMWR), the recommendations are the first issued by the federal government regarding the accuracy and proper use of DNA-based genetic testing.

The CDC, together with the Centers for Medicare & Medicaid Services (CMS), worked with the Clinical Laboratory Improvement Advisory Committee (CLIAC) to review quality concerns related to molecular genetic testing. The newly released report reflects CLIAC recommendations of "good laboratory practices for ensuring the quality of molecular genetic testing for heritable diseases and conditions."

History of Genetic Testing Recommendations

"CLIAC began discussion of genetic testing way back in 1997 and 1998," explained Nancy Anderson, MMSc, from the Laboratory Practice Standards Branch, Division of Laboratory Systems, National Center for Preparedness, Detection, and Control of Infectious Diseases, Coordinating Center for Infectious Diseases, CDC, in a phone interview with Medscape Pathology & Lab Medicine.

"There have been a variety of CLIAC meetings at which certain specific aspects of genetic testing have been considered and recommendations made or, in general, comprehensive good lab practices...discussed. So the topic has been on the minds of CLIAC for some time," said Ms. Anderson, who is also part of the group that prepared the MMWR report.

A work group consisting of experts, specialists, and geneticists first met in spring 2008 and presented their input to CLIAC in the fall; CLIAC recommendations were finalized in September 2008. CDC then drafted the report for submission to MMWR.

This report cites previous studies of clinical laboratory tests in general, showing that more errors originate during preanalytic and postanalytic phases of testing than during the analytic process itself. Inappropriate test selection underlies many preanalytic errors. For example, a study of testing for the adenomatous polyposis coli gene found that in 17% of the cases, testing was unwarranted.

"There's far less data available on genetic testing. And the data that are available...say that the error rates are not known to be any more than what is found in other areas of lab testing. Overall it's pretty low," said Ira Lubin, PhD, a geneticist in the Laboratory Practice Evaluation and Genomics Branch, National Center for Health Marketing, CDC.

"When you talk about errors in laboratory testing, you look at that in the broad sense, where errors can occur anywhere from what goes into the medical decision that the doctor uses in test selection. And that's the practice of medicine, which the guideline does not address," Dr. Lubin told Medscape Pathology & Lab Medicine by phone.

"Good laboratory practice, and also part of the general regulatory requirement, is for the laboratory to make available to users information about the tests that they provide," said Dr. Lubin. "So the laboratory may be able to influence correct decision-making if physicians work with the laboratory to understand what tests are provided and the characteristics of that test" in terms of answering their questions. "The guideline strives to emphasize that role for laboratory practice but does not cross over that line of medical practice."

The present recommendations draw on government reports, scientific publications, the CMS CLIA database, studies carried out by professional groups, state programs, and a variety of laboratory directories and databases. The CLIAC Genetic Testing Good Laboratory Practices Workgroup evaluated factors in genetic testing likely to affect quality and identified areas in need of quality assurance guidelines to comply with current CLIA requirements. The recommended practices relate specifically to testing for heritable diseases.

"We know that genetic testing is growing and is expanding, and so we can't really say for sure how many labs will pick up on it and where the testing will be done," noted Ms. Anderson. "But the recommendations were set up the way that they were so that, regardless of where this testing is done in the future, if labs are following these recommendations, we could be assured of the quality of the testing, that it was producing accurate results."

Preanalytic Testing Phase Guidelines

In the preanalytic testing phase, the report lists laboratory guidelines for providing information about molecular genetic tests to those who use their services. The information should include:

  • Selection of appropriate tests;

  • Information on proper methods for collecting, handling, transporting, and submitting specimens;

  • Patient information needed for proper testing and reporting of results;

  • Indication of potential implications of the results for family members; and

  • Availability of laboratory consultations regarding the issues mentioned earlier.

"What is included in this guidance are elements that users of laboratory services should look for when deciding to engage a laboratory, in terms of good laboratory practice," commented Dr. Lubin. "[This] would be the laboratory making available to the user of its services information about the test that is offered.... When [physicians] are trying to evaluate laboratory services,...use those criteria to determine whether the test that you are thinking of ordering is actually valid for the purposes that you are ordering it for," Dr. Lubin emphasized.

Additional concerns addressed by the work group in the preanalytic phase were informed consent, test requests, specimen handling, and establishing policies to assess and correct problems in the preanalytic phase.

Analytic Testing Phase Guidelines

The analytic phase of molecular genetic testing was already regulated by CLIA requirements to "establish or verify the analytic performance of all non-waived tests and test systems before introducing them for patient testing." Beyond adherence to the more general CLIA requirements, recommendations for molecular and genetic testing focus on the assurance of validity and reliability in the tests and proper interpretation of test results.

"Many labs have pretty rigorous assurance quality management systems in place. What's less controlled is the lab's capacity to control or influence what comes to them from clinical settings," explained Dr. Lubin. "So there are communication issues there sometimes, and [issues] in terms of interpreting results.... While errors are made, the data do suggest that [in] the vast majority of testing, there's not a significant problem that has really been documented."

Recommendations regarding establishing and verifying performance specifications for molecular genetic tests list 5 guidelines to be followed for each test:

  • "Conduct a review of available scientific studies and pertinent references

  • Define appropriate patient populations for which the test should be performed

  • Select the appropriate test methodology for the disease or condition being evaluated

  • Establish analytic performance specifications and determine quality control procedures using the appropriate number, type, and variety of samples

  • Ensure that test results and their implications can be interpreted for an individual patient or family and that the limitations of the test are defined and reported."

Studies of the analytic phase of DNA-based genetic testing have reported errors in specimen handling and analysis in only .06% to .12% of nearly 100,000 tests studied. Recommendations for the analytic phase devote significant attention to control procedures and emphasize the importance of proficiency testing in evaluating laboratory competence (as well as in providing education for lab personnel).

Postanalytic Testing Phase Guidelines

Postanalytic errors commonly reflect problems in preparing reports and interpreting results. Studies have shown that a major contributor to these problems is poor understanding among healthcare providers of the limitations of molecular genetic tests and their proper interpretation.

For postanalytic testing, the recommendations focus on content, completeness, and interpretation of test reports. In addition, CLIA requires that test reports, records, and even the tested specimens be retained for specified periods ranging from "as long as possible" for tested specimens to 2 years for test reports and 10 years for pathology test reports. These recommendations recognize the potential importance of these materials for family members and for future diagnostic use, as medical technology and knowledge progress.

Finally, the report speaks to issues of confidentiality, laboratory personnel, and the "quality management system" approach — a system widely adopted internationally, but less common in the United States. Development of a quality management system is likely to play an important role in the ability of labs to receive test referrals from international sources, as well as improving their quality management.

Genetic Counseling

"Talking about sequence testing, you find new mutations you've never seen before once you open the test up to a broader population," observed W. Andrew Faucett, MS, CGC, assistant professor and director of the Genomics & Public Health Program, Department of Human Genetics, Emory University, Decatur, Georgia. Mr. Faucett, who talked by phone with Medscape Pathology & Lab Medicine, was trained as a genetic counselor, has used genetic testing since 1987, and currently runs the Collaboration, Education, and Test Translation (CETT) program, funded by the National Institutes of Health Office of Rare Disease Research, which helps CLIA laboratories develop new testing.

"Since 2004/2005, the CETT program has approved — it depends on whether you count genes, or count tests, or count conditions — probably about 40 applications enhancing 50 or 60 genes and over 100 genetic conditions that weren't available before," he said.

In the early days of genetic testing, physicians would send a pedigree, family history, and clinical notes to genetic testing labs and the lab would make an assessment about the right test. However, "as testing has gotten more common, we think we know more about it, and for commercial competition reasons labs have backed off of that," said Mr. Faucett. "But particularly in tests that you're not used to ordering regularly, it's really helpful to have someone who can discuss the case with you.... I'm particularly concerned with patients getting a negative when it's really not a negative."

"The reality is that most genetic testing is pretty accurate, and there are a few studies to show that," Mr. Faucett continued. "But it's the preanalytical and the postanalytical where you get in trouble. Clearly in preanalytical, if you're doing the wrong test, you're going to get the wrong information.

"CLIA is a minimum set of regulations meant to protect us, kind of a safety issue. There are lots of other laboratory guidelines that good laboratories use...what this article is trying to do is saying that following some of those is a very good idea, and you want to look at a lab that is doing the CLIA minimum but maybe is doing more. And that's what the article is outlining — some of the 'more' things that can be done," Mr. Faucett added.

He also mentioned GeneTests, a resource widely used in the genetics community, to learn about testing in the laboratory. "It's not well known in the nongenetics community. So, that clinician who's ordering a test that he's never ordered before can speak to the laboratory, but he also can look it up on GeneTests to see what's available. And there may be other labs using other methodologies or doing more extensive testing. That's the resource the genetics community turns to," Mr. Faucett said.

Although the CLIAC recommendations have now been publicized, they are in no sense complete: "Continual monitoring of the practice and test performance of molecular genetic tests is needed to evaluate the effectiveness of these recommendations and to develop additional guidelines for good laboratory practices," concluded the report.

Ms. Anderson and Dr. Lubin have disclosed no relevant financial relationships. Mr. Faucett has worked with many laboratories regarding issues related to genetic testing and for several years has been scientific advisor for DNA Direct, a direct-to-consumer contracting company (not a laboratory).

Morb Mortal Wkly Rep. 2009;58(RR-6):1–37.

Full text of CDC recommendations