Email this page
Print this page
September 20, 2006
Washington, DC
On September 21-22, the Genetics and Public Policy Center hosted the National Genetic Policy Summit in Washington, D.C., drawing opinion leaders and policy makers from regulatory agencies, Congress, professional organizations, and genetic testing companies. Following is a summary of the Summit; the agenda, a video (including slide presentations), and photos also are available.
Preliminaries
Jim O’Hara, director of the Policy Initiatives and Health and Human Services Programs at the Pew Charitable Trusts, began the program with a brief welcome. Human genetics is an area “where the rapid progress of science has frankly outstripped policy expertise, and where public engagement has been hampered because we lack solid information to build the necessary policy dialogue and make the policy decisions we need to,” he said. This void creates a need for events like the Summit.
Center Director Kathy Hudson set the stage for the Summit, which she said would focus on genetic testing quality. “While genetic research is moving ahead on super-hyper-overdrive, genetic policy is struggling to keep up and sometimes seems to be in a state of suspended animation,” she said.
The questions driving the Summit, Hudson said, would be:
- Whether the Centers for Medicare and Medicaid Services (CMS) should create a genetic testing specialty, tailored standards for laboratories performing genetic tests;
- How to increase transparency vis-à-vis the characteristics of genetic tests and the quality practices of the labs conducting the tests; and,
- How to get more and better professional guidelines on the use of genetic tests to health care providers.
Plenaries
Janet Woodcock, deputy commissioner for operations at the Food and Drug Administration (FDA), explained FDA’s role in regulating genetic tests. FDA is increasingly attentive to identifying biomarkers, including genetic biomarkers, to detect an individual's current or future health. Studying genetic biomarkers revealed by genetic testing will provide a “much finer lens into an individual’s state of health,” she said.
The dissociation between the diagnostics industry and clinical, therapeutics and biologics industries is a considerable obstacle on the path to identify and validate biomarkers.
Woodcock discussed two new draft guidances put forth by FDA. The guidance addressing analyte-specific reagents (ASRs) doesn't represent a change in FDA's position, but rather addresses confusion in the marketplace about the FDA's rules governing the use of ASRs in laboratories. Another guidance related to in vitro diagnostic multivariate index assays (IVDMIAs) lays out boundaries – identifying a class of tests where FDA oversight will be key. Because many IVDMIAs are developed empirically by comparing gene expression in affected and unaffected populations and examining outcomes, the FDA plays a crucial role in governing their complex uses. “Unlike a single genetic test...in index tests, you have no idea what they are measuring,” she explained.
Senator Barack Obama (D-Ill.) shared his perspective of personalized medicine’s promise and the role of government in bringing it about. “Serious adverse drug reactions affect 2.2 million people, and kill an estimated 100,000 people per year in this country. Genomics could help change this,” he said, giving details on the Genomics and Personalized Medicine Act of 2006 that he introduced in the Senate in August 2006. The Act includes a number of genetics-related measures, including increased research funding, changes to the FDA approval process, and incentives for health care professionals to enter the genomics field.
“For more than a decade we’ve been on the verge of a new era in medicine, but scientific hurdles, market pressures and outdated regulations have blocked our progress, Sen. Obama said. “I think now is the time for leadership that’s going to break this impasse and jump-start the innovation that can save and improve millions of lives.”
Panel - Genetic Testing 2006: Focus on Quality
Mary Pendergast of Pendergast Consulting moderated the first panel, which began with a talk by Brad Popovich, president and CEO of Sirius Genomics. Tests originally performed in academic laboratories increasingly are done in large commercial labs, Popovich said, in a shift driven by the need for standardization and controls, by medical and legal standards, and by tests’ integration into clinical care. He pointed out some of the key differences between the two types of labs, but argued that the “quality endpoint” should be the same; that is, all labs that return results to patients should be subject to the same accrediting and licensing standards. He cautioned against enforcing standards through punitive measures, however, instead advocating for a collaborative, educational approach.
Technology that tests for multiple analytes at once is superior to single-analyte testing, both in terms of efficiency and patient safety, argued Tm Bioscience President and CEO Greg Hines. Products have a lower rate of failure if they test for the presence of either multiple gene variants or other analytes from a single patient, or for the presence of a single analyte in multiple samples – as opposed to a single analyte for a single patient, he said. The Food and Drug Administration has erected a barrier to the development of such tests, however, by regulating them as diagnostic devices, while exercising “enforcement discretion” for simpler tests devised by individual laboratories. Because of the extra time and expense required to develop and bring a so-called in vitro diagnostic test to market in the United States, he said, “The rule of thumb is: Don’t develop an IVD product unless you’re forced to.”
To find out what quality control practices genetic testing laboratories use, and their effects on error rates, the Center recently conducted a revealing survey, reported Director Kathy Hudson. Despite CMS testimony in July 2006 that genetic tests “are dispersed throughout various laboratory specialties and the requirements for those tests are encompassed by the current quality standards,” Hudson said, the Center’s survey showed that 16 percent of genetic testing laboratory directors reported their labs had no specialty certification, including a third of the highest-volume labs. Without a specialty certification laboratories have no specific requirements for proficiency testing, she noted – an important omission, since the survey also demonstrated a clear link between proficiency testing and a reduction in errors of which directors were aware.
During the discussion period, Hudson mentioned that CMS does not make public information on whether individual labs do proficiency testing, making it difficult for consumers or health care providers to gauge their quality. Audience member Annette Taylor of Kimball Genetics suggested that consumers could ask labs whether they are certified by the College of American Pathologists or by New York State, which would indicate whether they meet the high standards those entities require.
Asked whether enhanced regulation might have a negative effect on access to tests, Popovich said that laboratories carrying out “esoteric” tests should be held to different standards in view of the lack of formal proficiency testing programs for such tests. Hines responded that patient safety should come first; if a product isn’t proven to be safe, he said, patients shouldn’t have access to it.
Ann Willey from the New York State Department of Health said that there is some confusion as to what proficiency testing means – that in fact it only tests a lab’s ability to carry out a particular method. Pendergast asked her what would help labs get better, and Willey responded, “experience, and feedback about patient outcome.”
Thomas Hamilton of CMS said that the agency is stepping up its enforcement efforts. A specialty would not “magically” make formal proficiency-testing programs available, he said, since such programs must be developed by the private sector. Asked by Pendergast whether CMS could put out a call for submissions of proficiency-testing programs, Hamilton said the possibility was “interesting to think about.”
Jeffrey Kant of the College of American Pathologists said that laboratory errors are rare compared with the frequency at which health care providers deliver test results to patients in an inaccurate or incomplete way. Timothy O’Leary of the Department of Veterans Affairs stated that although he supports proficiency testing, there is no evidence that requiring it for cytology led to fewer errors. Results of proficiency testing should not be made public, as they could be unjustly used in litigation, he said. Hudson pointed out that “if a lab is routinely failing in proficiency testing, the public needs to know that.”
Tony Gregg of the American College of Obstetricians and Gynecologists pointed out that even given information about laboratory performance, providers and patients usually do not control where tests are done. He asked whether there are any data linking proficiency testing to improvements in patient safety. Buck Strom from Quest Diagnostics’ Nichols Institute said that technical validity was a small problem in the industry, and that false claims made about tests pose a bigger threat to patients. Small differences in proficiency testing results might be improperly used as marketing tools, he said, and intellectual property considerations are a barrier to laboratories working together on proficiency testing and other quality-control measures. Joe Leigh Simpson from the Baylor College of Medicine suggested a compromise on transparency of proficiency testing results, similar to that for physicians: Results would not be made public, but continuing bad actors could be sanctioned in some way.
Beth Jordan of the Association of Reproductive Health Professionals said that most doctors don’t understand statistics, and so would not ask laboratories about their accuracy rates. The Center’s Joan Scott asked Hines whether multi-variant testing would increase physicians’ difficulty in interpreting results; he responded that in the cystic fibrosis example, the data coming out would be the same whether single- or multi-variant testing was used.
Panel - Genes in Practice
Moderator Sharon Terry introduced the following session, relating how her experience with her sons’ illness led to her current position as president and CEO of Genetic Alliance, and to her developing a genetic test.
When Genetic Alliance recently polled its members about their experiences with genetic tests, Terry said, they reported a process rife with insufficient or outdated information; a lack of robust, informed decision-making ability; gaps in providers’ knowledge; test variability; poor or confusing reports; limited access; and concerns that safety is defined as a practical matter for the lab or the provider or the hospital, not by the risk and benefit for the patient. The first three panelists, all of them mothers who have needed tests for themselves or their family members, are “ordinary consumers,” she told the audience. Each of the panel members answered questions about specific aspects of the testing process.
Patricia Furlong’s son was born with Duchenne muscular dystrophy, and his doctor offered a genetic test to confirm the diagnosis and better predict the progression of the disease. Furlong did not realize until after the test came back negative that it only tests “hot spots” on the relevant gene. Since the mutation in her son’s gene was not in one of those spots, the test did not detect it. It was only through her own research that she discovered a laboratory in Europe that could completely sequence the affected gene, thus identifying the mutation. “I’m not sure the health care providers really understood the test at all,” she said.
By contrast, Jana Monaco’s family found their health care providers to be sympathetic and informed when her son was diagnosed with a metabolic disorder at age three-and-a-half. Once the genetic root of his condition was determined, the Monacos used prenatal testing to find that their unborn daughter had the same variant, and were able to treat her in time so that she avoided the brain damage suffered by her older brother.
For Rebecca Fisher’s family the path to useful test results was a long one: When they began working with researchers to get to the bottom of their hereditary high cancer rate, the culprit – the BRCA1 gene – was not yet known to science. Once a test did become available, Fisher’s two sisters kept getting “inconclusive” results, but decided to get prophylactic mastectomies in light of their family history. It later turned out that neither had BRCA1. Fisher herself got breast cancer at age 31, and hopes that early treatment can prevent her daughter from developing the disease.
Ned Calonge, chief medical officer and state epidemiologist at the Colorado Department of Public Health and Environment, discussed the use – or lack thereof – of genetic tests in clinical practice. Although most physicians collect information on a patient’s family history of disease, he said, few assemble it into a pedigree that could determine whether the patient would benefit from a genetic test. Barriers to integrating genetic tests into primary care, he said, include inadequate provider knowledge, lack of access to critical services and experts, reimbursement issues, an inadequate evidence base, and physicians’ discomfort with patient-driven care. Some remedies, he suggested, are to make routine the taking of a family health pedigree, to keep doctors abreast of new clinical research through continuing education and evidence-based guidelines, and to increase access to genetic counseling. “It’s important and vital that we have integration of genetics with traditional care, and appropriate reimbursement,” Calonge said.
Aetna’s senior medical director, Joanne Armstrong, rounded out the panel with a health insurer’s perspective on integrating genetics into the practice of medicine. This integration has the potential to lower costs for insurers, she said, but questions still to be addressed include the impact of testing on drug costs and the clinical utility of tests – that is, which patients would most benefit from which tests. Citing a “knowledge deficit” on genetics among primary care physicians, Armstrong said her company is exploring the prospect of contracting with genetic counselors, but has found them to be in short supply. Other hurdles to personalized medicine are consumers’ lack of knowledge about genetics and concerns about privacy, Armstrong said. To help alleviate the latter, her company has a policy of not basing coverage decisions on genetic test results, disclosing test results, or otherwise allowing members’ genetic information to be used against them.
During the discussion, Terry asked panelists what steps would be the “low-hanging fruit” in terms of improving the use of genes in practice. Armstrong responded that there needs to be a system for getting information to patients when they need it, where they need it. Rather than depending on physicians for this, she said, genetic counselors and other alternate providers of information would be a vital part of the system, and could be made more accessible through means such as phone interviews. Terry added that advocacy organizations are another good source of information for patients.
For Furlong, the low-hanging fruit is more training in genetics in medical and nursing schools, and more genetics counselors, she said. Calonge said that in order to better detect and treat rare genetic disorders, there should be a systematic link between medical information services, electronic medical records, personal heath records, and online family history reporting.
Terry’s next question for the panelists was whether, with advances in pharmacogenomics, lessons learned in the rare disease community will be applied to more common conditions. Calonge speculated that a big success story – such as a large number of people or a celebrity successfully treated with the help of a genetic test – could spur development of a systematized approach to collecting and utilizing genetic data.
“A celebrity would be great,” conceded Fisher, but said that data demonstrating that genetic tests were leading to preventative treatment, and hence to a drop in disease rates, would also help to push genomic medicine into the mainstream.
Audience member Marla Gilson said that many members of her organization (Hadassah) are hesitant to be tested for mutations in the BRCA1 gene because they fear losing their jobs or health insurance. She asked about efforts to pass legislation to protect genetic information from being used in this way. Terry responded that Genetic Alliance has been working for 10 years to get such a bill passed, and that it is important not only to individual patients but also to scientists, since fears of genetic discrimination sometimes make it hard to recruit study participants.
A genetic counselor in the audience said that reimbursement issues, and licensing of genetic counselors by the states, need to be addressed in order to increase the number of counselors. While a well-publicized story about a celebrity’s positive experience with a genetic test would move genomic medicine forward, she said, just one really bad story could bring about a widespread fear of testing. “We need to get there before the bad stories do,” she said.
Gail Geller of Johns Hopkins University noted that the test for BRCA1 is now being used in preimplantation genetic diagnosis, and asked Fisher how young is too young to be tested for the gene. Fisher responded that even 18 is too young, since “the person is still forming their vision of who they are.”
Cathy Wicklund from the National Society of Genetic Counselors asked Armstrong how insurers decide whether to cover genetic counseling. She responded that there is no equation to determine whether genetic counseling is cost effective (that is, whether it increases or reduces the number of genetic tests patients take). She added that because genetic counseling takes so much time, it doesn’t fit well into insurers’ calculations of how much reimbursement is warranted; one way to address this would be to gather health pedigree information electronically, thus reducing the time genetic counselors need to spend with patients, she said.
Susan Wolf of the University of Minnesota asked the panel whether inadequate consent had been a problem in their experiences as genetic research participants. Monaco said that her family’s doctors had shown genuine concern, and never made them feel like research subjects.
There is a strong need for further research to determine the utility of genetic tests, Calonge said, since paying for unnecessary tests increases the cost of health care and hence the number of people without insurance. Audience member Sherri Bale of Gene Dx, Inc., warned that obtaining rigorous evidence of clinical utility for tests for ultra-rare disorders would be impractical, since some such diseases only affect a few dozen people in this country.
Panel - Genetic Testing of Embryos
Preimplantation genetic diagnosis (PGD) “is a boutique medicine issue compared with the big issues we’ve been hearing about all day,” said panelist Mark Hughes, director of Genesis Genetics. While the market for PGD is very small, it attracts much attention because it touches on many hot-button issues, such as reproduction, the question of when life begins, and worries about “designer babies,” he said. The procedure was born out of the frustration of doctors who saw parents of children born with crippling genetic diseases and wanted to give those parents a way to ensure that subsequent children would not be similarly affected, Hughes related. He went through the steps involved in PGD, and gave examples of some of the inherited diseases for which it has been used. He predicted that use of the technology will increase as we learn more about our genomes and genetic testing becomes less expensive, and as the trend continues toward having children later in life. “These patients don’t want a perfect child—they know there’s no such thing,” Hughes concluded. “As public policy leaders, as genetic counselors, as physicians and scientists, and as entrepreneurs, maybe we can come together and move these powerful technologies from the bench to the bedside in an ethically responsible manner.”
Ethicist Eric Cohen, fellow and director of the Bioethics & American Democracy Program at the Ethics and Public Policy Center, was next to speak. While the motivation for PGD is to give children a healthy life, he said, it “necessarily requires a kind of genetic discrimination.” Specifically, he cited five areas where PGD poses moral challenges: the possibility of long-term safety risks, determining whether embryos are humans, non-medical applications such as sex selection, the fact that PGD is a form of eugenics, and the perhaps-unwelcome foreknowledge the procedure provides. The field needs better governance, he concluded, to set limits – for example, the use of PGD for non-medical sex selection or to avoid late-onset diseases should be prohibited.
GPPC’s Susannah Baruch summarized how PGD is being used today (based on the Center’s recent survey of in vitro fertilization clinic directors). She then surveyed the audience’s views on PGD using electronic keypads and compared the results to those from an earlier survey of the public. Among the results:
- 78 percent of Summit participants believed there should be limits set for acceptable and unacceptable uses of PGD.
- 96 percent agreed that PGD is acceptable to avoid a fatal or severe childhood illness in one’s offspring.
- 82 percent agreed that PGD is acceptable for HLA matching – to have a child to save the life of an existing ill sibling.
- 76 percent agreed that PGD is acceptable to avoid an adult-onset disease.
- 23 percent agreed that PGD is acceptable to select the sex of the future child.
- Asked who should set limits on acceptable and unacceptable uses of PGD, 43 percent responded that the decision should be left to patients and their doctors; 33 percent believed professional medical societies should set guidelines; 19 percent chose federal or state government; and 5 percent believed patient groups should devise guidelines.
- 98 percent agreed that there should be oversight of the safety and accuracy of PGD, with 48 percent in favor of the government providing such oversight, and 52 percent favoring professional societies.
During the question and answer session, moderator Joe Leigh Simpson of the Baylor College of Medicine asked Hughes whether he thought it would be feasible to do proficiency testing for aneuploidy (chromosomal abnormality) screening. Hughes replied that such proficiency testing would be “reasonable and straightforward,” adding that unqualified IVF clinics should not be allowed to do aneuploidy screening. However, he said, doing proficiency testing for single-gene disorders would be more difficult, since PGD providers usually have only one family’s DNA for quality-control purposes, and devise the test specifically for that family. PGD is complicated further by the fact that blastomeres are dividing quickly, sometimes exhibiting erratic cell cycles and mosaicism, he noted.
Audience member Yuri Verlinsky, director of the Reproductive Genetics Institute, commented that only specialists should do PGD, and gave examples of quality control procedures that lessen the chance that results will be tainted by the factors Hughes named. PGD accuracy is very high if done right, Verlinsky said. Baruch noted that only a small number of PGD procedures are done “in-house” at IVF clinics, and that most of these are for aneuploidy.
Robert Miller of the National Fragile X Foundation asked how the public can be made aware of information that contradicts common assumptions, such as that carriers of Fragile X syndrome are unaffected by that condition. Hughes said that undiagnosed Fragile X is common, and that this can pose a problem in IVF by causing ovaries to react unpredictably to synthetic hormones.
If a PGD database is established, what would be the markers of an “outlier” in terms of quality, asked Joanne Armstrong. Baruch responded that this was one of the questions being considered by those planning a database. It’s necessary to have more of a sense of what’s being done in the field before determining what an outlier looks like, she said. Simpson commented that it is important that outliers recognize they’ll be punished economically.
An audience member commented that the American Society for Reproductive Medicine and the Society for Assisted Reproductive Technologies have been successful in devising guidelines that members follow voluntarily – for example, the rate of multiple gestations declined after a guidance was issued recommending that no more than two embryos be implanted during an IVF cycle. David Granger said that all the measures being discussed are voluntary, although all cycles of IVF are reported to the federal government.
Perry Payne of Howard University asked whether anyone was tracking patients over time to determine the long-term outcome of PGD, and whether anyone had compared the cost of the procedure to that of treating the diseases it prevents. Hughes replied that long-term follow-up is difficult because most parents want to move on after PGD, but that studies in other countries have not shown an increased rate of birth defects following the procedure. He compared the removal of a cell from an embryo for PGD to the loss of cells when an embryo is frozen and later thawed, or divides spontaneously to form identical twins. While some insurance companies have concluded that PGD is cost-effective, he said, the comparative costs are hard to determine because of ever-improving treatments for the diseases PGD prevents.
State-by-state regulation of PGD providers would create a problematic patchwork, said Susan Crockin of the Susan L. Crockin Law Firm. She asked the panelists who should regulate the procedure. Baruch said that voluntary regulation might be best. Cohen said that while federal regulation would be preferable in some areas for reasons such as safety or data gathering, state regulation might be better in other arenas since different states have different values.
Hughes said that the number of couples getting PGD expressly for non-medical sex selection is very small; most non-medical sex selection, he said, is performed when a couple already is getting PGD for a medical condition.
Citing lack of data showing that aneuploidy screening increases the IVF success rate, Susan Wolf asked “how do we regard the fact that it’s so commonplace?” Verlinsky responded with his own question: Given that 80 percent to 90 percent of miscarriages are caused by aneuploidy, and that sometimes aneuploidy screening is the only way of choosing which embryo to implant, “why shouldn’t we do it?” he asked. Simpson commented that aneuploidy screening hasn’t been shown to improve the “take-home baby rate,” and that the techniques are still evolving. He predicted that future research will enable doctors to predict which women will benefit from aneuploidy screening and which will not.
Sara Katsanis of Johns Hopkins said that genetic testing should be regulated now so that it doesn’t become a market-driven field offering products of questionable value and validity. Simpson responded that it is premature to prohibit certain tests from being conducted, because the information needed to judge their clinical validity is not yet available. It’s currently only reasonable to regulate the safety and accuracy of tests, he said.
Beth Jordan asked whether genotype equals phenotype for conditions for which PGD is commonly performed. Hughes said no, giving the example of cystic fibrosis: Even siblings with the same mutation can have dramatic differences in the severity of the condition. Simpson said this imprecision is not a new problem, since uncertainty about the effects of genes is also an issue with amniocentesis and chorionic villus sampling.
Plenaries
Center Deputy Director Joan Scott opened the second day of the Summit by introducing speaker Francis Collins, director of the National Human Genome Research Institute. While the Human Genome Project provided a foundation for genetics, she said, identifying the variants responsible for a health condition is just a first step toward understanding the underlying biology of that condition and devising a treatment for it.
Collins also noted the nascent state of genetic knowledge. “What you see right now is but a tip of an iceberg,” he said, and cautioned against designing genetics-related policies that “don’t work for tomorrow.” One advance we can expect in the next few years will be the identification of genes linked to major diseases (such as diabetes, cancer, heart disease, autism, hypertension, bipolar illness, asthma, Alzheimer’s disease, and osteoporosis), he said, summarizing the recent scientific developments that that make such discoveries likely.
Collins also outlined some initiatives aimed at accelerating developments in genetics and their application to medicine. These include the Genetic Association Information Network, a public-private partnership that will provide genotyping for several large studies and ensure that the resulting discoveries are not patented. Another example is the Genes and Environment Initiative, an item in the proposed 2007 federal budget that would provide money to support 15 studies on genetic and environmental contributors to disease. The new Cancer Genome Atlas project uses 500 human specimens to identify the genetic and epigenetic basis of cancer, with the goal of identifying new drug targets. The National Institutes of Health is developing agency-wide policies on data sharing, data access, intellectual property, and protection of research participants in Genome-Wide Association Studies, Collins said.
Case-control studies are a poor way to identify the significance of gene variants in causing disease, he said. What’s needed, Collins asserted, is a large-scale, prospective cohort study with a budget comparable to that of the Human Genome Project.
To move forward smoothly, the field of genetics needs good policy in the areas of analytic and clinical validity of tests, direct-to-consumer testing, and nondiscrimination, Collins said. “Obviously you don’t want to introduce a test into the mainstream that’s not validated, but at the same time, if you insist on having prospective studies on every single test to actually demonstrate a clinically valid conclusion, then it’s going to greatly slow down the field,” Collins said.
The future of public health is an exciting one, he concluded, as genetics will enable medicine to grow more personalized and predictive.
Audience member Mike Watson of the American College of Medical Genetics asked about designing studies of rare diseases; Collins agreed that a database approach might be best suited to address the special research challenges these diseases present. He said that advocacy groups have an important role to play in this arena, and gave the example of the Cystic Fibrosis Foundation’s registry.
Kirsten Moore of the Reproductive Health Technologies Project asked what the intellectual property relationship should be between researchers and patients. Collins responded that the ability to patent genes linked to disease has been a disincentive to innovation, and therefore is not in the public interest.
Steve Phurrough, director of the Coverage and Analysis Group at the Centers for Medicare and Medicaid Services (CMS), spoke next. Some of the issues with genetic testing as far as that agency is concerned, he said, are whether to pay for treatment indicated by genetic tests, how to safeguard privacy, CMS’s role in large cohort studies, and ethical issues such as genetic determinism.
Phurrough focused primarily on reimbursement issues, however. He said that the agency can only pay for diagnosis and treatment, so if a person does not have symptoms of a disease, CMS cannot pay for them to be genetically tested for that disease. This applies even if the person has a family history of a condition. If a patient does exhibit symptoms of a genetic condition, CMS will pay for a test only if a treatment exists for that condition. In addition, the agency cannot reimburse genetic counselors directly, although it can pay physicians who employ genetic counselors for their services. These restrictions can be changed only through Congressional action, Phurrough said.
Most decisions about how much to pay are made by local contractors who may have different systems, he said, noting that one advisory committee has recommended determining the best local coverage models and making them the national standard.
“I think the long-term, overall solution is to have good quality information that says that these particular tests provide value in ensuring that patient outcomes are improved,” Phurrough concluded, but admitted, “that’s a pretty high bar to get to.”
Sherri Bale described the difficulty she’d had getting her diagnostic company listed as a provider with Medicare, and asked for Phurrough’s advice. “‘There are a whole host of things that Medicare perpetuates on the healthcare system that ought to be done differently,” he said. “Until the rules change, it’s very difficult for the people you get on the phone [at CMS] to respond.”
Another audience member asked whether a laboratory can bill Medicare for tests that were marketed as research-only test kits, but have been validated independently by the laboratory. Phurrough reiterated that Medicare pays for genetic tests only under “very limited circumstances… a particular label on a particular package really doesn’t determine whether that is coverable or not.”
Mike Watson asked whether physicians could be reimbursed for genetic counseling. Phurrough responded that while there is no regulation stating that genetic counseling is not a physician service (and therefore not reimbursable), local contractors may interpret the law differently.
The Washington Post’s Rick Weiss suggested that there are, in fact, instances where CMS can go beyond the limits imposed by Congress, and asked how Summit participants might go about precipitating changes in the agency’s genetics policies. Phurrough noted that this was not one of the areas he’d been asked to address, but said specific tests that would not otherwise be covered could possibly be reimbursed if done for data-gathering purposes.
As Center Director Kathy Hudson announced a break, she also told participants that Senator Edward Kennedy had released a draft version of a genetic testing bill he intended to introduce to the Senate. Summit attendees were encouraged to comment on the draft.
Panel - Direct-to-Consumer Genetic Testing: Empowerment or Endangerment?
Center Communications Director Rick Borchelt moderated the final panel of the Summit. “Among the issues that we’ve been discussing at the Summit,” he said, “probably none has galvanized media and public awareness more than direct-to-consumer genetic testing, or, as it’s sometimes called, at-home genetic testing.”
Gail Javitt, law and policy director at the Center, defined direct-to-consumer (DTC) tests as those marketed directly to consumers that can be taken without a physician’s involvement, and specified that the panel session would only deal with health-related tests. She outlined the players in DTC regulation, starting with the Federal Trade Commission, which prohibits false or misleading advertising but has not taken any actions against specific companies. The FDA only regulates genetic tests sold as test kits, but since most DTC tests are “home brews” developed by individual laboratories, FDA has not played much of a role in DTC testing. All clinical laboratories are required to be certified under the Clinical Laboratory Improvement Amendments of 1988 (CLIA), but some labs performing DTC tests are not, and the failure of CMS to create a genetic testing specialty under CLIA means that labs are subject to minimal scrutiny. An example of the outcome of this regulatory landscape is Baby Gender Mentor, which purports to detect fetal gender as early as five weeks into a pregnancy. “Based solely on what’s reported in the media, it gets the answer right less often than my grandmother used to when she’d look at our bellies and tell us it was a boy or a girl,” Javitt said.
Robert Nussbaum, professor of medicine and chief of medical genetics at the University of California, San Francisco, said that evaluating the validity of a test comes down to three “ities”: analytic validity, clinical validity, and clinical utility. Analytic validity is a “no-brainer,” he said: Companies that can’t get the right answer on genetic tests shouldn’t be operating. Clinical validity – the question of whether test results correlate with or predict phenotype or susceptibility – is a more complex question, and large cohort studies are the best way to determine a test’s predictive value, Nussbaum said. Measurements of clinical utility, or a test’s usefulness for guiding and improving health care, rely on outcomes, and are affected by factors such as the interpretation of the test and its cost. Advantages of allowing consumers to access tests directly might include raising awareness of genetics and of the interaction of genes and environment, and challenging physicians to be better-educated, he said. Real and potential harms, he said, include the “sleaze factor” of having companies market nutritional supplements supposedly based on a person’s genetic profile, false reassurance, and harming the field of genetic testing. “To me, the devil’s in the details,” Nussbaum opined.
Panelist Howard Coleman, CEO of Genelex Corporation, advocated a cooperative approach to developing quality requirements, with proficiency testing done in an educational way. If proficiency testing results are made public, he said, they could be used in an “inappropriate fashion” by competitors and lawyers. Coleman said he supports a genetic testing specialty under CLIA: “I view it as a natural expansion, a growth, of the quality assurance programs to meet the needs of the DNA testing industry,” he said. However, he said that in general the government should concentrate on enforcing existing regulations rather than on making new ones, and that there is a need for “harmonization” between regulation and the public’s access to tests. He defended nutritional genomic tests, saying that they can help people make long-term changes to their lifestyles, and said one reason there is a need for direct-to-consumer tests is that health care professionals frequently miss adverse drug reactions that can be predicted by genetic tests. Another reason, Coleman said, is privacy: People have a right to learn about their genomes, and to control who else has access to that information. He detailed the case of Warfarin, calling it the “low-hanging fruit” of genetic testing because severe adverse reactions to the drug can be predicted easily using a genetic test.
Peter Lurie, deputy director of Public Citizen’s Health Research Group, spoke mainly about DTC drug advertising, using it as “a bit of a metaphor for what’s going on in genetic testing.” DTC drug ads are effective and are increasing in number, he said. Most drugs advertised are for chronic, annoying conditions, since acute conditions disappear too quickly for advertising to be profitable. A survey has shown that most consumers believe DTC ads are pre-approved by the government, although they are not, and warning letters to companies running deceptive campaigns usually arrive after the ads have finished running. Advertising increases drug costs, mostly by pushing customers to move from less-expensive to more-expensive drugs, Lurie said. As examples of poor advertising in the genetics field, he showed Web sites marketing tests for athletic ability, aging, and depression risk. In many ways such sites pose more risks to consumers than drug advertising, he said, since consumers and doctors are less informed about genetic tests, the tests are not regulated by FDA, and the industry has no ethical code for marketing (unlike the pharmaceutical industry). He recommended FDA regulation of genetic testing and educational efforts by federal agencies.
Borchelt then surveyed the audience on their views of DTC. The questions (and responses) were:
Given the current regulatory environment for genetic testing, with which of the following statements do you most agree:
DTC is not appropriate for any genetic tests: 31%
DTC is appropriate for all genetic tests: 4%
DTC is appropriate for some genetic tests: 64%
If laboratories offering DTC testing employed a provider who approved test orders and reviewed test results sent directly to consumers, would your answer change?
More favorable: 39%
Less favorable: 3%
Same: 58%
If there were oversight for genetic testing that ensured analytic and clinical validity, with which of the following statements would you agree?
DTC is not appropriate for any genetic tests: 12%
DTC is appropriate for all genetic tests: 5%
DTC is appropriate for some genetic tests: 83%
What is the primary reason you think DTC may be inappropriate for at least some tests?
Promised benefits of testing may be exaggerated: 15%
Risks of testing may be minimized: 6%
A provider is needed to ensure appropriate ordering and to explain results: 42%
Tests may not be analytically valid: 5%
Tests may not be clinically valid: 20%
Companies offering DTC are inadequately regulated: 12%
What is the primary reason you think DTC may be appropriate for at least some tests?
Greater access to testing: 40%
Greater consumer involvement over decision making: 35%
Better informed consumers: 17%
Lower cost of testing: 8%
DTC should be available for genetic alterations that:
Indicate risk of adverse drug reactions: 56%
Indicate risk to develop blood clots: 51%
Indicate increased risk of developing Alzheimer disease: 27%
Indicate increased risk of developing breast cancer: 34%
Indicate increased risk of depression: 22%
Determine fetal gender by testing maternal blood: 34%
Do you think the Federal Trade Commission should take action against companies making false or misleading claims for genetic tests sold directly to consumers?
Yes: 99%
No: 1%
Do you think CMS should ensure that laboratories offering tests DTC are CLIA certified?
Yes: 97%
No: 3%
Do you think CMS should make public the CLIA certification of laboratories offering genetic tests?
Yes: 95%
No: 5%
Consensus: Where do we go from here?
Jonathan Ortmans, president of the Public Forum Institute, facilitated a discussion in the final session of the Summit with the aim of identifying common ground and innovative approaches to challenges in genetics policy. Initially, he asked participants to think of what can be done to improve genetic testing. Asked where it was possible to find common ground, audience members suggested the need for improved reimbursement for tests, a genetic testing specialty under the Clinical Laboratories Improvement Act of 1988 (CLIA), truth in labeling, guidelines for physicians to determine when tests are “ready for prime time,” more data on preimplantation genetic diagnosis, standardized reporting of test results, a clearinghouse of reliable information for consumers, and more proficiency-testing (PT) programs.
Asked how satisfied they were with the current mandatory system for laboratory quality is sufficient to ensure the analytic validity of tests, 38 percent of audience members indicated they were not at all satisfied, while 28 percent were slightly satisfied, 26 percent were moderately satisfied, and eight percent were very satisfied. Some elaborated on their reasons for dissatisfaction. To better ensure validity, participants suggested that genetic counselors should market themselves better, and that laboratories should perform proficiency testing. This led to Ortmans’s next question, “Should laboratories be doing proficiency testing, whether formal or informal (e.g. sample splitting or sample swap) on all tests they offer?” Ninety-eight percent of audience members answered affirmatively. However, some said that PT would be difficult or impossible for rare diseases. Ortmans observed that antagonism between academics and the private sector would need to be addressed in order for the genetic testing field to overcome challenges in PT and other areas.
He next asked whether proficiency testing should be required or rewarded. Seventy-eight percent of the audience indicated it should be required; 22 percent preferred rewards. Ortmans steered the conversation toward how to deal with “bad performers” in the genetic testing field. One audience member suggested a system like that already used for physicians: While problems licensing boards find with physicians are not made public, those who do not improve face sanctions that make them unable to practice medicine. Another said that CLIA certification could be denied to laboratories that repeatedly perform badly. Thomas Hamilton of CMS said his agency is increasing its inspections and educational activities.
Ortmans asked for suggestions other than PT that could improve genetic testing within the current statutory framework. Audience members named truth in advertising, development of evidence-based reviews of tests, clinical trials, professional guidelines, incentives for the diagnostics industry to run clinical trials, requiring test providers to perform analyses of the usefulness of genetic tests as a condition of gaining patents, incentives for benefit providers to do clinical studies of tests, and incentives for laboratories to use FDA-approved test kits rather than home brews.
“What about linking validation to reimbursement?” Ortmans asked. One person said organized data collection would be needed to make that link. Joanne Armstrong of Aetna said that strong evidence of tests’ clinical utility would show insurance providers whether the tests are cost effective and hence reimbursable. Another audience member warned that too-stringent requirements for proof of clinical validity could “temper further discovery.” A further suggestion was that reimbursement should be based on the research costs of demonstrating the clinical validity of a test, rather than on the mechanics of performing a test, once validated.
The next topic was whether enough information is available to the public and health care providers about genetic tests, and if not, how to better disseminate that information. No one disagreed that more information is needed. Joan Scott said that professional organizations are a logical means of getting guidelines to physicians and other providers. Another audience member said that evidence-based guidelines, while useful, hold tests to a very high standard, and other sources of information are needed to evaluate whether a test should be used. A representative from the American Society of Reproductive Medicine said that his society has sometimes found it necessary to generate “clinical opinions” when evidence was insufficient to come up with a guideline on a particular treatment. Armstrong asked whether professional societies have the resources to inform their members about developments in genetic testing; materials they distribute are more likely to be read than even major medical journals, she said. While two representatives of professional organizations said they were well-equipped for this job, others cited challenges with developing multiple guidelines for a multidisciplinary membership, or with resource limitations. One advocated using patient groups as a resource when developing guidelines.
Audience members offered final, more general ideas for improving genetic medicine: making comprehensive information available to the public on evaluating direct-to-consumer genetic tests, investigation into why genetic tests are rarely used in clinical practice, despite recent FDA label changes recommending a test before prescribing certain drugs, and expanding the ranks of genetic counselors.
Ortmans’s penultimate survey question was whether there should be a professional database for preimplantation genetic diagnosis. Participants unanimously agreed that there should be. He then asked whether results should be made public. Fifty-eight percent answered “yes, in aggregate,” 39 percent thought clinic-by-clinic results should be disclosed, and three percent said results should not be public at all. Asked who should pay for a database, various audience members named the government, those getting the procedure, and clinics.
Asked for parting thoughts, audience members offered that:
- resources are badly needed to educate current health care providers about genetic tests;
- advocacy groups have much to offer in the decision-making processes surrounding testing;
- there should be more government incentives for development of pharmacogenomic applications of tests; and,
- government agencies should explore ways in which they can help innovative new products get to market faster.
Video Link:
http://www.dnapolicy.org/video/2006summit/index.htm
