Physicians, plans and patients are discovering that the promise of genetic testing will be hard to fulfill. Even when a test can show predisposition toward a disease, performing it can't necessarily improve medical outcomes. Unfortunately, doing these tests can have some unintended negative effects.
Your patient is a young woman whose mother and aunt died of ovarian cancer at ages 47 and 48, respectively. Because of this strong family history, she probably has a 50-percent, rather than a 7- or 8-percent, risk of contracting breast or ovarian cancer. She elects to have a complete hysterectomy based on her "hereditary ovarian cancer-proneness."
What would you say? Blue Cross and Blue Shield of Nebraska said no, that familial breast-ovarian carcinoma syndrome was not an illness and that the operation was not medically necessary. Sindie Katskee, the patient, sued, lost, appealed and eventually had the surgery, but off her plan. End of story?
Not so fast. The Supreme Court of Nebraska took issue with Blue Cross and Blue Shield's interpretation of its contract language dealing with the terms "illness," "sickness" and "disease." Sindie's experts had testified that her surgery had been performed as a result of an at-risk condition resulting from "a genetic deviation from the norm." If one defines illness as any abnormal condition of the body or its components such that its natural progression would be expected to pose problems, the court asked, wouldn't breast-ovarian carcinoma syndrome qualify? In other words, an abnormal state resulting from a genetic deviation equaled illness, at least in Nebraska.
George J. Annas, J.D., M.P.H., director of the Health Law Department of Boston University's School of Public Health, has pointed out that genetic tests will be even more specific than the family history used in the Nebraska case. Specific, maybe, but not necessarily more predictive of a negative outcome, say many experts.
Genes as destiny
What started all this and is rapidly accelerating it is the Human Genome Project at the National Center for Human Genome Research at the National Institutes of Health in Bethesda, Md. As scientists there steadily map the human genetic code, they seem almost daily to be announcing the finding of a gene "for" colon cancer, breast cancer, prostate cancer, cystic fibrosis and even for less hardwired-seeming conditions such as anxiety and "risk taking." Once a gene is isolated, a test for it is not far behind. But interpreting those tests is quite another matter.
In what she calls "the diseasing of risk factors," Lynn Payer, author of Disease-Mongers: How Doctors, Drug Companies and Insurers Are Making You Feel Sick (John Wiley & Sons, 1992), quotes Paul Billings, M.D., Ph.D., chief of genetic medicine at California Pacific Medical Center: "As a medical geneticist, I know more than the average physician. Yet I don't know how to interpret a finding that the p53 gene [found in many breast and colon cancers] is present, and I'm not sure what finding the Fragile X means in an individual who is not noticeably retarded."
"In the best of all worlds, managed care has a tremendous opportunity to focus on prevention," says Reed E. Pyeritz, M.D., Ph.D., chairman of the Department of Human Genetics at Allegheny University of Health Sciences in Pittsburgh and Philadelphia, and president-elect of the American College of Medical Genetics. "If we could determine without a doubt that someone was susceptible and there was something we could do about it, we would have to make the most of our obligation. But, remember, I said 'perfect world.' We're not there yet."
John D. Reinhard, M.D., is physician manager of the Teminex (for Technology Management and Information Exchange) program of the New Brunswick, N.J.-based HMO Group (a national network of 18 staff- and group-model HMOs), as well as a physician at Health Care Plan in Buffalo, a staff-model/IPA hybrid with about 100 doctors. "Teminex is a group of M.D.s and Ph.D.s trained in epidemiology that provides our health plans with the evidence they need to make decisions," Reinhard explains. "We realized early on, a year ago, that genetic testing required a totally different approach. There was going to be lots of public demand, but very little evidence that testing was either helpful or harmful.
"No trials, no studies, nothing that said, 'Here are 200 women with the BRCA I breast cancer gene, and in two years, this is what happened with prophylactic surgery and this is what happened without it.'"
"We definitely have patients interested in knowing whether they are at increased risk for breast or ovarian cancer," confirms Jacob Reiss, a pediatrician and director of genetic services at Kaiser Permanente Northwest in Portland, Ore., and a member of a committee developing Kaiser's interregional physician clinical practice guideline for recommending whom to refer for genetic counseling for inherited susceptibility to breast and ovarian cancer.
"There are many pros and cons to providing this counseling and possible testing," says Reiss. "It's very controversial."
He notes that there is very little evidence that removing breasts or ovaries from women at high risk of developing cancer actually affects morbidity and mortality. "Such dramatic procedures may be medically indicated in rare instances," he says, "but patients and providers together have to weigh carefully the advantages and disadvantages of testing for these cancer susceptibility genes. At this time, there does not appear to be a big benefit to testing unless personal or family history indicates a risk."
Although Kaiser's physician guidelines are not out yet, plan doctors may be advised to recommend genetic counseling for interested patients in four broad categories: 1) those who have close blood relatives with known gene alterations, 2) those with a personal and strong family history of breast and/or ovarian cancer, 3) those without a personal history but with a very strong family history of breast and/or ovarian cancer and 4) those with a personal history of breast and/or ovarian cancer and an additional risk indicator such as early-onset cancer or membership in an ethnic group with higher chance of having one of these cancer susceptibility genes.
Counseling clearly is very important in this area, and it is no minor undertaking. A professional counselor, who might be a physician or might be another health professional with training in this area, is likely to spend three or four hours counseling an individual patient.
"There simply is no guarantee that a genetic disposition means you will get a disease," says Matthew Y. Biscan, J.D., who chairs the health care practice area of Hall & Evans, a Denver law firm. "Genetic testing has been around for quite a while. It's mandated in some cases in the neonatal area, and used to establish paternity or in sexual assaults. But extending it to the diagnosis and curing of disease is very new."
In northern California, Kaiser Permanente has four genetics centers, staffed by clinical geneticists, counselors, nurses and laboratory technicians who offer lectures and teleconferences in addition to individual genetic counseling and testing.
Physicians and other professionals must know what services are available and how to use the system best. At Kaiser, this is accomplished in part by a periodic short newsletter called The Screen, which has been expanded to all Kaiser regions and also offered via electronic mail. Each issue covers a topic such as neonatal hemoglobinopathy screening or triple-marker screening.
Kaiser does offer some presymptomatic counseling and testing, such as for Huntington's disease, and it tests for Fragile X, Prader-Willis syndrome, cystic fibrosis and sickling syndromes. Records are stored not only in a patient's chart but in a special on-line computer data base available to all Kaiser centers.
"We need to develop sophisticated pamphlets to help both physicians and patients understand the ramifications of genetic testing. If people decide they do want the testing after the counseling, it should be done under a protocol so we can track the results," recommends Reinhard.
Prudential Insurance Co. of America, with 15 million health care subscribers about evenly split between managed care and indemnity insurance, addresses genetic testing and follow-up directly in its technology assessment coverage policies. "Things are moving so fast in the genetic area, we are revising our guidelines," says Arthur L. Levin, M.D., vice president of the insurer's Technology and Clinical Practice Assessment Unit and a member of the Task Force on Genetic Testing of NIH's Working Group on Ethical, Legal and Social Implications (ELSI) of the Human Genome Project. (Look for the results of that group's two-year study later this year. Genetic Information and Health Insurance, a report by an earlier ELSI task force, is available as NIH pub. No. 93-3696.)
Prudential, Levin says, does not reimburse for tests and treatments it considers to be investigational. Prenatal testing, yes; BRCA I, no. "Some tests have been proven," Levin says. "We don't think BRCA I is ready for prime time."
Would Prudential consider offering the test and tracking all results in a formal study? "I don't know if we'd want to help a company prove or disprove a product," Levin says, adding: "ELSI may recommend that all testing go under protocols. Then, sure, that might influence us. In the meantime, we believe that we as a company ought to determine whether a test or procedure is useful or not before we offer it. It has to make a clinical difference."
An abnormal BRCA I gene confers on female carriers an 80- to 90-percent lifetime risk for developing breast cancer and may increase the risk for other cancers, but no single abnormality has been found to predominate. Whether all abnormal alleles confer the same risk is generally unknown. It is also unknown whether a BRCA I mutation in a woman with no family history confers the same risk that it would in a family with a history of breast cancer. Also — the test is expensive, around $2,400.
Tell me, doctor
Many physicians are not comfortable dealing with this issue. "It would be great if the M.D.s did insist on counseling. I've found that primaries and even specialists are just ordering the tests and aren't in a good position to explain what the results means or to determine who should be tested," agrees Pyeritz of the American College of Medical Genetics.
"Physicians' roles as diagnosticians and healers will evolve to much larger roles as patient educators and counselors," Reinhard has written. "This will require amplification and enhancement of previous skills and acquisition of new skills. The trend from physician-centered decision making to patient-centered decision making will accelerate.... Genetic counselors, with whom primary physicians have traditionally had little contact, will become important resources and allies."
Physician comfort levels aside, pressure to test is strong, Reinhard points out. But — again — a risk factor is not a disease, he stresses. Women from high-risk families probably are already being screened early and aggressively. There is nothing showing that genetic testing will decrease their risk of dying, he says. Chemoprevention trials are on the way — but years out. Put simply: Do the risks outweigh the benefits?
"In the absence of data, the HMOs are justified in not covering the tests or treatments," says the lawyer George Annas.
Risks all around
Let's use BRCA I as an example of risk factors. Women from high-risk families who screen negative are likely to benefit in that they will know that despite a horrific family history, they have no more chance than average of developing breast cancer. Might that lead them, though, to neglect their mammograms and figure they are home free? Some physicians have suggested that they might. And that's not even addressing the possibility of a false negative?
What if the test is positive? The most obvious effect is increased anxiety and possibly a desire for prophylactic surgery. The latter may or may not decrease morbidity and mortality, but surely will have emotional and medical consequences.
What about the woman's siblings and other close relatives? Should they be contacted? What if they don't want to consider testing or surgery? What if they don't even want to know that their sister carries the gene, thus putting pressure on them to be tested?
And what about a false positive? asks the lawyer Biscan. "You might find out you're predisposed to colon cancer and freak out. Do you have a claim? Informed consent becomes very important. But people can still sue for anything they want. Are HMOs to be liable for the quality of the lab? Think about that."
The diseasing of risk factors also may expose people to social discrimination or discrimination in employment. Would a predisposition to breast cancer cost someone a job with an employer who doesn't want higher health premiums?
More specifically, could a genetic finding — a risk factor — make it difficult to get health or life insurance? Many insurance experts insist that limitations on discrimination for pre-existing conditions (and you can't get more "pre" than the moment of conception) will prevent widespread abuses in this area.
Yet stories persist of people denied health care coverage. In one case cited by the earlier NIH task force, a woman was denied health care coverage because her nephew had been diagnosed as having cystic fibrosis and she wanted to be tested to see if she was a carrier. She was tested and found to carry the gene. The insurance company then told her that neither she nor any children she might have would be covered unless it was determined that her husband did not carry the CF gene.
In another example, a man diagnosed as having Huntington's disease was denied coverage because it was a "pre-existing condition," despite no previous diagnosis having being made. The health plan later reversed its decision under protest, but the man continued to have problems with his insurance and was denied custodial care.
We're quite a way from genetic tests being required for all people (in the manner, say, that syphilis tests are required to get a marriage license). For one thing, everyone has genetic defects — and pretty soon, everyone will have detectable genetic defects. It will behoove all of us, not just managed care organizations, to determine the actual significance and predictive value of genetic tests.
Such questions aside, insurance companies are in the business of quantifying risk. In a paper titled "Genetic Testing and Insurance" (Am J Hum Genet 56:327-331, 1995), Maimon M. Cohen, Ph.D., in the division of human genetics in the Department of Obstetrics and Gynecology at the University of Maryland Medical Center in Baltimore, writes: "An insurer might argue that the condition is present, albeit in latent form, as soon as the individual acquired — or knew that he or she had acquired — the abnormal gene."
Yet, writing in the Journal of Law and Medical Ethics (Spring 1993), N.S. Jecker declares that private health insurance companies have a "social responsibility" to avoid using genetic testing to discriminate among risk groups. "The reduction of uncertainty about individuals' future health status through genetic information," Jecker continues, "may weaken the risk-sharing principle upon which insurance is founded. If genetic discrimination is practiced by private health insurance companies and plans, public insurance may be the only answer."
The ELSI Task Force on Genetic Information and Insurance asserts that "The information about past, present or future health status, including genetic information, should not be used to deny health care coverage or services to anyone."
If these warnings are not heeded, there could be a return to community rating. "Once it's clear that all of us have a variety of genetic susceptibilities and are all potential risks, community rating will come back," predicts Pyeritz of the American College of Medical Genetics.
"Coverage and reimbursement depend on how the plan is written," says Biscan. "Genetic testing will have to be excluded if plans do not want to pay. Eighteen states have laws prohibiting discrimination on genetic grounds. There is federal legislation in the hopper. Such as happened with autologous bone marrow transplants for breast cancer, if the medical community decides genetic testing and consequent treatments have value, companies will have a tough time refusing. But, by the same token, when the tests become perfectly predictive, managed care plans may also want to revisit the insurability issue."
Don White, a spokesman for the American Association of Health Plans, disagrees. "Twenty years from now, people won't say genetic testing kept people off health plans. They'll say it changed the whole method of health care delivery."