Genetic advances could spawn incredible improvements in health care. Given public demand, they also pose what may be unmanageable issues of resource use.
At the dawn of the 21st century, we stand on the cusp of the fourth great age of healing: genetic medicine. There are striking similarities between this and the other three. Genomics possesses the seductive mystery of the first era, that of the shaman. It has potential to add years to life, as did the public health advancements of the first half of the 20th century. And it promises miraculous cures, such as a succession of powerful new pharmaceuticals delivered in the second half of the last century. No wonder genomics has caught the public's fancy.
The theory of genetic diagnosis and treatment turns the practice of Western medicine inside out — literally. Instead of starting with disease and searching for its origin, genomics begins with a genetic variation and relies on treatments that manipulate it, often before the gene can express itself in terms of illness. It bespeaks a change in the nature of health care from treating symptoms to predicting health status and taking steps accordingly.
That's a good fit with the disease-prevention goals of managed care. But it also requires highly individualized treatment, which nobody's been fooled into believing will come cheap. Add that to the ever-growing demand for health care services and tricky issues of confidentiality, and you have to wonder whether MCOs are prepared to handle it all.
The answer depends, in part, on how fast the transformation to genetic medicine will come. "Most of us missed the rate of adoption of digital technology by consumers," says Douglas Goldstein, a practical futurist, a recognized strategist, and president of eHealthcare.net. "There's going to be a tidal wave of consumer demand for diagnostic testing and DNA treatment that builds very fast because of accessibility to the knowledge being generated. Consumers drove Internet adoption, and they will demand access to DNA knowledge, tests, and therapies sooner than later."
To be sure, we're not in that new era yet, and some geneticists now are trying to play down the marvels they sold — maybe oversold — to an all-too-eager public. Nonetheless, plans, purchasers, providers, policy makers, and the public all smell change in the air, and are beginning to wrestle with the myriad implications of genetic medicine — the better to capitalize on them and provide access in a way that's fair to all.
When leaders of the Human Genome Project, a group of academic research centers, and its private rival, Celera Genomics, announced in February that the mapping of the human genome was more or less complete, they in essence prepared the medical world for a sharp turn toward the new age of healing. This "biologic periodic table," as HGP leader Francis Collins, M.D., described it, allows us to understand how genes function and interact. To oversimplify, we can use new technologies to analyze genes and unlock secrets of disease susceptibility. In turn, this creates opportunities for the development of gene-based treatments.
Now the action shifts from mapping the genome to finding applications. Some 4,000 single-gene disorders are known. Still more illnesses develop when multiple genes don't work together, meaning that having a single bad gene isn't necessarily a marker for disease. Knowing where to intervene in the biological chain of events that leads to illness is what Scott Gottlieb, M.D., called "disease-pathway medicine" earlier this year in American Medical News.
Complicating this picture, external factors, such as personal health behaviors, can influence recessive genes that can cause illness. We all carry perhaps 5 to 50 genetic variations, most of which will never manifest in illness — though some could if given the proper stimulus. This explains why identical twins, genetically alike, may not develop identical diseases: Expression of their genetic variations can be influenced by forces outside the body.
That will place an emphasis on prevention, says Charles Cantor, Ph.D., chief scientific officer for Sequenom, a San Diego company that develops and uses technologies to reveal one's genetic variations. "Think of how much prophylaxis there is now for gum disease. In the past, people waited until their teeth fell out, then dealt with it," he says. "We'll see that kind of shift in obesity, hypertension, and cancer, where environmental factors can be manipulated to compensate for genetic risk."
Cantor, who was one of the founders of the HGP, sees great promise in replacing nongenetic diagnoses with genetic ones, which ultimately could lead to better risk stratification. "Women are subjected to annual mammograms without regard to their genetic risk for breast cancer," he says by way of example. "If we knew some women had an insignificant risk, why undergo the mammogram? With this type of diagnosis, we can tell who needs vigilant surveillance from who doesn't need any at all.
"For managed care," he continues, "this is going to save money. A genetic test is done only once. If you can replace a recurring test with a genetic test, you'll see a rather quick impact."
Already, genetic tests are available for about 500 disorders; most are rare, but some are not — cancers, sickle cell disease, epilepsy, deafness, autism, and Alzheimer's, to name a few. More tests for common problems are in the pipeline. Digene, a Maryland company, is developing a test that, in recent National Cancer Institute-sponsored trials, was 14 percent more accurate than Pap tests in detecting cervical cancer.
As for treatments, GenVec, a Maryland biotech company, has begun human testing of a genetic drug that causes the body to grow new blood vessels around clogged arteries — a bio-bypass. Other applications, envisioned but not yet developed, are equally fascinating. An analysis in the journal Science in February reckoned that genomics will revolutionize treatment of mental illness, not by correcting chemical imbalances in the brain, as today's psychiatric medications do, but repairing damaged genes. This has far-reaching social implications: Prisons, the authors noted, are filled with the mentally ill, the addicted, and the antisocial.
"Very rapidly, we'll begin to learn about specific genes that predispose for heart disease, alcoholism, and more," says Kathy Hudson, Ph.D., director of policy and public affairs for the National Human Genome Research Institute. "I think we'll have a handle on those players in three to four years."
Hope or hype?
Such fantastic talk creates high public expectations — some, perhaps, premature. Even after the players are identified, effective treatments would be years away from approval. "You run the risk of raising false hopes," Hudson acknowledges.
Steve Jones, Ph.D., in a recent paper for the Milbank Memorial Fund, "Genetics in Medicine: Real Promises, Unreal Expectations," noted that while there have been some useful applications of genomics, such as prenatal diagnoses, few of the promises of the last 20 years have been fulfilled.
"It has been said that the four letters of the genetic code are H, Y, P, and E," wrote Jones, a professor of genetics at University College London, "and medical providers must realize that the molecular biology business is as adept at promoting its wares as any other."
In fairness, translating theory into practice can be analogous to Sisyphus pushing his rock. But in pointing out that the genetic errors that cause sickle cell anemia, cystic fibrosis, Huntington's disease, and muscular dystrophy have long been understood, Jones observed that such discoveries have not led to treatments. "To complete the map of DNA was a triumph of genetics as a science. Its success as a technology, however, has yet to be established."
And so, some say, sea change in medical practice may not be so sudden. "The thought of going for your annual physical and having your blood drawn for every genetic disease is probably still more science fiction than reality," says David Shulkin, M.D., chairman and CEO of DoctorQuality. Shulkin, former chief quality officer/chief medical officer for the University of Pennsylvania Health System, foresees a gradual increase in the number of conditions that can be screened with genetic tests over the next 5 to 10 years. "Once people at risk are aware of that, they'll go to their doctors for testing."
Others aren't certain this discussion will be academic for long. And with consumer demand being what it is ("Just give me the pill, doc"), it's not a stretch to suggest that there will be a clamor once treatments that keep people from dying of what ailed their parents, or that end their back pain, or that restore their natural hair color, are available.
"There will be a huge demand, driven by the baby boom population — which has multiple reasons to move behind this: their own health, the health of their parents, and the health of their kids," says Goldstein. "The baby boomers have defined every major event that has driven our economy."
As it is, allocation of health care resources is a field strewn with land mines. Are the promises of genomics propelling us toward a very nasty public discussion about rationing?
"We're always having that conversation when you talk about the relative importance of tests; this is no different," says Reed Tuckson, M.D., senior vice president for consumer health and medical care advancement at UnitedHealth Group. "We should not assume that new technology is inappropriate and wasteful, but we have to learn what makes sense. That's not a burden — it's exciting to anyone who relishes the intellectual challenge."
For employers and those who set policy at health plans, the dilemma of gauging the medical necessity of genetic tests and treatments is very real. To illustrate, Helen Darling, senior consultant at Watson Wyatt Worldwide, uses the example of a woman who doesn't have breast cancer but has close relatives who have had it.
"You can decide whether to get tested for the gene; if you test positive, there's still a high probability that you will not get the disease," she says. "You can have mammograms more often. You can be put on tamoxifen prophylactically, though even that's controversial. As a further step, you could have a preventive radical mastectomy. Would this be considered medically necessary? Technically, no."
Just to carry the example further, Darling turns around the hypothetical probability of disease. "Let's say testing indicates there's a 50 percent chance you'll develop cancer, and your doctor agrees that could be true. You go to your carrier and say, 'I want a radical mastectomy. You can't tell me that's not medically necessary.' To what extent do you treat the probability as a condition? Does all the treatment around that become medically necessary, and therefore covered?"
The potential for undertreatment poses equally vexing questions. Most breast cancer patients had no known risk factors. "Will talk about relative risk falsely lull people into complacency, failing to be tested on a reasonable schedule?" Darling wonders.
Tough questions, and health plans and employers are waiting for some guidance. So much of this is new that there is little evidence to suggest when genetic services are appropriate. A search of the U.S. Agency for Health Care Research and Quality's National Guideline Clearinghouse in late April yielded just nine protocols for genetic testing — three for colorectal cancer, and six for other conditions.
Kaiser Permanente, which runs one of the largest genetics programs in the world, contributed two of those guidelines — breast cancer and prenatal care. Ronald Bachman, M.D., chief of the Department of Genetics at Kaiser Permanente Medical Center in Oakland, hopes that groups like the American College of Medical Genetics will take initiative in developing guidelines, but he understands the difficulties.
"I was on the national team that developed the breast cancer guideline. It's a very time-consuming task to develop them, get agreement, and educate people about them," he says. "You have to keep up with all that's going on in the human genome project, then find the time to write those guidelines."
That may explain why Bachman has yet to see any protocols for use of predictive testing (the two KP guidelines excepted). "We have a comprehensive, agreed-upon approach to Huntington's disease. We've come to some agreement on hemachromatosis. Some other predisposition testing is determined by the disease. We meet frequently, we jawbone these things, and come up with good agreement as to how we'll approach them, but I must admit, we haven't put them in writing."
Darling thinks determinations about whether something meets evidence criteria — and thus might be considered for coverage — should be made by people without a stake in the outcome.
"It ought to be considered not just by scientists directly involved in the technology — they're the true believers — but by outside experts with appropriate training who might say, 'It may work, but there's not evidence that it works.'" That's important, she says, because "Some things come online that turn out to be harmful. It's tough to decide how quickly to let things get paid for, because once they're in practice, it's hard to get them out, even if they do harm."
If there was a time to craft guidelines, it's now — before the crunch. In their absence, health plans could be forced to decide coverage on factors other than evidence — such as consumer demand or political mandates.
It can be argued that judicious application of tests offers potential for significant cost reduction.
There are thousands of genetic variations, the sheer number of which will, in all likelihood, force physicians who order genetic tests to rely on computer databases for medical decision-making support when interpreting results. While that will make testing itself relatively expensive, consider two examples of how it can save expense in the system.
First, 5 to 10 percent of people carry a gene that prevents metabolism of azathioprene, an immunosuppressant. "You can have side effects that can land you in the hospital or kill you," says Bachman. "It's cost-effective to give somebody a $260 test for the genetic variation before you start therapy."
Second, it takes trial and error to learn which drugs some patients will respond to. "To use the shotgun approach — doing combination treatment after combination treatment — doesn't make sense," says Rob Koska, vice president for marketing at Vysis, an Illinois genetic-test manufacturer. "Your genetic profile can determine which treatment will be most beneficial. If you're going to be most responsive to a third-line therapy, you may want to start with that one."
Much of what we pay for in medicine is inefficiency — waste that genetic testing can help to eliminate. Koska mentions a Vysis genetic assay, under review by the FDA, that helps clinicians select appropriate candidates for trastuzumab, which is used to treat some metastatic breast cancers.
"There's a lot of controversy as to how to select those patients," he says. "A regimen of trastuzumab costs $30,000. To the extent that you can identify patients up front who won't respond, or will suffer unacceptable side effects, you're delivering better health care. The cost of a genetic assay, in comparison, is minor."
Any pharmacy directors out there who see potential here, raise your hands. Lots of hands. OK, now the lab guys. Uh, huh — not so many. It's this notorious fiefdom mentality within HMOs that Koska runs into time and again.
"You can't find the person who makes the decision for an entire system — everyone looks at his specific department. Medical laboratory has little interest in a diagnostic modality that is more expensive but will save pharmacy the drug costs. It's critical to the adoption of these technologies that they understand the total medical economic picture."
Greater Food and Drug Administration oversight of test manufacturers may boost insurers' willingness to cover certain genetic tests. The FDA requires clinical trials of products that are marketed to others, such as those by Vysis, but not of so-called home-brew outfits that develop proprietary tests and contract with health systems for their services. Last year, the Secretary's Advisory Council on Genetic Testing, a Health and Human Services panel, recommended that home-brew tests developed with private funds be subject to FDA regulations.
Tuckson, of UnitedHealth, sits on the SACGT. He says plans and policy makers walk a fine line between preserving access to care and preventing costs from escalating needlessly.
"We will be bombarded by a lot of snake-oil salesmen over the next few years. There will be 1,001 tests for things like what color your baby's eyes will be," he says. "The question for society becomes, at what level should we be paternalistic to protect ourselves from the environment?"
It works two ways
Darling thinks that while genetic services will eliminate a certain amount of waste in medicine, they will create a new breed of inefficiency. "On a net basis, it will cost more," she asserts. "There will be people who take a lifetime of genetic drugs to keep the switch [the expression] from occurring. We'll be treating everybody prophylactically, and the sheer value of that will drive costs up."
As of last month, 369 genetic drugs were in development in the United States. Pharmaceutical companies are pumping billions into compounds that can turn genetic receptors on or off and that can be tailored to address one's personal genetic variations. Once successful products come to market, their price tag will be anyone's guess.
"The big-money decisions will involve treatments, not tests," says Robert Field, Ph.D., J.D., M.P.H., director of the graduate health policy program at University of the Sciences in Philadelphia. "As we customize medications, the market for them shrinks. It's almost as if every drug will become an orphan drug with a market of one. The question becomes, how does the financing work?"
Cantor has an idea. When genetic treatments finally do get to clinical trials, the genetic variations they can affect will be studied. "When the therapy emerges on the market, it will be bundled with diagnostic tests," he predicts. The test results will determine coverage: "You will have side effects to this medication, so you shouldn't take it," or "You're a nonresponder, so there's no reason to pay for you to take this drug — it won't do you any good.'"
That would turn health plans' business model on its ear. But then, genomics has the potential to blow out the whole concept of insurance as we know it.
"If genetic testing ... sheds light on the predisposition of individuals to disease, then the concept of experience-rated health insurance ... becomes seriously lunatic," futurist Ian Morrison, Ph.D., wrote on the GeneSage web site last year.
Field puts it more delicately. "Insurance depends on the notion of fortuity — that the event insured against occurs by chance. If, through genetic screening, you know what your likely conditions will be, that element of fortuity is eliminated. People with favorable genetic profiles may opt out of insurance, leaving only those likely to become sick." That forces rates up, which in turn causes more people to forgo coverage — and sending the insurer into what the industry so elegantly calls a death spiral.
Cantor sees potential for either the insurer or the insured to gain an unfair advantage if one has genetic information that the other doesn't. To resolve that, and to preserve health coverage in general, he suggests merging medical and disability benefits with life insurance and retirement vehicles (such as pension plans), to create a single security blanket he thinks is fair to all: quality-of-life insurance.
Under his idea, the cost of maintaining a desired quality of life is computed, then premiums are adjusted up or down, based on genetic risks, personal health behaviors, lifestyle, and willingness to choose various levels of insurance. It's the kind of thing that would make privacy advocates squeal, but as Cantor says, "I can't see any alternative."
If the last four paragraphs haven't sent you into shock, there are strategies plans can enact to prepare for the new era of healing — if not thrive in it.
Develop policies on underwriting. Field suggests that MCOs do this to prevent getting burned by members who already know their genetic susceptibility to disease. HIPAA would limit this for most group policies, but, he says, "With regard to individual coverage, insurers may be concerned that the person has information that they don't."
Think collaboration. Field sees opportunities for health plans to work with drug companies to develop mutually advantageous treatments (based, for instance, on an insurer's population).
Consider marketing strategies. Goldstein thinks a few leading-edge plans will want to be perceived as leaders, and will proactively offer tests and certain high-cost treatments. "If we end up with a genetic-based treatment to change insulin production, and if diabetes involves 3 percent of a health plan's members but drives 12 to 15 percent of its costs, you have a home run," he says. "It's risky, but it can have a lot of reward, because it recognizes that the patient is the center of the universe."
Smart organizations, says Goldstein, will adopt a strategy he calls "4D health care" — one that analyzes the four major drivers of change in health care: DNA science, digital technology, "doing self-care by patients," and the "D-force, which harnesses dedicated leadership and creative energy for change." Understanding the characteristics of these can help health care companies develop a service portfolio that capitalizes on them.
Select qualified providers. A shortage of medical geneticists and genetic counselors will eventually force primary care physicians to offer at least some genetic services. But genetic medicine requires expertise that most physicians do not have.
Shulkin's company, DoctorQuality, contracts with health plans to provide information about physicians to members. "Consumers are going to need to know how to find somebody who's up to date on these services and is using them in a way that is safe and consistent with best practices," he says. DoctorQuality, through its web site, provides members with information about physicians' advanced training in genetics (among other specialties), as well as descriptions of their experience and evidence of adherence to professional standards.
Until recently, it had been thought that genetics accounted for, perhaps, a third of illness. But as we learn more about genetics, it appears to be a far greater factor. "You take all the major chronic diseases of adults, they're in there: heart disease, atherosclerosis, schizophrenia, Alzheimer's, diabetes, and how people respond to medications," says Bachman. "They're all intertwined with genetics."
"Ten years from now, much of this discussion will be academic," says Matthew Wynia, M.D., M.P.H., director of the AMA's Institute for Ethics. "Diseases that are now thought of in the nongenetic bucket will be moved over to the genetic bucket."
"There are no buckets," Cantor replies with certainty.
And while that may seem daunting, it provides plans with an opportunity to get physicians and members to think about resources and prevention.
"This is exceedingly compelling stuff," says Wynia. "The fact that people find this interesting is a great thing — it's a unique opportunity to get people involved in a policy discussion."