Limits of the Genetic Revolution FREE

In the mid 20th century, antibiotic discoveries revolutionized medicine. Diseases that once killed thousands could be cured simply and effectively. In a short time, treatment of infection became part of the everyday practice of medicine. While the field of infectious disease remains a strong independent discipline, all physicians—from dermatologists to orthopaedists—have an antibiotic armamentarium tailored to the needs of their patients.

In essence, this same revolution and infiltration is happening with genetics. Some medical specialties, like oncology and pediatrics, have felt the tug of genetics for quite a while because those areas have the largest burden of easily recognizable genetic disease. Oncologists use cytogenetic indicators to diagnose and prognosticate for their patients. In pediatrics, 5% of all patients have a birth defect, and birth defects are the leading cause of infant mortality in the United States.¹

Of course, genetics is not limited to chromosome aberrations, Mendelian conditions, and congenital anomalies. Any medical condition, except trauma, has a genetic component. Work within the Human Genome Project has demonstrated the genetic contribution to common adult conditions like cardiovascular disease,² Alzheimer disease,³ and hypertension.⁴ Thus, physicians who treat these patients will have to become familiar with genetics, genetic counseling, and the mode of action of new genetic treatments.

The genetic revolution will have its turn at reinventing medicine. However, just as the advent of antibiotics did not cure all ills, neither will genetics. It is possible that some conditions will never respond to a genetic remedy, just as there are afflictions not touched by antibiotics. For those conditions amenable to genetic manipulation, there are 3 considerations when creating effective genetic treatment of disease: (1) the relative individuality of genetic problems, (2) the financial disincentive to develop treatment for rare disease, and (3) the need to affect people's behavior. Each of these issues is a stumbling block to the effectiveness of genetic medicine.

For some conditions, the genetic contribution is obvious. These generally follow Mendel's laws of inheritance: achondroplasia, neurofibromatosis, and sickle cell anemia. At the other end of the spectrum are the purely acquired problems like concussion and amputation. Everything else is somewhere in the middle. In Figure 1, those conditions to the far right of the scale are genetic, those on the far left are acquired, and the middle 98% are multifactorial. Some multifactorial conditions will not happen without the genetic contribution, some will remain silent without the environmental exposure.

In Western adult society today, the most common medical conditions are those on the left half of Figure 1. As such, the usefulness of genetic medicine will probably come to fruition first in some types of the common, non–Mendelian diseases such as heart disease and cancer. These conditions have the force of numbers—more patients needing the treatment—even though they are less "genetic" than Mendelian conditions. Treatment may involve direct change of someone's genome, or it may be indirect like stimulating cells to increase (or decrease) production of a particular protein.⁵ Certainly the pharmaceutical companies will be working from many angles. It is likely that much good will be done in this area using information from the Human Genome Project.

In decades of study, genetics has become more rather than less complex. Mendel knew exactly what a gene was, what it did, and how it interacted with other genes. That certainly is not true today. Genetic manipulation or targeted vaccines are not going to be "magic bullets". Complexity will work against genetic treatment: what succeeds for one race, family, or type of disease may very well have no action outside that context.

Genetic heterogeneity must be dealt with at the population level and at the disease level. Differences within a disease population may result from locus heterogeneity, that is, different mutated genes causing the same phenotype, without even considering environmental variation. It is already clear that causes of hypertension vary among races.⁴ One might expect, then, that any genetic-based hypertensive treatment that works for blacks will not work for whites and vice versa. Again, the more ubiquitous the phenotype is in the general population, the greater the possibility that it is there for myriad reasons.

Heterogeneity of genetic conditions may lead to treatments that are either very broad or very specific and the problems echo those of antibiotic therapy: too broad and some patients will be overtreated or have untoward affects; too specific and many different treatment options will be needed. Treatments may need to be tailored to ethnic or population groups. An unfortunate reality is that fear of discrimination or mistreatment has led some groups to eschew participation in medical research in general⁶ and genetic research specifically.⁷ The resulting catch-22 is that directed genetic treatment for those groups will be delayed, which is discrimination and mistreatment.

What about diseases that are relatively uniform across groups? These are more likely to be the Mendelian conditions in which all people affected with the same condition share a common mutated gene. They are going to affect smaller overall numbers of persons and the molecular genetic differences will be fewer, although there will still be mutation heterogeneity. A single gene may have deletion or duplication mutations, insertions, splice junction, or nonsense mutations. In sickle cell anemia, the vast majority of patients have the same exact mutation, but in cystic fibrosis, there are more than 800 mutations documented to date.⁸ A gene-level genetic treatment would have to account for all possible mutations of the gene.

There is also the problem of pleiotropy (a single gene mutation affecting multiple organ systems). The genome is ubiquitous in the body: all genes are present everywhere although they may only be active or relevant in some tissues. When gene mutation in only one organ causes the significant manifestations of disease, it may be possible to treat that organ and ignore the others. This becomes a stickier problem when the organs involved are multiple, as in storage diseases, or difficult to access, like the brain.

The most straightforward conditions to treat will be those with few causative mutations and a single most-commonly involved organ. There are some conditions that are both common and accessible like sickle cell anemia (bone marrow). Some that are common and, though multisystemic, have a primarily involved tissue that is easily accessible such as cystic fibrosis (lung). Others are common but with diffuse tissue involvement and no single target organ like neurofibromatosis. The more organs of the body involved, the more difficult the management and treatment of disease.

So, the molecular complexity of the genome may itself be an impediment to providing genetic therapies. Within a single disease, 2 persons may have different genes involved, different mutations of the same gene, or different manifestations of the same mutation. Treatments that account for all of this will be overly broad; tailoring treatments to the individual patient too cumbersome. As with other types of disease, the most common cause will engender the first and most accessible treatments, which may not be useful for the remainder of the affected population.

Most Mendelian genetic conditions and birth defects are uncommon to rare when considered within the whole population; therefore, for many conditions the patient population may be too small to prompt investment by a pharmaceutical company. For patients with these rare diseases, the promise of genetic therapy may be just as distant today as it was before February 2001 when the first draft of the human genome was announced as completed.

Government-funded research is invaluable and serves as the basis for research and development in both the public and private realms. It is, however, the pharmaceutical companies that have the machinery and money to develop effective treatments. Unfortunately, private development must be rewarded with a financial gain and not threatened by litigation. If the target population is deemed too small, the procedure too laborious, or the treatment too dangerous, the private sector is unlikely to be interested. This is as much a political-financial problem as it is a scientific one.

The Orphan Drug Act of 1983 established market incentives to pharmaceutical companies for development of drugs that treat rare diseases.⁹ The Orphan Drug Act considers pharmaceutical profits and provides government subsidy ($14 million in 1990) to pharmaceutical companies to produce effective medications.¹⁰ The Orphan Drug Act defines "rare disease or condition" as:

(1) in the case of a drug, any disease or condition which (A) affects less than 200,000 persons in the United States, or (B) affects more than 200,000 in the United States and for which there is no reasonable expectation that the cost of developing and making available in the United States a drug for such disease or condition will be recovered from sales in the United States of such drug, [Pub L No. 97-414, 21 USC §360ee]¹⁰

This rather broad definition means that rarity is not solely measured by how many people are afflicted. The first point supports public health needs, particularly those outside the United States. Depending on the interpretation, it may or may not apply to conditions that are more common in the United States, but rare overall.¹¹ The second sense of "orphan" relates to the economics of researching, producing, and marketing the medication. European criteria for the definition of an orphan drug are similar.¹¹ There is concern by a major US consumer rights organization that the Orphan Drug Act protects pharmaceutical profits at taxpayers' expense and that it decreases competition among pharmaceutical companies.¹²

Table 1 and Table 2 list diseases currently targeted by the Food and Drug Administration–designated as orphan drugs.¹³ Those conditions that are Mendelian are reported separately for clarity. Using the argument of the previous section, some conditions in Table 1 do have genetic components but are only now being thought of in that light. Not all drugs initially designated as orphans go on to get marketing approval. Zidovudine for treatment of acquired immunodeficiency syndrome– related complex and acquired immunodeficiency syndrome is designated as an orphan drug.

As of this writing, there are 227 orphan drug designations. Designations and approvals are assigned by effectiveness, so some drugs have more than one target disease or more than one designation within that disease. For example, a single drug may be designated once for prophylaxis and separately for treatment. Somatotropin has 21 separate designations based on treatment goals under different clinical circumstances (one for growth stimulation in growth hormone deficiency, another for growth stimulation in Turner syndrome, and others). (See the first footnote for Table 1.) It is important to note that the majority of conditions for which there are medications covered under the Orphan Drug Act are not numerically rare diseases. Even most of the Mendelian genetic disorders are common within the population.

Certainly the current orphan drug situation is better than it was 30 years ago, and there is hope that research in one set of diseases will lead to treatments in others.¹⁴ However, many drugs being marketed today under the Orphan Drug Act are not for truly rare diseases. Genetic diseases are truly rare. Some genetic disorders affect only a few hundred people. Relative to all the other technological and pharmacological advances, developing a treatment for such a small group is unlikely to happen because (1) there simply are not enough people demanding the treatment, (2) pharmaceutical companies are unlikely to allocate time and resources when they will not realize a profit, and (3) a drug produced for company profit would be so expensive that patients and/or their insurance companies may be unable to afford it. Currently, orphan drugs tend to be covered by insurance companies; however, there is no way to be sure this will continue in the constantly changing atmosphere of cost cutting.

There are already recommendations promulgated to decrease risk for common adult-onset conditions and some congenital ones: do not smoke, exercise daily, drink alcohol in moderation and not at all when pregnant, and others. These reflect only the first steps in our understanding of gene-environment interaction. Smoking is a risk for lung cancer and heart disease, but we do not understand why one smoker is affected very young while another lives past 100 in apparently good health.¹⁵ Likewise, we recognize that risk for spina bifida can be reduced by maternal intake of folic acid.¹⁶ However, neural tube defects are multifactorial conditions, so the risk reduction differs among ethnic groups.¹⁷ The discussion in the "Heterogeneity" section showed the difficulty in addressing the genetic causes of multifactorial conditions. The discoveries and debates about folic acid represent the difficulty in dealing with environmental causes of medical conditions.

Folate is a naturally occurring B vitamin found in legumes and leafy green vegetables. Folic acid is the manufactured synthetic form in multivitamins and enriched foods. In the body, folate plays a role in remethylation of homocysteine to methionine; the biosynthesis of nucleosides; the methylation of DNA, proteins, and lipids; and the levels of homocysteine and methionine. The major circulating form of folate is 5-methyltetrahydrofolate (5-MTHF). The enzyme that produces 5-MTHF is methyltetrahydrofolate reductase (MTHFR). Complete or severe deficiency of MTHFR causes a rare form of homocystinuria.¹⁸ (The common form of homocystinuria results from cystathionine B synthetase deficiency.) The enzyme and its gene are being studied for their role in folate deficiency–related diseases.

The MTHFR gene mutation C677T converts a cytosine to a thymine at base pair 677 that, in the protein, changes an alanine to a valine at amino acid 222.² This mutation is associated in some studies with increased risk for neural tube defects. Homozygous C677T persons have impaired folate metabolism, decreased levels of 5-MTHF, and increased levels of homocysteine.¹⁹ However, this biochemical imbalance can be overridden. Someone homozygous for C667T who takes adequate amounts of folic acid will have normal blood folate levels.²⁰ Thus, we have an environmental compensation for a genetic inadequacy.

Folic acid dietary supplementation has received a large amount of well-deserved publicity in the past decade. It is increasingly apparent that low blood folate levels are associated with increased risk of some birth defects, most notably neural tube defects,^12,21- 22 and of homocysteinemia-associated heart attacks²³ and strokes.²⁴

Herein lies another aspect of disease cure that the Human Genome Project cannot address: preventative behavior. The US Public Health Service, the Institute of Medicine, and the American Academies of Pediatrics and of Obstetrics and Gynecology recommend that all women of childbearing age take 400 µg/d of synthetic folic acid and eat a varied diet with folate-containing foods.^18,25- 26 For various reasons, the Institute of Medicine does not specifically recommend increased folic acid intake for all adults but does acknowledge possible cardiovascular health benefits of the vitamin.

One of medicine's anecdotes is that people do not want to work for improved health, they just want to take a "magic pill." For some conditions and birth defects, folic acid appears to be exactly that magic pill, but, despite that, its use is not increasing dramatically. One difficulty is education. State public health entities, the March of Dimes Birth Defect Foundation, and numerous medical societies have ongoing outreach to both physicians and the public, but these resource-limited agencies have to prioritize their audiences. The second difficulty is availability. Women of lower socioeconomic class, as always at greatest risk of poor nutrition, may be unable to afford vitamin supplements. The value of a $6- to $8-bottle of tablets may be outweighed by the value of a few pounds of hamburger or gallons of gasoline. Food stamps cannot be used to purchase vitamins or medicine.²⁷

Both education and availability are resource issues. That does not make them less daunting, but it does offer the prospect that increased allocation or redirection of money may solve the problems. The third hurdle to folic acid supplementation is the hardest to overcome: turning knowledge into action. Repeated studies show that even people who understand the health benefits of folic acid still do not avail themselves of it.^28- 30 The Texas Behavioral Risk Factor Surveillance Study³¹ data reveal that, of women of childbearing age who knew that folic acid supplementation reduces the risk of birth defects, only 49.1% took folic acid daily. The study also asked women why they do not take vitamins or supplements. The answers were varied and included both predictable answers such as an inability to afford the supplements, or forgetting to take them, as well as unpredictable answers like believing the vitamins cause weight gain and a physician's advising against taking them (Table 3). So, aside from resource issues, there is a mythology or behavior pattern that has to be overcome for the benefit to occur.

A recent survey of psychosocial factors involved in the use of prenatal care indicates that unwanted pregnancy is the single largest factor involved in keeping high-risk mothers out of prenatal care.³² When the pregnancy was wanted, women attended prenatal care, even if there were transportation or scheduling hardships. This suggests strongly that access to and understanding of the importance of the care are insufficient to induce patients to get care, or, in the case of folic acid, to take their vitamins.

Currently available vaccines, vitamins, and medical care are not used to their full extent. Sequencing the genome, and the discoveries to follow, will not decrease smoking and speeding or increase mulitvitamin and sunscreen use. Indeed, one wonders if a misunderstanding about genetics will lead people to believe that they have neither control over nor responsibility for their own health.

The genetic revolution will have a great influence on everyone's practice of medicine. However, like the antibiotic revolution of the last century, genetic diagnosis and treatment will have to overcome obstacles that are scientific, economic, and social. Genome complexity will abrogate simple universal treatments. The cost of producing intricate medications for small populations will frustrate either pharmacutical development or a patient's ability to pay for therapy. It will still be up to the patient to seek out care, participate in treatment, and finish a course of medicine.

Lastly, one must not expect the sudden appearance of cures announced in the next journal edition. Cancer therapy has been under development for decades and, while large advances have been made, and many cures realized, we have not won that war. The mutation in sickle cell disease has been known for 30 years, but treatments are still largely supportive rather than preventive. The first wave of studies in gene therapy were disappointing and no second wave has really started yet. Certainly change and advances will happen faster now than they did at the beginning of the last century, thanks to our new knowledge of genetics, but they will not be happening overnight.

Accepted for publication June 4, 2001.

This article was corrected November 14, 2001.

Corresponding author: Angela Scheuerle, MD, PA, Genetics, Teratology, and Ethics Consulting, 9702 Vinewood Dr, Dallas, TX 75228-3772 (e-mail: angela.scheuerle@tdh.state.tx.us).