CHAPTER 39 PRINCIPLES OF GENETICS
39 PRINCIPLES OF GENETICS BRUCE R. KORF
The elucidation of the structure and function of the genome is one of the great scientific triumphs of the 20th century. The relevance of inheritance to health and disease probably has been recognized throughout history, but it is only during the last century that the rules governing inheritance and the mechanisms whereby genetic information is stored and used have come to light. The application of this knowledge to medical practice so far has focused on relatively rare monogenic and chromosomal disorders. Major contributions have been made in these areas in the form of approaches to genetic counseling, genetic testing, prenatal diagnosis, newborn screening, carrier screening, and, to a limited extent, treatment. As important as these contributions are, however, their impact has been limited by the rarity of these disorders. Powerful tools resulting from the Human Genome Project are changing this situation (Chapter 42). Genetic factors that contribute to common and rare disorders are being identified, leading to new approaches to diagnosis, prevention, and treatment. Genetics and genomics are increasingly occupying center stage in medical practice, guiding treatment decisions and preventive strategies. This chapter reviews the paradigm whereby genetics is being integrated into the routine practice of medicine.
GENETIC CONTRIBUTION TO DISEASE
It may be argued that no disorder is either completely determined genetically or completely determined by nongenetic factors. Even monogenic conditions, such as phenylketonuria, are modified by the environment, in this case by dietary intake of phenylalanine. Genetically determined host factors are known to modify susceptibility to infection or other environmental agents. Even individuals who are victims of trauma may find themselves at risk in part because of genetic traits that affect behavior or ability to perceive or escape from danger.
Complex traits that are important for both health and disease are the result of an interaction of multiple genes with one another and with the environment (Fig. 39-1). In some cases, individual genes or environmental factors contribute overwhelmingly to the cause of a disorder, as with a genetic condition, such as neurofibromatosis or Marfan syndrome, or an acquired disorder, such as bacterial infection or trauma. Other times, there may be interplay among many factors, making it difficult to dissect out the specific genes or environmental exposures. From a medical perspective, it is helpful to divide the genetic contribution to disease into three categories: high-penetrance monogenic or chromosomal disorders, monogenic versions of common disorders, and complex, mul tifactorial disorders. Each of these has an impact on medical practice in distinctive ways.
High-Penetrance Monogenic or Chromosomal Disorders
High-penetrance monogenic or chromosomal disorders are the disorders that most clinicians think of as “genetic conditions.” They include rare but familiar single-gene disorders, such as neurofibromatosis, Marfan syndrome, and cystic fibrosis, and chromosomal abnormalities, such as trisomy 21 (Down syndrome). Several thousand distinct human genetic disorders have been described and cataloged in Mendelian Inheritance in Man (available at www.ncbi.nlm.nih.gov/Omim/). These include mendelian dominant or recessive disorders, sex-linked disorders, and conditions that are due to mutations within the 16.6-kilobase mitochondrial genome. They also include major chromosomal aneuploidy syndromes and syndromes associated with duplication or deletion of small regions of the genome that result in either reproducible syndromes, such as Williams syndrome (deletion of contiguous loci from a region of chromosome 7), or nonspecific mental retardation.
Role of the Nonspecialist
Because of the rarity of many of these conditions, most practitioners have limited experience with a given disorder and are likely to need to refer the
patient to an appropriate specialist for assistance with diagnosis and management. Nevertheless, the nonspecialist has many distinct roles in the care of these patients. These roles begin with the recognition of the fact that the patient may have a disorder and arrangement for appropriate diagnostic evaluation. Many genetic disorders produce obvious signs or symptoms that at least prompt referral even if they are not immediately suggestive of a diagnosis. Others can be more subtle, with nevertheless significant consequences if the diagnosis is missed. An example is Marfan syndrome (Chapter 268). The physician needs to be alert to the physical characteristics of patients with Marfan syndrome because life-threatening aortic dissection can be avoided with appropriate monitoring and treatment. Table 39-1 lists examples of some adult-onset monogenic conditions with which the internist should be familiar.
Treatment of Patients with Genetic Disorders
The treatment of patients with genetic disorders may require the assistance of a specialist, but the nonspecialist is likely to be the first contact when an affected individual is ill. The primary care physician needs to be familiar with the disorder and major potential complications. For example, the patient with neurofibromatosis who experiences chronic back pain may be presenting with a malignant peripheral nerve sheath tumor, requiring more aggressive evaluation than would be typical for an unaffected individual with back pain. Formation of a good working relationship between the specialist and nonspecialist is crucial to ensure effective care. The nonspecialist also has an important role in supporting the patient and helping to explain the difficult choices that may be offered for management. This includes providing support for patients who have disorders that cannot be treated and for the emotional impact that accompanies knowledge that a disorder may be transmitted to one’s offspring or shared with other relatives. Most patients have little understanding of the mechanisms of genetics and genetic disease. Although the responsibility to explain these issues may reside with specialists and counselors, the primary care provider has an important supportive role.
Advances in Genetics
Many of the disorders in this group have been known for a long time, but more recent advances in genetics have had a substantial impact on approaches to diagnosis and management. Genetic testing has been refined with the advent of molecular diagnostic tests that detect mutations within individual genes. Even rare disorders may be amenable to diagnostic testing; a database of testing laboratories can be found on the Internet (available at www.genetests.org). Whole-genome scanning for small deletions or duplications is revealing mutations in patients with disorders such autism, for which standard chromosomal analysis had previously been unrevealing. Population screening for carrier status of some disorders now is offered routinely. Some tests are targeted to particular ethnic groups, such as Ashkenazi Jews (Tay-Sachs disease, Canavan disease, cystic fibrosis, Gaucher’s disease) or individuals of African, Mediterranean, or Asian ancestry (hemoglobinopathies) (Table 39-2). Pan-ethnic screening is now being made available for cystic fibrosis, although risks differ in different ethnic groups. Newborn screening is being expanded beyond inborn errors of metabolism such as phenylketonuria and galactosemia, with the advent of tandem mass spectrometry and the availability of a standardized panel of tests. Finally, treatment of some monogenic disorders is becoming feasible. Life expectancy for patients with cystic fibrosis has been increasing gradually with better treatments for chronic lung disease; dietary therapy is available for many inborn errors of metabolism; novel therapies that use either pharmaceuticals or gene or enzyme replacement strategies are in use or being tested for many conditions. The principles of management of genetic disorders are evolving rapidly so that care of patients increasingly requires active partnership of specialists and primary care providers. Moreover, individuals with congenital disorders such as Down syndrome are routinely surviving to adulthood and require primary care providers who are familiar with their special needs.
Monogenic Versions of Common Disorders
Not all monogenic disorders produce obscure phenotypes, and not all common disorders are due to complex multifactorial causes. Some common disorders occur in some families as single-gene traits (Table 39-3). This is usually true for only a proportion of affected individuals, but in some cases it is a significant proportion and represents an important group of patients to be recognized.
CHAPTER 39 PRINCIPLES OF GENETICS
An example is breast cancer (Chapter 204). About 10% of cases of this common form of cancer can be attributed to mutation in one of two genes, BRCA1 or BRCA2. Women who inherit a mutation in one of these genes face a high risk of eventually developing breast or ovarian cancer—more than 80% by age 70 years for breast cancer. Women at risk because of mutation do not look different from women with sporadic breast cancer but can be distinguished by many features, including family history of breast or ovarian cancer in multiple relatives, early age at onset of cancer, and multifocality of the cancer (e.g., bilateral breast cancer or breast and ovarian cancer).
Colon Cancer and Other Common Disorders
Another example from cancer genetics is colon cancer (Chapter 199). Two syndromes, familial adenomatous polyposis and hereditary nonpolyposis colon cancer, are autosomal dominantly inherited and convey a high risk of colon cancer. Other noncancer examples are hemochromatosis (Chapter 219), in which cirrhosis, cardiomyopathy, diabetes, joint disease, and other problems ensue from excessive iron absorption; 10% of whites carry an allele that predisposes to this recessive disorder. Mutations in the factor V gene or the prothrombin gene occur commonly and predispose to deep venous thrombosis (Chapter 179). Rarer examples include inherited forms of cardiomyopathy, hypertension, and familial hypercholesterolemia.
Presymptomatic Environmental factors
FIGURE 39-1. Multifactorial etiology of disease. An individual is born with a genetic liability but remains in a presymptomatic state for some time until additional events occur, including exposure to environmental factors, that result in crossing a threshold that is identified as disease. In instances of high-penetrance monogenic disorders, the genetic liability may be overwhelming. In other instances, genetic factors may contribute only slightly to disease risk.
The physician may be called on to address these disorders in many ways. There is a compelling reason to make an early diagnosis of hemochromatosis because the complications can be prevented, but not reversed, by phlebotomy and subsequent monitoring of iron stores. Individuals at risk of colon cancer can be offered surveillance with colonoscopy or surgical resection of the colon to reduce the risk of cancer. Individuals at risk of breast and ovarian cancer likewise can be offered surveillance, chemoprevention, or surgery. The benefits of knowledge of genetic risks are less clear in some instances. Carriers of the factor V Leiden mutation would not be treated with anticoagulation until after an event of thrombosis, and the treatment may not be different for a carrier versus a noncarrier. In some cases, however, knowledge of carrier status might help ensure prompt diagnosis or avoid situations of high risk.
TABLE 39-2 MAJOR RECESSIVE DISORDERS FOR WHICH CARRIER SCREENING COMMONLY IS OFFERED IN THE UNITED STATES DISORDER
MAJOR AT-RISK POPULATION
TABLE 39-1 HIGH-PENETRANCE SINGLE-GENE DISORDERS THAT MAY PRESENT IN ADULTHOOD, WITH SOME MAJOR MEDICAL IMPLICATIONS*
White Ashkenazi Jewish
Sickle cell anemia
Southeast Asian, Chinese
Ashkenazi Jewish French Canadian
MAJOR MEDICAL IMPLICATIONS
Risk of aortic dissection; lens dislocation
Long QT syndrome
Arrhythmia, sudden death
Adult polycystic kidney disease PULMONARY
α1-Antitrypsin deficiency AR
Benign and malignant nerve sheath tumors, gliomas
Schwannomas (especially vestibular), meningiomas
Hemangioblastoma of cerebellum, brain stem, eye; pheochromocytoma; renal cell carcinoma
Movement disorder, psychiatric disorder, dementia
Stroke, iron overload
Tumors of thyroid and parathyroid, pheochromocytoma
Globin disorders ENDOCRINE
*See Table 39-3 for examples of lower penetrance disorders. AD = autosomal dominant; AR = autosomal recessive; MEN = multiple endocrine neoplasia; NF = neurofibromatosis.
TABLE 39-3 SINGLE-GENE DISORDERS WITH INCOMPLETE PENETRANCE THAT MAY ACCOUNT FOR INHERITED FORMS OF SELECTED COMMON DISORDERS DISORDER
MAJOR MEDICAL IMPLICATIONS
Cirrhosis, cardiomyopathy, diabetes mellitus
AD, AR: multiple genes
Deep venous thrombosis
Breast and ovarian cancers
AD: BRCA1, BRCA2
Breast and ovarian cancers
Familial adenomatous polyposis
Multiple colonic polyps, colon cancer
Hereditary nonpolyposis colorectal cancer
AD: DNA mismatch repair genes
Colorectal cancer, endometrial cancer
Maturity-onset diabetes of the young
AD: multiple genes
AD: genes involved in cardiac contractile apparatus
Arrhythmia, heart failure
AD = autosomal dominant; AR = autosomal recessive.
CHAPTER 39 PRINCIPLES OF GENETICS
As with other medical tests, the physician should carefully consider risks, benefits, and clinical utility in deciding to use a genetic test. Some distinct ethical and legal risks may apply to some genetic tests. These may include anxiety, stigmatization, guilt, and possibly discrimination for insurance or employment. Some of these risks may be addressed by legislation to maintain privacy of genetic information, such as the Genetic Information Nondiscrimination Act, but the risks of anxiety, guilt, and stigmatization cannot be legislated away. To some extent, further research may improve the basis for surveillance or lead to effective treatments. For now, many of these disorders present a double-edged sword of potentially useful knowledge and potentially harmful information.
Role of the Physician
The role of the physician in dealing with monogenic disorders includes recognition of individuals at risk and participation in formulation of a care plan. Individuals at risk cannot be identified by physical appearance and usually are not evident from medical history or physical examination findings. The most valuable screening tool is the family history. Directed questioning about a family history of major monogenic disorders, especially breast, ovarian, and colon cancer, as well as hypercholesterolemia, hypertension, deep venous thrombosis, cirrhosis, and diabetes, can identify the occasional patient with mendelian segregation of these common disorders. Even if the information is of uncertain reliability, eliciting a family history can prompt referral for further evaluation, documentation of the family history, and consideration for genetic testing. The physician’s job is not simply to identify individuals at risk; some people believe they are at high risk even in the absence of welldocumented risk factors. Addressing these misconceptions can bring peace of mind and usually does not require genetic testing.
Complex, Multifactorial Disorders
provided through a secure Internet site, sometimes with an option for genetic counseling. Although the concept of genomic risk assessment would appear to be an attractive paradigm, many questions may be raised about its practicality and implementation. First, predictive testing is useful only insofar as it guides further management. This is likely to be a moving target because ability to test for risk can be developed more quickly than ability to modify that risk. The utility of interventions may be valued differently by different people. This already has been the case for testing of disorders such as breast cancer. Some women at risk choose not to know their BRCA status because the options, including surveillance or prophylactic surgery, are unacceptable to them. If there were a low-cost, safe, and effective treatment that would neutralize any risk the decision to test would be simple, but short of that, there are reasonable arguments on both sides of the issue of whether to test. For many disorders, it will take a long time to show the efficacy of any intervention because there may be a period of many years between the test and the onset of a disorder. Unless surrogate markers can be identified and followed, the task of proving a benefit to predictive testing may require years to decades in some instances.
Predictive Value of Genetic Testing
A second issue surrounds the degree to which genetic testing would be predictive. Most genetic tests are likely to involve detection of relatively common polymorphic alleles that account for small increments of relative risk of disease. The predictive value of these tests would be modest, perhaps too low to induce an individual to modify behavior or to take medication. Here, again, much depends on the efficacy of any intervention that can be offered. There may be some disorders for which testing would have substantial predictive value and clinical utility and others for which testing would not be justified.
Social and Ethical Issues
Understanding the genetics of common disorders is one of the great challenges of modern medicine, with the promise of major returns in terms of prevention, diagnosis, and treatment. The etiology of these disorders is complex in that they result from an interaction of multiple genes with one another and with environmental factors. The specific genes that are relevant may be different from one person to the next. Identification of these genes is difficult given this heterogeneity and the relatively small impact that any particular gene may have in a particular person.
A third concern relates to social and ethical issues. Will people use test results as an excuse to pursue self-destructive behaviors, having received what may be false reassurance of “immunity”? Will genetic testing further exacerbate the divide between individuals who can afford to pay for their care and those who cannot? Will people misinterpret results of testing in terms of a simplistic notion of genetic determinism, erroneously believing that their futures have been written, leaving them no recourse but to meet their fate? The rapid pace of technologic change is going to challenge the ability of the social and legal systems to keep pace.
Dissection of the genetic contribution to common disease cannot be accomplished by the standard genetic approaches involving study of rare variants or family-based linkage studies. Most recent efforts have focused on study of large groups of patients, comparing the prevalence of particular genetic markers in case patients and control subjects. The availability of markers has been boosted by the identification of single-nucleotide polymorphisms (SNPs) (Chapter 42). These are differences in single DNA bases between individuals that occur every several hundred bases. Some of these account for common genetic differences between people, including differences that may contribute to disease. The catalog of SNPs currently includes several million variants; it has been found that the genome has evolved as blocks of clusters of genes, making it possible to use only a limited number of SNPs within a given region to determine whether there is a gene in that region that is associated with a disease. Since completion of the HapMap Project, there has been a dramatic increase in the number of SNPs found to be associated with common disorders. As genetic risk factors for common disorders come to light, it is likely that there will be advances in risk assessment, disease stratification, and developing new approaches to treatment.
Genetic Risk Assessment
The goal of genetic risk assessment is the identification of individuals at risk of disease before the onset of signs or symptoms. In principle, the genetic factors could be identified at birth, or any time in life, by testing a DNA sample. Individuals found to be at risk might be offered treatment in advance of onset of the disease to avoid complications or might be advised to modify their lifestyle to avoid exposure to environmental factors that might increase their risk of disease. Several companies have begun to offer “personalized genomic testing.” Tests are usually accessed through an Internet site without intervention of a health professional and involve analysis of hundreds of thousands of SNPs in a clinical laboratory. Results are
Finally, there are questions of the ideal context in which to offer such testing. The personal genomics companies provide their services directly to the consumer in most cases. This creates the obvious risk for incorrect interpretation of results by the patient, although it is not clear that the health care work force is otherwise prepared to deal with the challenges of interpretation of genome-wide studies. The challenges will only increase if whole-genome sequencing at modest cost (e.g., <$1,000) becomes available, as is widely anticipated.
A second application of genomics in medical practice entails stratification of disease. Even if genetic testing is not used to predict individuals at risk, it may well be used to determine the most appropriate treatment for a clinically diagnosed disorder. Most common disorders, such as hypertension and diabetes, are symptom complexes that probably result from a variety of causes. The particular combination of causes may differ in different individuals and may respond to different types of treatments. Choice of antihypertensive drug may come to depend on genetic testing to determine the specific cause of hypertension in a patient. There are already examples of genotypes that predict response to drugs, for example, in lung cancer. It is possible that genetic tests eventually will accompany many if not most treatment decisions.
Effects and Identification of Drugs
Aside from helping to choose the most efficacious drug, genetic testing may play a role in avoidance of side effects and in appropriate dosing. Many drugs are known to be associated with rare side effects, some of which are sufficiently severe as to lead the drug to be withdrawn from use. Some of these side effects may occur only in individuals who are susceptible on the basis of having a particular allele at a polymorphic locus. An example is the
TABLE 39-4 GENES IN WHICH COMMON POLYMORPHISMS AFFECT RATES OF DRUG METABOLISM OR ACTION GENE
MEDICATIONS (EXAMPLES) Phenytoin, warfarin
Debrisoquin, β-blockers, antidepressants
association of polymorphisms in certain sodium or potassium channel genes with risk of arrhythmia on exposure to specific drugs. Absorption and metabolism of drugs are largely under genetic control. Several polymorphisms are known to lead to particularly rapid or slow meta bolism, accounting for individuals who experience dose-related side effects or lack of efficacy at standard dosages (Table 39-4). Detection of these polymorphisms would allow customization of drug dosage to an individual’s pattern of metabolism, increasing the likelihood of efficacy without a prolonged period of trial-and-error dosing. There has been a major interest in applying this paradigm to warfarin dosing, although questions remain about the cost-effectiveness and practicality of genetic testing in routine use of this drug. The greatest gift of genetics and genomics to medicine may be in the ability to identify new drug targets and develop new approaches to treatment. Identification of genes that contribute to common disorders is revealing the cellular mechanisms that lead to disease. This knowledge offers the opportunity to develop new pharmaceutical agents that would target the physiologic mechanisms more precisely, leading to drugs that work better and cause fewer side effects. New approaches to gene replacement or insertion of genes into cells as localized drug delivery systems also may be developed. The treatment of common disorders likely would entail the use of approaches developed as a result of genomics even in cases in which genetic testing is not used to predict individuals who are at risk.
The Human Genome Project began after most practicing physicians completed their medical training, and few are familiar with the methods and approaches of medical genetics and genomics. Nevertheless, physicians will be using the products of the genome project increasingly in their day-to-day practice during the coming years. Whether they are providing care for a patient with a rare genetic disorder or for a patient with a common condition not usually regarded as genetic, management choices increasingly will be informed by tests and treatments that in some way are based on information from the genome sequence. The essence of the encounter between a physician and a patient can be distilled to two questions: Why this person? Why this time? A person who seeks medical care is doing so as the product of human evolution, having an ancestry associated with certain genetic vulnerabilities, because of inheritance of certain familial risk factors, because of exposure to some environmental factors, because of a particular physiologic process gone awry, because of behavioral traits that lead the person to seek medical care, because of prompting by family or friends to go to the doctor, because society makes medical services available, and because the person can afford to seek care. Genetics cannot answer all of these questions, but it is providing the key to addressing many of the biologic questions that underlie the medical mysteries that have puzzled humankind for generations. SUGGESTED READINGS Ginsburg GS, Willard HF. Genomic and personalized medicine: foundations and applications. Transl Res. 2009;154:277-278. Overview of the role genetics and genomics increasingly will play in medical practice. Guttmacher AE, McGuire AL, Ponder B, et al. Personalized genomic information: preparing for the future of genetic medicine. Nat Rev Gen. 2010; 11:161-165. Four experts with different insights into the field of genomic medicine. Khoury MJ, McBride CM, Schully SD, et al. The Scientific Foundation for personal genomics: recommendations from a National Institutes of Health-Centers for Disease Control and Prevention
multidisciplinary workshop. Genet Med. 2009;11:559-567. Report of a workshop on issues related to personal genomic testing. Limdi NA, Veenstra DL. Expectations, validity, and reality in pharmacogenetics. J Clin Epidemiol. 2009;63:960-969. Review of the use of pharmacogenetics in medicine. Rotimi CN, Jorde LB. Ancestry and disease in the age of genetic medicine. N Engl J Med. 2010;363:15511558. Review.