Genetic Epidemiology



Genetic Epidemiology


Ruth Ottman

Melodie R. Winawer



Introduction

A genetic contribution to the epilepsies has been suspected for centuries, but until recently, progress in elucidating the specific genetic influences was relatively slow. This slow progress was attributed in part to methodologic problems in research design and analysis, and in part to inherent complexity in the role that genetic factors play in increasing the risk for epilepsy. With the rapid development of research tools in molecular genetics, however, significant advances have been made over the last 10 years, and a number of genes that influence risk for either idiopathic or symptomatic epilepsy syndromes have been identified. Despite these very important advances, most epilepsy is not explained by mutations in these genes, and efforts to identify other genetic influences are continuing.

Identification of genes that influence risk for epilepsy has great implications for public health. It is a first step in investigating the physiologic effects of the susceptibility genes, which can lead to a better understanding of pathogenesis of epilepsy and to new strategies for treatment and prevention. It also facilitates early identification and treatment of susceptible individuals, and perhaps someday, even prevention of epilepsy in some people. See Chapters 17 and 18 for further discussion of these and other related issues.

Gene discoveries also raise important ethical and social issues. For example, when is it appropriate to offer genetic testing, who should offer it, and how should the results be presented to patients (e.g., by the treating physician or a genetic counselor)? These issues have only begun to be considered as the pace of discoveries in genetic research on the epilepsies increases. Some of these issues are discussed in Chapter 19.


Genetic Influences on the Epilepsies: Current Knowledge

So far, almost all of the progress in epilepsy gene identification has come from analysis of rare families with mendelian modes of inheritance (autosomal dominant, autosomal recessive, or X-linked). As of December 2005, 12 genes had been identified in autosomal dominant forms of eight idiopathic epilepsy syndromes (Table 1). All but two of these genes encode voltage-gated or ligand-gated ion channels. In families with benign familial neonatal seizures, mutations have been found in the potassium channel genes KCNQ2 and KCNQ3.18,106 Mutations in the genes encoding three sodium channel subunits, SCN1A, SCN1B, and SCN2A, have been found in different families with generalized epilepsy with febrile seizures plus (GEFS+),30,107,111,112,126 and mutations in SCN2A have also been found in families with a different phenotype, benign familial neonatal-infantile seizures.12,43 Mutations in GABRG2, the gene encoding the γ-2 subunit of the γ-aminobutyric acid subtype A (GABAA) receptor, have been found in families with GEFS+10,38 and in families with childhood absence epilepsy with febrile seizures.50,125 In a large French Canadian family with an autosomal dominant form of juvenile myoclonic epilepsy (JME), a mutation was identified in GABRA1, encoding the α-1 subunit of the GABAA receptor.21 Mutations in EFHC1, encoding a protein with an EF-hand motif that appears to influence calcium currents, were identified in another set of families with JME.114 In three families with an autosomal dominant form of idiopathic generalized epilepsy (IGE) with a range of different syndromes, mutations were identified in the chloride channel gene CLCN2.40 Mutations have been found in the genes encoding two subunits of the neuronal nicotinic acetylcholine receptor (CHRNA4 and CHRNB2) in families with autosomal dominant nocturnal frontal lobe epilepsy.22,108 In families with autosomal dominant partial epilepsy with auditory features (ADPEAF), mutations have been found in the leucine-rich glioma inactivated 1 gene (LGI1), which encodes a leucine-rich repeat protein.49,62,88 The mechanism by which LGI1 influences epilepsy risk is still not well understood, but based on protein homology, it appears likely to be involved in development of the central nervous system.49

Genes have also been identified in a number of mendelian symptomatic epilepsy syndromes. These include progressive myoclonic epilepsies (e.g., Unverricht Lundborg disease, Lafora disease, and the neuronal ceroid lipofuscinoses104), X-linked myoclonic epilepsy with mental retardation,109 and cortical malformation syndromes such as polymicrogyria, pac-hygyria, and periventricular nodular heterotopia.35,61 In addition, mutations in SCN1A have been identified in many patients with severe myoclonic epilepsy of infancy (SMEI).19,69,110,124

Despite the clear importance of these gene discoveries, they apply to only a small proportion of people with epilepsy. Most people with epilepsy have no affected relatives, and only a tiny fraction come from families with mendelian modes of inheritance. In the Epilepsy Family Study of Columbia University (EFSCU),79,83,86,87 we collected family history information on 1,957 people with epilepsy, ascertained from voluntary organizations for epilepsy without regard to their family histories. The proportion of subjects with a positive family history (with one or more first-degree relatives with epilepsy) was 15% in those with IGE and 12% in those with cryptogenic localization-related epilepsy (LRE). Moreover, most of those with a family history had just one affected relative (probands with IGE 77%, cryptogenic LRE 79%), and very few families appeared consistent with a mendelian mode of inheritance.81








Table 1 Mendelian Idiopathic Epilepsy Syndromes with Genes Identified by Positional Cloning (as of December, 2005)















































































Epilepsy syndrome Gene Chromosomal location References
Benign familial neonatal seizures KCNQ2 20q13 106
  KCNQ3 8q24 18
Generalized epilepsy with febrile seizures plus SCN1B 19q13 126
  SCN1Ab 2q24 30,111
  SCN2Aa 2q24 112
  GABRG2a 5q31 10,38
Benign familial neonatal-infantile seizures SCN2Aa 2q24 12,43
Childhood absence epilepsy with febrile seizures GABRG2a 5q31 50,125
Autosomal dominant juvenile myoclonic epilepsy GABRA1 5q34 21
  EFHC1 6p12 114
Autosomal dominant idiopathic generalized epilepsy CLCN2 3q26 40
Autosomal dominant nocturnal frontal lobe epilepsy CHRNA4 20q13 108
  CHRNB2 1q21 22
Autosomal dominant partial epilepsy with auditory features LGI1 10q24 49,62
a Mutations identified in more than one epilepsy syndrome.
b Mutations (many of which are de novo) also identified in severe myoclonic epilepsy of infancy.

In the large group of people with nonmendelian forms of epilepsy, the genetic influences on risk probably consist mainly of “complex” disease genes—that is, genes with only a small effect, which act additively to raise risk, possibly in combination with environmental factors.70 Research is under way to identify these complex epilepsy genes, but progress has been slow and few findings have been confirmed.116 Given that most of the genes identified in families with mendelian inheritance so far have encoded voltage-gated or ligand-gated ion channels, variants in ion channel genes may well contribute to risk for genetically complex epilepsies also.


Other types of genetic effects may also play a role in some cases. First, some “sporadic” epilepsies (i.e., those occurring in the absence of a family history) may be caused by de novo mutations. This mechanism is important in SMEI, where many of the mutations identified in SCN1A have been de novo.19,69,110,124 Second, some epilepsies may be caused by somatic mutations occurring in critical brain regions. Third, mitochondrial genetic defects have been demonstrated to underlie disorders in which epilepsy is a significant part of the phenotype (myoclonus epilepsy with ragged red fibers [MERRF] and mitochondrial encephalomyopathy with lactic acidosis and strokelike episodes [MELAS]),24 and could be involved in other forms of epilepsy as well.82 Finally, some of the genetic influences on epilepsy may involve either genomic imprinting, in which expression of a genotype is influenced by the sex of the parent from whom it is inherited,36 or trinucleotide repeat expansion, involving an amplification of the number of tandem repeats in a DNA sequence rather than a change in the nucleic acid sequence per se.115


Complexities in the Genetics of the Epilepsies

The search for genetic contributions to epilepsy is complicated by a number of factors. The relations between genotype (i.e., the genes that influence risk) and phenotype (i.e., the detectable clinical signs and symptoms) in the epilepsies are not at all straightforward. Even in the genes identified in families with mendelian forms of epilepsy, most of the identified mutations have reduced penetrance: Some of those who inherit the mutation are unaffected. This implies that some other factor—an environmental exposure or genotype at a different locus—is required for phenotypic expression of the mutation. Consequently, in families with multiple affected individuals, one cannot assume that unaffected individuals do not carry a susceptibility gene. A special case of this is age-related penetrance: Because risk for epilepsy increases with age, gene carriers may be unaffected if studied at young ages.

Another complication is etiologic and genetic heterogeneity. The epilepsies are extremely clinically heterogeneous, varying by seizure types, ages at onset, brain localization, electrophysiologic and neuroanatomic abnormalities, response to treatment, and many other factors. These differences are so striking that most epileptologists view the epilepsies as a collection of different disorders, or syndromes, with different etiologies. But to what extent do the different clinical entities also differ with respect to their genetic contributions? Which features can best be used to separate the epilepsies into subgroups likely to share susceptibility genes? The answers to these questions are still unknown. Discovery of clinical features that distinguish between epilepsies with larger and smaller genetic contributions and investigation of shared and distinct genetic contributions to different types of epilepsy are important research goals. Such distinctions could aid in the design of studies aimed at gene identification and greatly refine classification of syndromes.

With locus heterogeneity, different genes influence the risk for the same epilepsy syndrome; hence, different families with the same syndrome carry mutations in different susceptibility genes. This phenomenon is well documented in the epilepsies. Multiple autosomal dominant susceptibility genes have been identified in four syndromes: Benign familial neonatal seizures (KCNQ2 and KCNQ3), autosomal dominant nocturnal frontal lobe epilepsy (CHRNA4 and CHRNB2), GEFS+ (SCN1A, SCN1B, SCN2A, and GABRG2), and autosomal dominant JME (GABRA1 and EFHC1). Moreover, different genetic mechanisms—single gene versus complex—can produce the same syndrome in different families, making it impossible to classify syndromes according to genetic mechanisms. For example, the IGEs are genetically complex in most cases, but some families have autosomal dominant inheritance.21,40,114 Although mutations in LGI1 have been found in 50% of families containing two or more subjects with temporal lobe epilepsy with ictal auditory symptoms,88 most patients with these symptoms are sporadic and do not have LGI1 mutations.15,32

Another complication is variable expressivity, in which a mutation in a single gene produces different epilepsy phenotypes in different individuals. For example, in GEFS+, the seizure disorders in family members who have inherited the same SCN1A mutation can vary from simple febrile
seizures, febrile seizures plus (in which febrile seizures persist beyond age 6 or are accompanied by afebrile generalized tonic–clonic seizures), idiopathic generalized epilepsy, temporal lobe epilepsy, or myoclonic-astatic epilepsy.107 This variable expressivity within families suggests that other genes or environmental factors are involved in the causal pathway leading to a particular epilepsy phenotype. Further, in three of the four genes found to be mutated in GEFS+ families, mutations have been identified in other syndromes also: SCN1A mutations (many of which are de novo) in patients with SMEI,19,69,110,124 SCN2A mutations in families with benign familial neonatal infantile seizures,43 and GABRG2 mutations in families with childhood absence epilepsy with febrile seizures.50,125 Again, this variable expression across families probably reflects the involvement of other genes or environmental factors in the phenotypes under study, although variation in the types of mutations in the gene involved may also play a role.

Finally, the effects of some genotypes on epilepsy may involve gene–environment interaction, that is, the joint influence of genetic and environmental factors in a causal pathway leading to disease.71,72,95 Gene–environment interaction might explain some of the reduced penetrance observed in the epilepsy genes discovered so far. For example, some genotypes might not affect risk directly, but instead might increase susceptibility to the effects of environmental factors. In this case, individuals who inherit the risk-raising genotype would not develop epilepsy unless they were also exposed to the environmental factor; hence, some susceptibility genes might contribute to remote symptomatic epilepsy and even to acute symptomatic seizures as well. In a study by Schaumann et al.,98 seizure risk was increased in the relatives of people who had seizures associated with heavy alcohol consumption (either unprovoked seizures associated with chronic alcohol abuse or acute symptomatic seizures associated with alcohol intoxication), suggesting that some genotypes may interact with alcohol exposure to raise risk. In the same study, however, risk was not increased in the relatives of people with posttraumatic epilepsy.


Research Approaches in Genetic Epidemiology

In the study of a complex disorder such as epilepsy, genetic epidemiologists use a series of study designs to elucidate the genetic contributions on the population, family, and molecular levels. These studies begin with the assessment of familial aggregation: To what extent is the risk of epilepsy (or other disorders) increased in the relatives of people with epilepsy? Evidence of familial aggregation has only limited utility in evaluating genetic hypotheses; significant familial clustering can arise from shared environmental exposures (e.g., air pollutants) or high-risk behavior practices (such as diet) in members of the same family, in addition to genetic factors. Thus, the next step is to use special designs such as twin studies or adoption studies to ask: To what extent is the familial aggregation due to shared genes as opposed to shared environment? If a significant genetic effect is observed, additional studies must be carried out to determine what types of genetic effects underlie susceptibility. At one extreme, the genetic effects could involve single genes with a major effect on susceptibility, whether autosomal or X-linked, dominant or recessive. At the opposite extreme, some genetic influences could be polygenic; that is, they could be a consequence of the effects of a large number of genes at different loci, each of which individually contributes only slightly to the risk. Between these two extremes, some influences might involve pairs or small groups of genes, possibly with interactive (epistatic) effects. One method that can be used to distinguish among these possibilities, segregation analysis, involves examining the distribution of disease occurrence in families and testing its consistency with various mendelian models (autosomal dominant, autosomal recessive, X-linked dominant, X-linked recessive, X-linked dominant). This method has been used in only a few studies of human epilepsies.34,47,68,81,92 Finally, studies must be designed to investigate the molecular mechanisms by which the genes influence risk. At this stage, the specific genes involved need to be identified, and their basic pathophysiologic effects studied. So far, most studies aimed at gene identification have employed linkage analysis and subsequent positional cloning; many newer studies are using allelic association designs. Below we review these approaches and what has been learned from each in genetic studies of the epilepsies.


Collection of Family History Data

Accurate information about seizure occurrence in family members is essential for almost all designs used to test genetic hypotheses about epilepsy and other seizure disorders. In most genetic studies of epilepsy, data are obtained indirectly, in “family history” interviews in which the proband with epilepsy (or a parent or other caregiver, if the proband is a child) is interviewed about seizures in other family members. For many disorders, family history data obtained in this way have low sensitivity—many truly affected relatives are misclassified as unaffected. This problem can usually be remedied by using a “family study” design, in which each relative is examined or given a diagnostic interview directly. Study of the genetic epidemiology of epilepsy presents a unique problem, however, because the diagnosis in both probands and relatives cannot usually be made on the basis of physical examination or laboratory testing. It is essentially historical, based on a description of seizure events that occurred prior to visiting the physician.

Ottman et al. evaluated the validity of family history data on parents and siblings, collected in semistructured family history interviews with adults with epilepsy.83 In this study, many of the parents and siblings of these subjects were also interviewed about their own seizure histories and those of their other family members. The relatives’ self-reports or their mother’s reports were used as the “gold standard” in deciding whether or not they had seizure disorders. The results suggested that adults with epilepsy can report reasonably accurately about epilepsy in their parents and siblings, but isolated unprovoked seizures and acute symptomatic seizures are underreported. Sensitivity for epilepsy (i.e., the proportion of relatives with epilepsy who were correctly reported to have had seizures in the family history interview) was 87% assuming the mother’s report was correct, and 93% assuming the self-report was correct. For other seizure disorders in relatives, sensitivity was only 32% assuming the mother’s report was correct, and 18% assuming the self-report was correct.

Evidence also suggests that family history information on epilepsy is less accurate for older relatives than for younger relatives. One study based on family history reports found an apparent “cohort effect” in the familial risks for epilepsy, with a 50% increase in the proportion of relatives reported to be affected, for each 20-year increase in birth year of the relatives.85 This effect could not be attributed to a real change in risk over time, because the age-specific incidence rates of epilepsy among persons younger than 40 years did not increase during the time periods investigated.42 Instead, the apparent cohort effect was probably an artifact of better recall of recent events than of past events. In the older relatives, a diagnosis of childhood onset epilepsy would have occurred many years before the family history interviews were done, whereas in the younger relatives such a diagnosis would have occurred more recently. Thus, the subjects who were interviewed about their family histories were probably less likely to remember (or even to know
about) epilepsy in their older relatives than in their younger relatives.


Familial Aggregation Studies

Most early studies of the genetics of epilepsy were devoted to assessing familial aggregation, that is, an increased risk for epilepsy in the relatives of affected persons compared with risks in the general population (or relatives of unaffected persons). In familial aggregation studies a sample of people with epilepsy (probands) is ascertained, and then the risk of epilepsy is examined in the relatives of the probands and compared with the risk in the population or in the relatives of controls without epilepsy. The probands may also be divided into subsets based on syndromes or other clinical features (etiology, age at onset, seizure type, etc.), and the risks in the relatives of the different subsets compared. Also, the risks of specific clinically defined subsets of epilepsy (or other disorders) may be examined in the relatives and compared with the risks of the same outcomes in the relatives of controls (or the general population).

Several problems of epidemiologic design and analysis have impeded progress in these studies. Many studies used highly selected populations, possibly introducing bias in the estimates of the magnitude of increased risk. Few studies used comparison groups, rigorous methods for obtaining information about family history, or standardized interview methods. Many studies failed to specify which classes of relatives were included. In the analysis of risks in relatives, age adjustments were seldom made. Definitions of epilepsy were often ambiguous. The outcome of interest in the relatives was seldom clearly defined. Some studies defined only those with epilepsy as affected, whereas others included those with any type of seizure disorder or those with only electroencephalogram (EEG) abnormalities.

The best estimates of familial aggregation are derived from the work of Annegers et al., using population-based data from the Rochester Epidemiology Project.3,5,6,7,76,77,78 The Rochester Epidemiology Project is a unique exception to the usual methods used in familial aggregation studies of the epilepsies and other disorders. It takes advantage of the records linkage system of the Mayo Clinic, which includes essentially all medical, surgical, and pathologic diagnoses of residents of Olmsted County from 1922 to the present, and therefore provides an excellent resource for epidemiologic studies of epilepsy and other disorders.8,59 This system was adapted for collection of genetic information using a three-step procedure that avoids the use of interviews completely. First, probands with epilepsy were identified by searching the records of the Mayo Clinic to identify all children aged younger than 16 years with diagnoses of idiopathic or cryptogenic epilepsy or isolated unprovoked seizures while residing in Rochester after 1935. Second, the records were used to identify the parents of these probands, and all of the other descendants of the parents (i.e., the probands’ siblings, children, nieces and nephews, and grandnieces and grandnephews). Third, the medical records of these relatives at the Mayo Clinic and all other medical facilities serving southeastern Minnesota were reviewed for evidence of seizure dis-orders.

This study design offers several unique advantages for genetic studies. The problem of selection bias is avoided because all incident cases of epilepsy during a specified time period were included; the data on seizure disorders in relatives have high validity, because they are obtained by careful, page-by-page review of the relatives’ medical records rather than by proband interviews; and the clinical detail on both probands and relatives is extensive. This approach would be impossible in most studies, because family members often live in different areas and access to their medical records is very difficult to obtain.80 The only major disadvantages are the limitation in sample size imposed by the relatively small population of Rochester and the restriction to probands with childhood onset, idiopathic or cryptogenic epilepsies, which limited some of the comparisons that could be done in the analysis.

In the Annegers et al. study, the cumulative incidence of epilepsy to age 20 years was 3.6% in siblings and 10.6% in offspring of probands with idiopathic or cryptogenic epilepsy beginning before age 16, compared with 1.7% in the Rochester population.4 The standardized incidence ratios (SIRs) for epilepsy in the relatives of individuals with idiopathic or cryptogenic, childhood-onset epilepsy were 2.5 (95% confidence interval [CI], 1.3 to 4.4) in siblings and 6.7 (95% CI, 1.8 to 17.1) in offspring.5 Risk for unprovoked seizures was not increased in more distant relatives (e.g., nieces and nephews or grandchildren). In a later study of the same population with additional live births and follow-up, Ottman et al.76 found that the SIR for offspring was lower than in the Annegers et al. study: 3.4 (95% CI, 2.1 to 5.1). The lower risks in the more recent study were partly due to a larger number of offspring, leading to greater precision in the risk estimates than in the earlier study. Another possible explanation relates to a difference in study design: The probands in the Annegers et al. study were restricted to incident idiopathic or cryptogenic epilepsy cases with onset prior to age 16, whereas those in the Ottman et al. study were all prevalent epilepsy cases during a specified time period (regardless of etiology or age at onset).






Table 2 Risk of Epilepsy in Siblings, by Etiology of Epilepsy in the Probanda

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Aug 1, 2016 | Posted by in NEUROLOGY | Comments Off on Genetic Epidemiology

Full access? Get Clinical Tree

Get Clinical Tree app for offline access