|Home | About | Journals | Submit | Contact Us | Français|
Assessment of the penetrance of disease-causing mutations is extremely important for developing clinical applications of gene discovery, such as genetic testing and counseling. Mutations in the leucine-rich, glioma inactivated 1 gene (LGI1) have been identified in about 50% of families with autosomal dominant partial epilepsy with auditory features (ADPEAF), but estimates of LGI1 mutation penetrance have ranged widely, from 50 to 85%. The current study aimed to provide a more precise estimate of LGI1 mutation penetrance.
We analyzed data from all 24 previously published ADPEAF families with mutations in LGI1. To estimate penetrance, we used the information from the published pedigree figures to determine the proportion of obligate carriers who were affected. We assessed whether penetrance was associated with the total number of affected individuals in each family, or mutation type (truncating or missense) or location within the gene. We also compared penetrance in males and females, and among different generations within the families.
Overall penetrance was 67% (95% CI 55–77%), and did not vary according to mutation type or location within the gene. Penetrance was greater in families with more affected individuals, but this trend was not significant. Penetrance did not differ by gender but increased with advancing generation, probably because of limited information about early generations.
Our results suggest that about two-thirds of individuals who inherit a mutation in LGI1 will develop epilepsy. This probably overestimates the true penetrance in the population because it is based on data from families containing multiple affected individuals.
Autosomal dominant partial epilepsy with auditory features (ADPEAF) is an idiopathic focal epilepsy syndrome with auditory symptoms or receptive aphasia as major ictal manifestations.1–6 The most common auditory symptoms are simple unformed sounds such as humming, buzzing, or ringing; less common forms are distortions (e.g., volume changes) or complex sounds (e.g., specific songs or voices). Ictal receptive aphasia, consisting of a sudden inability to understand language in the absence of general confusion, has occurred prominently in some families,7–10 and some affected individuals have seizures precipitated by sounds such as the telephone ringing.3,11 The prominent auditory symptoms and aphasia strongly suggest localization of the epileptogenic zone to the lateral temporal lobe. Accordingly, the syndrome is also called autosomal dominant lateral temporal epilepsy.2
Mutations in the leucine-rich, glioma inactivated 1 gene (LGI1) on chromosome 10q have been identified in approximately 50% of families with ADPEAF,5,6,12–15 but not in families with other forms of familial temporal lobe epilepsy.15,16 More than 20 disease-causing mutations have been reported,17 almost all of which have been unique to an individual family. Germline mutations in LGI1 are seldom found in individuals with symptoms consistent with ADPEAF who do not have a family history.18,19 Two de novo mutations have been identified among about 77 isolated cases screened (2.6%).20,21
Inheritance follows an autosomal dominant pattern and most cases have an affected parent. Children of affected individuals who have a mutation in LGI1 have a 50% chance of inheriting the mutation, but penetrance is reduced, and thus the probability that an offspring who inherits a mutation will develop epilepsy is less than 100%. Previous estimates of LGI1 mutation penetrance have ranged widely, from 50% to 85%. Three studies of individual large families reported estimates of 71%,1 80%,2 and 60–84%.8 In an analysis of eight families, we previously estimated a penetrance of 54% after controlling for ascertainment bias by including only family members that did not lead to family selection.6 In another study of six of these families, we estimated a much higher penetrance (85%) using a statistical model that also attempted to control for ascertainment bias, this time by looking at gene carriers among unaffected individuals.22
Each of these previous penetrance estimates was limited by small sample size, which led to imprecision in the estimates. In the current study, we aimed to derive a more precise estimate by analyzing data from all 24 previously published families with ADPEAF and mutations in LGI1. We also attempted to assess the impact of biased selection of families for study, by examining the effect on the penetrance estimate of the number of affected individuals in the family. Finally, we examined whether penetrance was related to the location of the mutation within the gene, the predicted impact of the mutation on the encoded protein (truncating or missense), or sex or generation, within the published pedigrees.
We reviewed the literature to identify families with ADPEAF caused by mutations in LGI1.1,2,4–6,8,12–15,23–29 To estimate penetrance in the previously published families, we used the information from the published pedigree figures to determine the proportion of obligate carriers who were affected. Obligate carriers were defined as individuals who were inferred to have carried an LGI1 mutation because they had one or more affected offspring. Our rationale for selecting obligate carriers for analysis was that they were likely to have been included in the pedigrees regardless of their disease status, unlike other family members who were more likely to be included if they were affected. Restriction of the analysis to obligate carriers therefore avoided the selection bias that would have occurred if all family members had been included. It also ensured that the individuals included in the analysis had passed through the age period of risk, since all of the obligate carriers had had an affected child. However, our method did not eliminate the bias resulting from inclusion of only multiple-case families for analysis.30
Some of the families had been included in more than one publication (table 1); when this occurred, we selected the pedigree figure with more complete clinical information for analysis. We abstracted information from each pedigree and compiled it into two databases. The first database (table 1) contained family-level variables: 1) the total number of individuals in the family as shown in the published pedigree, 2) the total number of affected individuals (defined as any form of idiopathic epilepsy), 3) the total number of obligate carriers, 4) the number of obligate carriers who were affected, and 5) the specific mutation in the family, classified according to the exon containing the mutation and the predicted effect of the mutation on the gene’s protein product (truncating or missense). The second database (table e-1 on the Neurology® Web site at www.neurology.org) contained individual-level variables for each obligate carrier: 1) gender, 2) affection status (idiopathic epilepsy, yes or no), and 3) generation within the pedigree.
We calculated penetrance as the proportion of obligate carriers who were affected, individually for each family and for all families combined (table 1). The 95% CI for each penetrance estimate was computed as the simple CI for a proportion. We assessed whether penetrance in different families was associated with either the total number, or proportion, of individuals with idiopathic epilepsy in the family, or with the mutation type or location within the gene. Finally, using the individual data for obligate carriers in all families, we estimated penetrance separately for males and females, and for different generations within the families.
We included data from 24 published pedigrees containing a total of 574 individuals, excluding married-in individuals, of whom 177 (31%) had idiopathic epilepsy. Eighty-seven (15%) of the 574 family members were obligate carriers of LGI1 mutations, of whom 58 were affected, yielding an overall penetrance estimate of 67% (58/87) (table 1). The 95% CI for this estimate was 55–77%. Penetrance estimates for individual families ranged from 0 to 100%, and averaged 69% ± 6% (SEM) (table 2).
Penetrance was greater among families with more affected individuals (table 2), but this trend was not significant. The results were similar when we analyzed penetrance according to the proportion (rather than the total number) of affected individuals in the family (data not shown). We also found no difference in penetrance between families with truncating and missense mutations (p = 0.98, analysis of variance [ANOVA]), or among families with mutations in different exons (p = 0.72, ANOVA) (table 2).
In the analysis of the individual-level data (table 3), we examined gender in two ways. First, we assessed the distribution of gender among obligate carriers, to test for a parent-of-origin effect, i.e., a difference in the likelihood that an affected offspring had inherited the gene from his or her mother vs his or her father. Second, we examined penetrance in male and female obligate carriers, to test for differences between males and females in the impact of inheriting a mutation. For 27 of the 87 obligate carriers, gender was unknown because either the published pedigree did not show it, or it was impossible to tell whether the father or mother had transmitted the mutation to the affected offspring. Among the remaining 60 obligate carriers with known gender, females comprised 47% and males 53% (p = 0.61, χ2 test of equal proportions), and penetrance did not differ between females and males (71% vs 75%; table 3).
Penetrance increased from earlier to later generations (table 3, p = 0.04, χ2 test for trend). Most of this effect appeared to be due to lower penetrance in the first generation.
Our analysis of data from 24 published families with LGI1 mutations yielded a penetrance estimate of 67%, implying that an individual who carries an LGI1 mutation has a 67% chance of developing idiopathic epilepsy in his or her lifetime. This estimate is comparable to those of earlier LGI1 mutation penetrance reports, which have ranged from 50% to 85%, but has a narrower 95% CI (55–77%), reflecting its improved precision. This is invaluable information to individuals in families affected by the disorder, and should help in their decisions about whether or not to seek presymptomatic genetic testing. Although a risk of 67% may appear to be high, it also indicates that approximately one-third of mutation carriers will remain unaffected. Thus inheriting an LGI1 mutation is not sufficient to cause epilepsy, and some as-yet-unidentified genetic or nongenetic factors must interact with LGI1 mutations to influence risk. Future research should focus on identifying these genetic or nongenetic cofactors.
Through analysis of the family-level data, we were able to investigate whether the number (or proportion) of affected individuals in each family was associated with LGI1 mutation penetrance. While the difference was not significant, penetrance was higher among families with a larger number of affected individuals. (In this analysis, affected individuals were not limited to obligate carriers, but included all affected individuals independent of carrier status.) This result is consistent with previous data showing inflation of penetrance estimates when they are based on data from families selected because of multiple affected individuals.30 If a mutation is insufficient to cause disease by itself, penetrance will vary according to the frequency of the genetic and nongenetic cofactors that interact with mutations to cause disease. Families with more affected individuals are expected to have a higher prevalence of these cofactors than families with fewer affected individuals, and thus to have higher penetrance. This implies that our penetrance estimate of 67% is an overestimate of the true penetrance in the population. However, because of the rarity of mutations in LGI1, it would be very difficult to obtain an unbiased estimate in a sample unselected by family history of the disorder.
The published mutations have occurred throughout the gene without any notable clustering in any region or exon. (The greater number of mutations in exon 8 than in other exons is probably explained by the larger size of exon 8.) Almost all of the families had unique mutations in LGI1; the only exceptions were two Basque families reported to be from the same geographic region (which might have been related),2,5,13 and two families of unknown relationship from different regions (one Norwegian, the other Italian).8,14,28,29 Thus it would have been impossible to evaluate the effects of any specific mutation on penetrance. However, we did examine whether penetrance varied according to the mutations’ predicted effect on the protein (truncating or missense) or the mutations’ location within the gene, and found no relationship of penetrance to either.
To investigate the etiology of ADPEAF further, we sought to determine if gender might modify the relationship between LGI1 mutations and disease. We found no evidence of a parent-of-origin effect in the transmission of mutations to affected offspring. Also, penetrance was nearly identical in males and females.
Our estimates of penetrance increased significantly from earlier to later generations. We do not believe this finding reflects genetic anticipation in these families. Rather, it is primarily explained by a lower penetrance estimate in the first generation (54%, table 3) than in subsequent generations, which probably resulted from lack of information on disease status in founding individuals in the families. Selective inclusion of later generations in the published families if they contained affected individuals may have also contributed to the effect.
Address correspondence and reprint requests to Dr. Ruth Ottman, G.H. Sergievsky Center, Columbia University, 630 W. 168th Street, P&S Box 16, New York, NY 10032 ro6/at/columbia.edu
Supplemental data at www.neurology.org
Supported by NIH grants R01NS036319 and R01NS043472 (to R.O.) and by a fellowship at the Center for Advanced Study in the Behavioral Sciences at Stanford (R.O., 2007–08).
Disclosure: The authors report no disclosures.
Received March 18, 2008. Accepted in final form May 6, 2008.