0
Get Alert
Please Wait... Processing your request... Please Wait.
You must sign in to sign-up for alerts.

Please confirm that your email address is correct, so you can successfully receive this alert.

1
Reviews and Overviews   |    
New Ethical Issues for Genetic Counseling in Common Mental Disorders
Elliot S. Gershon, M.D.; Ney Alliey-Rodriguez, M.D.
Am J Psychiatry 2013;170:968-976. doi:10.1176/appi.ajp.2013.12121558
View Author and Article Information

The authors report no financial relationships with commercial interests.

Supported by NIH grant 1R01MH094483-01A1 to Dr. Gershon and a NARSAD Young Investigator Award to Dr. Alliey-Rodriguez; by the Eklund family; and by the Geraldi Norton Foundation.

From the Department of Psychiatry and Behavioral Neuroscience, University of Chicago, Chicago.

Address correspondence to Dr. Gershon (egershon@yoda.bsd.uchicago.edu).

Copyright © 2013 by the American Psychiatric Association

Received December 13, 2012; Revised March 06, 2013; Accepted March 11, 2013.

Objective  Recent genetic findings of high-impact genetic variants in bipolar disorder, schizophrenia, and autism spectrum disorder (ASD) must lead to profound changes in genetic and family counseling. The authors present risk calculations, discuss the ethical implications of these findings, and outline the changes now required in the risk counseling process.

Method  The authors use data from recent mega-analyses and reviews of common and rare risk variants in bipolar disorder, schizophrenia, and ASD to calculate risks of illness based on genetic marker tests. They then consider new ethical issues in mental disorders presented by these risks, including within-family conflicts over genetic testing; effects of genetic discoveries on stigma, abortion, preimplantation procedures, and population screening for susceptibility; and genetic tests as a factor in marital choice.

Results  New structural mutations (de novo copy number variants [CNVs], which are chromosomal microdeletions and microduplications) are present in 4%−7% of patients with bipolar disorder, schizophrenia, or ASD and can occur almost anywhere in the genome. For a person with a de novo CNV, the absolute risk of bipolar disorder, schizophrenia, or ASD is 14%, much higher than the population risk. Rare CNVs have also been identified that are generally not new mutations but constitute very high-effect risk factors, ranging up to 82%.

Conclusions  A substantial minority of patients with bipolar disorder, schizophrenia, and ASD have high-impact detectable genetic events. This greatly changes psychiatric genetic counseling for these patients and families. A psychotherapeutic approach may be needed as a routine part of risk counseling, particularly for resolution of ethical issues and for within-family stigma and conflicts over genetic test results.

Abstract Teaser
Figures in this Article

Public acceptance of genome-wide screening tests is rapidly increasing, spurred on by recent recommendations that prenatal screening be generally performed by genome-wide methods (1). Expert counseling will be needed to interpret genome-wide test reports, and recent research results will undoubtedly extend this need into the genetics of the major psychiatric disorders. Based on recent research, the risks of bipolar disorder, schizophrenia, and autism spectrum disorder (ASD) are greatly increased for individuals with specific genetic markers, and this applies to a significant minority of all patients. Psychiatric risk counseling has thus changed from risk estimates based on family history to estimates based on test results in specific individuals.

New ethical issues may be expected to arise in the course of this counseling. Ethics is the system of moral principles that serve to guide judgments and behavioral choices. The ethical issues that may arise include abortion and preimplantation selection of embryos based on likelihood of psychiatric disease; rights of family members to genetic information on other members, as well as stigma and conflicts within families over who is a carrier; stigma resulting from community- and ethnicity-based genetic information; population screening and prevention based on genotype; and the role of genetic tests in marital choice.

With these issues arising, the process of genetic counseling in common diseases may need to be revised to go beyond presenting risk estimates. Counseling would normally take into account the ethical and community or family contexts, as well as the perceptions of disease burden, of the persons seeking consultation. Risk counselors would now need to be prepared to offer short-term psychotherapy as part of the consultation process to address the narcissistic injuries and interpersonal issues that may arise over genetic test results.

The illness risk that is generally quoted to families seeking genetic counseling is actually a familial risk—the risk that was estimated from epidemiological observations of parent-to-child transmission before genetic analyses were available. For example, for a mother with schizophrenia, the risk that her child will have schizophrenia is about 10%, which is elevated compared with the general population prevalence of a little less than 1%. For a woman who is not ill but has a sibling with schizophrenia, the risk that her child will have schizophrenia is about 3%.

Current research efforts have been directed toward the use of various human genome maps, incorporating variations in DNA sequence, to map the genes responsible for the familial inheritance. Common single-nucleotide polymorphisms (SNPs) are changes in DNA that are used for mapping in genome-wide association studies (GWAS). In GWAS, as most recently implemented, approximately 1 million SNPs distributed throughout the human genome are tested for association with illness, using microarrays of markers. Each marker is chosen to be one of a set of markers that are inherited together, so it “tags” many nearby markers. When these sets of SNPs were developed for genetic studies of disease over the past decade, it was expected that common diseases would be associated with common genetic variants (that is, with polymorphisms whose disease-associated variant was common in the population). This expectation was developed as a result of theoretical modeling of the evolutionarily rapid expansion of the human population over the past 100,000 years (2); this model became known as the “common disease–common variants” model.

As GWAS results became available for many common diseases, the discoveries of SNP associations with disease turned out to be rather weak. One measure of effect size of a genetic association is the odds ratio, a relative frequency ratio measure of a marker in patients compared with controls. In bipolar disorder and schizophrenia, these are most often in the range of 1.1–1.3, which is a weak effect size, whereas in rare diseases such as Huntington’s disease, the odds ratio is thousands of times higher. Also, the fraction of heritability of illness accounted for by all the GWAS markers in a microarray was disappointing. Lee et al. (3, 4) estimated the total heritability that could be accounted for by every GWAS-based genetic marker in published studies as of 2011. Their analyses define heritability to include all additive genetic effects of the markers on disease, including multiple markers whose effects could not be individually demonstrated but which collectively had an impact on disease risk (polygenic inheritance). They found that 23% of the phenotypic variance in schizophrenia could be accounted for by the genome-wide markers, 40% in bipolar disorder, and 22% in inflammatory bowel disease. Thus, most of the disease variance in the population is not accounted for by genome-wide SNP markers, even though these diseases have genetic components.

At a recent scientific meeting, a mega-analysis of tens of thousands of patients and similar numbers of controls was presented (5). In that analysis, the number of schizophrenia-associated markers increased to 62, but the proportion of disease variance detected by associated markers was only 6%. The reason is that very large samples using current platforms (with common polymorphisms) uncover weaker effect associations than smaller samples (6, 7). This percentage may increase somewhat as the number of markers increases, as less common variants become tested, and when complete genome sequences become available for large numbers of individuals, but it is unlikely that there will be clinical predictive value for specific common SNP polymorphisms in risk counseling.

ASD has recently been considered a common disease with growing prevalence (8). The evidence for common frequency of ASD has been criticized (9). It has been proposed that reported diagnoses for individuals with the same clinical features have been shifting from mental retardation to autism in recent years (10). In any case, ASD is frequently the subject of genetic counseling inquiries, and it shares overlapping genetic events (rare and de novo copy number variants [CNVs]) with bipolar disorder and schizophrenia.

In the research literature, there is a relative paucity of GWAS data on ASD as compared with bipolar disorder and schizophrenia, and there have been few reports of associations that meet current GWAS thresholds of significant probability values (p<5×10−8). The only significant associations, reported by Wang et al. on behalf of a large consortium (11), showed five SNPs between CDH10 and CDH9 on 5p14.1 with significant case-control p values. The odds ratios were between 1.15 and 1.19. Devlin et al. (12) have proposed that there may be additional common loci with similar odds ratios and that these have not been discovered so far because the sample size in the case-control studies has been marginal. Devlin et al. also noted that there were three large case-control studies as of the time of their review, with no replication among them.

Thus, with this method, individual genes cannot yet be identified that can account for the familial transmission of risk for these three major psychiatric illnesses in most families. It may be that “private” mutations in single families are responsible. Some of these will be SNPs, but others may be part of a new class of mutations, the CNVs.

The advanced genomic technologies applied to bipolar disorder, schizophrenia, and ASD to detect variation in DNA as SNPs also detect a second kind of variation, which has proven to be more directly linked to psychopathology. Large chromosomal microdeletions and microduplications, ranging in length of disruption from 50,000 to several million DNA base pairs, are found as rare variants or new mutations in a small but substantial minority of patients. They are called “copy number variations” because they result in the individual no longer having the usual two copies of each gene, one on each of his or her paired autosomal chromosomes (chromosomes 1–23). Instead, a microdeletion on one chromosome results in the individual having only one copy of 1 to 20 or more genes, because they have been deleted on that chromosome and are present only on the other chromosome. Similarly, a microduplication results in the individual having three copies, two on the chromosome with the microduplication and one on the other chromosome.

In the mid-2000s, it became apparent that CNVs are responsible for a significant proportion of human genomic variation (13, 14). These events occur largely in regions of segmental duplication, an extended DNA sequence that is repeated once or a few times within a genome (1517). As a result, many mutations generating CNVs are recurrent mutations at segmental duplication sites. Karayiorgou et al. (18), in 1995, discovered the first rare CNV associated with schizophrenia, a large CNV with millions of DNA base pairs on chromosome 22. This discovery languished for a decade, until it became possible to efficiently scan the entire genome for CNVs, including by reanalysis of the data contained in GWAS microarrays. Because there were many thousands of individuals in GWAS data sets, associations with rare CNVs could be detected. It then became apparent that rare CNVs were associated with many diseases (19), most particularly with diseases of the CNS (2022). These rare CNVs have much higher effects on illness risk than common SNPs (Table 1), and several are associated with more than one of the three psychiatric diseases discussed here.

 
Anchor for Jump
TABLE 1.Rare CNVs: Risks of Illness for Bipolar Disorder, Schizophrenia, and ASDa
Table Footer Note

a ASD=autism spectrum disorder; CNV=copy number variant. Risks of illness are based on Bayesian probabilities. Data are from the comprehensive review of Malhotra and Sebat (22). The criterion for inclusion in this table was association of CNV with illness at p<0.003.

Furthermore, the mutation rate for appearance of new CNVs (de novo CNVs) is orders of magnitude greater than for SNPs. As originally reported by Sebat et al. for autism (23), Xu et al. for schizophrenia (24), and later confirmed and extended by Malhotra et al. (25) and others as reviewed elsewhere (22), this mutation rate is elevated in bipolar disorder, schizophrenia, and ASD. The rate of individuals with a de novo CNV anywhere in the genome is 4.3% in cases of bipolar disorder and 6.1% in schizophrenia, compared with 0.9% in controls (25). The odds ratios are 4.8 for bipolar disorder and 6.3 for schizophrenia, which represent much higher genetic effect sizes than odds ratios found in common SNP associations. Taken together (Tables 1 and 2), the CNV findings represent an important class of risk factors for bipolar disorder, schizophrenia, and ASD, and generate a profound change in the genetic counseling for them, as discussed below. Although each specific CNV mutation is rare, the presence of a mutation anywhere in the genome is common, so de novo CNVs are actually a common cause of bipolar disorder, schizophrenia, and ASD. The fact that the de novo CNVs associated with psychiatric diseases can occur almost anywhere in the genome suggests that they disrupt what are otherwise minor polygenic susceptibility locations.

 
Anchor for Jump
TABLE 2.De Novo CNVs: Attributable Risk and Risks of Illness for Schizophrenia, Bipolar Disorder, and ASDa
Table Footer Note

a ASD=autism spectrum disorder; CNV=copy number variant. Exposed attributable risk is the attributable risk in exposed individuals (i.e., ill individuals who have a de novo CNV), which is the proportion of the ill person’s risk attributable to the CNV.

Table Footer Note

b Based on Bayesian probability. Data are from Xu et al. (24), Malhotra et al. (25), and Sebat et al. (23). Computation of illness risk of any disorder is 1−(1−P1)(1−P2)(1−P3), where Pi is the risk for each disorder. This calculation indirectly accounts for the probability of the occurrence of more than one disorder in any individual as a product of the probability of each diagnosis.

In ASD, de novo CNVs have been reported by comparisons of case families and control families (23), as well as by comparisons of de novo CNVs in case subjects and unaffected sibs from the Simons Simplex collection and separately in the international Autism Genome Project Consortium collection (2629). The calculations presented here on risk associated with de novo CNVs for ASD are based on the data of Sebat et al. (23), because those are the only ASD data with independent estimates of patient and control rates of de novo CNVs (7.2% compared with 1.0%).

The CNV data reviewed here on ASD include some CNV frequencies derived from lymphoblastoid cell lines, which can have artifactual CNVs. In the first autism study of de novo CNVs, some patient and control data were taken from cell lines (23), whereas all the published de novo data for bipolar disorder and schizophrenia are from whole blood. The sample source would not bias the association findings of CNV burden in case-control studies that use the same sample source for both groups. The de novo CNV rates observed in controls are about the same in all the studies of each of the three diseases, so there does not appear to be inflation of CNV rates in the autism study.

From an epidemiological and counseling viewpoint, the major impact on disease risk is from de novo CNVs. However, the most spectacularly high risks are associated with certain specific rare CNVs that have replicated associations with the major psychiatric disorders. In population genetics, a rare variant represents a mixture of mutational events in recent generations that have been incorporated into the genome and transmitted, along with new mutations. A summary of these events is presented in Table 1. Several rare CNVs, such as the 16p11 CNV, have overlapping associations with bipolar disorder, schizophrenia, and ASD (22, 30).

One caveat in all reports of rare CNVs is that the number of individuals at risk is smaller than that of individuals with common SNP risk alleles, and this would reduce the precision of frequency estimates. Nonetheless, statistical bias is not present, and the association findings with disease are statistically significant, so the counseling and ethical issues they raise are timely for discussion.

Counseling individuals about their offspring’s genetic risk must take into account both the more common but less pathogenic SNP risk and the less common but more pathogenic CNV risk. However, no genetic variant is associated with disease every time it appears, and therefore the prediction of risk necessarily involves statistical probabilities. Table 3 presents the probability of illness given a particular SNP marker in mega-analyses of GWAS studies, which are analyses of all available data pooled together. This probability is referred to as the “genetic penetrance.” Penetrance is calculated as a Bayesian posterior probability (see the data supplement that accompanies the online edition of this article). For associated SNPs in GWAS studies of bipolar disorder or schizophrenia, the actual risk of illness to persons who have the risk alleles is small, ranging from 1.01% to 1.10%, compared with the population risks of 1% (Table 3). This is not a meaningful risk difference to a person receiving genetic counseling.

 
Anchor for Jump
TABLE 3.Risk of Illness for GWAS-Significant SNP Markers in Bipolar Disorder and Schizophreniaa
Table Footer Note

a GWAS=genome-wide association study; SNP=single-nucleotide polymorphism.

Table Footer Note

b Bipolar disorder data from mega-analysis of the Psychiatric GWAS Consortium Bipolar Disorder Working Group (43) (cases, N=11,977; controls, N=51,672). Schizophrenia data from mega-analysis of Steinberg et al. (44) (cases, N=18,206; controls, N=42,536). Note that for schizophrenia we include only one marker from the major histocompatibility complex (MHC).

Table Footer Note

c Based on Bayesian probability (see the online data supplement).

However, if there is reason to believe that a CNV is involved, then the risk calculation changes. In individuals who have de novo CNVs, the risks are 4.3% and 6.1% for bipolar disorder and schizophrenia, respectively, which are considerably greater than the SNP risks, although still modest (Table 2). For the rare CNVs associated with these disorders, the risks of illness are higher (Table 1), ranging up to an 82% risk that bipolar disorder, schizophrenia, or ASD will develop in a person with the 22q11 deletion. In the aggregate, de novo CNVs are not rare events. If the results of additional studies in bipolar disorder and schizophrenia replicate the reported results, one may expect that genotyping of patients and relatives will become a standard procedure for counseling. Screening for CNVs has already become a standard procedure in ASD, where the data on de novo and rare CNVs are comparable to those for bipolar disorder and schizophrenia (31). From a somewhat longer-term perspective, one may anticipate that the cost of whole-genome sequencing will continue to decrease rapidly, leading to the discovery of many more rare variants strongly associated with common disease. This would not have an impact on the ethical questions considered here, but it would make resolution of these questions important to more individuals.

In a counseling situation, the epidemiological concept of “exposed attributable risk” is potentially useful. This is the proportion of an individual’s risk that is accounted for by a specific risk factor that he or she has. Exposed attributable risk ranges up to 98% for the 22q11 deletion CNV (see Table S1 in the online data supplement). The exposed attributable risk due to any de novo CNV is 79% in bipolar disorder, 84.1% in schizophrenia, and 86.7% in ASD (Table 2). That is, for an affected person with a CNV event known to be associated with one or more of these diseases, most of that person’s risk is due to the CNV, with a corresponding reduction in risk for relatives without the CNV.

Exposed attributable risk thus has immediate effects on genetic counseling in the three disorders considered here, for the not-negligible proportion of cases with de novo CNVs or rare associated CNVs (32). Consider a family in which there is a patient with bipolar disorder and no other known individual with the disorder. His or her siblings may be concerned about their own risk of illness and that of their offspring. If the bipolar patient has a de novo CNV mutation, the risk to siblings who do not have that CNV mutation would not be appreciably different from the population risk, since 79% of the patient’s attributable risk is due to the CNV (Table 2). In this instance, the risk of illness in relatives, based on their genotypes, is comparable to the population risk (1%) and not the usual risk in sibs (which would be 6% for bipolar disorder [33]).

Over many years of practice and research, we have received questions from genetic counselors and from individuals who were not prepared to seek formal genetic counseling, such as patients and relatives attending national meetings of the National Alliance on Mental Illness. The majority of requests have been about the risk of mental illness in children of healthy relatives of persons with a mental illness, particularly children of healthy sibs. In the absence of genetic tests, the counseling consisted of empirical recurrence risks and education on prevalence of mental illness in the population and on reducing stigma in the patients and family members. However, with the new findings of potent risk factors that can be identified by testing, it is necessary to anticipate how counseling might be affected.

We discuss here a series of topics in which there are ethical issues concerning tests and genetic information, as well as psychodynamic challenges and interpersonal issues that arise in the people seeking consultation. We propose that counseling is more than risk prediction. The counselor must appreciate and work with the person seeking counsel to understand his or her perception of risk and burden of particular illnesses, and to help that person put the risks and issues into the appropriate ethical, psychodynamic, and social contexts for his or her own self.

+

Prenatal Screening, Preimplantation Procedures, and Abortion

It is not unusual today for a medically sophisticated couple planning or expecting a child to consider having a GWAS performed on themselves and on a fetal or embryonic genome. If routine prenatal genome screening were implemented in a population, the persons seeking counseling on a finding would want to know what it meant for their baby—that is, what the consequences might be for any disease or within broad groups of disease, such as mental illness or cancer. They would not necessarily have concerns centered on a particular diagnosis. Tables 1 and 2 list Bayes’s theorem probabilities for risk of bipolar disorder, schizophrenia, or ASD given a de novo CNV, and the probability of any of these disorders given a particular type of rare CNV. These probabilities give one pause; in Table 1 the risk of any of the three illnesses can be as high as 82%.

Abortion for a disorder that would be evident and impairing at birth is controversial, but for later-onset diseases and variable-penetrance diseases, it has barely been considered in public debate. It is difficult to conceive of a utilitarian calculus of risks and benefits that could be developed to inform this debate, or to inform the family facing this question. One shudders to think of the losses to human knowledge and culture that would develop if a high probability of these disorders were routinely accepted as grounds not to carry a fetus to term. And historically, this approach would be reminiscent of the widespread sterilizations and murders committed in the name of eugenics and racial hygiene in the 20th century.

Nonetheless, if abortion is a personal decision and not a social or political imposition, there will be people for whom the horror of particular illnesses or traits, or of a chronic disability, will justify abortion or the decision not to implant certain embryos. One of us (E.S.G.) is often reminded of a woman in an audience of patients and families with mental illness whom he addressed some years ago. He told that audience that the empirical risk of schizophrenia for children of a healthy sib of a patient with schizophrenia was 3%. She said that she heard the same number from Franz Kallmann, the founder of psychiatric genetics, 50 years earlier, and that she subsequently had three children, all of whom developed schizophrenia. Her comment was that if she had known then what she knew now, she never would have had children.

Let us imagine the same question arising today. If the woman sought counseling for schizophrenia risk before she had children, she, her spouse, her affected sib, and their parents might be offered whole-genome screening by GWAS. If the GWAS was unrevealing, the empirical risk estimates would be the same as they were 50 years ago, but the woman would be cautioned that we now know that there also exist high-risk genetic predispositions and that not all of them may be identified at this point. Let us further imagine that GWAS results revealed that the sibling has a high-penetrance rare CNV with a risk of schizophrenia and other disorders, and that the woman is a well carrier of the same CNV. Her prospective children who had the CNV would be at high risk. The risks and choices they now have could be specified to her and her prospective coparent. The reproductive choices the couple would make in response would depend on their personal ethical orientations and their responses to the family history and test results. Informing their decisions would be a range of reproductive technologies available today that were not available in the past. Preimplantation screening of embryos, for example, could avoid the CNV risk. On the other hand, if the test results were that her ill sib had the rare CNV and she did not, then the sib’s risk (exposed attributable risk) would be mainly attributable to his or her CNV, and the woman herself would not face an elevated risk of schizophrenia in her children.

+

Family Members’ Rights to Genetic Information and Conflicts and Stigma Within Families

It has been argued that family members who may be at risk of illness or face burdens of care for a relative at risk have a right to genetic test results of another member, and that this is grounds for breach of confidentiality (34). An ethical issue arises from revealing or implying carrier status in a relative through a test of someone else in the family. In fragile X syndrome, for example, which can cause autism and mental retardation, we have seen families with myths about what caused the known case (or cases), such as that the mother was exposed to rubella. These myths protect relatives from feeling they are “responsible” for the illness or that they or a revered ancestor may be a carrier. But this protection can lead to the hiding of information from persons who may be affected by it. Within families, conflicts can arise around genetic tests of a patient and of persons at risk of transmitting illness. Narcissistic injury (a negative impact on self-image) and changes in status, image, and perceived burdens within the family can follow a test result (34).

Confidentiality of the fact that a test was performed is not a solution to the ethical issue of whether relatives have a right to interfere with testing or a right to know test results. To continue with the example of fragile X syndrome, in which we have experience, there may be relatives who would very much want to know that they are at risk for carrier status, because they would then be concerned about their present or future children, or because they may be aware of medical problems of the carrier state, such as ataxia and ovarian failure in carriers of the premutation of fragile X. Several ethical issues are thus raised—whether relatives have a right to know about risks they may have and whether the physician has a duty to warn family members about their risks (34). For known diseases, these within-family controversies could be largely avoided if there were widespread population screening rather than screening of families of affected individuals. For “private mutations” that exist in only one family, however, population screening might never become applicable, as discussed below.

Within-family conflicts do not require a rational basis. Blaming a relative or a branch of the family for an illness, or for disclosing information about an illness, might occur. In the case of CNVs and other rare events with high attributable risk in exposed individuals, as in conditions with dominant inheritance, there is also the potential for conflict based on correct statistical understanding of actual or potential genetic test results when there are strong emotional implications. A carrier can be blamed; one branch of a family can shun another branch and claim that its own branch is genetically unaffected. It cannot be left to better education of the public to prevent or resolve such conflicts. Psychotherapeutic interventions may be required.

For the practicing mental health professional, narcissistic injury and within-family conflicts of this kind are not unusual psychotherapeutic challenges. The overt content of genetic test results is particular to the counseling offered on risk, but the psychodynamic and interpersonal issues that arise are not unlike other responses to misfortune and illness, and the treatment is not different either. Counseling based on genetic tests is new to the mental disorders, and the particular stigma of mental disorders is a unique aspect of this process for the psychiatric disorders, but even this does not present insuperable psychotherapeutic technical challenges.

+

Population Screening

The ethical issues of population screening include that it may be nonconsensual as part of a public health mandate, that it can lead to stigma, and, for a substantial part of the public, that it can promote abortion. The consent issue can be dealt with by preserving confidentiality of results and offering an option for tested persons to decline to receive results of their tests.

However, population screening can lead to stigmatization and discrimination against whole communities (35). This is particularly sensitive for mental retardation and for intelligence. In recent decades there have been great public controversies over intelligence differences among races (36, 37), and the historical stigmatization of different races or social classes over the frequency of idiocy and feeble-mindedness is well known. It would not be surprising if similar stigma arose over genetic susceptibility to mental disorders. Ethnic-targeted genetic screening is currently considered ethically questionable, and offering universal screening has been advocated (38).

Many of the ethical conundrums and conflicts within families are avoidable by replacing family-based screening with population screening. Population screening is technologically feasible for all known common and most rare genetic causes of illness, although it is not yet economically feasible. Arguably, the screening can be done ethically without consent of the individuals screened, if screened persons must give consent in order to receive their test results.

However, as noted above, there are “private” mutations, which occur once within one family and are essentially never seen again. Such mutations may reveal biological factors in disease, and such information may have therapeutic implications for other family members. The variant may not be immediately interpretable as a disease variant. These mutations can be verified and screened only by extending pedigrees (that is, by obtaining information on family members). The current ethically acceptable practice in the United States is that individuals can provide information about the health of family members, but only that individual can retain identifying information or contact the relatives and obtain consent to be tested. This preserves the privacy of the relatives but greatly impairs the effectiveness of studying extended families, as many investigators who have experience with pedigree studies know.

A new societal consensus on family study for research and diagnostic purposes is needed. One possibility is to apply the principle described above, of overriding confidentiality of results within a family because of the interests of persons who may be at risk. This overriding could be applied to allow clinical geneticists studying an illness in a family to have access to identifying information on relatives, provided by a cooperating family member, and to contact those relatives. Alternatively, licensed entities might be set up that are empowered to conduct family screening for all medical diseases and that do not reveal to each person contacted what disease it is studying and which relative is ill, thus preserving privacy to a significant extent.

+

Genetic Tests and Marital Choice

Although it might seem absurd to consider genotypes of a prospective spouse in a marital decision, family history as a surrogate for genetic tests is an important factor in marital choice in contemporary cultures with arranged marriages. Multiple communities worldwide practice arranged marriage, including countries such as India and Pakistan, and ultra-Orthodox Jews. Within these communities, the stigma of mental illness in a family is strong. Would the stigma be reduced if there were a known high-risk variant in the family but not in the prospective person to be married? At present, among ultra-Orthodox Jews, there is stigma even on having a genetic examination. It is to be expected that genetic testing for neuropsychiatric disorders will eventually come into use within these communities, and it remains to be seen whether this will have an overall beneficial effect on families who are currently stigmatized.

It is possible that premarital genome screening could become widespread for prenatal screening, as well as to anticipate vulnerability to illness in a prospective spouse. Presumably there would not be a consent problem; genetic disclosures would be part of the negotiations and disclosures that generally accompany marital decisions. The ethical issue is whether it is right to reject an otherwise desirable person because of their genetic test results, and whether human values are trampled upon in so doing.

+

Improved Prediction

The principal known risks of rare variants and new mutations are the risks of CNVs, but this is only the beginning. Recent studies in which every DNA base pair was sequenced have discovered additional risks from base pair mutations and from small insertions and deletions (indels) that are too small to detect by the microarray methods used to date to detect CNVs (39, 40). New mutations and rare variants (which by the principles of population genetics must be recent mutations) may become, in the aggregate, even more common and potent causes of disease of the CNS.

If all the polygenic risk in an individual could be measured using GWAS markers or whole-genome sequencing, the predicted risk could approach the estimates of total risk contained in GWAS markers generated by Lee et al. (3, 4). Similarly, subsets of SNP markers, such as markers known to affect gene expression, methylation, or translation into proteins, may become useful in risk prediction (41, 42). As new methods of risk estimation develop and risk becomes more predictable for a wider group of patients, the principles discussed here would remain valid, but requests for counseling may become much more widespread, and the ethical issues may become more of an issue for public debate than heretofore.

+

Improved Prevention and Treatment Based on Genetic Discoveries

The main motivation for genetic investigation of diseases is not risk prediction. It is the development of treatment based on the biology of associated genes, according to the genetic risk factors of each ill individual, and including prevention for persons at risk. Although this hope has not often been fulfilled in common diseases so far, when such progress does occur, it could make abortion or in vitro preimplantation tests for a disease no longer worth considering. And regret over past actions could then become considerable, including regret over decisions to avoid testing or to abort a fetus. Ethics is focused on the rightness or wrongness of decisions taken by people, and this judgment should include consideration of the likelihood of medical progress.

Wapner  RJ;  Martin  CL;  Levy  B;  Ballif  BC;  Eng  CM;  Zachary  JM;  Savage  M;  Platt  LD;  Saltzman  D;  Grobman  WA;  Klugman  S;  Scholl  T;  Simpson  JL;  McCall  K;  Aggarwal  VS;  Bunke  B;  Nahum  O;  Patel  A;  Lamb  AN;  Thom  EA;  Beaudet  AL;  Ledbetter  DH;  Shaffer  LG;  Jackson  L:  Chromosomal microarray versus karyotyping for prenatal diagnosis.  N Engl J Med 2012; 367:2175–2184
[CrossRef] | [PubMed]
 
Reich  DE;  Lander  ES:  On the allelic spectrum of human disease.  Trends Genet 2001; 17:502–510
[CrossRef] | [PubMed]
 
Lee  SH;  Wray  NR;  Goddard  ME;  Visscher  PM:  Estimating missing heritability for disease from genome-wide association studies.  Am J Hum Genet 2011; 88:294–305
[CrossRef] | [PubMed]
 
Lee  SH;  DeCandia  TR;  Ripke  S;  Yang  J;  Sullivan  PF;  Goddard  ME;  Keller  MC;  Visscher  PM;  Wray  NR; Schizophrenia Psychiatric Genome-Wide Association Study Consortium (PGC-SCZ); International Schizophrenia Consortium (ISC); Molecular Genetics of Schizophrenia Collaboration (MGS):  Estimating the proportion of variation in susceptibility to schizophrenia captured by common SNPs.  Nat Genet 2012; 44:247–250
[CrossRef] | [PubMed]
 
O’Dushlaine C, Ripke S, Moran J, Chambert K, Sklar P, Purcell S, Hultman C, McCarroll SA, Sullivan P: Genome-wide association study of 32,143 individuals reveals several novel associations in schizophrenia, in Twentieth World Congress of Psychiatric Genetics: Confronting the Complexity of Brain and Behavior: Presentation Abstracts. Brentwood, Tenn, International Society of Psychatric Genetics, 2012, p 62
 
Manolio  TA;  Collins  FS;  Cox  NJ;  Goldstein  DB;  Hindorff  LA;  Hunter  DJ;  McCarthy  MI;  Ramos  EM;  Cardon  LR;  Chakravarti  A;  Cho  JH;  Guttmacher  AE;  Kong  A;  Kruglyak  L;  Mardis  E;  Rotimi  CN;  Slatkin  M;  Valle  D;  Whittemore  AS;  Boehnke  M;  Clark  AG;  Eichler  EE;  Gibson  G;  Haines  JL;  Mackay  TF;  McCarroll  SA;  Visscher  PM:  Finding the missing heritability of complex diseases.  Nature 2009; 461:747–753
[CrossRef] | [PubMed]
 
Sullivan  PF;  Daly  MJ;  O’Donovan  M:  Genetic architectures of psychiatric disorders: the emerging picture and its implications.  Nat Rev Genet 2012; 13:537–551
[CrossRef] | [PubMed]
 
Kim  YS;  Leventhal  BL;  Koh  YJ;  Fombonne  E;  Laska  E;  Lim  EC;  Cheon  KA;  Kim  SJ;  Kim  YK;  Lee  H;  Song  DH;  Grinker  RR:  Prevalence of autism spectrum disorders in a total population sample.  Am J Psychiatry 2011; 168:904–912
[CrossRef] | [PubMed]
 
Charman  T:  The highs and lows of counting autism.  Am J Psychiatry 2011; 168:873–875
[CrossRef] | [PubMed]
 
Shattuck  PT:  The contribution of diagnostic substitution to the growing administrative prevalence of autism in US special education.  Pediatrics 2006; 117:1028–1037
[CrossRef] | [PubMed]
 
Wang K, Zhang H, Ma D, Bucan M, Glessner JT, Abrahams BS, Salyakina D, Imielinski M, Bradfield JP, Sleiman PM, Kim CE, Hou C, Frackelton E, Chiavacci R, Takahashi N, Sakurai T, Rappaport E, Lajonchere CM, Munson J, Estes A, Korvatska O, Piven J, Sonnenblick LI, Alvarez-Retuerto AI, Herman EI, Dong H, Hutman T, Sigman M, Ozonoff S, Klin A, Owley T, Sweeney JA, Brune CW, Cantor RM, Bernier R, Gilbert JR, Cuccaro ML, McMahon WM, Miller J, State MW, Wassink TH, Coon H, Levy SE, Schultz RT, Nurnberger JI, Haines JL, Sutcliffe JS, Cook EH, Minshew NJ, Buxbaum JD, Dawson G, Grant SF, Geschwind DH, Pericak-Vance MA, Schellenberg GD, Hakonarson H: Common genetic variants on 5p14.1 associate with autism spectrum disorders. Nature 2009; 459:528–533
 
Devlin  B;  Melhem  N;  Roeder  K:  Do common variants play a role in risk for autism? Evidence and theoretical musings.  Brain Res 2011; 1380:78–84
[CrossRef] | [PubMed]
 
Iafrate  AJ;  Feuk  L;  Rivera  MN;  Listewnik  ML;  Donahoe  PK;  Qi  Y;  Scherer  SW;  Lee  C:  Detection of large-scale variation in the human genome.  Nat Genet 2004; 36:949–951
[CrossRef] | [PubMed]
 
Sebat  J;  Lakshmi  B;  Troge  J;  Alexander  J;  Young  J;  Lundin  P;  Maner  S;  Massa  H;  Walker  M;  Chi  M;  Navin  N;  Lucito  R;  Healy  J;  Hicks  J;  Ye  K;  Reiner  A;  Gilliam  TC;  Trask  B;  Patterson  N;  Zetterberg  A;  Wigler  M:  Large-scale copy number polymorphism in the human genome.  Science  2004; 305:525–528
[CrossRef] | [PubMed]
 
Eichler  EE:  Recent duplication, domain accretion, and the dynamic mutation of the human genome.  Trends Genet 2001; 17:661–669
[CrossRef] | [PubMed]
 
Human Genome Structural Variation Working Group, Eichler  EE;  Nickerson  DA;  Altshuler  D;  Bowcock  AM;  Brooks  LD;  Carter  NP;  Church  DM;  Felsenfeld  A;  Guyer  M;  Lee  C;  Lupski  JR;  Mullikin  JC;  Pritchard  JK;  Sebat  J;  Sherry  ST;  Smith  D;  Valle  D;  Waterston  RH:  Completing the map of human genetic variation.  Nature 2007; 447:161–165
[CrossRef] | [PubMed]
 
Lupski  JR:  Genomic disorders: structural features of the genome can lead to DNA rearrangements and human disease traits.  Trends Genet 1998; 14:417–422
[CrossRef] | [PubMed]
 
Karayiorgou  M;  Morris  MA;  Morrow  B;  Shprintzen  RJ;  Goldberg  R;  Borrow  J;  Gos  A;  Nestadt  G;  Wolyniec  PS;  Lasseter  VK  et al:  Schizophrenia susceptibility associated with interstitial deletions of chromosome 22q11.  Proc Natl Acad Sci USA 1995; 92:7612–7616
[CrossRef] | [PubMed]
 
Lupski  JR:  Genomic rearrangements and sporadic disease.  Nat Genet 2007; 39(suppl 7):S43–S47
[CrossRef] | [PubMed]
 
Lander  ES:  Initial impact of the sequencing of the human genome.  Nature 2011; 470:187–197
[CrossRef] | [PubMed]
 
Bassett  AS;  Scherer  SW;  Brzustowicz  LM:  Copy number variations in schizophrenia: critical review and new perspectives on concepts of genetics and disease.  Am J Psychiatry 2010; 167:899–914
[CrossRef] | [PubMed]
 
Malhotra  D;  Sebat  J:  CNVs: harbingers of a rare variant revolution in psychiatric genetics.  Cell 2012; 148:1223–1241
[CrossRef] | [PubMed]
 
Sebat  J;  Lakshmi  B;  Malhotra  D;  Troge  J;  Lese-Martin  C;  Walsh  T;  Yamrom  B;  Yoon  S;  Krasnitz  A;  Kendall  J;  Leotta  A;  Pai  D;  Zhang  R;  Lee  YH;  Hicks  J;  Spence  SJ;  Lee  AT;  Puura  K;  Lehtimaki  T;  Ledbetter  D;  Gregersen  PK;  Bregman  J;  Sutcliffe  JS;  Jobanputra  V;  Chung  W;  Warburton  D;  King  MC;  Skuse  D;  Geschwind  DH;  Gilliam  TC;  Ye  K;  Wigler  M:  Strong association of de novo copy number mutations with autism.  Science 2007; 316:445–449
[CrossRef] | [PubMed]
 
Xu  B;  Roos  JL;  Levy  S;  van Rensburg  EJ;  Gogos  JA;  Karayiorgou  M:  Strong association of de novo copy number mutations with sporadic schizophrenia.  Nat Genet 2008; 40:880–885
[CrossRef] | [PubMed]
 
Malhotra  D;  McCarthy  S;  Michaelson  JJ;  Vacic  V;  Burdick  KE;  Yoon  S;  Cichon  S;  Corvin  A;  Gary  S;  Gershon  ES;  Gill  M;  Karayiorgou  M;  Kelsoe  JR;  Krastoshevsky  O;  Krause  V;  Leibenluft  E;  Levy  DL;  Makarov  V;  Bhandari  A;  Malhotra  AK;  McMahon  FJ;  Nöthen  MM;  Potash  JB;  Rietschel  M;  Schulze  TG;  Sebat  J:  High frequencies of de novo CNVs in bipolar disorder and schizophrenia.  Neuron 2011; 72:951–963
[CrossRef] | [PubMed]
 
Itsara  A;  Wu  H;  Smith  JD;  Nickerson  DA;  Romieu  I;  London  SJ;  Eichler  EE:  De novo rates and selection of large copy number variation.  Genome Res 2010; 20:1469–1481
[CrossRef] | [PubMed]
 
Levy  D;  Ronemus  M;  Yamrom  B;  Lee  YH;  Leotta  A;  Kendall  J;  Marks  S;  Lakshmi  B;  Pai  D;  Ye  K;  Buja  A;  Krieger  A;  Yoon  S;  Troge  J;  Rodgers  L;  Iossifov  I;  Wigler  M:  Rare de novo and transmitted copy-number variation in autistic spectrum disorders.  Neuron 2011; 70:886–897
[CrossRef] | [PubMed]
 
Pinto  D;  Pagnamenta  AT;  Klei  L;  Anney  R;  Merico  D;  Regan  R, et al:  Functional impact of global rare copy number variation in autism spectrum disorders.  Nature 2010; 466:368–372
[CrossRef] | [PubMed]
 
Sanders  SJ;  Ercan-Sencicek  AG;  Hus  V;  Luo  R;  Murtha  MT;  Moreno-De-Luca  D;  Chu  SH;  Moreau  MP;  Gupta  AR;  Thomson  SA;  Mason  CE;  Bilguvar  K;  Celestino-Soper  PB;  Choi  M;  Crawford  EL;  Davis  L;  Wright  NR;  Dhodapkar  RM;  DiCola  M;  DiLullo  NM;  Fernandez  TV;  Fielding-Singh  V;  Fishman  DO;  Frahm  S;  Garagaloyan  R;  Goh  GS;  Kammela  S;  Klei  L;  Lowe  JK;  Lund  SC;  McGrew  AD;  Meyer  KA;  Moffat  WJ;  Murdoch  JD;  O’Roak  BJ;  Ober  GT;  Pottenger  RS;  Raubeson  MJ;  Song  Y;  Wang  Q;  Yaspan  BL;  Yu  TW;  Yurkiewicz  IR;  Beaudet  AL;  Cantor  RM;  Curland  M;  Grice  DE;  Günel  M;  Lifton  RP;  Mane  SM;  Martin  DM;  Shaw  CA;  Sheldon  M;  Tischfield  JA;  Walsh  CA;  Morrow  EM;  Ledbetter  DH;  Fombonne  E;  Lord  C;  Martin  CL;  Brooks  AI;  Sutcliffe  JS;  Cook  EH  Jr;  Geschwind  D;  Roeder  K;  Devlin  B;  State  MW:  Multiple recurrent de novo CNVs, including duplications of the 7q11.23 Williams syndrome region, are strongly associated with autism.  Neuron 2011; 70:863–885
[CrossRef] | [PubMed]
 
Vacic  V;  McCarthy  S;  Malhotra  D;  Murray  F;  Chou  HH;  Peoples  A;  Makarov  V;  Yoon  S;  Bhandari  A;  Corominas  R;  Iakoucheva  LM;  Krastoshevsky  O;  Krause  V;  Larach-Walters  V;  Welsh  DK;  Craig  D;  Kelsoe  JR;  Gershon  ES;  Leal  SM;  Dell Aquila  M;  Morris  DW;  Gill  M;  Corvin  A;  Insel  PA;  McClellan  J;  King  MC;  Karayiorgou  M;  Levy  DL;  DeLisi  LE;  Sebat  J:  Duplications of the neuropeptide receptor gene VIPR2 confer significant risk for schizophrenia.  Nature 2011; 471:499–503
[CrossRef] | [PubMed]
 
Manning  M;  Hudgins  L; Professional Practice and Guidelines Committee:  Array-based technology and recommendations for utilization in medical genetics practice for detection of chromosomal abnormalities.  Genet Med 2010; 12:742–745
[CrossRef] | [PubMed]
 
St Clair  D:  Copy number variation and schizophrenia.  Schizophr Bull 2009; 35:9–12
[CrossRef] | [PubMed]
 
Gershon  ES;  Hamovit  J;  Guroff  JJ;  Dibble  E;  Leckman  JF;  Sceery  W;  Targum  SD;  Nurnberger  JI  Jr;  Goldin  LR;  Bunney  WE  Jr:  A family study of schizoaffective, bipolar I, bipolar II, unipolar, and normal control probands.  Arch Gen Psychiatry 1982; 39:1157–1167
[CrossRef] | [PubMed]
 
Doukas  DJ;  Berg  JW:  The family covenant and genetic testing.  Am J Bioeth 2001; 1:3–10
[CrossRef] | [PubMed]
 
Fulda  KG;  Lykens  K:  Ethical issues in predictive genetic testing: a public health perspective.  J Med Ethics 2006; 32:143–147
[CrossRef] | [PubMed]
 
Fraser  S:  The Bell Curve Wars: Race, Intelligence, and the Future of America .  New York,  Basic Books, 1995
 
Herrnstein  RJ;  Murray  C:  The Bell Curve: Intelligence and Class Structure in American Life .  New York,  Simon & Schuster, 1994
 
Grosse  SD;  Rogowski  WH;  Ross  LF;  Cornel  MC;  Dondorp  WJ;  Khoury  MJ:  Population screening for genetic disorders in the 21st century: evidence, economics, and ethics.  Public Health Genomics 2010; 13:106–115
[CrossRef] | [PubMed]
 
Sanders  SJ;  Murtha  MT;  Gupta  AR;  Murdoch  JD;  Raubeson  MJ;  Willsey  AJ;  Ercan-Sencicek  AG;  DiLullo  NM;  Parikshak  NN;  Stein  JL;  Walker  MF;  Ober  GT;  Teran  NA;  Song  Y;  El-Fishawy  P;  Murtha  RC;  Choi  M;  Overton  JD;  Bjornson  RD;  Carriero  NJ;  Meyer  KA;  Bilguvar  K;  Mane  SM;  Sestan  N;  Lifton  RP;  Günel  M;  Roeder  K;  Geschwind  DH;  Devlin  B;  State  MW:  De novo mutations revealed by whole-exome sequencing are strongly associated with autism.  Nature 2012; 485:237–241
[CrossRef] | [PubMed]
 
Xu  B;  Roos  JL;  Dexheimer  P;  Boone  B;  Plummer  B;  Levy  S;  Gogos  JA;  Karayiorgou  M:  Exome sequencing supports a de novo mutational paradigm for schizophrenia.  Nat Genet 2011; 43:864–868
[CrossRef] | [PubMed]
 
Gamazon  ER;  Badner  JA;  Cheng  L;  Zhang  C;  Zhang  D;  Cox  NJ;  Gershon  ES;  Kelsoe  JR;  Greenwood  TA;  Nievergelt  CM;  Chen  C;  McKinney  R;  Shilling  PD;  Schork  NJ;  Smith  EN;  Bloss  CS;  Nurnberger  JI;  Edenberg  HJ;  Foroud  T;  Koller  DL;  Scheftner  WA;  Coryell  W;  Rice  J;  Lawson  WB;  Nwulia  EA;  Hipolito  M;  Byerley  W;  McMahon  FJ;  Schulze  TG;  Berrettini  WH;  Potash  JB;  Zandi  PP;  Mahon  PB;  McInnis  MG;  Zollner  S;  Zhang  P;  Craig  DW;  Szelinger  S;  Barrett  TB;  Liu  C:  Enrichment of cis-regulatory gene expression SNPs and methylation quantitative trait loci among bipolar disorder susceptibility variants.  Mol Psychiatry 2013; 18:340–346
[CrossRef] | [PubMed]
 
Nicolae  DL;  Gamazon  E;  Zhang  W;  Duan  S;  Dolan  ME;  Cox  NJ:  Trait-associated SNPs are more likely to be eQTLs: annotation to enhance discovery from GWAS.  PLoS Genet 2010; 6:e1000888
[CrossRef] | [PubMed]
 
Psychiatric GWAS Consortium Bipolar Disorder Working Group:  Large-scale genome-wide association analysis of bipolar disorder identifies a new susceptibility locus near ODZ4.  Nat Genet 2011; 43:977–983
[CrossRef] | [PubMed]
 
Steinberg  S;  de Jong  S; Irish Schizophrenia Genomics Consortium; Andreassen  OA;  Werge  T;  Børglum  AD;  Mors  O;  Mortensen  PB;  Gustafsson  O;  Costas  J;  Pietiläinen  OP;  Demontis  D;  Papiol  S;  Huttenlocher  J;  Mattheisen  M;  Breuer  R;  Vassos  E;  Giegling  I;  Fraser  G;  Walker  N;  Tuulio-Henriksson  A;  Suvisaari  J;  Lönnqvist  J;  Paunio  T;  Agartz  I;  Melle  I;  Djurovic  S;  Strengman  E; GROUP; Jürgens  G;  Glenthøj  B;  Terenius  L;  Hougaard  DM;  Ørntoft  T;  Wiuf  C;  Didriksen  M;  Hollegaard  MV;  Nordentoft  M;  van Winkel  R;  Kenis  G;  Abramova  L;  Kaleda  V;  Arrojo  M;  Sanjuán  J;  Arango  C;  Sperling  S;  Rossner  M;  Ribolsi  M;  Magni  V;  Siracusano  A;  Christiansen  C;  Kiemeney  LA;  Veldink  J;  van den Berg  L;  Ingason  A;  Muglia  P;  Murray  R;  Nöthen  MM;  Sigurdsson  E;  Petursson  H;  Thorsteinsdottir  U;  Kong  A;  Rubino  IA;  De Hert  M;  Réthelyi  JM;  Bitter  I;  Jönsson  EG;  Golimbet  V;  Carracedo  A;  Ehrenreich  H;  Craddock  N;  Owen  MJ;  O’Donovan  MC; Wellcome Trust Case Control Consortium 2, Ruggeri  M;  Tosato  S;  Peltonen  L;  Ophoff  RA;  Collier  DA;  St Clair  D;  Rietschel  M;  Cichon  S;  Stefansson  H;  Rujescu  D;  Stefansson  K:  Common variants at VRK2 and TCF4 conferring risk of schizophrenia.  Hum Mol Genet 2011; 20:4076–4081
[CrossRef] | [PubMed]
 
References Container
Anchor for Jump
TABLE 1.Rare CNVs: Risks of Illness for Bipolar Disorder, Schizophrenia, and ASDa
Table Footer Note

a ASD=autism spectrum disorder; CNV=copy number variant. Risks of illness are based on Bayesian probabilities. Data are from the comprehensive review of Malhotra and Sebat (22). The criterion for inclusion in this table was association of CNV with illness at p<0.003.

Anchor for Jump
TABLE 2.De Novo CNVs: Attributable Risk and Risks of Illness for Schizophrenia, Bipolar Disorder, and ASDa
Table Footer Note

a ASD=autism spectrum disorder; CNV=copy number variant. Exposed attributable risk is the attributable risk in exposed individuals (i.e., ill individuals who have a de novo CNV), which is the proportion of the ill person’s risk attributable to the CNV.

Table Footer Note

b Based on Bayesian probability. Data are from Xu et al. (24), Malhotra et al. (25), and Sebat et al. (23). Computation of illness risk of any disorder is 1−(1−P1)(1−P2)(1−P3), where Pi is the risk for each disorder. This calculation indirectly accounts for the probability of the occurrence of more than one disorder in any individual as a product of the probability of each diagnosis.

Anchor for Jump
TABLE 3.Risk of Illness for GWAS-Significant SNP Markers in Bipolar Disorder and Schizophreniaa
Table Footer Note

a GWAS=genome-wide association study; SNP=single-nucleotide polymorphism.

Table Footer Note

b Bipolar disorder data from mega-analysis of the Psychiatric GWAS Consortium Bipolar Disorder Working Group (43) (cases, N=11,977; controls, N=51,672). Schizophrenia data from mega-analysis of Steinberg et al. (44) (cases, N=18,206; controls, N=42,536). Note that for schizophrenia we include only one marker from the major histocompatibility complex (MHC).

Table Footer Note

c Based on Bayesian probability (see the online data supplement).

+

References

Wapner  RJ;  Martin  CL;  Levy  B;  Ballif  BC;  Eng  CM;  Zachary  JM;  Savage  M;  Platt  LD;  Saltzman  D;  Grobman  WA;  Klugman  S;  Scholl  T;  Simpson  JL;  McCall  K;  Aggarwal  VS;  Bunke  B;  Nahum  O;  Patel  A;  Lamb  AN;  Thom  EA;  Beaudet  AL;  Ledbetter  DH;  Shaffer  LG;  Jackson  L:  Chromosomal microarray versus karyotyping for prenatal diagnosis.  N Engl J Med 2012; 367:2175–2184
[CrossRef] | [PubMed]
 
Reich  DE;  Lander  ES:  On the allelic spectrum of human disease.  Trends Genet 2001; 17:502–510
[CrossRef] | [PubMed]
 
Lee  SH;  Wray  NR;  Goddard  ME;  Visscher  PM:  Estimating missing heritability for disease from genome-wide association studies.  Am J Hum Genet 2011; 88:294–305
[CrossRef] | [PubMed]
 
Lee  SH;  DeCandia  TR;  Ripke  S;  Yang  J;  Sullivan  PF;  Goddard  ME;  Keller  MC;  Visscher  PM;  Wray  NR; Schizophrenia Psychiatric Genome-Wide Association Study Consortium (PGC-SCZ); International Schizophrenia Consortium (ISC); Molecular Genetics of Schizophrenia Collaboration (MGS):  Estimating the proportion of variation in susceptibility to schizophrenia captured by common SNPs.  Nat Genet 2012; 44:247–250
[CrossRef] | [PubMed]
 
O’Dushlaine C, Ripke S, Moran J, Chambert K, Sklar P, Purcell S, Hultman C, McCarroll SA, Sullivan P: Genome-wide association study of 32,143 individuals reveals several novel associations in schizophrenia, in Twentieth World Congress of Psychiatric Genetics: Confronting the Complexity of Brain and Behavior: Presentation Abstracts. Brentwood, Tenn, International Society of Psychatric Genetics, 2012, p 62
 
Manolio  TA;  Collins  FS;  Cox  NJ;  Goldstein  DB;  Hindorff  LA;  Hunter  DJ;  McCarthy  MI;  Ramos  EM;  Cardon  LR;  Chakravarti  A;  Cho  JH;  Guttmacher  AE;  Kong  A;  Kruglyak  L;  Mardis  E;  Rotimi  CN;  Slatkin  M;  Valle  D;  Whittemore  AS;  Boehnke  M;  Clark  AG;  Eichler  EE;  Gibson  G;  Haines  JL;  Mackay  TF;  McCarroll  SA;  Visscher  PM:  Finding the missing heritability of complex diseases.  Nature 2009; 461:747–753
[CrossRef] | [PubMed]
 
Sullivan  PF;  Daly  MJ;  O’Donovan  M:  Genetic architectures of psychiatric disorders: the emerging picture and its implications.  Nat Rev Genet 2012; 13:537–551
[CrossRef] | [PubMed]
 
Kim  YS;  Leventhal  BL;  Koh  YJ;  Fombonne  E;  Laska  E;  Lim  EC;  Cheon  KA;  Kim  SJ;  Kim  YK;  Lee  H;  Song  DH;  Grinker  RR:  Prevalence of autism spectrum disorders in a total population sample.  Am J Psychiatry 2011; 168:904–912
[CrossRef] | [PubMed]
 
Charman  T:  The highs and lows of counting autism.  Am J Psychiatry 2011; 168:873–875
[CrossRef] | [PubMed]
 
Shattuck  PT:  The contribution of diagnostic substitution to the growing administrative prevalence of autism in US special education.  Pediatrics 2006; 117:1028–1037
[CrossRef] | [PubMed]
 
Wang K, Zhang H, Ma D, Bucan M, Glessner JT, Abrahams BS, Salyakina D, Imielinski M, Bradfield JP, Sleiman PM, Kim CE, Hou C, Frackelton E, Chiavacci R, Takahashi N, Sakurai T, Rappaport E, Lajonchere CM, Munson J, Estes A, Korvatska O, Piven J, Sonnenblick LI, Alvarez-Retuerto AI, Herman EI, Dong H, Hutman T, Sigman M, Ozonoff S, Klin A, Owley T, Sweeney JA, Brune CW, Cantor RM, Bernier R, Gilbert JR, Cuccaro ML, McMahon WM, Miller J, State MW, Wassink TH, Coon H, Levy SE, Schultz RT, Nurnberger JI, Haines JL, Sutcliffe JS, Cook EH, Minshew NJ, Buxbaum JD, Dawson G, Grant SF, Geschwind DH, Pericak-Vance MA, Schellenberg GD, Hakonarson H: Common genetic variants on 5p14.1 associate with autism spectrum disorders. Nature 2009; 459:528–533
 
Devlin  B;  Melhem  N;  Roeder  K:  Do common variants play a role in risk for autism? Evidence and theoretical musings.  Brain Res 2011; 1380:78–84
[CrossRef] | [PubMed]
 
Iafrate  AJ;  Feuk  L;  Rivera  MN;  Listewnik  ML;  Donahoe  PK;  Qi  Y;  Scherer  SW;  Lee  C:  Detection of large-scale variation in the human genome.  Nat Genet 2004; 36:949–951
[CrossRef] | [PubMed]
 
Sebat  J;  Lakshmi  B;  Troge  J;  Alexander  J;  Young  J;  Lundin  P;  Maner  S;  Massa  H;  Walker  M;  Chi  M;  Navin  N;  Lucito  R;  Healy  J;  Hicks  J;  Ye  K;  Reiner  A;  Gilliam  TC;  Trask  B;  Patterson  N;  Zetterberg  A;  Wigler  M:  Large-scale copy number polymorphism in the human genome.  Science  2004; 305:525–528
[CrossRef] | [PubMed]
 
Eichler  EE:  Recent duplication, domain accretion, and the dynamic mutation of the human genome.  Trends Genet 2001; 17:661–669
[CrossRef] | [PubMed]
 
Human Genome Structural Variation Working Group, Eichler  EE;  Nickerson  DA;  Altshuler  D;  Bowcock  AM;  Brooks  LD;  Carter  NP;  Church  DM;  Felsenfeld  A;  Guyer  M;  Lee  C;  Lupski  JR;  Mullikin  JC;  Pritchard  JK;  Sebat  J;  Sherry  ST;  Smith  D;  Valle  D;  Waterston  RH:  Completing the map of human genetic variation.  Nature 2007; 447:161–165
[CrossRef] | [PubMed]
 
Lupski  JR:  Genomic disorders: structural features of the genome can lead to DNA rearrangements and human disease traits.  Trends Genet 1998; 14:417–422
[CrossRef] | [PubMed]
 
Karayiorgou  M;  Morris  MA;  Morrow  B;  Shprintzen  RJ;  Goldberg  R;  Borrow  J;  Gos  A;  Nestadt  G;  Wolyniec  PS;  Lasseter  VK  et al:  Schizophrenia susceptibility associated with interstitial deletions of chromosome 22q11.  Proc Natl Acad Sci USA 1995; 92:7612–7616
[CrossRef] | [PubMed]
 
Lupski  JR:  Genomic rearrangements and sporadic disease.  Nat Genet 2007; 39(suppl 7):S43–S47
[CrossRef] | [PubMed]
 
Lander  ES:  Initial impact of the sequencing of the human genome.  Nature 2011; 470:187–197
[CrossRef] | [PubMed]
 
Bassett  AS;  Scherer  SW;  Brzustowicz  LM:  Copy number variations in schizophrenia: critical review and new perspectives on concepts of genetics and disease.  Am J Psychiatry 2010; 167:899–914
[CrossRef] | [PubMed]
 
Malhotra  D;  Sebat  J:  CNVs: harbingers of a rare variant revolution in psychiatric genetics.  Cell 2012; 148:1223–1241
[CrossRef] | [PubMed]
 
Sebat  J;  Lakshmi  B;  Malhotra  D;  Troge  J;  Lese-Martin  C;  Walsh  T;  Yamrom  B;  Yoon  S;  Krasnitz  A;  Kendall  J;  Leotta  A;  Pai  D;  Zhang  R;  Lee  YH;  Hicks  J;  Spence  SJ;  Lee  AT;  Puura  K;  Lehtimaki  T;  Ledbetter  D;  Gregersen  PK;  Bregman  J;  Sutcliffe  JS;  Jobanputra  V;  Chung  W;  Warburton  D;  King  MC;  Skuse  D;  Geschwind  DH;  Gilliam  TC;  Ye  K;  Wigler  M:  Strong association of de novo copy number mutations with autism.  Science 2007; 316:445–449
[CrossRef] | [PubMed]
 
Xu  B;  Roos  JL;  Levy  S;  van Rensburg  EJ;  Gogos  JA;  Karayiorgou  M:  Strong association of de novo copy number mutations with sporadic schizophrenia.  Nat Genet 2008; 40:880–885
[CrossRef] | [PubMed]
 
Malhotra  D;  McCarthy  S;  Michaelson  JJ;  Vacic  V;  Burdick  KE;  Yoon  S;  Cichon  S;  Corvin  A;  Gary  S;  Gershon  ES;  Gill  M;  Karayiorgou  M;  Kelsoe  JR;  Krastoshevsky  O;  Krause  V;  Leibenluft  E;  Levy  DL;  Makarov  V;  Bhandari  A;  Malhotra  AK;  McMahon  FJ;  Nöthen  MM;  Potash  JB;  Rietschel  M;  Schulze  TG;  Sebat  J:  High frequencies of de novo CNVs in bipolar disorder and schizophrenia.  Neuron 2011; 72:951–963
[CrossRef] | [PubMed]
 
Itsara  A;  Wu  H;  Smith  JD;  Nickerson  DA;  Romieu  I;  London  SJ;  Eichler  EE:  De novo rates and selection of large copy number variation.  Genome Res 2010; 20:1469–1481
[CrossRef] | [PubMed]
 
Levy  D;  Ronemus  M;  Yamrom  B;  Lee  YH;  Leotta  A;  Kendall  J;  Marks  S;  Lakshmi  B;  Pai  D;  Ye  K;  Buja  A;  Krieger  A;  Yoon  S;  Troge  J;  Rodgers  L;  Iossifov  I;  Wigler  M:  Rare de novo and transmitted copy-number variation in autistic spectrum disorders.  Neuron 2011; 70:886–897
[CrossRef] | [PubMed]
 
Pinto  D;  Pagnamenta  AT;  Klei  L;  Anney  R;  Merico  D;  Regan  R, et al:  Functional impact of global rare copy number variation in autism spectrum disorders.  Nature 2010; 466:368–372
[CrossRef] | [PubMed]
 
Sanders  SJ;  Ercan-Sencicek  AG;  Hus  V;  Luo  R;  Murtha  MT;  Moreno-De-Luca  D;  Chu  SH;  Moreau  MP;  Gupta  AR;  Thomson  SA;  Mason  CE;  Bilguvar  K;  Celestino-Soper  PB;  Choi  M;  Crawford  EL;  Davis  L;  Wright  NR;  Dhodapkar  RM;  DiCola  M;  DiLullo  NM;  Fernandez  TV;  Fielding-Singh  V;  Fishman  DO;  Frahm  S;  Garagaloyan  R;  Goh  GS;  Kammela  S;  Klei  L;  Lowe  JK;  Lund  SC;  McGrew  AD;  Meyer  KA;  Moffat  WJ;  Murdoch  JD;  O’Roak  BJ;  Ober  GT;  Pottenger  RS;  Raubeson  MJ;  Song  Y;  Wang  Q;  Yaspan  BL;  Yu  TW;  Yurkiewicz  IR;  Beaudet  AL;  Cantor  RM;  Curland  M;  Grice  DE;  Günel  M;  Lifton  RP;  Mane  SM;  Martin  DM;  Shaw  CA;  Sheldon  M;  Tischfield  JA;  Walsh  CA;  Morrow  EM;  Ledbetter  DH;  Fombonne  E;  Lord  C;  Martin  CL;  Brooks  AI;  Sutcliffe  JS;  Cook  EH  Jr;  Geschwind  D;  Roeder  K;  Devlin  B;  State  MW:  Multiple recurrent de novo CNVs, including duplications of the 7q11.23 Williams syndrome region, are strongly associated with autism.  Neuron 2011; 70:863–885
[CrossRef] | [PubMed]
 
Vacic  V;  McCarthy  S;  Malhotra  D;  Murray  F;  Chou  HH;  Peoples  A;  Makarov  V;  Yoon  S;  Bhandari  A;  Corominas  R;  Iakoucheva  LM;  Krastoshevsky  O;  Krause  V;  Larach-Walters  V;  Welsh  DK;  Craig  D;  Kelsoe  JR;  Gershon  ES;  Leal  SM;  Dell Aquila  M;  Morris  DW;  Gill  M;  Corvin  A;  Insel  PA;  McClellan  J;  King  MC;  Karayiorgou  M;  Levy  DL;  DeLisi  LE;  Sebat  J:  Duplications of the neuropeptide receptor gene VIPR2 confer significant risk for schizophrenia.  Nature 2011; 471:499–503
[CrossRef] | [PubMed]
 
Manning  M;  Hudgins  L; Professional Practice and Guidelines Committee:  Array-based technology and recommendations for utilization in medical genetics practice for detection of chromosomal abnormalities.  Genet Med 2010; 12:742–745
[CrossRef] | [PubMed]
 
St Clair  D:  Copy number variation and schizophrenia.  Schizophr Bull 2009; 35:9–12
[CrossRef] | [PubMed]
 
Gershon  ES;  Hamovit  J;  Guroff  JJ;  Dibble  E;  Leckman  JF;  Sceery  W;  Targum  SD;  Nurnberger  JI  Jr;  Goldin  LR;  Bunney  WE  Jr:  A family study of schizoaffective, bipolar I, bipolar II, unipolar, and normal control probands.  Arch Gen Psychiatry 1982; 39:1157–1167
[CrossRef] | [PubMed]
 
Doukas  DJ;  Berg  JW:  The family covenant and genetic testing.  Am J Bioeth 2001; 1:3–10
[CrossRef] | [PubMed]
 
Fulda  KG;  Lykens  K:  Ethical issues in predictive genetic testing: a public health perspective.  J Med Ethics 2006; 32:143–147
[CrossRef] | [PubMed]
 
Fraser  S:  The Bell Curve Wars: Race, Intelligence, and the Future of America .  New York,  Basic Books, 1995
 
Herrnstein  RJ;  Murray  C:  The Bell Curve: Intelligence and Class Structure in American Life .  New York,  Simon & Schuster, 1994
 
Grosse  SD;  Rogowski  WH;  Ross  LF;  Cornel  MC;  Dondorp  WJ;  Khoury  MJ:  Population screening for genetic disorders in the 21st century: evidence, economics, and ethics.  Public Health Genomics 2010; 13:106–115
[CrossRef] | [PubMed]
 
Sanders  SJ;  Murtha  MT;  Gupta  AR;  Murdoch  JD;  Raubeson  MJ;  Willsey  AJ;  Ercan-Sencicek  AG;  DiLullo  NM;  Parikshak  NN;  Stein  JL;  Walker  MF;  Ober  GT;  Teran  NA;  Song  Y;  El-Fishawy  P;  Murtha  RC;  Choi  M;  Overton  JD;  Bjornson  RD;  Carriero  NJ;  Meyer  KA;  Bilguvar  K;  Mane  SM;  Sestan  N;  Lifton  RP;  Günel  M;  Roeder  K;  Geschwind  DH;  Devlin  B;  State  MW:  De novo mutations revealed by whole-exome sequencing are strongly associated with autism.  Nature 2012; 485:237–241
[CrossRef] | [PubMed]
 
Xu  B;  Roos  JL;  Dexheimer  P;  Boone  B;  Plummer  B;  Levy  S;  Gogos  JA;  Karayiorgou  M:  Exome sequencing supports a de novo mutational paradigm for schizophrenia.  Nat Genet 2011; 43:864–868
[CrossRef] | [PubMed]
 
Gamazon  ER;  Badner  JA;  Cheng  L;  Zhang  C;  Zhang  D;  Cox  NJ;  Gershon  ES;  Kelsoe  JR;  Greenwood  TA;  Nievergelt  CM;  Chen  C;  McKinney  R;  Shilling  PD;  Schork  NJ;  Smith  EN;  Bloss  CS;  Nurnberger  JI;  Edenberg  HJ;  Foroud  T;  Koller  DL;  Scheftner  WA;  Coryell  W;  Rice  J;  Lawson  WB;  Nwulia  EA;  Hipolito  M;  Byerley  W;  McMahon  FJ;  Schulze  TG;  Berrettini  WH;  Potash  JB;  Zandi  PP;  Mahon  PB;  McInnis  MG;  Zollner  S;  Zhang  P;  Craig  DW;  Szelinger  S;  Barrett  TB;  Liu  C:  Enrichment of cis-regulatory gene expression SNPs and methylation quantitative trait loci among bipolar disorder susceptibility variants.  Mol Psychiatry 2013; 18:340–346
[CrossRef] | [PubMed]
 
Nicolae  DL;  Gamazon  E;  Zhang  W;  Duan  S;  Dolan  ME;  Cox  NJ:  Trait-associated SNPs are more likely to be eQTLs: annotation to enhance discovery from GWAS.  PLoS Genet 2010; 6:e1000888
[CrossRef] | [PubMed]
 
Psychiatric GWAS Consortium Bipolar Disorder Working Group:  Large-scale genome-wide association analysis of bipolar disorder identifies a new susceptibility locus near ODZ4.  Nat Genet 2011; 43:977–983
[CrossRef] | [PubMed]
 
Steinberg  S;  de Jong  S; Irish Schizophrenia Genomics Consortium; Andreassen  OA;  Werge  T;  Børglum  AD;  Mors  O;  Mortensen  PB;  Gustafsson  O;  Costas  J;  Pietiläinen  OP;  Demontis  D;  Papiol  S;  Huttenlocher  J;  Mattheisen  M;  Breuer  R;  Vassos  E;  Giegling  I;  Fraser  G;  Walker  N;  Tuulio-Henriksson  A;  Suvisaari  J;  Lönnqvist  J;  Paunio  T;  Agartz  I;  Melle  I;  Djurovic  S;  Strengman  E; GROUP; Jürgens  G;  Glenthøj  B;  Terenius  L;  Hougaard  DM;  Ørntoft  T;  Wiuf  C;  Didriksen  M;  Hollegaard  MV;  Nordentoft  M;  van Winkel  R;  Kenis  G;  Abramova  L;  Kaleda  V;  Arrojo  M;  Sanjuán  J;  Arango  C;  Sperling  S;  Rossner  M;  Ribolsi  M;  Magni  V;  Siracusano  A;  Christiansen  C;  Kiemeney  LA;  Veldink  J;  van den Berg  L;  Ingason  A;  Muglia  P;  Murray  R;  Nöthen  MM;  Sigurdsson  E;  Petursson  H;  Thorsteinsdottir  U;  Kong  A;  Rubino  IA;  De Hert  M;  Réthelyi  JM;  Bitter  I;  Jönsson  EG;  Golimbet  V;  Carracedo  A;  Ehrenreich  H;  Craddock  N;  Owen  MJ;  O’Donovan  MC; Wellcome Trust Case Control Consortium 2, Ruggeri  M;  Tosato  S;  Peltonen  L;  Ophoff  RA;  Collier  DA;  St Clair  D;  Rietschel  M;  Cichon  S;  Stefansson  H;  Rujescu  D;  Stefansson  K:  Common variants at VRK2 and TCF4 conferring risk of schizophrenia.  Hum Mol Genet 2011; 20:4076–4081
[CrossRef] | [PubMed]
 
References Container
+
+

CME Activity

There is currently no quiz available for this resource. Please click here to go to the CME page to find another.
Submit a Comments
Please read the other comments before you post yours. Contributors must reveal any conflict of interest.
Comments are moderated and will appear on the site at the discertion of APA editorial staff.

* = Required Field
(if multiple authors, separate names by comma)
Example: John Doe



Related Content
See Also...
Articles
Books
DSM-5™ Clinical Cases > Chapter 2.  >
DSM-5™ Clinical Cases > Chapter 18.  >
Dulcan's Textbook of Child and Adolescent Psychiatry > Chapter 52.  >
DSM-5™ Handbook of Differential Diagnosis > Chapter 2.  >
DSM-5™ Handbook of Differential Diagnosis > Chapter 2.  >
Topic Collections
Psychiatric News
APA Guidelines
PubMed Articles