Schizophrenia: Family Prevalence Studies, Twin, Adoption

Family Prevalence Studies

Wide agreement now exists that the rate of schizophrenia among first-degree family members of persons with schizophrenia is higher than in control families. The chance of occurrence is approximately 10 times greater among these individuals than in individuals with no first-degree relatives with schizophrenia. There is approximately six times and two times greater chance of developing schizophrenia in second- and third-degree relatives of individuals with schizophrenia respectively. In addition, the higher prevalence of schizophrenia spectrum disorders among family members of individuals with schizophrenia, such as schizoaffective disorder and schizoid and schizotypal personality disorders, provides support for a common genetic basis for this family of schizophrenia-like illnesses.

Adoption Studies

Adoption studies constitute a powerful experimental strategy for examining the role of genetic versus environmental factors. In these studies, the rates of schizophrenia are compared in relatives of adoptees with and without schizophrenia. Danish adoption studies conducted in the 1960s and 1970s provided compelling evidence that adoptees with schizophrenia had higher rates of schizophrenia in their first-degree relatives than control adoptees. A reanalysis of these data in the late 1980s confirmed the original finding that biological relatives of schizophrenia adoptees had significantly higher rates of schizophrenia (4.1%) than biological relatives of nonschizophrenia (control) adoptees (0.5%). Several methodo-logical issues are important in interpreting the data from adoption studies, including the diagnostic status of biological fathers and levels of psychopathology in adoptive families. Additional factors to consider are the intrauterine environment, birth complications and length of time from birth to adoptive placement.

Twin Studies

Another approach to examining genetic contributions to schizo-phrenia involves concordance studies of dizygotic (nonidentical) and monozygotic (identical) twin pairs. Available data indicate that the concordance of schizophrenia among dizygotic twins is approximately 8 to 12%. This is much greater than the 1% rate found in the general population and comparable to the rate of concordance of schizophrenia among first-degree siblings. The concordance of schizophrenia among monozygotic twins is ap-proximately 50%. Even though the high rate of concordance among monozygotic twin pairs is compelling evidence for ge-netic contributions, the fact that it is not higher than 50% suggests a role for additional, perhaps nongenetic, factors in the etiology of schizophrenia. The number of sets of adopted-away, monozy-gotic twin pairs affected with schizophrenia is relatively small. However, data available on the limited number of pairs meeting these criteria support the strong concordance of schizophrenia in monozygotic twins.

Linkage and Association Studies

The Human Genome Project with its 3 billion base pairs and approximately 35 000 genes has ushered us into the “Genomic Era”. However, systematic genome scans done recently have not resulted in strong evidence for linkage to any chromosomal region. At the time of this writing, no genetic linkage or asso-ciation related to schizophrenia has been discovered. There have been reports of suggestive linkages but there has been a failure to replicate these findings. It has become painfully clear that the replication studies are, in many ways, more important to estab-lishing linkage than the initial report. Genes that have been found not to be associated with schizophrenia include the dopamine D₂ and D₄ genes.

The initial enthusiasm for these strategies has waned as no gene has yet been isolated for schizophrenia or bipolar dis-order. Impediments that have limited the success of linkage and association studies are etiologic and phenotypic heterogeneity of schizophrenia, lack of power and high false positive rates. Thus, newer approaches such as multi-investigator collaborative stud-ies to increase the power have already been implemented.

Meanwhile, modern functional genomic approaches such as DNA microarrays, based on the principles of nucleic acid hy-bridization, can check a tissue sample for presence of thousands of genes simultaneously. For example, Mirnics and colleagues (2000) employed cDNA microarrays and compared transcrip-tomes in schizophrenia and matched control subjects and found that only a few gene groups consistently differed between sub-jects and controls. In all subjects with schizophrenia, the most changed gene group was related to presynaptic group secretory function (PSYN) gene group and in particular the “mechanics” of neurotransmitter release.

Weinberger and colleagues (2001) suggest that the gene that encodes the postsynaptic enzyme catechol-o-methyl trans-ferase (COMT) is preferentially involved in the metabolism of dopamine in frontal lobe. Dopamine is hypothesized to underlie aspects of cognition in frontal lobe such as information process-ing. Based on animal studies, family-based association studies and fMRI studies in schizophrenia patients and general popula-tion, Weinberger and colleagues propose an interesting hypoth-esis that the COMT genotype with valenine allele (val/val type) may increase the risk of developing schizophrenia due to its ef-fect on dopamine-mediated prefrontal information processing.

Researchers are urgently searching for schizophrenia phenotypes for subgroups or dimensions that may define etio-logically or genetically distinct subtypes. Similarly, the field is yearning for endophenotypes with simpler architecture than schizophrenia possibly to guide to newer leads in research. Latent genetically influenced traits, which may be related only indirectly to the classic disease symptoms defined in major clas-sification systems, are known as “endophenotypes”. They reflect an underlying susceptibility to the disease phenotype (or some form of it). In schizophrenia we are interested in endophenotypes that are measurable by neurophysiological or neuropsychological means. Crucial characteristics of any endophenotype include the fact that it can be measured before the explicit onset of the illness, and that it represents the genetic liability of nonaffected relatives of probands with the disorder.