Individualizing Recurrence Risks for Severe Mental Illness: Epidemiologic and Molecular Genetic Approaches

The meta-analysis conducted by Rasic et al.1 expands on established empiric recurrence risks for severe mental illness (SMI; schizophrenia, bipolar disorder, and major depressive disorder) in offspring of affected individuals and represents a valuable addition to the existing literature. The advantages (and caveats) of these new findings for application in clinical practice deserve special consideration, particularly in light of recent molecular genetic discoveries. Concern about the heritability of SMI is common.2–4 Genetic counseling may be a therapeutic intervention capable of addressing and attenuating such concerns.2,3,5 The results of this meta-analysis will be of assistance in addressing direct questions from patients and family members with respect to a broader phenotype than heretofore possible with the long-standing individual risk estimates for SMI. For schizophrenia, the benefits of forewarning could include a sense of relief for those found to be at a risk that they perceive to be low,2,3 and an increased likelihood of early diagnosis and a shortened duration of untreated psychosis (with concomitant benefits with respect to functional outcomes)6 for those at higher risk.7 These new risk estimates could have been useful in our studies of genetic counseling, which were recently published in the Bulletin.2,3 There remain many limitations inherent in the existing risk estimates available for patients with SMI and their family members, however. Empiric recurrence risks are averaged, aggregate numbers computed from families with widely disparate histories, and may be relatively meaningless on an individual level. Clinical heterogeneity is one important confounder. Familial recurrence risks of SMI in the case of depression, eg, might depend on factors such as psychosis, recurrence, electroconvulsive therapy/other biological treatments, and hospitalizations; there are also known cohort effects for depression.8 In another example, by definition all affected parents are reproductively fit, hence the smaller number of parents with schizophrenia in this meta-analysis, and the potential enrichment within parents with schizophrenia for women and higher functioning individuals. The context of the high prevalence in the general population of many of the psychiatric conditions studied in offspring of individuals with SMI is not reflected in their relative risk estimates and should be provided as background in any counseling. Family history is important on both sides of the family, and as briefly acknowledged by the authors, assortative mating could play a major role in their findings. Finally, there also remains little to no ability to adjust recurrence risk estimates to account for additional nonparent relatives with SMI or spectrum conditions. Psychiatrists and other mental healthcare providers should therefore exercise caution when quoting recurrence risks in clinical practice and consult with/refer to genetic counselors or other genetics professionals as needed. In regard to the meta-analysis by Rasic et al.1 and the studies considered therein, what additional insights are provided by recent molecular genetic studies of SMI? First, not mentioned by the authors are the main clinically relevant molecular findings of recent years, ie, large rare copy number variants (CNVs). The variable expression of large rare CNVs has revealed a genetically related spectrum of disease that embraces conditions outside of those usually considered part of the schizophrenia spectrum. This molecular genetic spectrum includes developmental delay/intellectual disability, autism spectrum disorder (ASD), epilepsy, and congenital anomalies.9 As the authors point out for ASD,1 limited familial recurrence risk data have been collected with regard to this other spectrum. The finding of shared underlying susceptibilities for multiple neuropsychiatric and developmental disorders is consistent with discoveries in other fields, eg, cancer and autoimmune conditions.10,11 These other fields provide a glimpse of what might be possible with respect to diagnostics and interventions in the coming years for psychiatry, as a consequence of an improved understanding of etiopathogenesis. Second, presymptomatic risk prediction in schizophrenia is likely to be fundamentally linked to the underlying genetic architecture.7 Polygenic scores12 or other methods of universal risk prediction based on common sequence variants (single nucleotide polymorphisms) may be incremental and of small effect. In contrast, the identification of specific clinically informative genetic variants may be rare events but of large effect. In the latter case, the 22q11.2 deletions found in up to 1 in 100 individuals with schizophrenia provide a model for what might be increasingly possible in clinical practice: the ability to stratify and modify recurrence risk estimates on an individual basis using the results of clinically available molecular genetic tests (figure 1). Eventually, general risk stratification may be possible with other large rare CNVs that underlie emerging genetic subtypes of schizophrenia.7,13 For these and other reasons, individuals with such clinically significant CNVs should be excluded from studies of general genetic counseling for schizophrenia.2,3 Fig. 1. Discovery of a 22q11.2 deletion facilitates recurrence risk stratification for severe mental illness (SMI) in offspring. The 22q11.2 deletion syndrome (22q11.2DS) is associated with a 20%–25% lifetime risk for schizophrenia.14 Knowledge of an ... For the foreseeable future, the family history and empiric recurrence risk estimates will remain a cornerstone of disease prediction.16 The results of the meta-analysis by Rasic and colleagues1 therefore represent an important contribution and will help to inform genetic counseling. Diagnosis and treatment of schizophrenia and other SMI remain primarily a reactive process, even in the presence of major risk factors such as a family history of SMI in a first-degree relative. Accurate prediction of risk remains a major goal in anticipation of future preventative or palliative presymptomatic measures. Future studies that combine epidemiologic and molecular genetic approaches will bring us closer to the ideal: a coherent, overarching risk prediction model for schizophrenia and related SMI that incorporates complete family history and personal genotype information.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.