Schizophrenia usually emerges in late adolescence or early adulthood (
1,
2). If cognitive deterioration occurs, it generally appears early and with relatively stable impairment over the next 5–10 years. Later in the illness, psychotic symptoms may become less intense, and there can be modest improvement in function later in life (
3–
5). Mr. E’s early development was abnormal and showed multiple indications of a genetic syndrome such as 22q11 deletion syndrome (22q11DS) (e.g., cleft palate, dysmorphic features, and developmental delay). Mr. E’s initial psychiatric presentation was consistent with idiopathic schizophrenia, but treatment resistance was notable. He subsequently developed profound dementia with a hyperkinetic movement disorder. Genetic analysis, clinical evaluation, and neuropathology provided definite diagnoses of 22q11 deletion syndrome and Huntington’s disease.
To our knowledge, this is the only reported co-occurrence of 22q11DS and Huntington’s disease. The prevalence of the 22q11DS is around 1 in 4,000 live births (
6–
9), and a pathogenic trinucleotide expansion in
HTT occurs in approximately 1 in 10,000 live births (
10). These genetic variants are on different chromosomes, and we are not aware of a mutational mechanism that could predispose to both events. If these were independent events, the probability of co-occurrence would be around 1/40 million, or fewer than 10 similar cases in the United States. The co-occurrence of multiple genetic syndromes (referred to as “blended” phenotypes) has been reported in
∼4% of individuals referred for clinical genetic testing (
11,
12), particularly in people like Mr. E, with multisystem disease.
22q11DS is caused by a hemizygous deletion (i.e., the presence of one instead of two copies) of a region on chromosome 22 (
13). The deleted region is usually
∼3 million bases (chr22:18.6–21.7 Mb, hg19), contains over 50 genes, and is a de novo event in
∼90% of cases (
7,
14). 22q11DS has been observed worldwide because the fine structure of this region increases the chances of its occurrence (the deleted region is flanked by low copy repeat regions that predispose to nonallelic homologous recombination). The clinical presentation of 22q11DS is variable and can include developmental delay, intellectual disability, cardiac anomalies, palatal defects, immunodeficiency, thrombocytopenia, and hypocalcemia (
15). The neuropsychiatric concomitants of 22q11DS are diverse. It is the strongest known genetic risk factor for schizophrenia (with a prevalence of 0.3% in cases, and a genotypic relative risk >20) (
16,
17). While 25%−30% of individuals with 22q11DS will develop psychosis, it is also associated with autism, attention deficit hyperactivity disorder, pervasive developmental delay, and early-onset Parkinson’s disease (
18).
Huntington’s disease is a dominantly inherited neurodegenerative disorder (
19). Most cases (
∼90%) are inherited (
20). It is caused by a mutation in the first exon of the huntingtin gene (
HTT), where three DNA bases (cytosine-adenine-guanine, or CAG) are repeated multiple times (
21,
22). The number of CAG repeats is normally in the range of 11–35, but if there are ≥40 CAG repeats, Huntington’s disease virtually always occurs, as the resulting protein becomes selectively toxic to neurons in the caudate, putamen, and deep layers of frontal and parietal cortex (
23,
24). Key manifestations of Huntington’s disease include chorea, cognitive decline or dementia, and psychiatric symptoms (
25). Although choreiform movements are its diagnostic hallmark, cognitive and psychiatric manifestations often predate motor signs (
26,
27). Although mood symptoms are common, psychotic symptoms occur in 6%−25% of cases of Huntington’s disease (
25,
28–
30).
With the benefit of hindsight and a genotype-driven review of all available medical records, it is apparent that Mr. E manifested multiple physical signs (e.g., hernia, cleft palate, facial dysmorphism, thrombocytopenia, and hypocalcemia), developmental signs (developmental delay, lowered IQ), and psychosis, collectively strongly suggestive of 22q11DS (
31). Mr. E subsequently experienced an early-onset progressive dementia due to Huntington’s disease. The onset of cognitive decline in Huntington’s disease usually occurs in the mid-40s, but it can be earlier in individuals with longer CAG repeats (
32), as was the case with Mr. E. Mr. E had dyskinetic as well as choreiform movements in the context of prolonged antipsychotic use, which may be why Huntington’s disease was not suspected earlier. Mr. E’s chronic psychotic symptoms during most of his third decade support a diagnosis of schizophrenia. Informed clinicians disagree as to whether the presence of 22q11DS is sufficient to negate DSM-5’s criterion E (“not attributable to … another medical condition”), as it is a probabilistic but not deterministic risk factor. In either case, Huntington’s disease led to a profound neurocognitive disorder, which had an overwhelming impact during Mr. E’s last decade of life.
We speculate that Mr. E was affected both independently and interactively by this extremely rare combination of genetic insults. Mr. E’s abnormal development and predisposition to schizophrenia can be reasonably attributed to 22q11DS. The early onset and rapidly progressive dementia were certainly due to pathologically confirmed Huntington’s disease. These two features of Mr. E’s clinical course were probably independent, given their distinctive ages at onset. We speculate that these two genetic conditions may have interacted, particularly later in his illness. Brain regions jointly affected by Huntington’s disease, schizophrenia, and antipsychotic treatment converge on a limited set of neuronal cell types (e.g., medium spiny neurons of the ventral striatum and cortical pyramidal neurons) (
23). This could have been most evident in Mr. E’s atypical chorea, where an overt movement disorder may have been masked. The combination of a hypokinetic movement disorder (22q11DS increases the risk of early-onset Parkinson’s disease, associated with overactivity in the indirect striato-pallido-thalamo-cortical pathway, which predisposes to hypokinesia) (
33) and a hyperkinetic movement disorder (due to Huntington’s disease, associated with underactivity of this pathway) as well as the extrapyramidal side effects of antipsychotic treatment may have resulted in the observed atypical motoric presentation.
Recommendations and Conclusions
Given that the diagnostic yield of genetic testing for copy number variants (like 22q11DS) in people with schizophrenia is on the order of 2%−3% (
34), we advocate broader and even routine testing. Many psychiatrists evaluate people with a new-onset psychotic disorder with tests to exclude rare causes of psychosis (e.g., brain imaging, endocrine, metabolic, viral exposure, and autoimmune studies), although the diagnostic yield is considerably lower than evaluating genomic structural variants. Psychiatrists need not become experts in genetic testing but should have sufficient familiarity with and working knowledge of medical genetics to identify patients with signs and symptoms suggestive of a genetic condition that would benefit from referral. Psychiatrists can work closely with specialists in medical genetics to identify which patients should be tested, what clinical circumstances warrant closer evaluation, how to interpret results and incorporate them into clinical care, and how to convey important results to patients in a way that augments and informs the clinical process.
Certain clinical features raise the index of suspicion for the presence of a cryptic copy number variant in people with schizophrenia. Many copy number variants associated with schizophrenia also increase the risk for childhood psychiatric disorders, including intellectual disability, autism spectrum disorder, attention deficit hyperactivity disorder, and pervasive developmental delay. Because these copy number variants change the dosages of many genes in every cell, clinical features can include multisystem abnormalities. For example, as a group, copy number variants linked to schizophrenia are also associated with craniofacial dysmorphisms (e.g., cleft palate or atypical facial appearance), macro- or microcephaly, neurological disorders (e.g., epilepsy), congenital heart disease, immune and hematological dysfunction (e.g., thrombocytopenia and thymic aplasia), and renal disease. However, these additional somatic features can be subtle or absent, underscoring the importance of having a low threshold for screening for copy number variants.
In many health care systems, a clinician can order a “postnatal whole genome copy number variation by array comparative genomic hybridization.” A peripheral venous blood sample is the source of DNA for this test. The test can take several weeks. If abnormalities are found, the report will list the nature and significance of the finding and usually point to additional resources for the clinician and patient. Referral to clinical genetics may be indicated to ensure optimal care. Many copy number variants associated with schizophrenia have medical comorbidities. For instance, clinical management of people with 22q11DS requires care across medical specialties (
15). In Mr. E’s case, there was extensive clinical investigation of long-standing thrombocytopenia (deemed “idiopathic” in life), and his hypocalcemia was untreated.
Identifying copy number variants may become therapeutically important. Multiple academic groups and companies are pursuing development of therapies for people with psychiatrically important structural variants. Although there are no approved medications, the situation could change in the next decade, and knowing which patients have a large structural variant could prove important therapeutically. This can also be important to families, to connect them to support networks that are available for many rare genetic syndromes. Knowledge of recurrence risk can be important for reproductive planning.
As genomic evaluation becomes more common clinically, phenotypes associated with “classic” genetic disorders will likely expand, and additional features may be recognized as part of a syndrome (including neuropsychiatric conditions). This may be particularly important for people with long-standing psychiatric disorders who develop new signs and symptoms (e.g., dementia and a movement disorder, as with Mr. E). Although individually rare, dozens of single gene disorders can initially present with a clinical portrait confusable with idiopathic psychosis (
35), and many are detected only when new neurological or medical symptoms develop.
This is by no means a unique recommendation, but a strong case can be made for the rigorous and structured reevaluation of complex cases like Mr. E’s. In routine clinical care, important therapeutic and diagnostic decisions are often constrained by a lack of data. Systematic reassessment of medical records, psychiatric and medical reexamination, consideration of a broader differential diagnosis, and diagnostic testing can lead to new etiological insights.
In summary, this case report illustrates the value of genetic testing in psychiatry. What we call schizophrenia is a complex disorder caused by a heterogeneous disease process. The diagnosis and treatment of some proportion of people with schizophrenia will be influenced by the presence of rare, mechanistically potent variation.
Acknowledgments
The authors thank Dale Adair, M.D., Medical Director, Office of Mental Health and Substance Abuse Services, Commonwealth of Pennsylvania, for his encouragement and support for this project. The authors also thank NIMH for support in sample collection and processing. Sequencing and SNP genotyping were performed by the SNP&SEQ Technology Platform (Uppsala, Sweden) within the National Genomics Infrastructure Sweden and Science for Life Laboratory. The SNP&SEQ Platform is also supported by the Swedish Research Council and the Knut and Alice Wallenberg Foundation. Phlebotomy services were provided by Marice Davis and Kelly Bingaman.