We frequently teach what we learned as residents. This ensures the transmission of information from one generation to the next, creating a pedigree for knowledge in psychiatry that can be tempered by experience. But what happens when there is a sudden intellectual shift in a field? Will residency training be a generation behind? Two articles in this issue (1, 2) suggest that psychiatric training has missed a major intellectual innovation and remains apparently unaware of the recent genomic revolution.
Genomics has been transforming medicine over the past decade. In 2003, the successful completion of the first phase of the human genome project (3) provided a map of the 3 billion bases of the human genome. This was critical for identifying all 20,000+ genes, essentially describing a biological blueprint for humans. In 2005, the first comprehensive map of human variation was completed. This “Hap Map,” named for the location of haplotypes or common patterns of human genetic variation in DNA sequence, demonstrated more variation than expected, with roughly 3 million points of common variation in the genome. Over the next 2 years, association of this common genomic variation with a plethora of clinical phenotypes transformed our understanding of macular degeneration, inflammatory bowel disease, diabetes, and many other common disorders (4). Although it has been known for many decades that mental disorders are often more heritable than other medical disorders, the search for common variations yielded only a few success stories in psychiatry. The surprise for psychiatry came in 2007 when several rare, large-scale variations were detected in people with schizophrenia and autism (5, 6). In contrast to the common, single base changes in other diseases, copy number variations often involve a million bases of DNA duplicated or quadruplicated. These copy number variations, sometimes called structural variations, were found 10 times more often in psychiatric patients. Although each copy number variation may be rare, many such variations have been identified. In aggregate, these structural changes may occur in 15% of children with autism and potentially 5% of patients with schizophrenia (7).
Mapping the human genome, human variation, and structural variation might seem like enough excitement for 1 decade, but perhaps the most important breakthrough for psychiatry has occurred in the past year with the first maps of the human epigenome (8). If the genome is the text of the DNA sequence, the epigenome describes the modifications of this text due to proteins binding to the DNA. Protein modifications are essentially ways of highlighting or redacting large stretches of DNA, determining which parts of the genome will be translated into RNA. This is important for psychiatry because epigenomics is thought to be the mechanism by which experience, especially experience during development, can have long-term effects on gene expression and behavior.
How are these revolutionary discoveries being taught to psychiatric residents? The two surveys published in this issue (1, 2) present a sobering picture—little exposure to training in genomics, a lack of understanding of important concepts in genomics, and a dearth of faculty with adequate expertise in teaching the subject. Among the most disconcerting findings of the study by Winner et al. (1) was that 56% of survey responders had no or very little genomic training as residents, and the vast majority of residents did not know which patients could benefit from genetic counseling. In the study by Hoop et al. (2), residents ranked the importance of genomics training last on a list of qualitative skill sets. Furthermore—and disturbingly—only 29% of residents felt competent to interpret papers on psychiatric genetics, and among educators this proportion only increased to 42%. Not surprisingly, educators assessed the proportion of faculty with adequate expertise in genomics as very low.
Genomic medicine is a young and quickly evolving field that currently holds more promise for paradigm-changing translational breakthroughs than practical applications for clinicians of all specialties. This may have led to a feeling of unfulfilled promise and a belief that genomic medicine is not applicable to the practice of psychiatry among many faculty and residents despite efforts by the research (9) and educator (10) communities to emphasize the importance of this area for training of the next generations of psychiatrists. Notwithstanding demands on time, competing priorities, and the overall daunting task of preparing trainees to successfully practice psychiatry, we think it would be a mistake to forgo rigorous training in the concepts of genomics for three reasons.
First, genomic research is changing our understanding of several mental disorders. Although most of these discoveries are premature for practice, they are shifting our understanding of pathophysiology. For instance, nearly all of the genes associated with autism code for synaptic proteins, and many of the genes for schizophrenia are critical for brain development. Although approaching autism as a synaptic disease or schizophrenia as a developmental brain disorder may not translate into a treatment recommendation today, this is the information that will be the basis of treatments in the next generation. The intellectual foundation of psychiatry now needs to include this kind of information in addition to theories about the mechanisms of antipsychotic drug action or models of metapsychology. If we are training residents for tomorrow and not yesterday, they will need fundamental skills for understanding the literature and critiquing reports of new findings based on the biology of mental illness.
Second, some aspects of genetic information are relevant for practice today. Mental disorders are the result of complex interactions among multiple genetic and nongenetic (e.g., environmental) factors that have a powerful proxy in family history (11). In the absence of a mechanistic understanding of these disorders, family history remains the strongest predictor of risk for the development of a psychiatric disorder (12) and can be invaluable for diagnosis and treatment. Two examples may help. Behavioral changes in a 55-year-old man with a family history of Fragile X syndrome suggest the need for genetic evaluation for potential Fragile X permutation, which is associated with symptoms in late life (13). Or consider the case of 22q11 deletion syndrome, or velocardiofacial syndrome (VCFS), which is the most common microdeletion syndrome and includes nearly a 50-fold increased risk of developing schizophrenia (14). Recent discoveries of many similar but rare structural variants suggest that schizophrenia and autism are heterogeneous disorders with potentially overlapping pathogenic molecular lesions. Psychiatrists treating this patient population need to have an in-depth understanding of genetic risk that includes experience with complex issues ranging from genetic testing and counseling to treatment in this patient population.
A third reason for psychiatrists to be well versed in the language of genomics and its implication for clinical practice is to keep up with their patients. Newly formed companies like 23andMe or Psynomics built their business models around direct-to-consumer marketing of genetic testing and personalized information products providing genetic risk assessments from individual genetic data. Relative and absolute risk increases that are reported to users of these services are often small and of little or no clinical significance. Nevertheless, these companies report genetic risks for schizophrenia, bipolar disorder, and drug response to treatment for depression to the patient without physician involvement. Psychiatrists need to be prepared for patients who come to them with this information requesting counseling and, ultimately, clinical care and advice.
Genomic medicine is here to stay and will soon take center stage in many clinical disciplines. A modern psychiatry that strives for a mechanistic understanding, diagnosis, and treatment of clinical disorders and their underlying phenotypes needs strong foundations in the basic sciences with an emphasis on molecular neuroscience and genomics. This needs to be reflected in the curricula of psychiatric residency programs and in the expertise of the teaching faculty. We believe that psychiatric residency programs should include teaching the genetics of complex disorders with an emphasis on mental disorders, concepts of genetic risk and genetic counseling and the ethical, legal, and social issues of genomic medicine in their core curriculum. Most important, genomic medicine is an evolving concept and requires continual learning and training that has to transcend a residency program and become part of the continuing education of the psychiatrist of the 21st century.
At the time of submission, the authors reported no competing interests.