Quantitative, diagnosis-independent traits, known as intermediate phenotypes or endophenotypes, may identify more homogeneous patient groups than do psychiatric diagnoses. The Bipolar-Schizophrenia Network on Intermediate Phenotypes (B-SNIP) examined potential intermediate phenotypes in probands with schizophrenia, bipolar disorder with psychosis, or schizoaffective disorder and in their relatives and healthy subjects. Tamminga et al. (p. 1263) found that symptoms and psychosocial functioning were generally worse in schizophrenia probands than in probands with the other diagnoses, but there was considerable overlap. Likewise, many families contained both schizophrenia and bipolar diagnoses. Hill et al. (p. 1275) report increasingly severe cognitive impairment from bipolar disorder to schizoaffective disorder to schizophrenia. Family rates of cognitive deficits were comparable in schizophrenia and bipolar disorder, but in bipolar disorder the deficits mainly affected relatives with cluster A or cluster B personality traits. Measurements of gray matter volume by Ivleva et al. (p. 1285) showed differences between psychotic and nonpsychotic individuals and between diagnoses: schizophrenia and schizoaffective probands had overlapping areas of low gray matter volume in numerous cortical and subcortical regions, whereas bipolar probands had fewer low-volume areas. The editorial by Frangou (p. 1223) notes that structural imaging reflects the static brain configuration, whereas cognitive performance is an expression of a dynamic configuration.
Clinical Guidance: Personalizing Panic Disorder Treatment
Many patients diagnosed as having panic disorder with agoraphobia benefit from cognitive-behavioral therapy (CBT) even if they do not respond fully, but functional brain imaging provides clues about who is likely to improve most. Lueken et al. (p. 1345) compared pretreatment responses to a fear-conditioning task in CBT responders and nonresponders. Nonresponders had greater neural responses to a safety signal, raising the possibility that CBT may not be as effective for patients with a bias toward detecting threat. Successful treatment was associated with a posttreatment increase in hippocampal activation during processing of stimulus contingencies. This suggests awareness of stimulus-outcome associations, another quality that may help identify patients who will benefit the most from CBT. Editorialist Henn (p. 1235) cites deep brain stimulation for depression as an example of how knowledge of brain circuits can inform treatment.