Velo-cardio-facial syndrome (VCFS), also known as 22q11.2 deletion syndrome and DiGeorge syndrome, is caused by an interstitial deletion of approximately 40 genes at the q11.2 locus of chromosome 22 (1
). Children with VCFS have physical anomalies, intellectual deficits and psychiatric disorders. Common childhood disorders include attention deficit hyperactivity disorder, phobias, generalized anxiety disorder and major depression (3
). As youth with VCFS reach adulthood, up to 32% develop psychotic disorders, including schizophrenia and schizo-affective disorder (2
). Accordingly, VCFS is the highest known genetic risk factor for schizophrenia next to having two parents with the illness (9
). The onset of psychosis in young adults with VCFS is accompanied by a decline in verbal IQ (2
The high co-occurrence of schizophrenia and VCFS has increased interest in identifying biological risk factors for psychosis in VCFS individuals. Since neuroanatomic anomalies in the non-VCFS population have been reported [albeit inconsistently, see review by Pantelis (10
)] in youth at familial (11
) and clinical (15
) high risk for schizophrenia, and in first episode patients with schizophrenia (16
), several studies have focused on the brain structure of individuals with VCFS. Like youth at risk for schizophrenia, youth with VCFS display reductions in volumes of the hippocampus (18
), inferior frontal lobe (19
) and anterior cingulate gyrus (20
). In addition, associations have been reported between volumes of temporo-occipital cortex and corpus striatum, and schizotypy in VCFS (21
). However without longitudinal data, it is not clear whether these neuroanatomic anomalies are early markers for psychosis or characteristics of VCFS.
Only one longitudinal study of volumetric alterations (22
) and another of cortical thickness (23
) in youth with VCFS have been published. Gothelf and colleagues (22
) examined neuroanatomic trajectories in nineteen individuals with VCFS and eighteen controls. Youth with VCFS displayed significant longitudinal reductions in amygdala volumes, and longitudinal increases in superior temporal gyrus gray matter, cranial white matter and cerebellar white matter. Although longitudinal decline in verbal IQ was associated with psychotic disorder at Time 2, longitudinal alterations in neuroanatomic regions of interest were not. Similarly, Schaer and colleagues (23
) did not observe an association between longitudinal alterations in cortical thickness and psychosis, although they observed cortical thickness reductions in superior and inferior temporal gyri in a cross-sectional sample of adults with VCFS and psychosis relative to VCFS-affected adults without psychosis.
Herein, we report a longitudinal study of youth with VCFS, their unaffected siblings and community controls. Based on the literatures cited above, we measured several neuroanatomic regions of interest including lobar gray and white matter (22
), frontal lobe subregions ((25
), amygdala (27
), superior temporal gyrus (24
), and cerebellum (24
). By recruiting over 70 youth with VCFS and administering dimensional and categorical measures of psychosis, we address limitations of previous longitudinal studies of VCFS, which were underpowered to detect associations between neuroanatomic trajectories and psychosis.
We hypothesized: 1) relative to controls, volumetric changes between Time 1 and Time 2 would be observed in amygdala, superior temporal gyrus, cranial white matter and cerebellum; 2) associations would be observed between a decline in verbal IQ score and prodromal symptoms; and 3) volumetric changes between Time 1 and Time 2 in hippocampus, temporal gray matter, and inferior frontal lobe would predict to prodromal symptoms.