A multilevel analysis of cognitive dysfunction and psychopathology associated with chromosome 22q11.2 deletion syndrome in children

In recent years, a great deal has been learned about a disorder that is one of the most common genetic causes of developmental disability, mental retardation, and psychopathology. Resulting from a 1.5- to 3-Mb microdeletion on the long (q) arm of chromosome 22 (Driscoll, Budarf, & Emanuel, 1992; Driscoll et al., 1992), this disorder is most accurately characterized as the “chromosome 22q11.2 deletion syndrome” (DS22q11.2). Thus defined, the disorder encompasses previously described phenotypes including DiGeorge (1965), velocardiofacial (Shprintzen, Goldberg, Lewin, Sidoti, Berkman, Argamaso, & Young, 1978), and conotruncal anomaly face (Burn, Takao, Wilson, Cross, Momma, Wadey, Scambler, & Goodship, 1993) syndromes, and some cases of Cayler cardiofacial syndrome (Giannotti, Digilio, Marino, Mingarelli, & Dallapiccola, 1994) and Opitz G/BBB syndrome (McDonald–McGinn, Driscoll, Bason, Christensen, Lynch, Sullivan, Canning, Zavod, Quinn, & Rome, 1995). A molecular fluorescence in situ hybridization probe for the deletion set the prevalence at 1 in 4,000 live births (Burn & Goodship, 1996), an estimate currently thought to be quite conservative (e.g., Shashi, Muddasani, Santos, Berry, Kwapil, Lewandowski, & Keshavan, 2004). Furthermore, several factors point to a significant growth in the identified population of individuals with DS22q11.2 in the near future. One is that the major cause for mortality in the syndrome, congenital heart defects, is now routinely resolved surgically. Another is that, because this is a contiguous gene deletion syndrome with no effect on reproductive fitness, adults with this deletion will have a 50% chance of having an affected child. A third is that growing knowledge of the disorder is increasing rates of early identification and diagnosis. Given that DS22q11.2 typically produces developmental disability and frequently results in serious psychopathology it is essential to develop a deep understanding of the cognitive and behavioral phenotype. That should improve clinical management of the disorder and may provide important clues into gene–brain–behavior relationships. Investigation of this syndrome from a cognitive neuroscience perspective can provide valuable information for basic research scientists and for clinicians. For example, studies of DS22q11.2 are likely to help explicate the neural bases of basic cognitive processes that are disturbed in other developmental disorders that produce characteristic impairments in visuospatial and numerical cognition (e.g., fragile X, Turner, and Williams–Beuren syndromes). Such studies involving children with DS22q11.2 are also likely to shed light on developing brain structure/function relationships and to what extent these can account for the observable impairments. Furthermore, investigations of disorders like DS22q11.2 can complement research into the role of genetic variation in individual differences in cognitive function (for review, see Parasuraman & Greenwood, 2004), as well as the genetic etiology of complex neuropsychiatric disorders. In contrast to traditional psychiatric genetics studies, in which one begins with a complex cluster of symptoms (i.e., psychiatric diagnosis) and attempts to map these traits to specific genetic loci, here we are investigating the phenotypic manifestations of a disorder with a known, homogeneous genetic etiology. The effects of a known alteration in a system with an identified function (e.g., the ability to produce the catechol-O-methyltransferase [COMT] enzyme, which affects regulation of synaptic dopamine particularly in the prefrontal cortex), can thus be studied in terms of its effect on a theoretically motivated research question (i.e., the genetic mediation of dysfunction in executive cognition). As we shall make clear below, a multiple levels approach to such investigations is likely to prove fruitful as a way to integrate the findings of investigations of the neurocognitive basis of developmental disability and psychopathology. Most of the early progress in understanding the effects of the deletion was related to its physical manifestations (see Figure 1 for an image of a child with the deletion). These typically involve cleft palate and/or velopharyngeal insufficiency, immune deficiencies, neonatal hypocalcemia, the aforementioned congenital heart defects, and facial dysmorphisms (Emanuel, McDonald–McGinn, Saitta, & Zackai, 2001; McDonald–McGinn et al, 1999). Although many of these symptoms are at present well understood and can be effectively treated and managed, far less is known about the apparent changes in brain structure and function, and their relation to the range of cognitive impairments and psychiatric disorders that have been reported. The first reports of these behavioral aspects of the phenotype came from standardized tests of intellectual function, academic achievement, and behavioral parent-report measures. For example, a study of 33 children and adults with DS22q11.2 reported a mean full-scale IQ in the borderline range of intellectual functioning (71.2 ± 12.8), with a mean Verbal IQ of 77.5 (±14.9), and a mean performance IQ of 69.1 (±12.0; Moss, Batshaw, Solot, Gerdes, McDonald–McGinn, Driscoll, Emmanuel, Zackai, & Wang, 1999). Similar results were reported in a Belgian sample that extended the age range from infancy to older adults (Swillen, Devriendt, Legius, Prinzie, Vogels, Ghesquiere, & Fryns, 1999). With regard to academic achievement, Moss et al. also reported that math composite scores (from the Wechsler Achievement Scales) were significantly lower than spelling and reading composite scores (80.1 ± 15.2 vs. 88.3 ± 16.4 and 86.7 ± 18.2, respectively). Analyses on enlarged samples of this same study population generally replicated and extended this pattern (e.g., Woodin, Wang, Aleman, McDonald–McGinn, Zackai, & Moss, 2001). In addition, within-subject comparisons of memory function indicated that rote verbal memory scores were significantly higher than those for visuospatial memory (Bearden, Woodin, Wang, Moss, McDonald–McGinn, Zackai, Emannuel, & Cannon, 2001). This was consistent with the findings of Wang, Woodin, Kreps–Falk, and Moss (2000) who, in a study of the basis of arithmetical cognitive impairments in this syndrome, found poorer performance on a test of visuospatial short-term memory than a test of auditory number recall. Figure 1 A 4-year-old girl with DS22q11.2. With regard to psychosocial and behavioral functioning, Swillen, Devriendt, Legius, Eyskens, Dumoulin, Gewillig, and Fryns (1997) reported results from the Child Behavior Checklist, indicating significantly elevated scores for the “social problems,” “attention problems,” and “withdrawn” sub-scales. A study focusing on younger children (13 to 63 months) with DS22q11.2 (Gerdes et al., 1999) not only reported the typical findings of delayed language, speech, and motor development, but also found that 75% of the toddlers studied exhibited behaviors during testing judged to be “highly active, emotional and disorganized,” as measured by the Bayley Scale of Infant Development (Bayley, 1969), whereas one-quarter of the toddlers in the sample were judged to be “behaviorally inhibited.” It is possible that early childhood behavioral problems in DS22q11.2 do develop into more chronic, severe mental illnesses in adolescence and early adulthood. Indeed, this syndrome is one of the highest known risk factors to date for schizophrenia, with prevalence rates on the order of 25–30% in adult patients (Bassett & Chow, 1999; Bassett, Hodgkinson, Chow, Correia, Scutt, & Weksberg, 1998; Murphy, Jones, & Owen, 1999). However, it is not known at present whether these cases are related to the childhood behavioral disorders that have been observed in DS22q11.2. In addition, elevated rates of bipolar disorder (Papolos, Faedda, Veit, Goldberg, Morrow, Kucherlapati, & Shprintzen, 1996), anxiety disorders and attention-deficit/hyperactivity disorder (ADHD; Niklasson, Rasmussen, Oskarsdottir, & Gillberg, 2001; Swillen, Devriendt, Legius, Prinzie, Vogels, Ghesquiere, & Fryns, 1999) are frequently reported. ADHD (particularly the inattentive or combined type, rather than the hyperactive type) appears to be the most common psychiatric morbidity in children and adolescents with DS22q11.2, found in 35–55% of patients. There are estimates that the prevalence of autistic spectrum disorders is between 14 and 31% (Fine, Weissman, Gerdes, Pinto–Martin, Zackai, McDonald–McGinn, & Emannuel, in press; Niklasson, Rasmussen, Oskarsdottir, & Gillberg, 2001). Oppositional defiant disorder is reported in 8–43% of children with the syndrome (Arnold, Siegel–Bartelt, Cytrynbaum, Teshima, & Schachar, 2001; Feinstein, Eliez, Blasey, & Reiss, 2002; Niklasson, Rasmussen, Oskarsdottir, & Gillberg, 2001, 2002; Papolos et al., 1996), and one study found that about one-third meet criteria for obsessive–compulsive disorder (OCD; Gothelf et al., 2004). The reason for these elevated rates of psychopathology is not known at present.