Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited. AUTISM IN CHILDREN WITH 22Q11.2 DELETION SYNDROME To the Editor: In their recent article, Vorstman et al. (2006) present a psychiatric assessment of 60 individuals ages 9 to 18 affected by 22q11DS, also known as velocardiofacial syndrome. The following results characterize their findings: (1) high rates of autism spectrum disorder (ASD; 50%), based on the Autism Diagnostic Interview-Revised, and (2) high rates of psychotic symptoms (26.7%). During the past decade, high rates of psychotic symptoms in affected individuals have been commonly reported in the literature (Baker and Skuse, 2005); however, the high rates of ASD are unprecedented. Although I appreciate the importance of detecting commu- nication problems in persons with the syndrome, I am concerned about the implications of the results of Vorstman et al. on the treatment of individuals affected by 22q11DS. Although both are heterogeneous disorders, research reveals similarities and differences between the typical developmental trajectories of language and communication in 22q11DS and autism. Similar to autism, language disorders represent a common challenge to young children with 22q11DS, and 90% of individuals experience communication disorder caused by velopharyngeal insufficiency. Accordingly, affected children often demonstrate delayed language, saying their first words around 2 to 3 years of age and their first sentences at 3 to 6 years. Nevertheless, later in development, and often subsequent to surgical and therapeutic interventions, children with 22q11DS recover some of their language delay, and verbal reasoning skills are typically stronger than nonverbal reasoning. In contrast, individuals with autism demonstrate weaker verbal than nonverbal intellectual profiles, and disordered language remains a hallmark feature throughout development. Of the more than 300 children affected by 22q11DS whom I have seen clinically, approximately 10 were nonverbal, usually as a consequence of hypoxia or a severe heart malformation. This rarity is in stark contrast to what is reported in individuals with autism. Moreover, social deficits in 22q11DS differ from autism. Affected individuals often start out overly familiar with others and then become withdrawn during adolescence. Ideas of concern, as well as feeling threatened by social contact, are associated with the onset of schizophrenia in 30% of affected individuals. Thus, fulfilling diagnostic criteria on the Autism Diagnostic Interview-Revised may simply be indicative of prepsychotic psychological difficulties in 22q11DS. Although there exist multiple case studies of children with infantile autism who later develop psychotic symptoms, comorbidity in large samples is typically less than 5% (Leyfer et al., 2006; Sverd, 2003). Comorbidity may, however, be more frequent in individuals with childhood-onset schizophrenia because of premorbid symptoms of schizophrenia affecting early com- munication and socialization skills. Indeed, a previous study found that 56% of children with childhood-onset schizo- phrenia met criteria for pervasive development disorder (PDD; Watkins et al., 1988). Like the authors, I would argue that the PDD criteria represent premorbid symptoms of psychosis rather than PDD itself and that affected individuals should be treated accordingly. These developmental differences have important implica- tions for treatment in 22q11DS. Programs for individuals with ASD often use strict behavioral techniques to target gaze avoidance, delayed verbal skills, and approaches to social situations. These programs are generally not geared toward the subtle communication problems typically associated with 22q11DS and require a large time and financial commitment from families with affected children. In my experience, with resource help, children with 22q11DS usually can be included in regular school programs and benefit from exposure to mainstream curricula with supplementary vocational training during adolescence. Although high rates of autism, as described by Vorstman and colleagues, are interesting to researchers studying 22q11DS, they should not be interpreted clinically. We are still characterizing the developmental psychopathology associated with 22q11DS to permit the definition of effective syndrome-specific treatments. Assigning an additional label of Bautism[ could greatly impede this progress. Stephan Eliez, M.D. Service Me ´dico-Pe ´dagogique University of Geneva School of Medicine Geneva, Switzerland Disclosure: The author has no financial relationships to disclose. Baker KD, Skuse DH (2005), Adolescents and young adults with 22q11 deletion syndrome: psychopathology in an at-risk group. Br J Psychiatry 186:115Y120 LETTERS TO THE EDITOR 433 J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:4, APRIL 2007