Autism, Regression, and the Broader Autism Phenotype

J. E. Lainhart, S. Ozonoff, H. Coon, L. Krasny, E. Dinh, J. Nice and W. McMahon, American Journal of Medical Genetics, 113: 231 - 237 (2002).

The broader autism phenotype (BAP) is a subclinical set of personality and other features that is thought to index familiarity and/or genetic liability to autism. Eighteen parents of autistic probands with a history of language regression and 70 parents of autistic probands without regression were assessed for features of the BAP and compared with published rates in parents of nonautistic subjects. Parents of probands with regressive and nonregressive autism demonstrated similar rates of the BAP (27.8% vs. 32.9%; P = 0.33). The rate of the BAP was significantly higher in both groups of autism parents than in parents of nonautistic subjects (P ³ 0.01). Thus, this measure of genetic liability is increased equally in families with both forms of autism when compared with controls. Environmental events are therefore unlikely to be the sole cause of regressive autism in our sample. Environmental events, however, may act in an additive or "second-hit" fashion in individuals with a genetic vulnerability to autism. About 20% of children with autism appear to have relatively normal development during the first 12-24 months of life. This period of relative normalcy gradually or suddenly ends and is followed by a period of regression, characterized most prominently by a significant loss of language skills. Then the full autism "syndrome" soon becomes evident. If regressive autism is solely caused by environmental events, such as adverse reactions to vaccines, rates of the BAP in the relatives of children with regressive autism should be no greater than in the general population. If environmental events do not independently cause regressive autism, or if they act as "second-hit" phenomena in children who already have the genetic liability to autism, rate of the BAP should be similar in relatives of autistic children with and without regression. Signs of the BAP in parents were measured using instruments specifically developed for this purpose: the Modified Personality Assessment Schedule-Revised (MPAS-R) the Pragmatic Rating Scale (PRS) and the Friendship Interview. There were two interesting differences between the regressive and nonregressive autistic probands. The first difference was in head size. In the nonregression group, there were 10 probands with macrocephaly and 1 proband with microcephaly. All probands in the regression group were normocephalic. The second interesting difference between the regressive and nonregressive autism probands was in dysmorphology. All of the regression subjects were classified as nondysmorphic. Because the rates of the macrocephaly and dysmorphology were both increased in the nonregression group, we explored the relationship between them. The rate of autistic subjects who were equivocally in the subjects with macrocephaly, compared with the subjects who were normocephalic. Further, liability to autism, as measured by the BAP, is increased to the same degree in the parents of children with regressive and nonregressive autism. Environmental events are therefore unlikely to be a sole cause of regressive autism in our sample. Our data cannot rule out that environmental events may act in an additive or "second-hit" fashion in individuals with a genetic vulnerability to autism. We have observed that the BAP is often associated with functional impairment, particularly affected social relationships with other at home and at work. Though our clinical observation needs to be further systematically tested, it seems that when features of the BAP, as measured in this study, occur in combination, they may make it difficult for at least some individuals to achieve a professional level commensurate with their IQ.