Autistic traits in simplex and multiplex autism families Focus on unaffected relatives.код для вставкиСкачать
LETTER TO THE EDITOR Autistic Traits in Simplex and Multiplex Autism Families: Focus on Unaffected Relatives Rosa A. Hoekstra* and Sally Wheelwright Department of Psychiatry, Autism Research Centre, University of Cambridge, Cambridge, UK Received 4 December 2008; Accepted 10 February 2009 TO THE EDITOR: Virkud et al.  recently published findings from a very interesting dataset on the aggregation of autistic traits in families in which more than one individual is affected with autism (multiplex autism, MA) compared to families with only one affected individual (simplex autism, SA). Recent studies suggest that individuals from MA families may have a stronger genetic predisposition to autism compared to SA families, due to common genetic variants running in these families. Conversely, autism in SA families may be more often induced by relatively rare gene mutations of large effect or by copy number variations that are unshared with the other family members [Sebat et al., 2007; Weiss et al., 2008]. Advancing our knowledge about how genetic transmission may be different in these two types of families is of utmost importance to our understanding of the etiology of autism spectrum conditions. We therefore read the article of Virkud et al.  with great interest and would like to share our thoughts on their results and offer some suggestions for future analyses in similar datasets. If the hypothesis that genetic predisposition for autism is different in MA and SA families holds true, not only would individuals from MA families be at increased risk for autism compared to SA family members, but they may also show increased expression of sub-threshold autistic traits measured on a dimensional scale. To test this hypothesis, it is of interest to see if unaffected members of MA families show increased expression of autistic traits compared to unaffected relatives of SA families. Some evidence for this is presented in Virkud et al.’s article. Mean quantitative autistic traits (QAT) scores as assessed using the Social Responsiveness Scale (SRS) seem to be somewhat higher (on a trend level, P ¼ 0.075) in unaffected male siblings from MA families compared to male siblings from SA families. This result is promising. However, it remains unclear whether this trend holds when the significantly higher QAT-scores in the affected brothers of MA families compared to SA brothers (Mean SRS score 104.7 vs. 87.8, P ¼ 0.0001) are taken into account. It is not unconceivable that the autism observed in the MA families is simply of a more severe kind (hence leading to increased QAT expression in siblings) than in SA families. A second test of the hypothesis would be to compare the resemblance between siblings from MA and SA families. Virkud et al. reported the intraclass correlation (ICC) between all male Ó 2009 Wiley-Liss, Inc. How to Cite this Article: Hoekstra RA, Wheelwright S. 2010. Autistic Traits in Simplex and Multiplex Autism Families: Focus on Unaffected Relatives. Am J Med Genet Part B 153B:356–358. siblings from SA families (of which per definition one is affected and the other is unaffected) and all male siblings from MA families (in which a substantial amount of the sibling pairs are both affected and only a small proportion consists of affected–unaffected sibling pairs). The sibling correlations were found to be non-significant (ICC ¼ 0.0) in SA families, and significant (ICC ¼ 0.46) in MA families. Although this result is promising, the more interesting question would be whether unaffected siblings from MA families resemble their affected sibling more closely than affected– unaffected sibling pairs do from SA families. To answer this, the ICC should be computed for affected–unaffected sibling pairs from MA families only. Unfortunately, this association was not tested by Virkud et al. . Thirdly, Virkud et al. present the distribution of QAT in respectively all boys from SA families, all boys from MA families, and all girls from MA families. The distribution of the scores in boys from MA families appears to follow a markedly different distribution than the scores from both other groups. Whilst the graphs depicting scores from SA siblings and from MA female siblings follow bimodal distributions, the distribution of scores from boys from MA families appears continuous. However it is important to realize that there are marked differences in the number of affected and Grant sponsor: Netherlands Organisation for Scientific Research (NWO Rubicon). Comment on ‘‘Familial aggregation of quantitative autistic traits in multiplex versus simplex autism’’ by Virkud et al., 2009 in the American Journal of Medical Genetics Part B: Neuropsychiatric Genetics 150B: 328–334. *Correspondence to: Dr. Rosa A. Hoekstra, Department of Psychiatry, Autism Research Centre, University of Cambridge, Douglas House, 18b Trumpington Road, Cambridge CB2 8AH, UK. E-mail: firstname.lastname@example.org Published online 14 April 2009 in Wiley InterScience (www.interscience.wiley.com) DOI 10.1002/ajmg.b.30947 356 ROSA A. HOEKSTRA AND SALLY WHEELWRIGHT unaffected siblings in each of the graphs. The graph for SA families (80 families in total) includes 80 affected males and 89 unaffected males. The affected:unaffected ratio is thus approximately 1 (which follows directly from the criterion for SA families set by Virkud et al. that each affected male should have at least 1 unaffected brother). This is in strong contrast to the graphs of MA families, which include many more affected than unaffected boys (ratio ¼ 262:29, or 9:1) and more affected than unaffected girls (ratio ¼ 81:45 or 1.8:1). Therefore, the distribution of QAT scores in MA male siblings may mainly look unimodal because there are so few unaffected siblings in these families. Moreover, the distribution in MA girls could appear to be bimodal because the unaffected MA girls have autistic brothers (who are left out of the graph) rather than sisters. To illustrate this point, we have amended the original Figure 1 from Virkud et al.  and displayed the distribution of QAT scores as the proportion of the number of available siblings (Fig. 1) and 357 FIG. 2. Distribution of raw social responsiveness scale scores for all boys and girls from multiplex autism families. Each line represents the total number of affected or unaffected siblings obtaining the score. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] we have grouped all siblings (both boys and girls) from MA families together (Fig. 2). What these figures illustrate is that, depending on the way you portray the data, the distribution looks bimodal (for graphs representing proportions) for both male and female siblings or unimodal (for graphs representing total number of individuals). We think more evidence is needed to establish whether the aggregation of autistic traits is attenuated in female siblings of autistic individuals. Of course it remains intriguing that the sibling recurrence rate for autism is found to be much higher in boys than in girls. To further study possible differences in vulnerability for autism between the sexes, it would be interesting to compare the resemblance in affected–unaffected sibling pairs from MA families for male–male sibling pairs with the resemblance in female–female sibling pairs and in opposite-sex pairs. In sum, studies exploring the differential aggregation of QAT in relatives from MA and SA families are of great value to further our understanding of the etiology of autism spectrum conditions. We suggest that in future comparisons between these two types of autism families the attention should focus on the aggregation of autistic traits in unaffected relatives from MA families. ACKNOWLEDGMENTS Dr. Hoekstra is financially supported by the Netherlands Organisation for Scientific Research (NWO Rubicon). FIG. 1. A–C: Distribution of raw social responsiveness scale scores for all boys from simplex autism families (A), for all boys from multiples autism families (B), and for all girls from multiplex autism families (C). Each line represents the proportion of affected or unaffected siblings obtaining the score. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] REFERENCES Sebat J, Lakshmi B, Malhotra D, Troge J, Lese-Martin C, Walsh T, Yamrom B, Yoon S, Krasnitz A, Kendall J, Leotta A, Pai D, Zhang R, Lee Y, Hicks J, Spence SJ, Lee AT, Puura K, Lehtim€aki T, Ledbetter D, Gregersen PK, Bregman J, Sutcliffe JS, Jobanputra V, Chung W, Warburton D, King M, Skuse D, Geschwind DH, Gilliam TC, Ye K, Wigler M. 2007. Strong association of de novo copy number mutations with autism. Science 316:445–449. 358 Virkud YV, Todd RD, Abbacchi AM, Zhang Y, Constantino JN. 2009. Familial aggregation of quantitative autistic traits in multiplex versus simplex autism. Am J Med Genet Part B 150B:328–334. Weiss LA, Shen Y, Korn JM, Arking DE, Miller DT, Fossdal R, Saemundsen E, Stefansson H, Ferreira M, Green T, Platt OS, Ruderfer DM, Walsh CA, AMERICAN JOURNAL OF MEDICAL GENETICS PART B Altshuler D, Chakravarti A, Tanzi RE, Stefansson K, Santangelo S, Gusella JF, Sklar P, Wu B, Daly MJ. 2008. The Autism Consortium. Association between microdeletion and microduplication at 16p11.2 and autism. N Engl J Med 358:667–675.