Elevated rates of testosterone-related disorders in women with autism spectrum conditions
Introduction
Autism is a spectrum of neurodevelopmental conditions characterized by difficulties in social development, abnormalities in communication, and the presence of repetitive behaviors/obsessive interests (APA, 1994). Asperger Syndrome (AS) shares these features, but children with AS do not show the delay in language acquisition or general intellectual impairment of classic autism. It has been argued that autism and AS are essentially the same condition but with varied degrees of language development or IQ (Wing, 1988). Together they constitute two major subgroups of autism spectrum conditions (ASC).
Four males are diagnosed with autism for every female, and AS males are nine times as common as AS females (Wing, 1981). This sex difference suggests that there may be a male vulnerability to developing ASC, a hypothesis supported by multiple lines of evidence. For example, individuals with ASC tend to display a hypermasculine profile on many cognitive tasks (Baron-Cohen, 2002). On the Embedded Figures Test and on measures of ‘intuitive physics’, they perform better than typical males, who perform better than typical females (Jolliffe and Baron-Cohen, 1997, Lawson et al., 2004). On tests involving empathy or ‘intuitive psychology’, they perform worse than typical males, who perform worse than typical females (Baron-Cohen et al., 1999, Happe, 1995, Baron-Cohen et al., 2001a). Typical males also have more autistic traits on average than typical females, as measured by the Autism-Spectrum Quotient (AQ) (Baron-Cohen et al., 2001b, Wheelwright et al., 2006). These findings have led to the idea that the autistic brain may be an ‘extreme’ of the typical male brain (Baron-Cohen, 2002), a theory that is also currently being explored at the level of neural connectivity (Baron-Cohen et al., 2005).
At the biological level, higher levels of fetal testosterone (FT), measured in amniotic fluid, are inversely correlated with amount of eye contact at 12 months of age (Lutchmaya et al., 2002a), vocabulary size at 18 and 24 months (Lutchmaya et al., 2002b), and quality of social relationships at 4 years (Knickmeyer et al., 2005). FT levels are correlated with number of autistic traits as measured on the Childhood Asperger Screening Test (CAST) and the Child Autism Spectrum Quotient (AQ-C) (Auyeung et al., submitted for publication), higher scores on the Child Systemizing Quotient (SQ) (Auyeung et al., 2006), and lower scores on the Child Empathizing Quotient (EQ) (Chapman et al., 2006). Individuals with ASC also have lower-than-expected 2nd to 4th digit (2D:4D) ratios (Manning et al., 2001), which is correlated with higher ratios of FT to fetal estrogen (Lutchmaya et al., 2004), as well as lower verbal and higher numerical intelligence (Luxen and Buunk, 2005).
Some neuroanatomical studies comparing the brains of individuals with and without ASC reveal structural differences associated with high levels of FT, including hemispheric asymmetries (Herbert et al., 2005). Finally, girls with abnormally high FT levels as a result of congenital adrenal hyperplasia (CAH) have a higher number of autistic traits than their unaffected sisters (Knickmeyer et al., 2006a). These findings have led to the androgen theory of ASC, which proposes that elevated FT contributes to differences in brain development that underlie the cognitive traits found in autism (Baron-Cohen et al., 2004, Geschwind and Galaburda, 1985).
Studies of women with ASC, though less common, are consistent with the androgen theory. Women with ASC score similarly to men with ASC on the EQ (Baron-Cohen and Wheelwright, 2004, Wheelwright et al., 2006), the SQ (Baron-Cohen et al., 2003, Wheelwright et al., 2006), and the AQ (Baron-Cohen et al., 2001b, Wheelwright et al., 2006). On average, girls with ASC show an 8-month delay in the onset of menarche (Knickmeyer et al., 2006b) and are more likely to display male-typical play preferences (Knickmeyer et al., submitted for publication). One study found elevated testosterone levels in a subgroup of children with ASC and aggressive behavior (Tordjman et al., 1997). The single pubertal female with ASC in this study had testosterone levels 24% higher and adrenal androgen levels 500% higher than control means.
There is also evidence for cognitive hyper-masculinization in the parents of individuals with ASC, who often display traits that reflect the ‘broader autism phenotype’. They score higher on the AQ than people without autism, though not high enough to be in the ASC range, and mothers score similarly to control males on items pertaining to social skills (Bishop et al., 2004). Additionally, having both a mother and a father who score in the upper quartile on the Social Responsivity Scale (SRS), another measure of autism spectrum traits, results in an elevenfold increase in the prevalence of meeting clinical criteria for an ASC (Constantino and Todd, 2005). Mothers and fathers of individuals with ASC perform as well as typical males on the Embedded Figures Test and poorer than control females on a test of reading facial expressions of emotions from the eyes (Baron-Cohen and Hammer, 1997), and findings from a preliminary neuroimaging study suggest that they display a masculine pattern of brain activity while performing these tasks (Baron-Cohen et al., 2006). There is also some evidence for elevated FT levels in parents of children with autism: like their children, they have lower 2D:4D ratios than expected (Manning et al., 2001).
Both the androgen theory and the extreme male brain theory predict that women with ASC might manifest physical masculinization, and be more vulnerable to conditions associated with elevated levels of androgens. In addition, because of the evidence of psychological hyper-masculinization in mothers of children with autism, they too might share some of these vulnerabilities. The survey reported here was developed to investigate whether this was the case.
Section snippets
The Testosterone-related Medical Questionnaire (TMQ)
The TMQ (see Appendix A) was developed by our research group based on a literature search of endocrine conditions and relevant traits or symptoms, particularly those that have documented associations with androgens. Participants responded either via e-mail or by filling out the survey on a secure Web site. Below we justify the inclusion of each of the 35 items (italicized here) in the TMQ.
Several conditions were included in the TMQ because of their direct effect on sex hormone levels.
Results
The reported rates of medical conditions investigated in this sample of controls were consistent with prevalence rates reported in the literature (Table 1). The exceptions to this were the rates for epilepsy and congenital adrenal hyperplasia (CAH), which were reported to be higher in this control sample than expected.
Compared to controls, women with autism reported higher rates of hirsutism (χ2 = 29.0, P < 0.0001), medical diagnosis of polycystic ovary syndrome (PCOS) (FET, P < 0.018), diagnosis of
Discussion
The results of the study are consistent with the androgen theory of autism and suggest that heightened testosterone levels and/or activity may exist in individuals with ASC into adulthood, manifesting in other conditions. Although some of these clinical conditions appear at increased rates in this sample of individuals with autism, there is no suggestion that one of these conditions (e.g., PCOS) causes autism (or vice versa). Rather, we assume that these conditions and autism share a common
Acknowledgments
The authors were supported by the Nancy Lurie Marks Family Foundation and the Medical Research Council during the development of this work. We are grateful to Rick Lathe, Joe Herbert, Lindsey Kent, Kevin Taylor, Ieuan Hughes, and Grant Hill-Cawthorne for discussions.
References (96)
- et al.
Hirsutism and acne in polycystic ovary syndrome. Best practice and research
Clin. Obst. Gynaecol.
(2004) Pathogenesis of polycystic ovary syndrome—the enigma unravels?
The Lancet
(1999)The extreme male brain theory of autism
Trends Cogn. Sci.
(2002)- et al.
fMRI of parents of children with Asperger Syndrome: a pilot study
Brain Cogn.
(2006) - et al.
A higher rate of hyperandrogenic disorders in female-to-male transsexuals
Psychoneuroendocrinology
(1997) - et al.
Intergenerational transmission of subthreshold autistic traits in the general population
Biol. Psychiatry
(2005) - et al.
Sexual behavior and sexual orientation of the female rat after hormonal treatment during various stages of development
Horm. Behav.
(1988) Premenstrual syndrome and premenstrual dysphoric disorder: definitions and diagnosis
Psychoneuroendocrinology
(2003)Psychoneuroendocrine aspects of temporolimbic epilepsy. Part 1. Brain, reproductive steroids, and emotions
Psychosomatics
(1999)Androgens and puberty
Best Pract. Res. Clin. Endocrinol. Metab.
(2002)