Within-syndrome variability

Behavioral phenotypes are probabilistic. Thus, among individuals with most genetic conditions, many but not all will show the syndrome’s “characteristic” behaviors. Indeed, rarely are etiology-related behaviors found in every person with a particular syndrome. Consider hyperphagia (or over eating), the behavior that most would identify as Prader–Willi syndrome (PWS)’s hallmark psychiatric symptom. Yet even this behavior does not occur in every child or adult with PWS. In a study of 4–19-year-old children, Dykens and Kasari (1997) showed that parents rated the item “overeats” for 80% of children with PWS. While issues of increasing chronological age and genetic subtypes all enter in, within-syndrome variability remains important to consider.

Total versus partial specificity

Total specificity means that a particular behavior is unique to a single syndrome; partial specificity means that two or more etiological groups show a higher-than-normal risk to have the same psychiatric behavior or behavioral profile (Hodapp, 1997). In the first, unique pattern, a genetic syndrome results in a particular outcome that is simply not seen in other genetic disorders. At present, the following behaviors seem unique to one and only one syndrome:

In contrast to this relatively short list, more instances occur in which partial specificity is operating. Within psychiatric symptoms (and compared to groups with ID in general), hyperactivity is more frequently found in children with 5p-syndrome (Dykens and Clarke, 1997) and in boys with Fragile X syndrome (Baumgardner et al., 1995). In a few genetic disorders, then, the same type of maladaptive behavior–psychopathology is found in higher percentages of individuals than is commonly noted among others with ID.

Finally, partially specific behavioral effects seem more in line with many areas of genetics, child psychiatry, and psychiatry. Across these different disciplines, researchers are now discussing the many pathways – both genetic and environmental – by which one comes to have one or another psychiatric disorder. As the clinical geneticist John Opitz (1985) noted, “The causes are many, but the common developmental pathways are few.”

Multiple domains, complex interactions

Behavioral phenotypes are not limited to maladaptive behavior or psychopathology; instead, they are also found across many different domains. Individuals with Down syndrome show particular deficits in grammar, articulation, and expressive (as opposed to receptive) language (Abbeduto et al., 2007). In terms of cognitive processing styles, patterns of “simultaneous over sequential processing” on the Kaufman Assessment Battery for Children (K-ABC) have been noted for most individuals with PWS (Dykens et al., 1992) and for boys with fragile X syndrome (Dykens et al., 1987; Kemper et al., 1988).

It is also the case that various etiology-related characteristics themselves inter-relate. To give some obvious examples, researchers in Down syndrome are increasingly appreciating the ways in which these children’s high levels of hearing problems (estimated at from 35% to 70%; Porter and Tharpe, 2010) tie to decreased levels of language, or even that the articulation levels of these individuals might be associated with the characteristic structure of the tongue (thickness, size, placement in mouth; Bunton and Leddy, 2011). Within the area of psychopathology, researchers have long discussed the connection between aggression and the inability to make one’s needs known verbally – there may also be more “etiology-linked” examples of such ties of physical, medical, and cognitive–linguistic characteristics and behavioral sequelae.

Development, context, and families

Just as behavioral phenotypes are seen in many areas and different behavioral and physical characteristics inter-relate, so too do other considerations enter in. One prominent issue concerns development, the ways in which etiology-related behaviors, profiles, or symptoms present differently in individuals who are of diverse ages. Fidler (2011) examined development in young children with Down syndrome, Fragile X syndrome, and Williams syndrome. In each case, developments in cognitive–linguistic profiles were described over the first few years of life, eventually producing in most children the etiology-related strengths and weaknesses found by the adult years. Age-related differences are also shown in maladaptive behavior–psychopathology. Comparing four age groups of children and adults with PWS, Dykens (2004) found a general mellowing as these individuals reached their 20s, 30s, and 40s, with externalizing problems decreasing across age groups. For other behaviors – including hoarding – the peak age-period was the 20s, with lower average scores before and after that period. Even when a particular behavior or symptom shows itself in most persons with a syndrome, the frequency and intensity of such symptoms often vary by age.

Genetic conditions also need to be embedded in a social and familial context, a context that often greatly affects the amount, type, and effectiveness of the services that individuals receive. Compared to other groups, for example, mothers of children with Down syndrome are more likely to be older (Grosse, 2010), and such mothers are more often highly educated, married, and have had prior children (Hodapp and Urbano, 2008). As such, mothers of children with Down syndrome may (as a group) be more skillful in knowing about their children’s development, as well as in locating, contacting, and effectively utilizing most every type of service (Hodapp et al., 2012).

By the time that the individual with Down syndrome has reached their mid 40s, however, having older parents (and experiencing earlier aging oneself) is detrimental and may lead to earlier and more complicated old-age care. Compared to same-aged adults with other forms of ID, 40–49-year-old adults with Down syndrome disproportionately showed that either they or their parents were experiencing health and/or functional declines, necessitating the increased caregiving roles (including legal guardianship) from one or more of the family’s non-disabled siblings. Something as simple as parental age, then, may greatly affect the care received by children or adults with Down syndrome.

Anxiety in Williams syndrome

Williams syndrome is a salient example of a genetic syndrome whose behavioral phenotype reflects increased propensity for the development of psychopathology. Individuals with Williams syndrome are at high risk for anxiety, both compared to typically developing populations and compared to ID groups of mixed etiology (Dankner and Dykens, 2012). This anxiety is largely manifest as fears and phobias, as well as generalized anxiety (Dykens, 2003; Cherniske et al., 2004; Leyfer et al., 2006).

Unlike in typically developing populations, in Williams syndrome the fears and phobias that appear in early childhood do not decline over age (Leyfer et al., 2006). Instead, phobias tend to cluster into a few discrete domains, as many of these individuals fear heights, uncertainty, loss, and loud noises, such as thunderstorms (Dykens, 2003). Generalized anxiety also appears to be a persistent aspect of the Williams syndrome phenotype. Cross-sectional work indicates that the risk for generalized anxiety problems in Williams syndrome begins to increase as children reach age 7–10 years (Leyfer et al., 2006). Longitudinal studies indicate that risk and severity remain stable across adolescence and young adulthood, although generalized anxiety may begin to decline in later adulthood (Einfeld et al., 2001; Woodruff-Borden et al., 2010).

Although problems with anxiety characterize Williams syndrome, they are not unique to the syndrome. Increased risk for anxiety is seen in other genetic syndromes and neurodevelopmental disorders, though often in different forms. Social anxiety is common in Fragile X syndrome (Cordeiro et al., 2011), while anxiety related to restricted interests and compulsions is often observed in PWS (Dykens et al., 1996). Anxiety and obsessions and restricted, repetitive, compulsive-like behaviors are also common features of autism spectrum disorders (ASD) (Moskowitz et al., 2013).

Moreover, anxiety is among the most common forms of psychopathology in the general population, and one of the best researched with regards to phenomenology and treatment. On one hand, then, this lack of specificity regarding anxiety in Williams syndrome makes it more difficult for professionals to use this phenotypic trait to glean information regarding the mechanisms or etiology of anxiety. On the other, the presence of anxiety across populations allows knowledge gained from typically developing individuals to improve outcomes in Williams syndrome and other high-risk populations.

Severe, adverse behavioral changes in Down syndrome

Although children with Down syndrome may less frequently show severe psychopathology, a small subset of children and young adults experience episodes of severe adverse behavioral changes. Such episodes, which Prasher (2002) described as Young Adults with a Disintegrative Disorder, have more recently been characterized as autistic regression (Worley et al., 2013), childhood depression (Stein et al., 2013), psychosis (Dykens et al., 2015), or young adult depression (Myers and Pueschel, 1991; Dykens, 2007).

Although we continue to know little about such adverse behavioral changes, several studies have begun to provide some basic information. Reviewing the medical records of 14 youths with Down syndrome who had experienced such adverse behavioral changes, Devenny and Matthews (2011) noted that about half experienced their first episode from age 4 to 12 years, half from age 18 to 21 years. In a larger, web-based parent survey examining 72 additional youths, Hodapp et al. (pers. comm.) found that half of all episodes began from when the child was age 4–9 years and most episodes lasted 3 years or less (about 20% lasted 4+ years).

Recent studies have also begun to describe the nature of and reactions to such events. For almost 90% of these youths and young adults, their episodes featured both substantial losses of adaptive–functional skills (e.g., regressions in their ability to communicate and in academic skills), as well as new or intensified problem behaviors. The most common problem behaviors – which either arose for the first time or increased greatly in either frequency or intensity – included anxiety; bizarre, inappropriate behaviors; attention seeking; obsessive-compulsive behaviors; aggressiveness; and depression. In response to the onset of such events, parents consulted a variety of professionals, usually beginning 2–6 months after the episode’s onset. These youths then mostly received a variety of behavioral and pharmacological treatments, usually for more than 18 months. Parents judged half of all treatments to be somewhat or very effective (Hodapp et al., pers. comm.).

Obviously, such studies barely scratch the surface of this rare but severe phenomenon. Why such changes occur remains a mystery, although risk factors may include major transitions or changes in the environment (Devenny and Matthews, 2011; Hodapp et al., pers. comm.), as well as co-occurring medical or neurological vulnerabilities (Hodapp et al., pers. comm.). Further studies are needed on this subgroup of atypical children with Down syndrome.

Hyperphagia, psychosis, and ASD in PWS

Prader–Willi syndrome is caused by a lack of paternally derived imprinted material in the 15q11–q13 region, either through a deletion in the genetic material contributed by the father (i.e., paternal deletion) or by having two chromosome 15s from the mother (i.e., maternal uniparental disomy, or mUPD). The syndrome is well known for its complex phenotype, with diverse medical, cognitive, emotional, and behavioral problems that do not cleanly fit into traditional psychiatric nosologies. Psychiatric problems in PWS are further complicated by developmental changes over time, the genetic subtypes associated with PWS, and (similar to the general population) a family history of psychiatric disorders and environmental stressors and supports. We here focus on three aspects of the PWS psychiatric phenotype: hyperphagia, severe psychiatric illness, and ASD.

Hyperphagia. As a salient feature of PWS, hyperphagia leads to high risks for morbid obesity. As a result, individuals with PWS need physical activities and strict dietary and food supervision in order to avoid life-threatening complications of obesity and early mortality (Dykens and Roof, 2008). In addition, as hyperphagia onsets in early childhood, many emotional and behavioral problems worsen (Dimitropoulos et al., 2001), including tantrums, aggression, rigidity, irritability, sadness, skin picking, low tolerance for change or different perspectives, insistence on sameness, arranging and rearranging, and hoarding. These problems may wax and wane over time, and lessen in some individuals in middle to late adulthood (Dykens and Roof, 2008).

Severe psychiatric illness. People with PWS have been variably described as being at high risk for obsessive-compulsive disorder; compulsivity that reflects arrested development; affective illness; bipolar illness with or without psychosis; mood disorder with or without a psychotic component; psychotic illness; and bipolar affective disorder (e.g., Dykens et al., 1996; Boer et al., 2002; Clarke et al., 2002; Descheemaeker et al., 2002; Dykens and Shah, 2003; Soni et al., 2007). Although those with mUPD are more prone to mood disorders or psychotic illness, especially as they transition into young adulthood, severe problems can also found in those with deletions, including those with positive family histories of psychiatric illness (Soni et al., 2007).

Unfortunately, psychiatric studies in PWS have relied solely on diagnostic labels, leaving specific psychiatric symptoms unclear or poorly described (Dykens and Shah, 2003). It remains unknown, for example, whether psychosis in PWS is best characterized by such positive symptoms as auditory or visual hallucinations, or delusional thinking, or by such negative symptoms as affective blunting, anhedonia, or withdrawal. Psychosis in PWS may shed light on genetic risks for psychosis in the general population, with recent work implicating the duplication and increased expression of maternally imprinted genes, especially UBE3A, in mUPD, as well as a differentially methylated region within the GABRA receptor genes located in the PWS 15q11–q13 region. Although two studies (Webb et al., 2008; Sharp et al., 2010) have implicated this locus or variation as a promising candidate for psychosis in mUPD, more targeted treatments and future genotype–phenotype research will require a precise characterization of specific psychotic symptoms.

Autism spectrum disorders. Studies find that approximately 40% of children with the mUPD subtype of PWS have co-occurring ASD, compared to 18% of deletion cases (e.g., Veltman et al., 2005; Descheemaeker et al., 2006). Even so, social communicative deficits are core features of ASD, yet remarkably little is known about social communicative functioning in PWS during infancy or early childhood (Dykens et al., 2011). Although the15q11–q13 region has been deemed an epigenetic “hotspot” for associations to ASD (Schanen, 2006; Hogart et al., 2010), the lack of detailed data on social or communicative functioning in PWS limits its usefulness as a genetic model for ASD. In addition, despite similarities across PWS and ASD in, for example, insistence on sameness or restricted interests, individuals with PWS also display hoarding and skin picking, whereas in ASD there may be increased instances of stereotypies and more diverse and severe self-injurious behaviors (Dykens et al., 2011). Thus, further work is needed on these different profiles of repetitive behaviors across PWS, ASD, and related 15q disorders (e.g., Angelman syndrome; maternal interstitial duplications of chromosome 15).

Finally, more studies are needed to explore the impact of hyperphagia on the types or severity of psychiatric symptoms in PWS. With impaired satiety, people with PWS are, as the motto of the PWS Association-US (PWSA-US) states, “always hungry, never full.” More formally, Holland et al. (2003) have described these individuals as being in states of “starvation that manifests as obesity in a food-rich environment.” Previously, Dykens et al. (2007) found that, compared to individuals with PWS who were overweight or obese, those who were lean and with a lower body mass index (BMI) exhibited greater levels of distress, disorganized thinking, sadness, and irritability. This finding, though counterintuitive, can perhaps be attributed to the psychological effort and stress of maintaining a low weight, to hormonal changes that come with weight loss, or to the distressed feelings we all sometimes experience when we are very, very hungry and in need of food.