Profiling endophenotypes in social anxiety disorder: A neurocognitive approach

Social anxiety (SA), an intense fear of social interactions and social
evaluation, is a relatively common problem. The life-time prevalence of social
anxiety disorder (SAD), the severe form of SA including avoidance behavior and
disturbance of general functioning, as described in the Diagnostic and
Statistical Manual (DSM IV, 1994), is between 7 and 13% in Western societies
(Furmark, 2002, Rapee & Spence, 2004). The year prevalence in the Netherlands
is 4.8% (Bijl et al., 1997). SA with a sub threshold level below the DSM-IV
diagnostic threshold of SAD, with subclinical levels of social anxiety
symptoms, shows a life-time prevalence rate up to 25% in the general population
(e.g., Davidson et al. 1994, Wittchen et al. 1999; Ruscio et al. 2008).
SAD patients show a high rate of impairment of social functioning, working- and
family-life and close relationships (Lampe et al. 2003). In addition, patients
with SAD are less likely to be in a relationship or marriage (Dingemans et al
2001). SAD is also associated with early leave of school (Stein & Kean 2000),
lower level of education (Dingemans et al., 2001; Wittchen et al., 1999), a
higher risk of being unemployed (Dingemans et al., 2001; Lampe et al 2003) and
engagement in jobs below the level of qualification (Dingemans et al 2001;
Katzelnick & Greist 2001). In addition, the economic costs of SAD are
relatively high: about ¤12,000.- per person per year, and about ¤5,000,- for
persons with sub threshold social phobia, compared to about ¤3,000,- for
persons with other psychological problems (Acaturk, Smit, De Graaf, Van
Straten, Ten Have, & Cuijpers, 2009).
Longitudinal studies indicate that SAD grows out of subclinical SA. The growing
fear of social interactions and social evaluation often goes unnoticed or is
noticed only after a long period of time, because socially anxious individuals
shy away from social situations or stay in the background. In contrast with
externalizing behavior, socially anxious persons are not an obvious burden on
society, another reason why this problem often goes unnoticed. Generally, the
age of onset of SAD is during late childhood or adolescence (Rapee and Spence,
2004). SAD is a life-long disease with a low likelihood of spontaneous
remission, intervention studies show that SAD is resistant to treatment in
comparison with other anxiety disorders. SAD shows a high rate of comorbidity
with other anxiety disorders, depression and substance use disorders, in
general SAD preceding the other disorders (Merikangas et al. 2002).
In order to reduce the development of SAD, to minimize individual suffering and
to reduce the substantial economic costs (as described above), it is imperative
to develop preventive interventions (Acaturk et al., 2009). Preventive
interventions, focused on subjects with increased SA or subclinical SAD, is
called indicated prevention. Until now, such interventions have not been
developed and studied in persons with SA, but only in adult populations with
subclinical panic attacks (Meulenbeek et al. 2010) and in children with
increased anxiety (Kendall, 1990). Insight into the risks and protective
factors that play a role in the development and maintenance of SAD form the
basis for the development of such preventive interventions.

INTERGENERATIONAL TRANSMISSION OF SAD
Family studies have shown that SAD runs in families. Stein et al. (1998) showed
that first-degree relatives of individuals with generalized SAD were 10 times
more likely to have SAD than relatives of individuals without SAD. This
increased risk of SAD was particularly found for the generalized type of SAD
and not for specific SAD (Mannuzza et al., 1995). Furthermore, Lieb et al.
(2000) demonstrated that 9.6% of the children of parents with SAD also met the
criteria of SAD, whereas 2.1% of the children of parents with no psychiatric
disorder had SAD. In contrast, there appears to be no significant association
between spouses, indicating a random mating for social anxious persons (Distel
et al 2008).
Behavior-genetic research (e.g., twin studies) supports both a hereditary and
an environmental contribution to the development of social phobia. Twin studies
yield heritability estimates of 20-50% (Distel et al 2008; Kendler et al 1992;
Middeldorp et al 2005; Nelson et al., 2000). Investigations of genetics (e.g.,
Genome Wide Association studies) contributed to the determination of a
SAD-specific genetic background (Gelernter, Page, Stein, & Woods (2004),
however, it remains unclear which genes are specifically involved in SAD.
Likewise, most environmental correlates of SAD seem to be risk factors for
internalizing disorders in general (see Rapee & Spence, 2004). Few specific
environmental effects have been identified and these effects are rather modest.
Recently, several authors have argued that investigations of gene-environment
interactions are crucial for a better understanding of the involvement of both
genetic and environmental factors in the development of psychopathology
(Beauchaine, Neuhaus, Brenner, & Gatzke-Koop, 2008). A special issue of
European Archives of Psychiatry and Clinical Neuroscience (2008) emphasized the
importance of studying gene-environment interactions in anxiety disorders such
as SAD. Our ultimate objective is to conduct investigations of gene-environment
interplay in the service of developing preventative interventions. However, to
investigate gene-environment interactions it is imperative to understand SAD at
the level of its endophenotypes. Allegedly, endophenotypes have a simpler
genetic architecture than that of the disorder, and this should enhance the
power to discover genes that are specific to the development of SAD (Walters
and Owen, 2007). A successful delineation of endophenotypes will facilitate the
investigation of gene-environment interactions, which will be the subsequent
phase of this research program. Profiling endophenotypes for SAD is the purpose
of the present study

PROFILING ENDOPHENOTYPES OF SAD
The SAD phenotype may be linked to the genotype through endophenotypes, which
refer to genetic trait markers of the disorder. Complementary to traditional
diagnostics, profiling endophenotypes of SAD will advance our understanding of
the genetic architecture of this disorder, and reveal its neurobiological and
neurocognitive abnormalities (Ritsner, 2009). The characterization of SAD
specific endophenotypes is of both clinical and scientific relevance.
Endophenotypes (1) help to improve diagnostic criteria, which are difficult to
interpret when signs and symptoms overlap with other disorders; (2) may
supplement psychiatric diagnosis solely based on signs and symptoms; (3) are by
definition present before the onset of an illness, and can be used to aid in
identifying persons who are at elevated risk for becoming ill and for the
delivery of preventive interventions; (4) may help to identify a biological
subtype of SAD, individualizing treatment and predicting therapeutic response;
and (5) may aid in the search for more precise genetic and environmental
determinants of the disorder (Ritsner, 2009).
For a marker to be considered an endophenotype, *it has to (1) be associated
with the phenotype (formal diagnosis), (2) be independent of clinical state,
(3) be highly heritable, and (4) the impairment must co-segregate with the
illness within a family, with non-affected family members showing impairment
relative to the general population* (Glahn et al., 2007) This calls for a
multigenerational family study that examines the affected individual with SAD,
his or her siblings and children, as well as the partners of each family
member. It is hypothesized that the SAD relevant trait characteristics of the
patient will also be observable in family members, however, in a milder form.
It is anticipated that these observed SAD traits will not be evident in
non-relatives (partners). The proposed family study is the first comprehensive
study on SAD endophenotypes.
The key question addressed in this study is whether the psychophysiological and
neurocognitive abnormalities often reported in SAD patients are heritable, and
can thus be found in family members of the SAD patients. Determination of
genetic kinship of these neurocognitive deficiencies is essential for
endophenotyping (Ritsner, 2009), and will result in a better understanding of
the constellation of trait markers that, together, will determine whether an
individual will develop social anxiety.
This study includes paradigms that tap into the two key dimensions relevant to
the SAD symptomatology: Attention and Social Cognition (Clark & McManus, 2002;
Hirsch & Clark, 2004; Miskovic & Schmidt, 2012a; Rapee & Spence, 2004; Stein &
Stein, 2008). The relevance of attention and social cognition to SAD will be
outlined below, but shortly, we expect that SAD individuals can be
characterized by both vigilance (a biased attentional orienting response) and
aberrant inhibitive processes, that is, impaired inhibition of attention to
social threatening information and impaired inhibition of negative social
cognitions, or rumination.

The major objective of this pilot study is to delineate SAD endophenotypes by
investigating information processing characteristics, brain activity and
connectivity, within the two key dimensions relevant to the SAD symptomatology:
attention and social cognitions. To this end, we employ neurocognitive
paradigms suitable for measuring event-related brain potentials (EEG) and
functional brain activity (functional MRI). The paradigms are discussed in more
detail in sections 3 and 6, but see Table 1 for an overview. The included
paradigms are representative of a weighted balance between the investigation of
both temporal and spatial characteristics of neural abnormalities in the
processing of social information (i.e., EEG and fMRI, respectively), attention
and social norm processing, as well as network connectivity methods.
The theoretical foundation of the proposed study is derived from the existing
literature that has demonstrated its potential to detect robust abnormalities
in social information processing corresponding to cognition and behavior
(Bokhorst, Westenberg, Oosterlaan, & Heyne, 2008; Miers, Blote, Bogels, &
Westenberg, 2008; Miers, Blote, & Westenberg, 2010, 2011; Westenberg, Drewes,
Goedhart, Siebelink, & Treffers, 2004; Westenberg, Gullone, Bokhorst, Heyne, &
King, 2007), as well as to the brain and network level (Blackford, Avery,
Shelton, & Zald, 2009; Blair, et al., 2010; Freitas-Ferrari, et al., 2010;
Killgore & Yurgelun-Todd, 2005; Klumpp, Angstadt, & Phan, 2012; Miskovic &
Schmidt, 2012b).

This study is a cross-sectional family study with two generations. Family
composition is depicted in Figure 2. Twelve families will be selected provided
that one adult (25-55 years of age) has the psychiatric diagnosis of SAD
(labeled as *Target* participant hereafter) and that one of the Target*s
children has social anxiety symptoms (or SAD). The target*s other children, the
target*s spouse, and the target*s siblings and their spouses and children will
be included as well. Families are included based on the age of the target*s
children: they should be between 8 and 18 years of age. Adolescence appears to
be a critical period for the development of clinical levels of SAD. Research
shows that SAD is rarely diagnosed below the age of 10 years and that the
prevalence increases between 10 and 20 years of age. In addition, retrospective
research comprising adult SAD patients indicates that the age at onset is in
the mid-teens (DSM-IV, 1994). Hence it is expected that the
genotype-endophenotype-phenotype connection is particularly visible in families
with children at this vulnerable age.

Due to the complexity of this study we will conduct a pretest and a pilot:
1. Pretest: Before we will recruit the families to participate in this study,
we will first recruit a subset of control participants (N = 10, e.g., first
year students) for preliminary tests of the procedures and effects (= pretest
phase). For these controls we provide modified informant consent forms and
information brochures (please see the attached documents).
2. Pilot: As this is the first time we are recruiting families for research, we
will conduct a pilot study in which the protocol discussed below will be
administered to 3 families. This pilot study will provide important information
about the feasibility of family inclusion, logistics, as well as power to
detect endophenotypes with the included paradigms. After testing these 3
families we will evaluate the procedures and report the findings to the METC.
Then we will test the remaining 9 families.

Schematic of the test phases
- Protocol will be administered to students (pretest)
- Protocol will be administered to 3 families (pilot)
- Protocol will be administered to the remaining 9 families

After completion of the pilot we will report on the feasibility of the protocol
to the Medical Ethics Committee (METC) of the LUMC. Any significant changes
made to the protocol will be notified to the METC with amendments. The
continuation of the study is dependent on a positive evaluation of the
amendment(s) by the METC.

*********************************
AMENDMENT March 17 2014
*********************************
(please see the research protocol and amendment letter for details)

- Changes to the experimental paradigms. These changes allow for improved
sensitivity to detect endophenotypes of social anxiety, without adding
significant measurement time or a burden on the participants.

- Debriefing protocol: participants will now be directly debriefed after the
EEG test session about the purpose of the social evaluation paradigm.

- Questionnaires: a view questionnaires have been excluded or replaced with
more appropriate questionnaires

- Duration of study per participant: due to the proposed amendments to the
protocol, the total duration of the experiment per participant has been reduced
from 7 hours to 6:30 hours.

- Changes to the "inclusion and screening protocol": Step 1: telephone
conversation with potential target and 'screen' for signs of social anxiety. If
signs are present, invite person with partner for a interview. Step 2: explain
research protocol to the potential target and partner, administer psychiatric
standardized interview to the potential target and partner. If potential target
meets the inclusion criteria and his/her child(ren) has(have) signs of social
anxiety, then the target is asked to invite his/her family members to
participate. After the target discusses the research with his/her family
members, we ask for the contact details of the family members . This way we are
certain that the target participant meets all inclusion criteria, before (s)he
contacts his/her family members about participating in this research (please
see page 15 of the protocol for details about inclusion).

- Minor changes to the information letters and informed consent forms.

Abnormalities in the MRI and EEG recordings are the only risks anticipated.
These negative observations will be communicated to the subject's general
practioner (medical doctor) by the neuro-radiologist. Separate consent forms
for this a procedure are attachted to this submission