The genetics of trichotillomania and excoriation disorder: A systematic review

Abstract

Background:

Trichotillomania (TTM) and excoriation disorder (ED) are impairing obsessive-compulsive related disorders that are common in the general population and for which there are no clear first-line medications, highlighting the need to better understand the underlying biology of these disorders to inform treatments. Given the importance of genetics in obsessive-compulsive disorder (OCD), evaluating genetic factors underlying TTM and ED may advance knowledge about the pathophysiology of these body-focused repetitive behaviors.

Aim:

In this systematic review, we summarize the available evidence on the genetics of TTM and ED and highlight gaps in the field warranting further research.

Method:

We systematically searched Embase, PsycInfo, PubMed, Medline, Scopus, and Web of Science for original studies in genetic epidemiology (family or twin studies) and molecular genetics (candidate gene and genome-wide) published up to June 2023.

Results:

Of the 3536 records identified, 109 studies were included in this review. These studies indicated that genetic factors play an important role in the development of TTM and ED, some of which may be shared across the OCD spectrum, but there are no known high-confidence specific genetic risk factors for either TTM or ED.

Conclusions:

Our review underscores the need for additional genome-wide research conducted on the genetics of TTM and ED, for instance, genome-wide association and whole-genome/whole-exome DNA sequencing studies. Recent advances in genomics have led to the discovery of risk genes in several psychiatric disorders, including related conditions such as OCD, but to date, TTM and ED have remained understudied.

1. Introduction

Trichotillomania (TTM) and excoriation disorder (ED) are debilitating psychiatric conditions classified as obsessive-compulsive and related disorders (OCRDs) in the Diagnostic and Statistical Manual of Mental Disorders Fifth Edition (DSM-5) [1]. In this edition, TTM was reclassified from an impulse control disorder to an OCRD, and ED was added as a diagnosis due to its phenomenological and biological similarities to obsessive-compulsive disorder (OCD) [2–4]. TTM and ED are often further grouped together as body-focused repetitive behaviors (BFRBs) due to their shared symptomology and presentation. TTM is characterized by repetitive pulling out of one’s hair from the scalp, eyebrows, eyelashes, or other hair-bearing parts of the body, leading to hair loss. ED is characterized by repetitive manipulation of the skin that causes tissue damage, often leading to sores and scabs. Both disorders occur despite efforts to stop and cause significant distress and impairment for the afflicted individual.

Though understudied, both conditions are relatively common with estimated pooled prevalence ratios (current or lifetime) of 1.14% for TTM [5] and 3.45% for ED [6] based on recent meta-analyses. These conditions typically onset in childhood with an average onset age of 11–15 years [7–11], although for some individuals they may present later in adulthood [10,11]. Several studies, but not all, suggest that these BFRB conditions may impact females more than males [reviewed in 5–6]. The natural history of these conditions is characterized by a chronic course of waxing-and-waning symptom severity throughout the lifetime [12–14]. Therefore, the impact of these conditions on children and adults may be substantial across the lifespan, and both TTM and ED have been associated with impairments in occupational, academic, and social functioning as well as financial burdens [15–17]. Studies have also demonstrated that individuals with either condition have high rates of psychiatric comorbidities, including depression (12–65%), anxiety (8–32%), and substance use disorders (14–36%) [18,19]. TTM and ED can also lead to medical consequences, including infections and repetitive hand-use injuries from pulling and picking for hours a day. Currently, available estimates indicate that 2 in 10 individuals with TTM engage in trichophagia, or eating of their hair [20], which puts them at risk for developing trichobezoars and other possibly serious gastrointestinal complications. The considerable prevalence of TTM and ED in the general population, their impacts on psychosocial functioning across the lifespan, and the plethora of psychiatric and physical conditions associated with these BFRBs underscore their public health importance.

Psychotherapeutic interventions such as cognitive-behavioral approaches with habit reversal training show efficacy in reducing TTM and ED symptoms [21–23]. However, identifying trained providers in evidence-based psychotherapeutic approaches is challenging, and many individuals do not respond or may relapse in the long-term [24,25]. Currently, there are no FDA-approved medications or pharmacologic agents with robust, clear evidence of efficacy to treat either disorder [21–23,26–28], highlighting the urgent need to better understand the underlying biological mechanisms to inform the development of effective treatments. Previous work on the pathophysiology of TTM and ED supports the critical role of genetic factors and provides evidence that the same genes may contribute risk to both disorders [4,29].

To our knowledge, there are no previous systematic reviews that have focused on the genetics of TTM and ED. Previous reviews [3,30–32] have acknowledged the genetic overlap between TTM and ED as being related to OCD but have not discussed separate, disorder-specific genetic findings for TTM or ED. In this systematic review, we summarize previous research on the genetics of these disorders and highlight gaps in the field that warrant further research that may advance our understanding of these common and impairing psychiatric conditions.

2. Methods

This study is reported in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA). A protocol was not pre-registered. One independent reviewer conducted all review stages (MR) and consulted regularly with senior researcher (EO) to discuss progress.

We searched select electronic databases (Embase, PsycInfo, PubMed, Medline, Scopus, Web of Science) up to June 2023 (summary of records identified in Table S1). The search strategies employed were tailored to each database and details of the search terms are provided in the supplement (Table S2). We restricted eligibility to studies in English. Titles and abstracts were initially screened followed by the full text to assess eligibility. Articles were considered for inclusion if they provided data on the genetics of TTM, ED, and/or more generally obsessive-compulsive related disorders. No restrictions regarding the definition or assessment of TTM, ED, and obsessive-compulsive-related disorders were made. Articles were excluded if they did not present genetic findings in either human or animal studies. Fig. 1 shows the PRISMA flowchart of article selection. After the final screening, 109 studies were included in this systematic review.

Fig. 1.

PRISMA Flowchart.

Systematic selection process for articles included in this review. Other sources include expert knowledge of the field and reference lists of other included studies. Articles designated out of scope include articles that did not discuss the genetics of trichotillomania or excoriation disorder.

3. Results

3.1. Family studies

For decades, studies have shown familial transmission of OCRDs, including TTM and ED. Familial case studies lack comparator groups but provide evidence that BFRBs, with a focus on grooming behaviors, may run in families. For instance, King et al. (1995) [33] found that in a small sample of 15 females with TTM, three subjects had family members with pathological grooming behaviors. Additionally, Ramot et al. (2013) [34] reported on three generations of a family (grandfather, father, and son) presenting with TTM, and Alireza et al. (2013) [35] reported on four sisters presenting with TTM. More recently, in a large sample of 145 participants with TTM and 120 with ED, Redden et al. (2016) [36] showed that 29% of the participants (n = 77) had a first-degree relative with either one of those BFRBs.

Comparative family studies examining individuals with TTM, ED, and OCD provided further evidence regarding the familial transmission of BFRBs (Table 1). For instance, Keuthen et al. (2014) [37] found that 6.3% of first-degree relatives of individuals with TTM (n = 128) had TTM compared to none of the first-degree relatives of control individuals without TTM (n = 50). Most recently, Zhang & Grant (2022) [38] showed that 17.1% of individuals with TTM (n = 152) reported a family history of OCD, TTM, and ED compared to 1.4% of controls (n = 71). Additionally, studies have also observed higher rates of OCD in relatives of individuals with TTM compared to controls. In a small sample of 16 individuals with TTM, Lenane et al. (1992) [39] reported that 6.4% of first-degree relatives of these TTM probands (n = 65) had OCD, compared to none of the first-degree relatives of controls (n = 90). Keuthen et al. (2014) [37] similarly showed that first-degree relatives of individuals with TTM were more likely than control first-degree relatives to have OCD, with 17.2% of first-degree relatives of individuals with TTM (n = 128) having OCD compared to none in the controls (n = 50).

Table 1.

Family studies examining how TTM, ED, and/or OCD impact first-degree relatives.

Author, Year	Assessment criteria	Sample size of individuals with OCD and OCRD (number of first-degrees relatives examined)	Mean Age Cases (Years)	Female Cases (%)	% of 1st-degree case relatives with OCD or OCRD vs % of controls	Statistical significance if provided in studies
Studies Analyzing TTM or ED Cases
Lenane et al. 1992 [39]	DSM-III-R	16 TTM (65 relatives), 65 non-TTM (90 relatives)	32	100%	6.4% (OCD) vs 0% (OCD)	N/A
King et al. 1995 [33]	Child Behavior Checklist	15 TTM	12	100%	20% (grooming behaviors) 53.3% (OCD)	N/A
Keuthen et al. 2014 [37]	DSM-IV-TR criteria, TDI-R, SPDI	110 TTM (128 relatives), 48 non-TTM (50 relatives)	23	94.5%	17.2% (OCD) vs 0% (OCD) 6.3% (TTM) vs 0% (TTM) 5.4% (ED) vs 4% (ED)	OCD: p < 0.0001 TTM: p = 0.0032 ED: p = N.S.
Redden et al. 2016 [36]	DSM-5 criteria	145 TTM, 120 ED	33	89.4%	29% (ED or TTM)	N/A
Zhang & Grant 2022 [38]	DSM-5 criteria	152 TTM, 71 non-TTM	30	N/A	17.1% (OCD) vs 1.4% (OCD) 17.1% (TTM) vs 1.4% (TTM) 17.1% (ED) vs 1.4% (ED)	OCD: p < 0.001 TTM: p < 0.001 ED: p < 0.001
Studies Analyzing OCD Cases
Bienvenu et al., 2000 [42]	DSM-IV criteria	80 OCD (343 relatives), 73 non-OCD (300 relatives)	N/A	N/A	1% (TTM) vs 0% (TTM) 6% (ED) vs 4% (ED)	TTM: N.S. ED: N.S.
Brakoulias et al. 2011 [40]	Self-Report ED/TTM	77 OCD	44	57.1%	63.6% (OCD) 1.3% (TTM)	N/A
Bienvenu et al. 2012 [43]	DSM-IV criteria	382 OCD (974 relatives), 73 non-OCD (233 relatives)	N/A	69%	4% (TTM) vs 0% (TTM) 17% (ED) vs 4% (ED) 55% (OCD) vs 7% (OCD)	TTM: p = 0.001 ED: p < 0.0005 OCD: p < 0.0005
Carmi et al. 2022 [41]	Self-Report (DSM-5) ED/TTM	222 OCD in Israeli cohort, 235 OCD in Australian cohort	Israeli cohort: 25, Australian cohort: 43	Israeli cohort: 48.6% Australian cohort: 52.8%	73% (OCD or OCRD)	N/A

Characteristics and percentages of case/control relatives from family studies of trichotillomania, excoriation disorder, and obsessive-compulsive disorder. Abbreviations: TTM = trichotillomania; ED = excoriation disorder; OCD = obsessive-compulsive disorder; OCRD = obsessive-compulsive related disorder; DSM-III-R = Diagnostic and Statistical Manual of Mental Disorders third edition revised; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders fourth edition; DSM-IV-TR = Diagnostic and Statistical Manual of Mental Disorders fourth edition text revised; DSM-5 = Diagnostic and Statistical Manual of Mental Disorders fifth edition; TDI-R = Trichotillomania diagnostic interview-revised; SPDI = Skin Picking Diagnostic Inventory; N/A = not applicable, N.S. = non-significant in family study with p > 0.05.

Studies focused on probands with OCD also showed familial aggregation of BFRBs in first-degree relatives. For instance, Brakoulias et al. (2011) [40] showed a lower, but still significant, TTM prevalence of 1.3% in first-degree relatives in a sample of 77 individuals with OCD. In a more recent study, Carmi et al. (2022) [41] found that 73% of their OCD cohort (335/457) had a first or second-degree family member affected by OCD or an OCRD, including TTM and ED. Bienvenu et al. (2000) [42] demonstrated a higher prevalence of TTM and ED in 343 first-degree relatives of 80 individuals with OCD compared to 300 first-degree relatives of 73 individuals without OCD. The same results were later found with a larger sample of 382 individuals with OCD and the same 73 controls, with 4% of first-degree relatives of individuals with OCD having TTM and 17% having ED [43]. Additionally, Hanna et al. (2005) [44] compared individuals with familial and sporadic OCD and examined the distribution of familial OCD phenotypes, not TTM or ED, but found that grooming behaviors as a general phenotype, especially skin picking (χ2 = 7.26, p = 0.006), were more frequently observed in families with multiple individuals affected by OCD rather than sporadic cases of OCD. Overall, these family studies suggest a potential genetic overlap between OCD, TTM, and ED.

3.2. Twin studies

Twin studies are a powerful tool to help determine the contribution of genetic influences and shared/unique environmental factors to a condition when a familial link is suspected. Monozygotic twins (MZ) share almost 100% of their genetic material while dizygotic twins (DZ) share approximately 50%; thus, MZ twins will have higher concordance than DZ twins if a disorder has large genetic underpinnings.

Only two twin studies focused on TTM have been published and both support differences in MZ and DZ concordance rates based on the DSM-IV definition of TTM (Table 2). Novak et al. (2009) [29] examined 24 MZ twin pairs and 10 DZ twin pairs and provided a heritability estimate of 76%. Monzani et al. (2014) [4] studied a population-based sample of 5409 females in the Twin-UK registry and estimated heritability to be 32% for TTM.

Table 2.

Heritability estimates of trichotillomania and excoriation disorder from twin studies.

Author, Year	Assessment instrument for ED or TTM	Sample Size (included in analysis)	Mean Age (Years)	Female (%)	Heritability Estimates (TTM or ED)
Novak et al. 2009 [29]	MGH Hairpulling Scale	24 MZ and 10 DZ twins	35	97.1%	76% (TTM)
Monzani et al. 2012 [45]	SPS	1303 MZ, 869 DZ, and 19 Unknown Zygosity twins	N/A	100%	40% (ED)
Monzani et al. 2014 [4]	MGH Hairpulling Scale, SPS	3042 MZ and 2367 DZ twins	N/A	100%	31.6% (TTM) 47% (ED)

Demographic characteristics and heritability estimates from trichotillomania and excoriation disorder twin studies. Abbreviations: TTM = trichotillomania; ED = excoriation disorder; SPS = Skin Picking Scale; MGH = Massachusetts General Hospital; MZ = monozygotic twins; DZ = dizygotic twins; N/A = not applicable.

For ED, Monzani et al. (2012) [45] showed a heritability estimate of ED to be approximately 40% using the Twin-UK registry. This study also showed a higher concordance rate in MZ twins rather than DZ twins, highlighting that both environmental and genetic factors play a role in developing ED. A follow-up study by this group showed a slightly higher heritability estimate of 47% [4].

Considering TTM and ED together, Monzani et al. (2014) [4] provided evidence for shared genetic factors across different OCRDs. In their analysis, they found that OCD and OCRDs may be influenced by two latent factors. The first factor was common to all OCRDs with a genetic overlap of 63.3% (95% CI, 59%–66%). The second factor was specific to TTM and ED with a genetic overlap of 73.7% (95% CI, 30%–99%). This suggests that some genetic factors may be shared across the OCD spectrum, while others seem specific to the BFRBs of TTM and ED.

3.3. Candidate Genes and Animal Models

The candidate gene approach allows researchers to investigate the validity of a hypothesized gene that is thought to be plausibly related to a disorder [46]. This approach involves assessing the association of a particular allele (or set of alleles) of a gene that may be involved in the disorder [47]. Additionally, candidate gene studies are typically performed with unrelated individuals and control subjects or with small families (e.g., a proband and parents) [47]. However, the small nature of candidate gene studies poses a significant limitation to their findings, especially when considering the small effect sizes of common variants and the polygenic nature of psychiatric disorders [48]. Importantly, many candidate genes have not replicated in large-scale genome-wide studies of other psychiatric disorders [49,50], so caution is warranted in interpreting these results.

3.3.1. SAPAP3/PSD95-associated protein 3

SAP90/PSD95-associated protein 3 (SAPAP3) is a postsynaptic scaffolding protein expressed in the striatum, an area of the brain that controls executive functions such as decision-making and inhibitory control. Welch et al. (2007) [51] showed that Sapap3 knockout (KO, −/−) mice had skin lesions when they were born and later had increased grooming behaviors and anxiety compared to wild-type mice. These phenotypes were thought to be similar to those found in humans on the obsessive-compulsive spectrum, especially TTM, leading to increased interest and studies in humans focused on SAPAP3.

Bienvenu et al. (2009) [52] genotyped a sample of 1618 individuals with OCD and their first- and second-degree relatives intending to assess for comorbid grooming behaviors. Out of six common single nucleotide polymorphisms (SNPs) found on the SAPAP3 gene, one was found to be nominally associated with TTM (rs4652869), one associated with ED (rs4652867), and one associated with both disorders (rs6662980) [52]. In another study, the polymorphism rs6662980 was also found to be associated with excessive hand washing and extreme contamination fears (p = 0.003) in individuals with OCD [53], suggesting an additional overlap of this polymorphism with OCD and OCRD symptoms. Zuchner et al. (2009) [54] looked at a sample of 165 individuals with TTM and/or OCD compared to 178 controls and found seven novel non-synonymous heterozygous SAPAP3 (+/−) variants. These variants were found in 4.2% of individuals with OCD or TTM compared to just 1.1% of controls, and three of the seven variants were predicted to be possibly damaging using PolyPhen (Polymorphism Phenotyping, http://genetics.bwh.harvard.edu/pph). Finally, a nominally significant association between the SAPAP3 rs11583978 common variant and TTM was found in a sample with TTM (n = 45) compared to controls (n = 153), but this association was not significant after correcting for multiple comparisons [55].

Candidate gene studies and animal models focused on Tourette disorder have also generated intriguing findings regarding SAPAP3. Although Tourette disorder is formally recognized as a neurodevelopmental disorder in the DSM-5, many researchers believe that there are important similarities between Tourette disorder and OCRDs, including TTM and ED [56]. Crane et al. (2011) [57] examined a sample of 289 parent-child trios with Tourette disorder and identified associations with common genetic variants in SAPAP3. Specifically, they identified associations between Tourette disorder and rs11264126 and two haplotypes containing rs11264126 and rs12141243 that could not be explained by co-occurring OCD or TTM. These results were not significant after correction for multiple hypothesis testing. The association between SAPAP3 and Tourette disorder has also been supported by a recent mouse behavioral study [58]. Sapap3 KO mice presented with tic-like behaviors that improved after administering aripiprazole, an evidence-based treatment for Tourette disorder [58,59]. However, it is also important to note that a recent study by Ehmer et al. (2020) [60] found that excessive grooming behaviors in Sapap3 KO mice may not be related to OCD-like compulsive behaviors. The authors observed that wild-type mice and Sapap3 KO mice did not differ in OCD-like compulsive behavior measured by a signal attenuation task, suggesting that grooming behaviors may be related to a different mechanism of compulsivity and not tied to OCD-like behaviors. However, genome-wide studies of Tourette disorder, including studies of common and rare sequence and structural variation did not implicate SAPAP3 [61–66].

Recent SAPAP3 research has also focused on rodent neuroimaging and functional brain activity. Neuroimaging and functional studies of Sapap3 wildtype and KO mice have led to findings implicating the striatum [67–82], the glutamatergic system [83–87], the monoaminergic system [88], the orbitofrontal cortex [89–91], the motor cortex [73], the medial prefrontal cortex [92], and dopamine receptors [93] to obsessive-compulsive symptoms and aberrant grooming behaviors. Additionally, some studies have shown Sapap3 KO mice present with impaired learning behaviors [92,94], behavioral flexibility deficits [95,96], and impaired valence processing [97]. These abnormalities are similar to those found in humans with OCD and OCRDs; however, more research is needed to determine the precise relationship between brain functioning, SAPAP3, and obsessive-compulsive-related symptoms (for a full summary of results see Table S3).

3.3.2. The SLIT and TRK-like family

The SLIT and TRK-like, or SLITRK family of proteins are neuronal transmembrane proteins involved in neurite outgrowth [98,99]. Previously, rare SLITRK gene variants have been associated with Tourette disorder [100–103], epilepsy [104], and OCD [105–108], however, there is some disagreement about the validity of these associations due to these findings not replicating in other samples [109–118].

Zuchner et al. (2006) [119] reported two rare non-synonymous variants in the SLITRK1 gene in 44 families with TTM, resulting in 4.5% of the families having a rare SLITRK1 mutation. Interestingly, these variants were in a different region of the gene than those reported in individuals with Tourette disorder [100]. Knocking out Slitrk1 in mice caused increased anxious behaviors, depressive behaviors, and aberrant grooming behaviors [99]. In addition, concentrations of norepinephrine and its metabolite 3-methoxy-4-hydroxyphenylglycol (MHPG) were increased in the prefrontal cortex, striatum, and nucleus accumbens of the mice, suggesting that norepinephrine mechanisms may be involved in the development of TTM and anxiety-like behaviors [99]. Hatayama & Aruga (2023) [120] also found that norepinephrine (noradrenergic) fiber growth and signaling were altered in Slitrk1 KO mice and hypothesized that altered norepinephrine signaling plays a role in the development of anxiety-like behaviors.

Another SLITRK member, SLITRK5, has been studied as a candidate gene for TTM. Slitrk5 homozygous knockout mice were shown to have an increased duration of grooming events compared to wild-type mice, which were rescued with fluoxetine treatment. These mice also had decreased striatum sizes compared to controls. Slitrk5 homozygous and heterozygous KO mice both showed facial hair loss and severe skin lesions due to excess grooming, but homozygous KO mice exhibited these symptoms at three months of age while heterozygous KO mice showed them seven to nine months later [121] (for a full summary of results see Table S4).

3.3.3. Homeobox B8

Homeobox B8 (HOXB8) is a member of the homeobox (HOX) gene family, a group of genes whose main function is to provide positional information during embryonic development along the head-tail axis [122]. However, HOX genes are also responsible for the formation of various organs and tissues in the body, and Hoxb8 was hypothesized to be involved in the neural circuit responsible for animal grooming and/or itch control [122,123]. Hoxb8 mutant mice were shown to spend almost twice as much time engaging in grooming behaviors compared to wild-type mice [124,125]. Because of this excessive grooming behavior, they often also develop self-inflicted open lesions [124].

In addition, these mice display behaviors characteristic of anxiety and social impairment [125]. These phenotypes were found to be rescued by fluoxetine treatment over one week [125] and transplantation of bone marrow from wild type to mutants [123]. Interestingly, Tränkner et al. (2019) [126] found that anxiety behaviors related to over-grooming behaviors and altered physiological stress responses were present in all female Hoxb8 mutant mice, but not in male Hoxb8 mutant mice. They also found that female sex hormones, including progesterone and 17β-estradiol, were linked to increased over-grooming and anxiety behaviors in the mice, with symptoms declining after reducing hormones.

While Greer and Capecchi (2002) [124] found the expression pattern of Hoxb8 in the central nervous system to be very broad, Chen et al. (2010) [123] identified that microglia were the only cells in adult mice that expressed Hoxb8, and only 40% of the microglia population expressed Hoxb8. De et al. (2018) [127], upon further investigation, approximated that only one-third of the microglia population expressed Hoxb8. They also noted that Hoxb8 microglia had a distinct spatial organization pattern, highly concentrated in areas associated with the cortico-basal ganglia network [127], which has previously been associated with OCD [128]. They hypothesized that this arrangement may play a role in the pathological grooming behaviors presented by Hoxb8 mutant mice. Tränkner et al. (2019) [126] found that ablating Hoxb8 lineage microglia in wild-type mice produced aberrant grooming and anxious behaviors similar to Hoxb8 mutant mice and that ablating Hoxb8 lineage microglia in Hoxb8 KO mice did not rescue the pathology. Because of these results, they hypothesized that loss of function in Hoxb8 lineage microglia may contribute to developing obsessive-compulsive behaviors and anxiety. In the most recent study of Hoxb8 microglia, Nagarajan & Capecchi (2023) [129] found that optogenetic stimulation of Hoxb8 microglia within the dorsomedial striatum or the medial prefrontal cortex induces excessive grooming behaviors, stimulation of Hoxb8 microglia in the basolateral amygdala or central amygdala increases anxiety, and stimulation of Hoxb8 microglia in the ventral CA1 region of the hippocampus induced both excessive grooming and elevated anxiety as well as freezing responses. Nagarajan & Capecchi (2023) [129] hypothesized from these results that Hoxb8 microglia may be among the first responding agents within the brain to evoke anxiety responses and may be involved in the development of many anxiety-like behaviors (for a full summary of results see Table S5).

Even though studies have been conducted on the mouse Hoxb8 gene, no studies have directly investigated the HOXB8 gene in humans concerning TTM or ED. Although the excessive grooming phenotype is reminiscent of TTM, ED, and other obsessive-compulsive spectrum disorders, further research in humans is necessary to understand the role of HOXB8 in OCRDs.

3.3.4. Additional candidate models examined in individual studies

In a study comparing 39 individuals with TTM, 250 individuals with OCD, and 152 controls, Hemmings et al. (2006) [130] hypothesized that the serotonin (5-HT) receptors may be involved in the development of TTM since the serotonin system may be involved in the development of OCD. They found that the common T102C variant of the 5-HT2A receptor differed between individuals with TTM and controls (p = 0.028 and p = 0.024), with the T102T-genotype possibly contributing to the development of TTM [130].

Zinc finger protein 462 (Zfp462) may play a role in embryonic development and neuronal differentiation [131]. Wang et al. (2016) [131] found that Zfp462 heterozygous KO (+/−) phenotype mice exhibited increased anxiety behaviors and excessive self-grooming similar to Hoxb8 mutant mice. The authors suggested that a deficiency in Zfp462 may contribute to the development of anxiety and aberrant grooming behaviors in mice [131].

Nitric oxide synthase 2 (NOS2) may be involved in the migration and maturation of neuronal cells in the cortex, especially the cortical-striatal-thalamus-cortical circuitry, which is hypothesized to be involved in TTM [132]. Casarotto et al. (2018) [132] found that Nos2 KO mice had elevated self-grooming behaviors and that partial Nos2 expression reduced grooming behaviors.

Parolari et al. (2021) [133] found that mutations on the paired like homeodomain 2 (Pitx2) and gamma-aminobutyric acid type A receptor subunit rho3 (Gabrr3) genes within the subthalamic nucleus of mice elicited highly repetitive self-grooming behaviors, which were then shown to be reduced after a trial of fluoxetine, a first-line treatment for OCD. Interestingly, they also found that Sapap3 KO mice presented with reduced grooming behaviors after modulated inhibition of Gabrr3 neurons. Parolari et al. (2021) [133] hypothesize that targeting GABRR3 may yield beneficial treatment for OCD and TTM behaviors.

None of the previous four candidate gene mouse models have been replicated. Additional research is necessary to determine the validity of these findings and how they may be related to TTM presentation in humans.

3.4. Copy Number Variants

Copy number variants (CNVs) are genome areas with duplications or deletions in comparison to the reference genome. Rare CNVs that are uncommon in the general population have been associated with multiple neuropsychiatric disorders [134]. Our review did not identify any CNV studies that focus specifically on TTM or ED, but a few CNV studies have been conducted showing associations with OCD. McGrath et al. (2014) [135] conducted a genome study of rare and large CNVs in 2699 patients (1613 ascertained for OCD, 1086 ascertained for Tourette disorder) and 1789 controls and did not find a global burden difference. They did find, however, an increased number of deletions in the 16p13.11 locus, which is associated with other psychiatric and neurodevelopmental disorders, and one individual with the 16p13.11 locus had TTM and ED in addition to OCD [135]. Gazzellone et al. (2016) [136] identified de novo CNVs in 2.3% of 307 unrelated pediatric probands with OCD (including 174 complete parent-child trios), and one individual with OCD with comorbid TTM had a deletion involving CDH7 and CDH19 [136]. More recent studies show additional success in identifying CNVs in individuals with OCD and provide support to previously identified CNVs [137,138], however, these studies did not specifically report CNVs in individuals with TTM or ED. Further studies are necessary to understand the role of CNVs in the pathogenesis of TTM and ED.

3.5. Genome-Wide Association Studies

Genome-wide association studies (GWAS or GWA studies) are observational studies that examine a genome-wide set of genetic variants and typically examine the association between common SNPs in individuals who have a condition compared to controls. To date, there have been no published GWA studies focused on TTM or ED. Of OCRDs, five GWAS have been published for OCD [139–143], two GWAS have been conducted for hoarding disorder [144–148], and no GWAS has been conducted focused on body dysmorphic disorder [31]. Stewart et al. (2013) [139] found in a case-control analysis of 1465 individuals with OCD, 5557 ancestry-matched controls, and 400 complete trios that the SNPs with the two lowest p-values were in an intron of DLGAP1. DLGAP1, also known as SAPAP1, is in the same gene family as SAPAP3 which has been implicated as a candidate gene for TTM and ED (see SAPAP3 section above). Similar findings have been reported by Mattheisen et al. (2015) [140] with a sample size of 1406 individuals with OCD with a total sample of 1061, and IOCDF-GC & OCGAS (2018) [141] with a sample size of 2688 individuals with OCD and 7037 controls.

Additionally, Mattheisen et al. (2015) [140] found a SNP near PTPRD that had the lowest p-value for their sample. PTPRD interacts with SLITRK3 [146], which is a member of the see SLIT and TRK family. Other members of the SLIT and TRK family, SLITRK1 and SLITRK5, are candidate genes for TTM (see SLITRK Family section above). IOCDF-GC & OCGAS (2018) [141] den Braber et al. (2016) [142] have also reported similar findings. Despite these recent strides in OCD GWAS findings, there is still a significant underutilization of GWAS for OCD and OCRDs, especially TTM and ED. More recently, Strom et al. (2021) [145] conducted the largest OCD GWAS to date identifying the first genome-wide significant locus, which may yield further insight into the genetics of OCD and related disorders.

3.6. DNA sequencing studies

Rare DNA sequence variants are uncommon in the general population, but also likely play a role in the genetic architecture of OCRDs, including TTM and ED [148]. To date, there have not been any published whole-exome or whole-genome DNA sequencing studies focusing on TTM or ED. However, five exome sequencing studies were identified for OCD [106,149–152]. Noh et al. (2017) [150] found variations in the protein-coding variant NRXN1 to be genome-wide significant in a sequencing study of 592 individuals with OCD. NRXN1 variations are associated with many other psychiatric disorders, including Tourette disorder [153]. Cappi et al. (2016) [149] found an increased burden of de novo non-synonymous variants in a sample of 20 parent-child trios where the child had OCD compared to unaffected trios. In a larger follow-up, Cappi et al. (2020) [151] identified two high-confidence risk genes, CHD8 and SCUBE1, in a whole-exome DNA sequencing study of 184 parent-child trios with childhood-onset OCD. Interestingly, one of the subjects with a CHD8 de novo gene-damaging mutation also had TTM [151]. Halvorsen et al. (2021) [106] conducted the largest whole-exome DNA sequencing study in OCD (1313 individuals with OCD, including 587 trios, 41 quads, and 644 singletons of affected individuals) and did not find any significant gene-specific mutations related to OCD. The most significant single gene result in the case-control analysis examining rare variant mutation burden was SLITRK5, which is a previous candidate gene for TTM (see SLITRK Family section above). Additionally, Halvorsen et al. (2021) [106] assessed TTM status, male sex, tics, and skin picking as predictors of having more damaging de novo mutations related to OCD, but only male sex was a significant predictor. Most recently, the first whole-genome sequencing study of OCD was published examining 53 families with 54 probands with OCD and identified four possible risk genes associated with de novo mutations [152]. These findings reinforce the potential contribution of rare variants in the development of OCRDs, including TTM and ED, and suggest an area for future research to identify risk genes for these conditions.

4. Discussion

In summary, this systematic review highlights the important role of genetic factors in TTM and ED from family and twin studies but provides limited evidence for specific high-confidence genetic risk factors. The majority of past research on the genetics of TTM and ED has focused on candidate gene association studies of common genetic variants and animal models. Replication of these intriguing findings in large-scale unbiased genomic studies is a critical next step. Importantly, no genome-wide studies of common (e.g., GWAS), rare sequence (e.g. DNA sequencing studies), or rare structural (e.g., CNV analysis) variants have been published focused specifically on TTM or ED as stand-alone disorders. The few findings that are available for these disorders were identified through studies that focused on related disorders, especially OCD.

The current state of affairs of genetic research on TTM and ED is in contrast with that of other psychiatric conditions, such as schizophrenia, depression, anxiety, autism, attention-deficit/hyperactivity disorder, and even obsessive-compulsive disorder where several specific genetic risk factors have been implicated in recent years. The lack of data available on the genetics of TTM and ED makes it difficult to draw conclusions about the genetic contributions to these disorders and related biological mechanisms. Hence, most of the genetic influence contributing to TTM and ED remains undiscovered, creating roadblocks to further mechanistic studies. While this may also be true for psychiatry in general [154], genetics research of TTM and ED clearly lags behind that of other psychiatric conditions, especially given the public health impact of these conditions. We suggest that additional research should be conducted on the genetics of TTM and ED, with a specific focus on genome-wide approaches, to bridge the gaps in TTM and ED genetics.

4.1. Limitations

Overall, the strengths of the present review include the search criteria as we comprehensively reviewed the available literature on the genetics of TTM and ED to examine the most updated genetic findings for these disorders. However, the strengths of this study should be considered in the context of its limitations. First, many of the genetic findings in this review were based on candidate gene approaches which are limited by the small effect sizes of common variants and the polygenic nature of psychiatric disorders [48], and historically often do not replicate in large-scale genome-wide studies [49,50]. Second, many of the genetic findings in this review were based on single studies of animal models, which have not been replicated in additional studies or human models. Lastly, we acknowledge that many of the studies included did not have TTM or ED as the primary focus of the genetic analyses, limiting our conclusions about the genetics of these disorders.

5. Conclusions

Despite these limitations, the present study contributes to our understanding of the current state of genetics research of TTM and ED as it attempts to comprehensively summarize genetic findings for these disorders. The most notable conclusions of our review is that there is a paucity of genetic risk factors known to be associated with TTM or ED, and few studies used genome-wide approaches to examine the genetics of these disorders as compared to other, related disorders [31]. In recent years, genome-wide analyses have contributed significantly to our understanding of the genetics of psychiatric conditions [155,156], and it is our suggestion that more research should leverage these approaches in TTM and ED.