Tracing Autism Traits in Large Multiplex Families to Identify Endophenotypes of the Broader Autism Phenotype

Families comprising many individuals with Autism Spectrum Disorder (ASD) may carry a dominant predisposing mutation. Our aim was to use rigorous phenotyping of the ‘Broader Autism Phenotype’ (BAP) in large multiplex ASD families to identify endophenotypes of the BAP for future genetic studies. We evaluated ASD/BAP features using standardised tests and a semi-structured interview to assess social, intellectual, executive and adaptive functioning in 109 individuals, including two large multiplex families (Family A: 30; Family B: 34) and an independent sample of small families (n=45). Our protocol identified four psychological endophenotypes of the BAP that were evident in both samples, and showed high sensitivity (97%) and specificity (82%) for individuals classified with the BAP. The patterns of inheritance of these endophenotypes varied in the two large families, supporting their utility for identifying genes in autism.

clinical research identified autistic traits in relatives of children with ASD, described as the 48 'broader autism phenotype' (BAP) 2,3 . The BAP sits at the mildest end of the ASD spectrum and 49 includes a range of subtle behavioural and cognitive features that reflect the two core domains of 50

ASD. The Diagnostic and Statistical Manual of Psychiatric Disorders -V (DSM-V) stipulates 51
that a significant degree of impairment must be present to qualify for a diagnosis of ASD 4 , 52 whereas BAP traits lie on a continuum of normal population behaviours 5 . Monozygotic twins 53 demonstrate 30% concordance for ASD 6 , increasing to 92% if the BAP is considered, while 54 dizygotic twin concordance is ~10% 7 . Prevalence of the BAP in the general population is 55 unknown, whereas several studies have demonstrated higher rates of BAP traits (20-50%) among 56 relatives of children with ASD compared to controls, particularly in the areas of pragmatic 57 language 8,9 , personality 10 , social cognition 9,11,12 and executive function 13,14 . Together these 58 findings support the notion of complex inheritance of ASD. 59 The overall diagnostic rate of ASD is now >30%, including monogenic and chromosomal 60 aetiologies 15,16 . The remaining ~70% are likely to have a genetic basis, with polygenic 61 architecture in some, and unique de novo mutations in others 17 . Complementary techniques will 62 be necessary to unravel aetiology in unsolved cases. Typically, family studies combine many small families (2-3 affected individuals), however, these are likely to be confounded by genetic 64 heterogeneity. Very large multiplex families (> 8 affected) where ASD traits appear dominantly 65 inherited are rare, but more genetically homogeneous. In other complex disorders, such as 66 epilepsy, phenotypic characterisation of such families has proved powerful in gene discovery 18 , 67 however this approach has received limited attention in ASD 1818 . In multiplex ASD families, the 68 identification of family members with BAP traits, or endophenotypes, may serve as markers of 69 carrier status 19,20 . In turn, this may facilitate gene identification 21 . 70 Endophenotypes are measurable features within a disorder that are proposed to reduce its 71 complexity into more quantifiable elements 22 . They have been hypothesised to reflect more 72 aetiologically homogeneous subgroups within genetically heterogeneous conditions. There are 73 several BAP traits that may be considered "endophenotypes" from within the domains of 74 language, executive function, and social cognition 21 . In the context of a single large family where 75 numerous individuals demonstrate ASD or the BAP, recognition of BAP endophenotypes should 76 allow granular identification of an autism gene of dominant effect. This study is the first known 77 to the authors to apply this approach in autism. 78 The aim of our study was to analyse autistic traits within large multiplex families to 79 examine inheritance of ASD by identifying endophenotypes of the BAP. We achieved this aim 80 using an iterative process, first rigorously phenotyping many members of large multiplex 81 families to delineate the full range of BAP traits for potential endophenotypes. We then assessed 82 these traits in a separate sample of 20 small families, each with at least one member with ASD, 83 to independently validate the endophenotypes. We then applied these endophenotypes to two of 84 our fully characterised large multiplex families (from step 1) to assess their utility for examining 85 inheritance patterns. We hypothesised that (1) multiple individuals in large families would 86 examination was conducted for dysmorphic and neurocutaneous features and growth parameters. 109 Standard genetic testing (karyotype, fragile X testing) and metabolic investigations were 110 performed on probands. 111 Across Family A and B, 65 individuals were recruited: 16 children (2 -12 years), 9 112 adolescents (13-17 years) and 40 adults (18-79 years) spanning 4 generations. Of these, 16/65 113 met criteria for a diagnosis of ASD. Family B also reported a deceased family member who had 114 a diagnosis of ASD (generation V) and an additional family member with ASD (generation 4) 115 who was not recruited. Scrambled pedigrees of affected status are presented to preserve 116 participant anonymity (Fig.1). In each family, a matriarch was identified. Individuals directly 117 related to each matriarch are classified as 'core family'; others are referred to as 'married-in'. 118 Family A comprised 30 individuals, including 7 diagnosed with ASD (6/9 children, 1/6 adults; 119 Fig.1a). Nineteen were core family; 11 were married-in. In Family B, we fully phenotyped 35 120 individuals, not including the matriarch (who was not assessed). Three children and three 121 adolescents participated in a limited range of phenotyping activities and as such, these 122 individuals were excluded from final analyses. Nine had ASD (5/7 children, 1/5 adolescents, 123 3/22 adults; Fig.1b We used an iterative process to characterise, refine and assess endophenotypes of the BAP in our 175 two separate samples, as summarised in Fig.2

. 176
Step 1: Identification of Potential BAP Endophenotypes in Large Multiplex Families 177 Using a grounded theory approach, BAP traits were initially identified from a detailed literature 178 review targeting the theoretical domains described in the seminal work of Bolton (1994), on 179 which the conceptualisation of the BAP is largely based. The domains included speech, literacy, 180 pragmatics, relationships, and circumscribed interests, which were explored in-depth using our 181 BAP phenotyping protocol (described above) in members of unrelated large multiplex families 182 primarily ascertained through the Collaborative Autism Study 33 . This in-depth characterisation 183 was phenomenologically based 37 , whereby the number of traits within each domain was fully 184 expanded through administration of the semi-structured interview (BAPI) with separate family 185 members until no further traits were identified (saturation) to capture the entire range of BAP 186 traits (Table 3). 187 Interpretation of cluster groupings was informed by the relative similarity and dissimilarity in the 196 linkage output combined with clinical judgement, leading to the initial identification of five 197 endophenotypes. Inspection of these endophenotypes revealed a consistent rating of 0 for two of 198 the 36 traits across all interviews, leading to their removal. One further trait reflecting 199 inflexibility to intentional errors was removed due to challenges reliably assessing it across 200 interviewers, resulting in a final set of 33 BAP traits (Table 3). 201 Step 2: Validation of BAP Endophenotypes in Small Families 202 In the small families sample, an independent expert in ASD assessment (CG) interviewed and 203 rated 45 participants on the 33 BAP traits based on all qualitative and quantitative data, with a 204 subset (9%) rated via consensus between CG, IES and SJW to ensure consistency in ratings 205 across both samples and to clarify borderline cases. As above, Ward's hierarchical cluster 206 analysis was used to examine natural trait groupings. This led to the identification of four 207 endophenotypes that showed a high degree of similarity to the initial five cluster solution. 208 To account for a variable number of traits in each cluster we computed proportional 209 scores, whereby scores on each trait (range 0-3) were summed and divided by the maximum total 210 score for that cluster, to produce four cluster scores for each individual. An ROC curve was 211 plotted for each cluster in the small families sample to identify optimum cut-off scores for 212 determining endophenotypic status using Youden's Index to allow mildly affected individuals to 213 be included 39,40 . The highest score was used to represent the most prominent endophenotype for 214 each individual, calculated as the difference between the observed endophenotype (i.e., cluster) 215 score and the threshold score for the endophenotype (i.e., cut-off score

Table 3. Identification of BAP traits through expansion of the BAP domains 219
Step 3: Assessment of BAP Endophenotypes in Family A and B 221 A team member who had not been involved in the phenotyping of Family A and B (Step 222 1) performed the endophenotype analysis (KT). Proportional scores for the four endophenotypes 223 were calculated, and family members classified as having the endophenotype if their 224 proportional score was greater than or equal to the cut-off scores identified in the small families 225 analysis (step 2). As above, the highest score (observed endophenotype score -threshold 226 endophenotype score) for any endophenotype was used to represent an individual's most 227 prominent endophenotype. A discriminant function analysis was then used to determine the 228 sensitivity and specificity of the endophenotype approach to identifying the presence of the BAP 229 in these families. In addition, endophenotype results were correlated with measures of intellect, 230 executive, social and adaptive functions using conservative non-parametric Spearman's 231 correlations (rs). 232

Hypothesis 1: Multiple individuals in large families demonstrate the BAP 234
Based on our rigorous protocol for phenotyping the BAP in large multiplex families, we 235 identified 32 members with the BAP across Family A and B. Of the 23 members in Family A 236 who did not have an ASD diagnosis, we detected the BAP in 17 (74%) individuals, with 6 237 individuals unaffected. In Family B, we detected 15 (63%) individuals with the BAP, with 9 238 individuals unaffected (Table 4). Although we note that Box's M was violated in the discriminant function analysis (likely due to 263 variation in the sample sizes), combined, the four endophenotypes captured 93% of cases 264 (Wilk's l=0.47, c 2 =27.83, p<.001). In particular, the endophenotypes showed high sensitivity for the BAP group (97%), characterised by higher proportional scores, and good specificity for 266 the unaffected group (82%), with lower proportional scores (

Hypothesis 3: BAP endophenotypes vary in large multiplex families 277
Applying the above endophenotype thresholds to the proportional scores of the 33 BAP traits for 278 members of Family A and B led to the identification of all individuals classified as having the 279 BAP. Two additional BAP cases were identified in Family B based on the presence of above 280 threshold endophenotype scores, indicating good utility of this approach (Fig.3). One individual 281 was excluded from this analysis due to incomplete data (III-7). Across both families, the aloof 282 endophenotype was most commonly observed (62%), followed by obsessive (60%), pedantic 283 (55%) and socially unaware (48%). Approximately one quarter of family members met criteria 284 for only one endophenotype, 15% met criteria for two, and the remainder met criteria for 3-4 285 (62%) (Fig.3). The dominant endophenotype across both families, as determined by the highest 286 score, was aloof (47%), followed by obsessive (26%), socially unaware (18%) and pedantic 287 (9%). 288 Family A appeared to have two endophenotype profiles, with one characterised by the 289 presence of a single endophenotype (35%) seen in individuals who were mostly married-in 290 (67%), contrasting with the second profile (41%) of all four endophenotypes, most evident in 291 core family members (72%) (Fig.4). Overall, the obsessive endophenotype occurred most 292 frequently (77%), followed equally by pedantic (65%) and aloof (65%), and then socially 293 unaware (53%). The co-occurrence of the obsessive and pedantic endophenotypes was relatively 294 common, seen in 29% of married-ins and core family members. Overall, there was a range of 295 dominant endophenotypes across individuals, with aloof the most frequent (35%) particularly in 296 core family members (83%). 297 Contrasting with Family A, Family B had more individuals (70%) with multiple 298 endophenotypes, in both married-in and core family members (Fig.4). All four endophenotypes were again most frequently observed in core family members, indicative of a more severe BAP 300 presentation. Unlike Family A, however, the aloof endophenotype occurred most frequently in 301 Family B (88%), followed by obsessive (71%), pedantic (65%), and socially unaware (65%). 302 The aloof endophenotype was also identified as dominant (59%), evident in 70% of core family 303 members. 304

Correlates of the BAP Endophenotypes 307
Across both families, no sex or age differences were observed for any of the 308 endophenotypes (all p<.200). Overall, a more severe BAP presentation (indicated by a greater 309 number of endophenotypes) was associated with reduced social adaptive functioning on both 310 self-report and objective measures of social communication (Table 6). In particular, a more 311 severe BAP presentation showed a strong correlation with more severe pragmatic language 312 difficulties, with scores for each endophenotype also significantly correlated. A similar 313 relationship was evident for the ability to detect a faux pas in social discourse and self-reported 314 social functioning, particularly for family members with the socially unaware endophenotype 315 (  For the cognitive measures, a more severe BAP presentation was associated with reduced 322 executive functioning, particularly for nonverbal measures of cognitive flexibility (switching and 323 fluency; Table 6). A pattern of weaker correlations was also evident for specific endophenotypes, 324 including lower IQ in the socially unaware and aloof endophenotypes (Table 6). 325

Discussion 326
We studied the BAP to highlight the phenotypic variation within and between high-risk ASD 327 families, to improve identification of individuals crucial for accurate molecular genetic analysis.

Endophenotypes of the BAP 348
Over the last 20 years, the BAP has emerged as strongly associated with ASD. The BAP is 349 considered a marker of carrier status of genes that may contribute to autism risk 21,45 . Here we 350 aimed to dissect the BAP into endophenotypes to understand the phenotypic variation within and 351 between families. Importantly, each endophenotype cluster was characterised by a combination 352 of communication, personality and behavioural indicators showing how specific traits across the 353 traditional BAP domains may group together to form distinct endophenotypes or 'profiles'. As 354 summarised in Table 7, these profiles capture identifiable 'personas' that have core characteristics with high face validity. These profiles also vary with functional correlates in 356 distinct ways, supporting their construct validity. For example, the aloof endophenotype was 357 characterised by a lack of innate social motivation or ability to meaningfully connect and 358 empathise with others, associated with decreased theory of mind, lower executive and 359 intellectual functioning. One individual dominant for the aloof endophenotype described social 360 interactions as "a means to an end". In contrast, the pedantic endophenotype was primarily 361  (Table 7). 366 Although multiplex families with ASD are genetically homogeneous, our phenotyping 387 analysis suggests possible bi-lineal inheritance of the BAP in both families. Therefore, multiple 388 risk alleles may contribute to ASD/BAP in later generations, consistent with recent genetic and 389 phenotyping evidence 47,48 . The importance of unique de novo genetic changes in both sporadic 390 (or 'simplex'), ASD 17 , and small multiplex ASD families 44 has become increasingly apparent. 391 However, with at least seven individuals with ASD and many more with the BAP in our families, 392 there is less likelihood of de novo changes contributing to each phenotype. It is much more likely 393 that there is a single genetic variant of major phenotypic effect in each family, with the possibility that there are additional de novo genetic changes in some individuals that contribute 395 to phenotypic severity. 396

Limitations 397
The intensive nature of the study meant that clinicians were not blinded to family relationships, 398 potentially leading to investigator bias. However, our diagnostic method of consensus between 399 experienced clinicians aligns with current best practice for ASD/BAP diagnosis and was 400 informed by quantitative and qualitative measures. We selected a relatively low threshold for 401 BAP classification, leading to the identification of many affected individuals. However, this 402 approach is justified in a family with a clear genetic liability for ASD and was validated by the 403 finding of consistent data-driven endophenotypes in the small families. Successful gene 404 identification in future work requires capture of all individuals who may carry the putative 405 variant, with the approach outlined here enabling more robust gene identification work. 406

Conclusion 407
Despite significant advances in unravelling the heterogeneity of ASD, in most cases, the 408 underlying genetic aetiology remains unknown in part due to difficulties identifying 409 endophenotypes and potential carriers. We used a rigorous phenotyping approach to characterise 410 the BAP in two large multiplex families with dominant inheritance of ASD and the BAP.  phenotypes and endophenotypes. All individuals with ³ 1 endophenotype had the BAP, with the 528 exception of two individuals from Family B (III-3 and IV-9) marked with an asterisk. These 529 individuals were clinically determined as unaffected (Family B III-3 and IV-9) but had above 530 threshold endophenotype scores based on ROC curves. Family members who were not 531 phenotyped are not shown to preserve the anonymity of these families. 532 Family A tend to have a single endophenotype, indicating a more mild BAP presentation, in 535 contrast with core family members who have multiple endophenotypes (obsessive most 536 frequent). In Family B, married-in and core family members tend to have more than one 537 endophenotype, with the aloof endophenotype most frequent. 538