*Article* **Pregnant Mothers' Medical Claims and Associated Risk of Their Children being Diagnosed with Autism Spectrum Disorder**

**Genevieve Grivas 1,2,3, Richard Frye 4,5 and Juergen Hahn 1,2,6,\***


**Abstract:** A retrospective analysis of administrative claims containing a diverse mixture of ages, ethnicities, and geographical regions across the United States was conducted in order to identify medical events that occur during pregnancy and are associated with autism spectrum disorder (ASD). The dataset used in this study is comprised of 123,824 pregnancies of which 1265 resulted in the child being diagnosed with ASD during the first five years of life. Logistic regression analysis revealed significant relationships between several maternal medical claims, made during her pregnancy and segmented by trimester, and the child's diagnosis of ASD. Having a biological sibling with ASD, maternal use of antidepressant medication and psychiatry services as well as non-pregnancy related claims such hospital visits, surgical procedures, and radiology exposure were related to an increased risk of ASD regardless of trimester. Urinary tract infections during the first trimester and preterm delivery during the second trimester were also related to an increased risk of ASD. Preventative and obstetrical care were associated with a decreased risk for ASD. A better understanding of the medical factors that increase the risk of having a child with ASD can lead to strategies to decrease risk or identify those children who require increased surveillance for the development of ASD to promote early diagnosis and intervention.

**Keywords:** autism spectrum disorder; medical claims; logistic regression analysis; retrospective analysis; associated risk

#### **1. Introduction**

Autism spectrum disorder (ASD) is an early onset neurodevelopmental disorder characterized by difficulties in social communication/interactions and by the presence of restricted and repetitive behaviors [1]. The prevalence of ASD has significantly increased over the last three decades [2] with the most recent estimate being 1 in every 54 eightyear-old children in the United States has been diagnosed with ASD with a 4.3 times higher occurrence in males than females [3]. The etiological understanding of ASD has also changed over the years, with current research suggesting a combination of genetic and environmental factors [4]. It is now generally acknowledged that investigation of environmental risk factors for ASD should not only be limited to the life of the child, but also include the prenatal and preconception period [5].

**Citation:** Grivas, G.; Frye, R.; Hahn, J. Pregnant Mothers' Medical Claims and Associated Risk of Their Children being Diagnosed with Autism Spectrum Disorder. *J. Pers. Med.* **2021**, *11*, 950. https:// doi.org/10.3390/jpm11100950

Academic Editor: Farah R. Zahir

Received: 12 June 2021 Accepted: 21 September 2021 Published: 24 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Numerous maternal body systems have been hypothesized to contribute to ASD including the gastrointestinal, immune, metabolic, and endocrine systems [6–10]. It is not surprising that research has extended these investigations to include the influence of maternal systems disorders. Maternal endocrine or hormonal disorders, such as polycystic ovary syndrome, show an increased risk of offspring developing ASD [11]. Maternal autoimmune disorders are notable ASD risk factors [12–15] with emphasis on hypothyroidism [16], psoriasis [17], and rheumatoid arthritis [18]. The presence of maternal infection during pregnancy significantly increases the risk of ASD in the offspring [19], with studies suggest this effect is specific for bacterial [20], viral [21], severe [22,23] or febrile [24,25] infections. In fact, the maternal immune activation (MIA) mouse model, a major animal model of ASD, induced ASD-like behavior in the offspring by activating the material immune system but also highlights the variability of this effect [26]. Though it may provide difficult to distinguish the confounding effects of the infection itself from the treatment for the infection, as some studies have shown antibiotic consumption is a risk factor, though there are discrepancies regarding the significance for antibiotics taken during the second or third trimester [21] or when taken for longer than 14 days [27].

It is well known that ASD is commonly linked to other cognitive or mental health disorders such as epilepsy, ADHD, and anxiety [28–31], and the role of brain development cannot be understated. Similarly, the influence of maternal mental disorders is crucial to understand and has been widely studied, with a heavy focus on maternal depression and antidepressant usage during pregnancy. Recent literature suggests the risk factor for ASD may be associated with prior antidepressant treatment or maternal psychological conditions rather than antidepressant consumption [5,32]. Another widely studied maternal pharmaceutical is prenatal vitamin supplementation, particularly folate (vitamin B9), which has been found to reduce the risk of offspring developing ASD by almost half [33–35].

Lastly, risk factors have also been found for delivery-related events such as preterm delivery [36,37] and cesarean delivery [38,39]. Though, these factors may be influenced by abnormal child development stemming from previously mentioned risk factors.

Given the large number of studies that have presented contradicting results, this work focuses on identifying ASD risk factors from a very large cohort of mothers in the United States. Specifically, this study is a retrospective analysis of maternal medical events that occurred during pregnancy and their effect on the risk of ASD in the child. These maternal events are reflected by diagnostic, procedural, and pharmaceutical claims from a private United States health plan.

#### **2. Materials and Methods**

#### *2.1. Mother and Child Cohort Identification*

This retrospective analysis used de-identified claims data with a family identifier and socioeconomic status information from the OptumLabs® Data Warehouse (OLDW), which included medical and pharmacy claims, laboratory results, and enrollment records for commercial and Medicare Advantage (MA) enrollees. The database contained longitudinal health information on enrollees and patients, representing a diverse mixture of ages, ethnicities and geographical regions across the United States [40]. As this study uses deidentified data, approval is exempt from the Institutional Review Board.

Children diagnosed with and without ASD, born between 1 January 2000 and 31 December 2010, were previously identified using the OLDW [10]. Vargason et al. (2019) used the children's diagnostic claims from their date of birth until five years of age to identify children diagnosed with ASD. This study identified the mother of these children through the use of family identifiers, policy holder relationship codes, and delivery claims within 10 days of the child's earliest enrollment date (assumed to be the child's date of birth) [41,42]. Diagnostic, pharmacy, and procedural claims (Table S1) were identified for each mother ten months prior to the birth of each child.

The processing steps used to identify children and their mothers, as well as the resulting number of women and children identified from the OLDW, are outlined in Table 1. Women were identified between the ages of 14 and 49 with commercial health coverage that included medical, pharmaceutical, and mental health coverage; this medical plan matched that of the children cohort. Step 4 outlined in Table 1 required that all children be labeled as "child" in relation to the policy holder. In Step 6, women were linked to children by having a delivery claim within 10 days of the child's first enrollment date. During this process, some children were found to be linked to multiple mothers, most likely due to different women under the same policy having birth claims at similar times; these mothers and children were excluded in Step 7. The final cohort sizes, shown in Step 8, identified various pairs of siblings (different children having the same mother) during the investigated time frame, resulting in a greater total number of children than women. This included siblings (single and multiple births) with different ASD outcomes, i.e., with and without an ASD diagnosis. For this study, each child was associated with the events that occurred during his or her gestational period, which were unique for siblings but identical for multiple births (i.e., twins and triplets). The final data set was comprised of 123,824 pregnancies identified using the OLDW data base; 1265 pregnancies resulted in children with ASD (ASD cohort), and 122,559 pregnancies resulted in children with no ASD diagnosis (population or POP cohort) during the investigated time frame. The ASD prevalence determined in this study agrees with the prevalence estimation performed by the Center for Disease Control during the same time period [43]. Within the dataset, 37,775 (30.5%) of all children had a sibling and of this 5616 (4.5%) were a part of multiple births.

**Table 1.** Data Attrition Steps for Identifying Mother Cohorts and Associated Children.


#### *2.2. Medical Claims Identification*

Medical claims were split up into the following three categories of variables: diagnostic claims identified by the International Classification of Diseases coding, version 9 (ICD-9), filled prescription claims determined by the National Drug Code (NDC) identifiers, and medical procedures claims denoted by their Current Procedural Terminology (CPT) coding. Total claims investigated included 478 ICD-9 diagnostic codes as variables, 10,810 NDC codes segmented into 132 pharmacy variables, and 3,808 CPT codes segmented into 122 procedural variables (Table S1). Pharmacy variables were created based on code descriptions embedded into the database. Procedural variables were based on CPT code descriptions. Table 2 shows the progression of variable selection from these categories, further outlined below.


**Table 2.** Progression of the Number of Variables for Diagnostic, Pharmacy, and Procedural Categories.

<sup>1</sup> Pearson correlation coefficient.

All variables had relatively similar percentages of claims between both cohorts (Table S1). A heuristic threshold was used to exclude variables due to uncommon claims, thereby eliminating small cell sizes and ensuring that all variables were present for both cohorts. If the number of claims fell below 2% for both cohorts combined (2476 of 123,824) or 2% for the ASD cohort only (25 of 1265) then the claim was excluded from further analysis. An example can be found in Figure S1, depicting the number of pregnancies that had a claim for the first 100 diagnostic variables (Table S1). Only variables with claims greater than 2% for the combined and ASD cohorts were further investigated (noted by the dashed lines in Figure S1, top and bottom, respectively). Thus, of the first 100 diagnostic variables, only 3 diagnostic variables were kept. Due to OptumLabs Data Policy, all cell values less than 11 are censored for de-identification purposes as noted by the *y*-axis starting at 11 for all figures.

Of the original 732 variables investigated, 156 remained after thresholding: 82 ICD-9 diagnostic codes, 27 pharmacy variables, and 45 procedural variables, see Table 2. These claims were then used to identify the mothers' gestational ages, as well as associated trimesters, using the same protocol presented in Li et al. [44]. Claims that fell outside of the identified gestational days were then removed and the remaining variables with claims above the threshold were kept. In addition, the following 6 maternal sociodemographic variables were included: race, home ownership, education level, income level, age, and a binary indicator for women who have had previous children with ASD. The latter variable, denoted as 'Previous ASD', refers to all subsequent children whose mother had a previous child diagnosed with ASD during the time frame of this study. This variable was included because women who have had previous children with ASD are at an increased risk of having another child with ASD [45].

For comparison, the sociodemographic data were analyzed based on individual women (referred to as "women cohorts") instead of pregnancies or resulting children (referred to as "pregnancy cohorts") to better represent the population. For this case, cohort separation is defined as women who have never had a child with ASD (population or POP) and women who have had one or more children with ASD (ASD). For ASD cohort women, age range refers to the age at which each woman had her first child with ASD, for POP cohort women it refers to the age at which she had her first child.

#### *2.3. Statistical Analysis*

A chi-square analysis was used to determine a statistically significant difference in proportions between the women's ASD and POP cohorts for the sociodemographic variables. For small cell sizes, a Fisher's exact test was used. For each age category, a Welch's t-test was used to determine statistically significant differences between mean age of the ASD and POP cohorts.

An F-test, with 5% significance level, was used to determine a statistically significant difference in variance between the ASD and POP cohort for total number of medical claims (diagnostic, pharmacy, and procedural combined), as well as each variable category individually. For categories that showed a statistically significant difference in variance, a Welch two-sample *t*-test was used to determine a statistically significant difference between the mean number of claims for the ASD and POP cohorts, at a 5% significance level. For equal variance, a standard t-test was used at a 5% significance level. Histograms were normalized in order to better compare the two cohorts.

Logistic regression was used to estimate the relationship between the presence of ASD in the child over the investigated time frame and the maternal medical claim (diagnostic, pharmacy, or procedural) made during pregnancy [46]. Pearson correlation analysis was conducted to identify claims that were highly correlated with any other claim, *r* > = 0.7, of which those of lower significance (denoted using *p*-values calculated from the unadjusted logistic regression) were removed from the analysis (see Table 2) [47,48]. Adjusted odds ratios (ORs) were used to quantify the effect of the medical claim and the associated risk of the child being diagnosed with ASD later, using a 95% confidence interval [49]. An initial logistic regression model showed the previous ASD variable was highly skewed towards the ASD cohort due to bias associated with multiple births; to correct for this, two adjusted logistic regression models were used, one with all variables and one with all variables except previous ASD. The statistically significant variables determined from these two models were then used for a third adjusted logistic regression model, which allowed for correction of multiple comparisons by reducing the number of variables included in the model as well as identifying false significance from the latter models. A schematic of the model development can be found in Figure 1. Due to computational restrictions, all logistic regression models were built using 10% of the POP cohort data, resulting in a ratio of approximately 10:1 POP to ASD, stratified based on sociodemographic variables. This analysis was then repeated for each trimester excluding sociodemographic variables since these are constant throughout the entire pregnancy. For brevity, all statistically significant findings as defined in this section and reported in this study are referred to as significant.

Some diagnostic variables were further specified based on their ICD-9 coding. ICD-9 codes are structured through a numeric system where a whole value code can also contain decimal values to elaborate on a diagnosis. For example, ICD-9 code 649 corresponds to the variable Other Conditions Complicating Pregnancy and can be broken down as follows: 649.0 tobacco use disorder complicating pregnancy, 649.1 obesity complicating pregnancy, 649.2 bariatric surgery complicating pregnancy, etc. Claims are made using either whole values or decimal points, at the discretion of the medical professional. All statistically significant final diagnostic variables are further evaluated by their whole and single digit ICD-9 coding in a separate logistic regression analysis, shown in Figure 1. Variables were only included in this analysis if they contain claims greater than 11, and due to this smaller threshold were evaluated with both unadjusted and adjusted ORs.

*J. Pers. Med.* **2021**, *11*, x FOR PEER REVIEW 6 of 24

**Figure 1.** Schematic of model development where N represents the number of variables included for each analysis on claims data obtained throughout the entire pregnancy. This procedure was also repeated for each trimester individually (not shown). **Figure 1.** Schematic of model development where N represents the number of variables included for each analysis on claims data obtained throughout the entire pregnancy. This procedure was also repeated for each trimester individually (not shown).

#### Some diagnostic variables were further specified based on their ICD-9 coding. ICD-9 **3. Results**

codes are structured through a numeric system where a whole value code can also contain decimal values to elaborate on a diagnosis. For example, ICD-9 code 649 corresponds to the variable Other Conditions Complicating Pregnancy and can be broken down as follows: 649.0 tobacco use disorder complicating pregnancy, 649.1 obesity complicating There were some key differences in sociodemographic data for the two women cohorts (Table 3). The percentages for race were for the most part comparable, with the majority being White, however, a significantly larger percentage of Asian pregnancies in the ASD

pregnancy, 649.2 bariatric surgery complicating pregnancy, etc. Claims are made using

cohort (10.4%) was found compared to the POP cohort (7.7%). In addition, the ASD cohort, as compared to the POP cohort, was significantly more educated (i.e., attained a degree higher than a Bachelor's degree), had higher income (i.e., income greater than USD 125,000) and was older (i.e., age 30 years or older), see Tables 3 and 4. Furthermore, the ASD cohort had a significantly smaller percentage of women having only a high school diploma, an income between USD 40,000–74,999, and being of age between 20–29 years old. The ASD cohort had a significantly higher percentage of previous ASD children, 245 (20.1%), compared to the 0 (0.0%) from the POP cohort. This was obviously expected since the women POP cohort is defined as women who have never had a child with ASD and thus will not have a previous ASD indicator.


**Table 3.** Sociodemographic Data on Women Cohorts.

<sup>1</sup> Number of women. <sup>2</sup> *p*-values are calculated using chi-squared analysis or Fisher's exact test for small cell values, significant *p*-values are shown in bold. <sup>3</sup> Values < 11 are censored for anonymity and *p*-values are calculated using Fisher's exact test.



<sup>1</sup> *p*-values are calculated using chi-squared analysis or Fisher's exact test for small cell values, significant *p*-values are shown in bold. <sup>2</sup> values < 11 are censored for anonymity and *p*-values are calculated using Fisher's exact test.

Correlation analysis for all variables during the entire pregnancy depicted six pairs of variables containing correlations of 0.7 or higher. These variables and their associated

unadjusted *p*-values are shown in Table S2. The variable with the larger *p*-value in each pairwise correlation was discarded from the adjusted logistic regression analysis; variables that remain in the analysis are bolded in Table S2. Three pairs of highly correlated variables all related to the same medical event of receiving a vaccination (variables: Vaccinations, Need for Prophylactic Vaccination against Viral Diseases, and Immunization Administration for Vaccinations). The remaining three pairs of variables were associated with a cardiovascular procedure, diabetic-related materials (such as test strips), and a thyroid disorder. pair-wise correlation was discarded from the adjusted logistic regression analysis; variables that remain in the analysis are bolded in Table S2. Three pairs of highly correlated variables all related to the same medical event of receiving a vaccination (variables: Vaccinations, Need for Prophylactic Vaccination against Viral Diseases, and Immunization Administration for Vaccinations). The remaining three pairs of variables were associated with a cardiovascular procedure, diabetic-related materials (such as test strips), and a thyroid disorder.

<sup>1</sup> *p*-values are calculated using chi-squared analysis or Fisher's exact test for small cell values, sig-

Correlation analysis for all variables during the entire pregnancy depicted six pairs of variables containing correlations of 0.7 or higher. These variables and their associated unadjusted *p*-values are shown in Table S2. The variable with the larger *p*-value in each

**ASD Women POP Women**

<sup>2</sup> values < 11 are censored for anonymity and *p*-values are

20–29 26.8 (27) 26.6 (27) 0.36 30–39 34.1 (34) 33.8 (34) **<0.001** 40–49 42.3 (42) 42.3 (42) 0.80

**Mean (Median)** *p***-Value <sup>1</sup>**

Normalized histograms for claims from both the ASD and POP cohort can be found in Figure 2 with associated descriptive statistics in Table 5. These data were generated by summing claims made throughout each entire pregnancy. Histograms for all medical claims (diagnostic, pharmacy, and procedural) and only diagnostic claims (Figure 2A,B, respectively) closely followed a normal distribution as shown by the similar mean, median, and mode values listed in Table 5. Pharmacy claims and procedural claims (Figure 2C,D) were right- and left-skewed, respectively, where most women had few (1–2) prescriptions and many (14–16) procedural claims. For all categories of variables (including total combination), the ASD cohort had a statistically significantly higher mean number of claims than the POP cohort. Normalized histograms for claims from both the ASD and POP cohort can be found in Figure 2 with associated descriptive statistics in Table 5. These data were generated by summing claims made throughout each entire pregnancy. Histograms for all medical claims (diagnostic, pharmacy, and procedural) and only diagnostic claims (Figure 2A,B, respectively) closely followed a normal distribution as shown by the similar mean, median, and mode values listed in Table 5. Pharmacy claims and procedural claims (Figure 2C,D) were right- and left-skewed, respectively, where most women had few (1–2) prescriptions and many (14–16) procedural claims. For all categories of variables (including total combination), the ASD cohort had a statistically significantly higher mean number of claims than the POP cohort.

*J. Pers. Med.* **2021**, *11*, x FOR PEER REVIEW 8 of 24

<20 <sup>2</sup>

**Table 4.** Age Statistics on Women Cohorts.

nificant *p*-values are shown in bold.

calculated using Fisher's exact test.

Age Range

**Figure 2.** Histogram data of normalized number of women belonging to the POP and ASD cohorts (shown in blue and red, respectively) with (**A**) any medical claim, (**B**) diagnostic claim, (**C**) pharmacy claim, or (**D**) procedural claim. Values associated with small cell sizes are not shown in order to be compliant with OptumLabs' de-identification policy. **Figure 2.** Histogram data of normalized number of women belonging to the POP and ASD cohorts (shown in blue and red, respectively) with (**A**) any medical claim, (**B**) diagnostic claim, (**C**) pharmacy claim, or (**D**) procedural claim. Values associated with small cell sizes are not shown in order to be compliant with OptumLabs' de-identification policy.


**Table 5.** Statistics on Number of Medical Claims during Entire Pregnancy.

<sup>1</sup> All *p*-values are calculated using random samples of 1000, significant *p*-values are shown in bold. <sup>2</sup> Standard deviation. <sup>3</sup> Confidence intervals. <sup>4</sup> Calculated using Welch two-sample *t*-test. <sup>5</sup> Calculated using two-sample *t*-test.

Normalized histograms for claims made in each trimester, along with descriptive statistics, can be found in Figures 3 and 4, as well as Table 6. The largest number of total medical claims was made in the third trimester with similar values for the first and second trimester. While the mean and median number of claims was greater in the first trimester compared to the second, the first trimester had a greater amount of zero-claims. In all trimesters, the majority of claims were made for procedures, followed by diagnostics and pharmacy. For all trimesters, the ASD cohort had a significantly higher mean number of diagnostics, pharmacy and procedural claims with one exception: diagnostic claims made in the first trimester showed no significant difference. Normalized histograms for claims made in each trimester, along with descriptive statistics, can be found in Figures 3 and 4, as well as Table 6. The largest number of total medical claims was made in the third trimester with similar values for the first and second trimester. While the mean and median number of claims was greater in the first trimester compared to the second, the first trimester had a greater amount of zero-claims. In all trimesters, the majority of claims were made for procedures, followed by diagnostics and pharmacy. For all trimesters, the ASD cohort had a significantly higher mean number of diagnostics, pharmacy and procedural claims with one exception: diagnostic claims made in the first trimester showed no significant difference.

*J. Pers. Med.* **2021**, *11*, x FOR PEER REVIEW 9 of 24

All Medical Claims

Diagnostic Claims

Pharmacy Claims

Procedural Claims

lated using two-sample *t*-test.

**Table 5.** Statistics on Number of Medical Claims during Entire Pregnancy.

**Cohort Max Mean Median Mode Stdev <sup>2</sup>**

POP 73 27.1 27 26 8.6 **<0.001 <0.001** <sup>4</sup> ASD 59 28.6 29 29 9.4 (1.10, 1.40) (1.13, 2.74)

POP 34 11.1 11 10 4.0 **<0.001 <0.001** <sup>4</sup> ASD 29 11.7 11 11 4.2 (1.09, 1.40) (0.34, 1.06)

POP 15 2.1 2 1 1.8 **<0.001 <0.001** <sup>4</sup> ASD 15 2.5 2 1 2.1 (1.20, 1.54) (0.29, 0.65)

POP 31 13.9 14 15 4.4 0.102 **<0.001** <sup>5</sup> ASD 26 14.4 15 16 4.7 (0.98, 1.26) (0.36, 1.18)

<sup>1</sup> All *p*-values are calculated using random samples of 1000, significant *p*-values are shown in bold.

Standard deviation. <sup>3</sup> Confidence intervals. <sup>4</sup> Calculated using Welch two-sample *t*-test. <sup>5</sup> Calcu-

**Difference in Distribution Variance Mean**

> *p***-Value <sup>1</sup> (95% CI 3)**

> > 2

*p***-Value <sup>1</sup> (95% CI 3)**

**Figure 3.** Histogram data of normalized number of women belonging to the POP and ASD cohorts (shown in blue and red, respectively) with any medical claim during the (**A**) first trimester, (**B**) second trimester, and (**C**) third trimester. Values associated with small cell sizes are not shown in order to be compliant with OptumLabs' de-identification policy. **Figure 3.** Histogram data of normalized number of women belonging to the POP and ASD cohorts (shown in blue and red, respectively) with any medical claim during the (**A**) first trimester, (**B**) second trimester, and (**C**) third trimester. Values associated with small cell sizes are not shown in order to be compliant with OptumLabs' de-identification policy.


**Table 6.** Statistics of Number of Medical Claims during Each Trimester.


**Table 6.** *Cont.*

<sup>1</sup> All *p*-values are calculated using random samples of 1000, significant *p*-values are shown in bold. <sup>2</sup> Standard deviation. <sup>3</sup> Confidence intervals. <sup>4</sup> Calculated using Welch two-sample *t*-test. <sup>5</sup> Calculated using two-sample *t*-test. *J. Pers. Med.* **2021**, *11*, x FOR PEER REVIEW 10 of 24

**Figure 4.** Histogram data of normalized number of women belonging to the POP and ASD cohorts (shown in blue and red, respectively) with any diagnostic (top row), pharmacy (middle row), and procedural (bottom row) claim during the first trimester (left column), second trimester (middle column), and third trimester (right column). Values associated with small cell sizes are not shown in order to be compliant with OptumLabs' de-identification policy. **Figure 4.** Histogram data of normalized number of women belonging to the POP and ASD cohorts (shown in blue and red, respectively) with any diagnostic (top row), pharmacy (middle row), and procedural (bottom row) claim during the first trimester (left column), second trimester (middle column), and third trimester (right column). Values associated with small cell sizes are not shown in order to be compliant with OptumLabs' de-identification policy.

The adjusted logistic regression models (with and without previous ASD) show a total of 20 significant variables (Table S3): 2 sociodemographic variables, 7 diagnostic variables, 3 pharmacy variables, and 8 procedural variables. When modeled by themselves, only 17 of these significant variables retained their significance (Table 7, full model results

can be found in Table S3). A majority of the variables (13 of 17) were associated with an increased risk of having a child diagnosed with ASD. While both sociodemographic variables showed an increased risk, having a child previously diagnosed with ASD was associated with the largest increased risk of all variables, OR 16.09 (8.27, 32.12). Three diagnostic, all three pharmacy, and five procedural variables were also associated with an increased risk. The remaining four variables (two diagnostic and two procedural) were associated with a significantly decreased risk. Results for the subcode logistic regression analysis on diagnostic variables from the entire pregnancy can be found in Table S4.

**Table 7.** Logistic Regression Analysis Results of Variables Identified as Highly Significant during the Entire Pregnancy.


<sup>1</sup> *p*-value and odds ratios are calculated using adjusted logistic regression analysis. <sup>2</sup> Confidence intervals.

Results from the adjusted logistic regression analyses for all trimester variables can be found in Table S3. The final significant variables identified for each trimester can be found in Table 8; some of these variables differed from the entire pregnancy analysis due to the different number of claims that occur in each trimester. A larger number of significant variables occurred for the first and third trimesters (15) compared to the second trimester (13). While all trimesters had a majority of variables associated with increased risk, the third trimester had the most variables associated with increased risk (11) while the first trimester had the most variables that were associated with decreased risk (5). Multiple Gestation, Antidepressants, and Procedure Services Psychiatry variables were consistently associated with increased risk for all three trimesters.

The variable Other Conditions Complicating Pregnancy was associated with a significantly increased risk during the third trimester, however further analysis showed that no subcode was significant for this occurrence and thus this study is not able to determine what event influenced this diagnosis (see Table S4). Similarly, Services Office or Other Outpatient was associated with a significantly increased risk during the second trimester. This variable corresponds to a new patient visit; however, this study is unable to determine if this visit was related to pregnancy or another maternal health-related event.


**Table 8.** Logistic Regression Analysis Results of Variables Identified as Highly Significant during each Trimester.


**Table 8.** *Cont.*

<sup>1</sup> *p*-value and odds ratios are calculated using adjusted logistic regression analysis. <sup>2</sup> Confidence intervals.

#### **4. Discussion**

The majority of this study's significant findings were associated with an increased risk of having a child with ASD. Many of these correspond to a single variable in the model such as having a previous child with ASD (Table 7, Previous ASD), first pregnancy over the age of 35 (Table 7, Other Indications Related to Labor), current cesarean delivery (Tables 7 and 8, Other Complications of Labor), prescription for antidepressants (Tables 7 and 8), psychiatric services (Tables 7 and 8), pre-existing diabetes (Table 7, Durable Medical Equipment Diabetic), urinary tract infection during the first trimester (Table 8, Other Disorders of Urethra and Urinary Tract), and premature pregnancy (Table 8, variables Normal Pregnancy and Surgical Procedures Maternity Care and Delivery). Some of these variables were grouped in order to identify a common theme associated with ASD such as variables corresponding to standard obstetrical procedures, non-pregnancy related procedures, or maternal immune dysfunction and allergens. Lastly, a few variables and their associations with ASD disagreed with current literature, such as being of Asian race, having a prescription for pre-natal vitamins, and having multiple gestations. These findings, and others, have all been further discussed below and a summary can be found in Table 9, listed as they appear in this section.

**Table 9.** Summary of Study Findings.


The data cohorts identified in this study found the highest percentage of ASD pregnancies among White children, followed by Asian, Hispanic, and Black children. This trend agrees with that reported in a CDC surveillance completed within the same time period as this study [43] except for those of Asian race, which was found to vary widely depending on location and where our study shows a significantly larger proportion in the ASD women's cohort (*p*-value < 0.001, Table 3). Asian race was associated with a 40% increased risk of having a child diagnosed with ASD (Table 7). The most recent CDC surveillance summary showed similar prevalence between Asian and White children within the United States [3]. Therefore, it is most likely that the increase in risk associated with Asian race found in this study was a result of sample bias as noted by the significantly greater percentage of Asian ASD women identified in Table 3.

It is well known that the recurrence risk for ASD in families is much greater than the risk for the general population, therefore, women who have a child with ASD are considered high-risk for having subsequent children diagnosed with ASD [45]. Our study confirmed and clarified this finding, suggesting a 16-fold increased risk associated with having another child with ASD when a previous child was diagnosed with ASD (Table 7). There was also evidence to suggest that this elevated risk increased with each additional child diagnosed with ASD [50,51]. However, having a child diagnosed with ASD may influence the parental decision of having subsequent children, known as reproductive stoppage, which is a confounding factor [51–53].

Sociodemographic trends for ASD noted a higher prevalence of the disorder among higher levels of education and income [54,55], similar to what was found in this study (see Table 3). While advanced maternal age was not reflected in the sociodemographic variables of the logistic regression analysis, it was reflected in the diagnostic subcode Elderly Primigravida (ICD-9 659.5, Table S4) of the variable Other Indications Related to Labor, ICD-9 659, which demonstrated an overall adjusted OR 1.28 (1.13, 1.45; Table 7). This subcode corresponds to women with their first pregnancy over the age of 35. Advanced maternal age has been associated with an increased risk of ASD, with studies showing the association for the highest age category [56], age greater than 35 [5,57], or age greater than 40 [58].

Our analysis showed that cesarean delivery (ICD-9 669.7, Table S4) was the significant contributing factor to the diagnostic variable Other Complications of Labor, ICD-9 669, OR 1.27 (1.09, 1.47), Table 7, overall as well as during the third trimester, OR 1.22 (1.05, 1.42), Table S4. Previous studies reported inconclusive results for associating ASD with cesarean delivery. Some studies showed a weak or no association [15,59,60] while others showed a significantly increased risk [39,61] though this may be correlated with the risk factors associated with the cause for cesarean delivery instead of the delivery itself [62,63]. Some studies find cesarean delivery with general anesthesia significantly increased the risk of ASD compared to cesarean delivery with regional anesthesia or other indications [38,64]. While some women elect to have a cesarean delivery, more commonly they occur due to complica-

tions that arise during pregnancy or delivery, which vary depending on maternal age [65]. Cesarean deliveries change the risk profiles for both the mother and newborn [66] and may directly affect the environment of newborns and possibly even their microbiome [67,68]. There are even long-term health risks associated with the delivery following a cesarean [69]. However, our study found that having a previous cesarean delivery (before the current pregnancy), code ICD-9 654.2, was associated with a decreased risk of ASD during the third trimester (Table S4), with an overall OR 0.86 (0.75, 0.99) denoted by Abnormality of Pelvis, ICD-9 654, Table 8. While currently there is little research on the effect of previous cesarean delivery or even vaginal birth after cesarean (VBAC) and having a child diagnosed with ASD, women with previous cesarean deliveries are more carefully managed, especially during labor [70,71]. It is possible that these extra precautions are a confounding factor as this finding contradicts other literature that associate prior cesarean delivery with an increased risk of adverse reproductive outcomes for subsequent pregnancies [72–75].

A prescription for antidepressants was significantly associated with an increased risk for ASD, with an overall risk greater than 40% and increased per trimester. Maternal antidepressant usage is a highly researched area as a potential risk factor for ASD. Many studies have shown that antidepressants, including the use of selective serotonin reuptake inhibitors (SSRIs), are associated with a significantly increased risk of the child developing ASD [5,76,77]. Contrary to this, other reviews find conflicting results [78,79] or no significant association [80]. However, recent investigations also examined the underlying mental illness, as many studies have shown that adjusting for depression attenuates the significant association of antidepressants while the association of mental illness remains strong [81–86]. Our analysis found a greater than 40% significantly increased risk associated with psychiatric services (Table 7), that increased to greater than 60% during the first trimester and decreased with each trimester (Table 8). We did not find a sufficient number of claims for Major Depression Disorder (ICD-9 311) to include in this analysis. As antidepressant medications are widely used for multiple psychiatric conditions, including anxiety, bipolar disorder and others, the data may suggest that an increased risk of ASD may be associated with a wider array of psychiatric conditions in the family, as have been documented in other studies [87]. However, it is clear that the risk associated with maternal antidepressant usage is heavily influenced based on study design [88], and that it is of great importance to acknowledge the underlying confounding effects of mental health disorders.

Our study did not have a sufficient number of claims to include the diagnosis of diabetes (ICD-9 250) and did not find a diagnosis for gestational diabetes (ICD-9 648) significant, but found that a prescription for Diabetic Durable Medical Equipment (DME, such as insulin needles) was associated with a significantly increased risk of ASD, OR 1.27 (1.00, 1.59), Table 7. This finding suggests a significant association with diabetes that may have been diagnosed before the time of our study. The risk of maternal diabetes associated with ASD remains unclear, with reviews suggesting a strong [89,90], moderate [15], or no [91] relationship. While some studies combine the effects of any type of diabetes, others suggest that familiar type 1 diabetes [13], gestational diabetes [92,93], or only diabetes in conjunction with obesity [94] are associated with an increased risk. Though, individuals who are predisposed to diabetes may act as a confounding factor. Extensive reviews have been conducted on how diabetes may relate to biological mechanisms involved in the development of ASD, specifically through the oxidative stress pathways [95].

Many studies show a significantly decreased risk associated with pre-natal vitamins [33,96], such as folic-acid supplements [34,97] or fatty acids [98]. However, our study showed an increased risk, overall OR 1.18 (1.04, 1.33), shown in Table 7, as well as a similar increased risk in the second and third trimesters, Table 8. This finding may not truly reflect the relationship between pre-natal vitamins and ASD as most vitamins are provided over the counter and therefore do not appear within an insurance claim and are not represented in this study. It is also possible that there are unknown reasons associated with receiving a prescription for nutritional vitamins that may be acting as a confounding factor such as economic concerns or medical conditions. For example, individuals with a previous

child with ASD may specifically request a prescription for vitamins. The fact that a large well-done study demonstrated that higher folate supplementation was associated with a decreased risk of ASD would suggest other confounding factors are possible [35]. In addition, our study did not quantify the type of vitamins prescribed, (i.e., vitamin D, vitamin B, multivitamin, dietary supplements, etc.). For example, prescribing folic acid, an oxidized folate that is poorly metabolized and poorly transported across the placenta in some women, as opposed to a reduced folate which has much higher bioavailability, can result in high levels of unmetabolized folate in the blood in those with poor folate metabolism [99]. This can lead some to make the wrong conclusions that too much folate supplementation during pregnancy can be associated with an increased risk for ASD [100], whereas the problem lies with providing the correct type of bioavailable folate [101].

One finding that did show a significantly decreased risk of having a child with ASD, was a procedure claim for receiving a vaccination, OR 0.58 (0.46, 0.71), Table 7. This was also significant in the third trimester, OR 0.68 (0.47, 0.96), as well as the second trimester (diagnosis claim Need for Prophylactic Vaccination against Certain Viral Diseases), OR 0.60 (0.41, 0.83), Table 8. This study was unable to determine the type of vaccination, however. There is a limitation of relating this paper's findings to influenza vaccination as many instances of this vaccination can occur within the community, outside of a doctor's office, and thus would not appear within the claim's data. Vaccinations have been recommended during pregnancy in order to prevent infections [102].

Other standard obstetrical procedures showed a decreased risk, specifically Uterine Size and Date Discrepancy (Table S4, ICD-9 649.6), Special Screening of Malignant Neoplasms (of Cervix) overall and during the first trimester (Tables 7 and 8, respectively), Cervical Incompetence (ICD-9 654.5 Table S4) during the first trimester, Antenatal Screening and Pregnancy Evaluation (ICD-9 V28.5-6 and V72.4, respectively, see Table S4) during the first trimester, Diagnostic Ultrasound Procedures during the second trimester (Table 8), physical therapy (Evaluations Physical Medicine and Rehabilitation) overall and during the second trimester (Tables 7 and 8, respectively), and Procedures Other Pathology and Laboratory during the first trimester (Table 8). It is well documented that obstetric complications increase the risk of having a child with ASD [103–105] and it is clinically recommended that women who are at high risk should be closely monitored throughout their pregnancy [102]. Thus, these findings suggest that women with earlier and more aggressive obstetrical care have a decreased risk of ASD.

Various hospital procedures showed an increased risk such as an in-hospital consultation (Services Consultation) overall and during the first and third trimester (Tables 7 and 9, respectively), surgical procedures that may require anesthesia (Anesthesia Procedures Lower Abdomen shown in Table 7, Surgical Procedures Female Genital System during the first trimester and Surgical Procedures Nervous System during the third trimester shown in Table 8), ventilation or breathing tests (Procedures Pulmonary, Table 7), and Procedures Diagnostic Radiology (Table 7). While the following claims may not have required hospitalization they also show an increased risk: Special Screening for Blood Disorders (Table 7), Special Screening for Endocrine Nutritional Metabolic and Immunity Disorders during the first trimester, Abdominal Pain (Other Symptoms Involving Abdomen and Pelvis) during the first trimester, Acquired Hypothyroidism during the third trimester, Consultations Clinical Pathology during the first trimester, and Services Office or Other Outpatient during the second trimester (all of which can be found in Table 8). These findings suggest that claims not relating to the pregnancy nor delivery are associated with an increased risk of ASD regardless of trimester.

Our study showed an increased risk of UTI (Other Disorder of Urethra and Urinary Tract) but only during the first trimester, OR 1.49 (1.19, 1.86), shown in Table 8. UTIs have been shown to be common during pregnancy but have inconclusive associations with ASD [19–22,24,25,60,106]. Urinalysis procedures showed an increased ASD risk during the third trimester, OR 1.18 (1.04, 1.33), Table 8. However, urinalysis procedures refer to any urine examination and is not only associated with diagnosing UTI but may include other

tests such as testing for pre-eclampsia. Pre-eclampsia has been shown to have an increased risk for ASD [5,107], but a diagnosis (ICD-9 642) was not found to be significant in this study. Maternal Antepartum Hemorrhage and Placenta Previa showed a decreased risk of ASD during the third trimester, OR 0.71 (0.55, 0.92), Table 8. Antepartum hemorrhage has been shown to be associated with intellectual disability but not ASD [60], while placenta previa is associated with a decreased incidence of pre-eclampsia [108].

Premature (pre-term) children, identified by the diagnostic variable Normal Pregnancy (ICD-9 V22) and the procedural variable Surgical Procedures Maternity Care and Delivery during the second trimester (Table 8), were associated with a 30% increased risk, consistent with Talmi et al. as well as other previous studies that found preterm to be a significant factor associated with ASD [37,109,110]. There is a higher prevalence of ASD among children born pre-term [111–116]. Though the risk has been shown to change depending on preterm gestational week cutoff [117]. Though, our study did not find any association with a diagnosis for Early or Threatened labor (ICD-9 642), a common diagnosis made at the discretion of medical personnel.

Other maternal prescriptions resulted in an increased risk of ASD, specifically Antiinflammatory Glucocorticoids during the first trimester, OR 1.44 (1.08, 1.89), and Respiratory Antihistamines during the second and third trimester, OR 1.39 (1.03, 1.83) and 1.62 (1.20, 2.15), respectively, shown in Table 8. These prescriptions were common treatments for maternal immune dysfunctions and allergens, respectively. However, this study is limited to antihistamine prescriptions that were prescribed and cannot take into account any over-the-counter remedies. Many reviews have shown that maternal inflammatory events in conjunction with maternal immune activation or autoimmune diseases are associated with ASD [8,12,102,118]. Specifically, anti-inflammatory glucocorticoids are a common treatment for psoriasis, which was found to be significantly associated with ASD in one case-control study [17], although a diagnosis for psoriasis (ICD-9 696) was not included in this study due to lack of claims.

Multiple Gestation was found to have a significant increased risk on ASD in all three trimesters (Table 8). However, the study was unable to determine if the twins identified are monozygotic or dizygotic. Previous studies have shown that multiple births have not been associated with ASD, instead the association can be explained by the higher rate of ASD in monozygotic twins compared to their siblings [119,120].

This study does have limitations; the diagnostic codes inputted into each claim were made at the discretion of the medical personnel and were subject to potential bias and all pharmacy claims represented prescription being filled. All variables investigated originate from maternal claims received through insurance and thus does not provide a full representation of all environmental factors that occur outside of insurance claims such as over-the-counter medicines or supplements. Paternal claims were not able to be identified and therefore their influence is unknown. The claims investigated occurred during each woman's pregnancy and thus do not consider pre-existing conditions that may have been diagnosed or treated prior. Lastly, being limited to claims during pregnancy also ignores the possibility of attenuating these factors through consistent proper treatment during or even after pregnancy.

#### **5. Conclusions**

Some environmental effects that influence the development of ASD might be identifiable as early as the gestational period. This study identified maternal medical claims made throughout women's pregnancies and determined risk factors associated with having a child diagnosed with ASD. Identifying these factors that either increase or decrease risk is essential especially for women who are at high risk of having children diagnosed with ASD. It is also beneficially for the child by allowing for early screening, leading to earlier diagnosis and the start of interventions. Early intervention is crucial in children with ASD and has been shown to save costs in the long-term [121,122]. Future research would benefit from exploring medical claims made throughout an individual's lifetime to truly evaluate

health trends and their influence on the risk of having a child with ASD. It would also be of interest to investigate paternal medical claims to emphasize the genetic influence in the development of ASD.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/jpm11100950/s1, Figure S1: Example of Variable Thresholding, Table S1: All Investigated Variables and Claims, Table S2: Highly Correlated Variables, Table S3: Logistic Regression Analyses Full Models, Table S4: Logistic Regression Analyses Subcodes.

**Author Contributions:** Conceptualization, J.H.; methodology, G.G.; software, G.G.; validation, G.G.; formal analysis, G.G.; investigation, G.G.; resources, J.H.; data curation, G.G.; writing—original draft preparation, G.G.; writing—eview and editing, G.G., R.F. and J.H.; visualization, G.G.; supervision, J.H.; project administration, J.H.; funding acquisition, J.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors gratefully acknowledge partial financial support from the National Institute of Health (Grant R01AI110642). Support for this research was also received from the Rensselaer Institute for Data Exploration and Applications. The authors express their gratitude to John Rodakis of the N of One: Autism Research Foundation for financially supporting the interactions with OptumLabs.

**Institutional Review Board Statement:** Ethical review and approval were waived for this study, due to use of de-identified data.

**Informed Consent Statement:** Patient consent was waived due to the use of de-identified data.

**Data Availability Statement:** Restrictions apply to the availability of these data. OptumLabs carefully manages access to its data to ensure appropriate use in accordance with its mission and values, policies and procedures, and prevention of re-identification. Users may, therefore, only access OLDW under an agreement with OptumLabs, and compliance with this policy.

**Acknowledgments:** The authors express their gratitude to the staff at OptumLabs for supporting the study design.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


### *Case Report* **Genomics as a Clinical Decision Support Tool: Successful Proof of Concept for Improved ASD Outcomes**

**Heather Way <sup>1</sup> , Grant Williams <sup>2</sup> , Sharon Hausman-Cohen 2,\* and Jordan Reeder <sup>2</sup>**


**Abstract:** Considerable evidence is emerging that Autism Spectrum Disorder (ASD) is most often triggered by a range of different genetic variants that interact with environmental factors such as exposures to toxicants and changes to the food supply. Up to 80% of genetic variations that contribute to ASD found to date are neither extremely rare nor classified as pathogenic. Rather, they are less common single nucleotide polymorphisms (SNPs), found in 1–15% or more of the population, that by themselves are not disease-causing. These genomic variants contribute to ASD by interacting with each other, along with nutritional and environmental factors. Examples of pathways affected or triggered include those related to brain inflammation, mitochondrial dysfunction, neuronal connectivity, synapse formation, impaired detoxification, methylation, and neurotransmitter-related effects. This article presents information on four case study patients that are part of a larger ongoing pilot study. A genomic clinical decision support (CDS) tool that specifically focuses on variants and pathways that have been associated with neurodevelopmental disorders was used in this pilot study to help develop a targeted, personalized prevention and intervention strategy for each child. In addition to an individual's genetic makeup, each patient's personal history, diet, and environmental factors were considered. The CDS tool also looked at genomic SNPs associated with secondary comorbid ASD conditions including attention deficit hyperactivity disorder (ADHD), obsessivecompulsive disorder (OCD), anxiety, and pediatric autoimmune neuropsychiatric disorder associated with streptococcal infections/pediatric acute-onset neuropsychiatric syndrome (PANDAS/PANS). The interpreted genomics tool helped the treating clinician identify and develop personalized, genomically targeted treatment plans. Utilization of this treatment approach was associated with significant improvements in socialization and verbal skills, academic milestones and intelligence quotient (IQ), and overall increased ability to function in these children, as measured by autism treatment evaluation checklist (ATEC) scores and parent interviews.

**Keywords:** autism spectrum disorder (ASD); genomics; personalized treatment strategy; single nucleotide polymorphisms; clinical decision support tool; ADHD; PANDAS; OCD; anxiety

### **1. Introduction**

Autism spectrum disorders (ASDs) are a group of neurodevelopmental syndromes characterized by deficits in social interaction and communication, as well as repetitive behaviors and restricted interests. ASD rates have increased tremendously over the last few decades from 3 per 1000 children in 1996 to 19 per 1000 children in 2016 [1,2]. While there are forms of ASD caused by pathogenic (disease-causing) genetic mutations, this represents only a small portion of individuals with ASD.

Considerable evidence is emerging that ASD is triggered by the interaction between a variety of single nucleotide polymorphisms (SNPs) and environmental factors such as toxicant exposures, changes to food supplies, and the gut microbiome [3]. In a 2020 study, only 19.7% of individuals with ASD were found to have rare pathogenic variants or copy number variants contributing to or causal of their ASD diagnosis [4]. This indicates that

**Citation:** Way, H.; Williams, G.; Hausman-Cohen, S.; Reeder, J. Genomics as a Clinical Decision Support Tool: Successful Proof of Concept for Improved ASD Outcomes. *J. Pers. Med.* **2021**, *11*, 596. https://doi.org/10.3390/jpm11070596

Academic Editor: Richard E. Frye

Received: 21 May 2021 Accepted: 22 June 2021 Published: 24 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

ASD is much closer to what is seen with other chronic illnesses, where a multitude of less common SNPs (found in 1–15% of the population) are likely the main contributors. Additional contributing SNPs may have a much higher population frequency. While individually these SNPs are not disease-causing, they can contribute in an additive manner to the manifestations commonly associated with ASD.

The use of genomics for clinical decision support is a novel approach to medicine that has become feasible only within the last few years. This is in part due to improvements in genetic testing technology as well as advances in the literature regarding the mechanisms of how supplements, nutrients, and other interventions interact with the genome and molecular pathways. This article illustrates how a genomically targeted and personalized medicine approach was successfully used at the Australian Centre for Genomic Analysis (TACGA). While a number of ASD treatment centers across the world incorporate a functional medicine and integrative approach, this is the first time to our knowledge that a specialized neurodevelopmental genomic clinical decision support (CDS) tool has been used systematically to achieve marked improvements in ASD related symptoms. Additionally, as a CDS, the SNPs that were presented and prioritized were actionable. For example, there have been many reports demonstrating the association between elevated tumor necrosis factor alpha (TNFa) and ASD [5], but few studies were identified that connect supplements known to cross the blood–brain barrier and lower TNFa-based inflammation to their usage in response to genetic predisposition to higher TNFa levels.

#### **2. Materials and Methods**

The four patients presented in this study are a subset of an ongoing pilot study composed of approximately 100 patients and are meant to be illustrative of and give insight into the process used for improving outcomes. Each of the four individuals received treatment for ASD at The Australian Centre for Genomic Analysis (TACGA), which began using genomics in a simplistic manner beginning in 2012. Children who came to TACGA prior to 2018 were evaluated using a basic 54 SNP "health and well-being" panel that included information relating to inflammation, oxidative stress, vitamin D, detoxification and methylation. In 2018, the original version of a neurodevelopmental report (designed to help with non-syndromic ASD) from IntellxxDNA™ (IXXD)—a clinical decision support tool—became available.

IXXD is a more specific CDS tool that offers various versions of its report, including one that focuses on pediatric developmental issues such as ASD, obsessive–compulsive disorder (OCD), pediatric autoimmune neuropsychiatric disorder associated with streptococcal infections (PANDAS), attention deficit hyperactivity disorder (ADHD), and anxiety. This tool was used to analyze DNA specimens from subjects, which consisted of both new and existing TACGA patients. For existing patients, IXXD was implemented as add-on therapy in order to obtain additional improvements in neurodevelopmental outcomes. DNA was collected via buccal cells and analyzed at Rutgers University Cell and DNA Repository (RUCDR) using a customized version of the Affymetrix precision medicine microarray.

Genomic information was presented to the clinician as formatted by IXXD, which was a curated collection of the genomic research. Discussions on gene and SNP function, as well as genomically targeted potential intervention strategies (including nutrients, supplements, and lifestyle modifications) were presented to the ordering provider. In keeping with being a clinical decision support tool, all information was referenced. IXXD reported a particular supplement, food, or nutrient as a potential intervention if it (1) mechanistically addressed both the gene function and SNP impact on the given pathway, and (2) had evidence for improvement of ASD and/or various neurodevelopmental parameters. IXXD nutritional handouts were also incorporated into treatment plans, so that genomics could be addressed with nutrition when possible. A variety of potential intervention options were presented by the CDS, but all treatment decisions were made by the ordering clinician.

The degree of improvements in ASD outcomes were gauged using autism treatment evaluation checklist (ATEC) scores, intelligence quotient (IQ) scores, behavioral improve-

ments, and other parental reporting. In addition to genetic profiling and behavioral observations, TACGA protocol also called for pre- and post-treatment measurement of vitamins, homocysteine, interleukins, and various additional blood markers.

Due to this being a pilot study, there was no specified control group. However, there were individuals evaluated who had previously been optimized with standard TACGA care, who were then given the opportunity to have their genomics evaluated to further improve outcomes. Supplements, nutrients, and dietary modifications were all items that had previously been discussed in the published medical literature and were available over the counter. Thus, consent in this pilot study was obtained via parental discussion.

A detailed discussion on how the CDS tool works is necessary due to this being the first presentation of the IntellxxDNA platform in the ASD literature. For illustration purposes, the Src homology 3 (SH3) and multiple ankyrin repeat domains 3 (SHANK3) variant discussed in the first case study below will be used as an example. Each sentence in quotations below and any accompanying information is linked to references in the live tool. To make the IXXD tool a useful resource for clinicians, genomic reports begin by discussing the gene function and how the variant of interest impacts protein functionality. For example, it would be relayed that SHANK3 "is part of the molecular scaffolding or platform where synapses, especially glutamate receptors of post synaptic nerves, are assembled." IXXD also provides clinicians with references supporting that the SNP conveys a decrease in SHANK3 expression, along with information linking the associated disorganization of synapses, pervasive developmental disorders, and ASD-like symptoms. Extensive discussions on how each of the SNPs can be modulated are also included. In this example, it is known that SHANK3 is destabilized and broken down by the extracellular signal-regulated kinase 2 (ERK2) protein. Therefore, inhibiting ERK2 with supplements such as curcumin, resveratrol, or a high butyrate diet are presented as potential interventions as they can help raise SHANK3 levels. Additionally, IXXD relays the cofactors that are needed in the molecular pathway, as they can also be modified to improve function. An excerpt from the referenced discussion of SHANK3 modulation in the CDS tool is as follows: "The protein encoded by SHANK3 is regulated by zinc, and zinc deficiency depletes synaptic pools of SHANK3. Melatonin increases SHANK3 protein concentrations. Blue light protection can be beneficial for maintaining proper melatonin levels. ERK2 inhibitors will increase SHANK3 protein indirectly by decreasing the degradation of SHANK3. Butyrate is an ERK2 inhibitor. A ketotic diet is one way to increase beta-hydroxybutyrate levels, but high butyrate foods (see patient dietary handout list), in addition to butyrate supplements, can also be beneficial. Resveratrol is also an ERK2 inhibitor and has data in children with ASD. Physical activity has also been shown to increase SHANK3 protein concentrations in the thalamus and cortex." This detailed information is given for every SNP in the report and differs in complexity, depending on the nature of the SNP itself. Discussions range from complicated SNPs such as SHANK3 and NAD(P)H quinone dehydrogenase 1 (NQO1), down to simple mechanisms such as the nutritional factor phosphatidylethanolamine N-methyltransferase (PEMT) for the choline pathway.

#### **3. Results**

#### *3.1. CJM Case Study*

3.1.1. Medical History and Background

The following is a case study of a male patient who initially made significant gains when following TACGA protocol, but later plateaued. His DNA was reevaluated using the IXXD CDS that targeted specific neurodevelopmental and neurobehavioral pathways.

This case study patient, referred to as CJM to protect his identity, was diagnosed ASD level 3 (highest level, requiring substantial support) at age three and intellectually impaired with an IQ of 54 at age seven. On a gluten free/dairy free diet since age five. First presented to the clinic at age 12 with an ATEC score of 117 (neurotypical ATEC score is about 10 or less). He was classified as non-verbal with some occasional rudimentary language in the form of two or three word strings when it suited him, and was unable

to follow multiple instructions. Behavioral issues included self-harm, aggressiveness to peers and family, running away, and bed wetting. Also noted was no desire to socialize, lack of attention (2–3 min), lethargy and very low mental energy (less than 2 min), and significant sensory defensiveness around noise, clothing and stimming. Additionally, he displayed hyperactivity at times, a lack of eye contact, and chronic constipation (permanently on laxatives).

In 2015, he was screened using the initial TACGA protocol. He had SNPs relating to interleukin 1 (IL1a and IL1b), vitamin D receptors (VDR) and detoxification pathways. Alongside dietary changes and a gut healing protocol, the interventions were as follows: fish oil (2 g eicosapentaenoic acid + docosahexaenoic acid), broccoli sprouts (releasing 16 mg sulforaphane), vitamin D (3000 IU), anti-inflammatory probiotics, fermented foods in diet, zinc, and D-ribose-L-cysteine (glutathione precursor).

Over the next 12 months, his family reported considerable improvements in both receptive and expressive language. Behavior significantly improved, stimming reduced, and constipation had resolved. His ATEC had reduced to 71, but he subsequently plateaued. Residual symptoms included lack of attention and focus (15 min), lethargy/became mentally tired very quickly (10–15 min), impaired cognitive abilities, bed wetting (still nightly), and some sensory issues (mostly when tired). Language skills were improved, and he was able to talk in phrases and understand most general words, but he struggled to have meaningful conversations. IXXD's neurodevelopmental report became available for alpha testing in November 2018, and his family decided to pursue this option shortly after in an attempt to break through the plateau.

#### 3.1.2. Genomic CDS Results and Interpretation

Various genomic pathways, including SNPs reported in the literature to contribute to neurodevelopment and cognitive dysfunction, were discovered and appropriately addressed. CJM was found to be homozygous for a relatively rare variant (c.1304 + 48C > T) in the SHANK3 gene, which is found in less than 4% of the population and is highly associated with increased ASD risk. Deletions and variations within the SHANK3 pathway have been associated with ASD [6], and this particular SNP has been associated with an odds ratio (OR) of 5.5 for ASD and an OR of 12.6 for pervasive developmental disorder [7]. This SNP appears to lead to decreased protein activity. SHANK3 variants (or deletions) causing decreased activity are associated with less ability to form glutamatergic nerve connections during brain development and throughout childhood [7]. Furthermore, SHANK3 contributes to delayed or absent speech, lower muscle tone, and altered social interactions [8]. Interventions targeted towards decreasing the breakdown of this scaffolding protein, as discussed in the genomic CDS, were introduced. Some of these interventions included increasing cofactors, such as zinc, that stabilized the SHANK3 protein [9]. Other interventions related to inhibiting ERK2, which is responsible for breaking down SHANK3, included melatonin [10], resveratrol [11], and using blue light filtering glasses (to block decreases in melatonin levels) [12].

This patient also had multiple SNPs that relate to memory and cognition, including mitochondrial membrane issues that predispose him to more oxidative stress and mitochondrial dysfunction. Additionally, CJM had SNPs that disrupt his natural ability to synthesize phosphatidylcholine, which is an essential nutrient for the synthesis of acetylcholine that is also involved in pathways relating to phospholipid membrane production [13]. He was started on citicoline for this PEMT variant, alongside and a variety of supplements for mitochondrial support that included a combined formulation of ubiquinol (UBQH) + pyrroloquinoline quinone (PQQ) and acetyl-L-carnitine. Additionally, targeted anti-inflammatory interventions were addressed with supplements, and dietary changes to support mitochondrial function were instituted (more coconut oil and mildly ketogenic).

ADHD is a frequent comorbidity to ASD [14]. CJM was homozygous for an ADHDassociated SNP found in approximately 7% of the population. This particular SNP can lead to higher glutamate and dopamine, and lower gamma aminobutyric acid (GABA) via serotonin dysregulation [15]. Variants are known to contribute to inattentive ADHD traits [16], reduced impulse control and increased impulsivity [17], and antisocial personality traits [18]. Additional variants were present in pathways that contribute to attention and focus, language delays, and difficulties with auditory processing. Targeted interventions including L-theanine, magnesium threonate, and magnesium citrate were introduced to address some of these additional variants. Over 600 clinically relevant SNPs were evaluated with the neurodevelopmentally focused genomic CDS. Due to the intended brief nature of this case study report, however, we will not go into each of these pathways in great detail.

#### 3.1.3. Effects of Implemented Interventions on CJM

Patient's bedwetting stopped completely, stimming ceased, cognition dramatically improved, and he is now fully conversational. These new interventions for the multiple mitochondrial related pathways markedly improved his mitochondrial function and energy to the point that he is now able to play tennis and attend the gym regularly.

The changes in this young man's life have been astonishing. His ATEC score decreased to 21 and IQ increased to 70. CJM was no longer officially classified as intellectually impaired and was legally, according to Australian guidelines, eligible to attend a mainstream school. The patient became class "captain", attended the end of year prom, passed his driver's license exam, and was even able to attain part-time employment, working in a gluten free café. Independence became a reality. CJM is now holding meaningful conversations with family, peers, teachers and employer, who are all thrilled with his progress.

#### *3.2. JD1 and JD2 Case Study*

#### 3.2.1. Medical History and Background

The next two case study patients, referred to as JD1 and JD2, are interesting in that they involve identical twins who presented to the clinic in 2019 at age six. Both patients were reported by the parents to have severely regressed following an early childhood vaccination. Clinically, the children appeared to have symptoms relating to mitochondrial dysfunction and had difficulties with verbal communication.

Although they were identical twins presenting clinically with ASD, one child had additional symptoms more characteristic of ADHD, while the other clinically suffered from severe anxiety and OCD. Both were prone to recurrent PANDAS/PANS flares and OCD symptoms were present and increased during these infections. Prior to genomic interventions, both patients were taking melatonin and low dose naltrexone. IXXD's neurodevelopmental report was used to elucidate and address some of the root causes not only of ASD, but also of PANDAS/PANS, attention and focus, and anxiety-related symptoms. A table of symptoms prior to and after treatment is presented below (Table 1).


**Table 1.** JD1 and JD2 symptoms before and after personalized treatment.

JD1-case study patient; JD2-case study patient; OCD-obsessive compulsive disorder; PANDAS-pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections. Note: ↑ = Significant symptom improvement after treatment; ∆ = slight changes/some improvement noted after treatment; − = no change noted after treatment.

#### 3.2.2. CDS Results and Interpretation

Genetic analysis identified the presence of many different SNPs that correlated with symptoms shared between the twins. The children had a SNP in the mannose-binding lectin 2 gene (MBL2), which plays a role in the complement pathway, a component of the immune system. The T allele of this MBL2 SNP has been associated with significantly reduced MBL2 levels [19]. This correlates with a lower capacity to recognize foreign invaders (such as Streptococcus infections) and a higher risk for PANDAS (OR = 4.15) [19]. The abnormal immune response from these same SNPs has also been associated with brain

autoimmune activity [20], reduced blood–brain barrier function [20], tics, and an increased risk of OCD symptoms [19]. To help address this, a combination of lignite to help tighten the tight junctions of the gut [21], vitamin D [22], probiotics [23], and prebiotics [24] were added to the treatment protocol of JD1 and JD2.

The OCD risk was believed to be exacerbated by the presence of two variants in the solute carrier family 1 member 1 gene (SLC1A1). This solute carrier SNP, particularly in homozygotes, appears to contribute to higher glutamate levels [25] and has been shown to be associated with increased risk of OCD behaviors such as hoarding, ordering and lining things up (OR = 2.01) [26]. Targeted interventions including N-acetylcysteine (NAC) [27], L-theanine [28], vitamin D [29] and vitamin B12 [29] were used to address this pathway. Additional SNPs associated with OCD comorbid with tic disorder and severe bed wetting issues (OR = 2.68) [30] were also discovered in JD1 and JD2. As presented in the genomic CDS, there was overlap in potential interventions between SLC1A1 and the additional SNPs (i.e., some of the same supplements could be used to address both pathways).

The twins were revealed to have some SNPs that are fairly uncommon, as is the case with most TACGA patients presenting with ASD. They were shown to be homozygous for a rare protein kinase SNP found in just 5% of the population. Variants have been associated with increased ASD risk (OR = 1.86) [31], and are involved in pathways relating to cell differentiation, autophagy and survival, and brain development and remodeling [32]. Resveratrol and NAD+ were incorporated into the treatment protocol since both have been shown in studies to help autophagy pathways [33,34].

SNPs contributing to mitochondrial dysfunction were present in JD1 and JD2 and were believed to impact severe fatigue, and may have also contributed to some of their muscle weakness, as evidenced by trouble holding own posture and pencil grip. Genetic analysis revealed that both children had a variant in NQO1 that is associated with an approximate 67% reduction in enzymatic activity [35]. This contributed to mitochondrial dysfunction, oxidative stress, and impaired ability to clear environmental toxins [35]. To combat this SNP's low NQO1-conveying effects, both patients were started on sulforaphane, which is known to upregulate NQO1 activity [36]. Additionally, NQO1 is needed to convert coenzyme Q10 (CoQ10) to its active form ubiquinol [37]. Therefore, ubiquinol was also used to address this pathway.

Variants relating to vitamins, ADHD, neurotransmitter balance and various other molecular pathways were also present in these patients. For the purposes of brevity, a comprehensive discussion of these additional pathways is not included in this case study report. To address some of these other pathways, the twins' personalized treatment plan included pycnogenol, ashwagandha, pyridoxal-5-phosphate (P5P) and specific soilborne probiotics.

#### 3.2.3. Post-Treatment Symptoms and Improvements

Highly significant improvements were seen in both JD1 and JD2. Gains in speech and socialization with family and peers were evident. Improvements in sleep were noted, bed wetting ceased, and increased energy levels were obvious. Fine and gross motor skills were improved; the children gained the ability to dress themselves and developed enough coordination to be able to ride scooters. The PANDAS flares decreased in frequency, and marked reductions were noticed regarding OCD and tics. JD10 s anxiety resolved, and he was no longer a picky eater. Parents relayed that he was "eating everything in front of him". These improvements led to an increase in his weight, and he was no longer considered malnourished. JD2, on the other hand, showed marked improvements relating to ADHD symptoms. ATEC scores in both children have noticeably improved; JD1 showed a 41% reduction in ATEC scores (from 85 to 50), and JD2 showed a 44% reduction (from 97 to 54). ATEC scores continue to improve with each passing month the twins remain on the protocol. Table 1 shows post-treatment symptoms and improvements.

#### *3.3. AD Case Study*

#### 3.3.1. Medical History and Background

This final case study patient, referred to as AD, will be briefly touched upon. In this case, the patient's mother chose to go very slowly with supplements—targeted potential interventions were added one at a time. Thus, even though treatment was initiated late in 2019 when his genomic results were initially received, his regimen is continuing to be optimized at time of publication. However, this case is also important in that the CDS allowed better prioritization of interventions, rather than the usual trial and error approach.

Prior to genomics, patient had an ATEC score of 54. Clinically, this male four-year-old exhibited significant language delays with only five words at the age of three, developed a stutter, very frequent hand flapping, tics, eye rolling, stimming triggered by excitement, seizures, inappropriate socializing, and found it hard to focus or concentrate. Child had many chest infections, adenoid surgery, and over 20 rounds of antibiotics and steroids in the previous 12 months. Patient was not on any supplements when he presented to the clinic.

#### 3.3.2. IXXD Genomic Results and Interpretation

AD had many variants known to be of clinical significance in the pathways discussed in both case study patients above, including those associated with the language center and mitochondrial pathways. In addition to the aforementioned pathways and SNPs, it was discovered that the patient had glutamate receptor SNPs as well as two copies of an alcohol dehydrogenase 5 (ADH5) SNP that has been associated with ASD (OR = 1.54) [38]. ADH5 is a glutathione dependent enzyme that is primarily responsible for removing formaldehyde and is also important for protecting natural lipids from peroxidation [38,39]. Formaldehyde is a natural by-product of white blood cells and myeloperoxidase, and when formaldehyde is not properly removed (as would be the case in individuals homozygous for this SNP) it can build up in the brain and become neurotoxic [40,41]. This child's high rate of infections likely contributed to high neutrophil/myeloperoxidase (MPO) activation and higher levels of formaldehyde. In addition to supporting glutathione levels, since this enzyme is glutathione dependent [42], a list of foods shown to upregulate ADH5 was given the patient's mother. This list included foods such as pomegranate, watermelon, and tomatoes [43]. AD's mother was also informed regarding foods that could exacerbate the negative effects of this genomic pathway. For example, it was recommended to avoid foods artificially sweetened with aspartame, since aspartame is converted to formaldehyde [44]. These types of food and supplement interventions were taken from information listed in the referenced IXXD CDS.

Genomic CDS testing revealed that this child was homozygous for the same NQO1 SNP discussed in the case above (but case above only had one copy). Two copies of this SNP are found in approximately 4% of the population and lead to a significant reduction in enzymatic activity (approximately 97%) [35]. This drastic impairment in NQO1 activity contributes to significant mitochondrial dysfunction, increased oxidative stress, and markedly reduced detoxification [37]. As discussed above, NQO1 variants can contribute to decreased levels of the activated form of CoQ10 [37,45]. Being homozygous for this SNP dramatically impaired his ability to detoxify benzene, solvents, and many other pollutants [46]. High levels of these toxicants have been shown to contribute to increased DNA damage when exposed to various pollutants [47]. Unsurprisingly, the patient had extremely high levels of gasoline additives detected in his GPL Tox screen results, which was addressed as well.

Patient was also found to be homozygous for a well-known haplotype in the brain derived neurotrophic factor (BDNF) gene, which is found in approximately 4% of the population. This growth factor has been shown to be very important for memory and mood [48]. These BDNF SNPs contribute to decreased ability to cleave pro-BDNF to the truncated, mature form of BDNF [49]. While the mature form of BDNF is synaptogenic, the pro-BDNF form induces neuronal apoptosis and is synaptoclastic [49]. Furthermore, higher

levels of pro-BNDF levels have been observed in patients with ASD [50]. Regular aerobic exercise was encouraged to help increase the conversion of pro- to mature BDNF [51]. A high butyrate diet and butyrate supplement was also implemented to address this pathway [52].

A personalized, genomically targeted treatment plan was developed. Regimen included moderately high dose UBQH-PQQ, sulforaphane, fish oil, L-theanine, butyrate, magnesium threonate, and a few other supplements. Regular aerobic exercise was also encouraged.

#### 3.3.3. Post-Genomic Testing Improvements

ATEC scores with above interventions improved by 54% (scores decreased from 54 to 25) and continue to improve (as per communication with mother). Regarding symptom improvements, speech and socializing improved very quickly upon reducing inflammation and oxidative stress, working on detox pathways, and addressing gut health. Additionally, and remarkably, after adherence to the personalized list of supplements discussed above for only a few weeks, his seizures stopped. Hyperactive behavior continues to decrease, and tics and stims are improving. Parents continue to notice improvements on a weekly basis and are very happy with the progress to date.

#### **4. Discussion**

Non-syndromic ASD is clearly due to a multitude of contributing genomic factors that interact with environmental factors. The CDS tool used in this study also looked at genomic SNPs associated with secondary comorbid ASD conditions, given that they are pervasive amongst individuals with ASD. Comorbid conditions investigated by IXXD include ADHD, OCD, anxiety, PANDAS/PANS, gastrointestinal issues, food intolerances and nutrient deficiencies. The genomic and environmental factors, however, significantly vary from person to person. Outcomes trials have shown benefit for methyl-B12 [53], sulforaphane [54], luteolin [55], quercetin [56], melatonin [57], vitamin D [58], omega-3s [59], L-theanine [60] and dozens of other supplements in the treatment of ASD and comorbid conditions. Determining which potential interventions would be the most likely to result in improved ASD outcomes in a particular individual, however, has been a difficult hurdle to clear. Evidence-based genomic clinical decision support tools that focus on variants associated with neurodevelopmental, nutritional, toxicant clearing, and inflammatory pathways can help in prioritization and choice of interventions.

These case studies demonstrate that a well-referenced genomic CDS can be used as a tool to aid in the understanding of some of the gene variants contributing to the patient's neurodevelopmental disorder. This enables clinicians to address root causes and truly personalize treatment strategies, allowing for the achievement of more robust improvements as well as potentially faster improved outcomes in children with ASD. Initial results from the Australian Centre for Genomic Analysis practice using the IXXD tool, as illustrated by these cases, have been extremely promising. This short case series provides optimism for the role of genomics in improving function and quality of life in children with ASD and neurodevelopmental disorders and suggests that genomics in the form of a CDS can decrease the burden of the trial-and-error method.

The first limitation of this case study report is that only four cases were discussed. It will be important to analyze the collective data (ATEC scores, IQ scores, behavioral observations, etc.) from the complete cohort of approximately 100 patients. In this future analysis it will be important to separate out the results from individuals with access to the IXXD neurodevelopmental genomics CDS tool from the beginning, versus individuals who were previously optimized using the center's previous treatment methods and then plateaued.

A second limitation of this method of addressing neurodevelopmental disorders is that some patients will respond better than others to genomic CDS tools. Additional research must therefore be conducted. Next steps, which are currently in progress, include being able to reproduce the ability to obtain significant improvements in ATEC or other ASD rating scales in private physicians' offices in ASD centers across the country. In further research, controlled trials comparing the use of genomics to traditional care in ASD would be beneficial. Another limitation of this method is that it is a relatively new field not taught in residency or fellowships, and thus in order for genomics to be used systematically on a larger scale, clinicians will require dedicated time for study and continuing education. Nonetheless, genomics as a CDS tool can shift the paradigm of care for individuals with non-syndromic ASD and allow for higher functioning and better, quicker outcomes.

An additional limitation to this type of personalized medicine is the treatment cost. Utilization of a tailored, genomically-targeted approach is an investment for the family or whomever else is covering the ASD-related expenses. The cost of the IXXD tool used in this study was \$900 per patient. The cost of working intensely with a clinical team that is experienced in functional or integrative medicine, genomic interpretation, and nutrition generally ranges from \$2000 to \$5000 per year. Currently, insurance coverage for genomic testing is most often limited to specific instances (cancer treatment, pharmacogenomics in some situations, whole genomic sequencing for diagnostic purposes). Therefore, the financial responsibility of this IXXD approach is borne by the families. This cost, however, pales in comparison to the multitude of fees that families of children with ASD incur (financing a caregiver, providing special education, loss of wages of family members, etc.). As illustrated in the cases above, there is potential for a significant financial, long-term benefit when a child can improve overall function, attend schools, and join the workforce rather than being fully reliant on caregivers. Ultimately, as additional studies are published showing the benefit of this precision medicine approach, the potential for this type of CDS targeted treatment to become mainstream and covered by insurance is likely to increase.

**Author Contributions:** Conceptualization, S.H.-C. and H.W.; methodology, H.W.; software, S.H.-C. and G.W.; writing—original draft preparation, H.W. and G.W.; writing—reviewing and editing, G.W., J.R. and S.H.-C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Ethical review and approval were waived for this study, due the study being a case report only with no double-blind or placebo control and with explicit parental consent obtained as below. All interventions in the study consisted of foods and supplements available over the counter.

**Informed Consent Statement:** Written parental consent was obtained for all subjects involved in the study. Consent discussed the risk, benefits, and limitations of genomics when genomics were ordered. Written informed parental consent was also obtained for each participant to participate in the study and publish this paper.

**Data Availability Statement:** Relevant genomics and data presented in paper. Full access to genomics is part of an online resource available to ordering clinicians and is not available in downloadable or printable form.

**Acknowledgments:** Amanda Mullard at TACGA for both administrative support and other support in collecting genomic information and assisting with implementation of plan.

**Conflicts of Interest:** Heather Way declares no conflict of interest. She was the sole investigator and clinician responsible for collection and interpretation of data as well as determination and implementation of the personalized plan for each child. Grant Williams and Jordan Reeder are employees at IntellxxDNA™, which was the genomics clinical decision support tool used in this study, but have no financial interests. Sharon Hausman-Cohen is the medical director of IntellxxDNA™ and does have ownership interest.

#### **References**


### *Case Report* **The Temple Grandin Genome: Comprehensive Analysis in a Scientist with High-Functioning Autism**

**Rena J. Vanzo 1,\* ,†, Aparna Prasad 1,†, Lauren Staunch <sup>1</sup> , Charles H. Hensel <sup>1</sup> , Moises A. Serrano <sup>1</sup> , E. Robert Wassman <sup>1</sup> , Alexander Kaplun <sup>2</sup> , Temple Grandin <sup>3</sup> and Richard G. Boles <sup>4</sup>**


**Abstract:** Autism spectrum disorder (ASD) is a heterogeneous condition with a complex genetic etiology. The objective of this study is to identify the complex genetic factors that underlie the ASD phenotype and other clinical features of Professor Temple Grandin, an animal scientist and woman with high-functioning ASD. Identifying the underlying genetic cause for ASD can impact medical management, personalize services and treatment, and uncover other medical risks that are associated with the genetic diagnosis. Prof. Grandin underwent chromosomal microarray analysis, whole exome sequencing, and whole genome sequencing, as well as a comprehensive clinical and family history intake. The raw data were analyzed in order to identify possible genotype-phenotype correlations. Genetic testing identified variants in three genes (*SHANK2*, *ALX1*, and *RELN*) that are candidate risk factors for ASD. We identified variants in *MEFV* and *WNT10A*, reported to be disease-associated in previous studies, which are likely to contribute to some of her additional clinical features. Moreover, candidate variants in genes encoding metabolic enzymes and transporters were identified, some of which suggest potential therapies. This case report describes the genomic findings in Prof. Grandin and it serves as an example to discuss state-of-the-art clinical diagnostics for individuals with ASD, as well as the medical, logistical, and economic hurdles that are involved in clinical genetic testing for an individual on the autism spectrum.

**Keywords:** autism spectrum disorder; genetic testing; chromosomal microarray analysis; whole exome sequencing; whole genome sequencing; clinical utility; polygenic risk scores; Temple Grandin

#### **1. Introduction**

Autism spectrum disorder (ASD) is one of the most common neurodevelopmental disorders characterized by impairments in communication and social interaction and the presence of restrictive and repetitive behaviors [1]. The American College of Medical Genetics and Genomics (ACMG) recommends genetic evaluation for individuals with ASD [2]. Discovering the underlying genetic cause for ASD can improve the care and management by personalizing services and treatment, including addressing the medical risks that are associated with the genetic diagnosis [3]. We performed chromosomal microarray (CMA), as well as whole exome and genome sequencing (WES, WGS), on our co-author, Prof. Temple Grandin (T.G.), a widely recognized animal scientist and woman with high-functioning ASD, who is renowned for her insights on the condition. While using this report of the genomic findings in T.G., we create a discourse on state-of-the-art diagnostics for individuals with ASD. Predictably, we found many variants of uncertain significance (VUSs).

**Citation:** Vanzo, R.J.; Prasad, A.; Staunch, L.; Hensel, C.H.; Serrano, M.A.; Wassman, E.R.; Kaplun, A.; Grandin, T.; Boles, R.G. The Temple Grandin Genome: Comprehensive Analysis in a Scientist with High-Functioning Autism. *J. Pers. Med.* **2021**, *11*, 21. https://doi.org/10.3390/ jpm11010021

Received: 13 November 2020 Accepted: 24 December 2020 Published: 29 December 2020

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ( https://creativecommons.org/ licenses/by/4.0/).

However, we also identified the variants that were previously reported in the literature as pathogenic/disease-causing and overlap with her clinical features. Furthermore, many of these variants lie in genes that will personalize medical management and guide potential therapeutic options, which underscores the importance of clinical genetic testing in those with ASD.

#### **2. Materials and Methods**

#### *2.1. Case Report*

T.G. was a full-term female infant born in 1947 after an uncomplicated pregnancy, labor, and delivery. She had normal muscle tone and early motor milestones. However, she did not make eye contact and had touch sensitivity and aversion, including stiffening up when held by others. Even based on these specific features, ASD was not suggested as the diagnostic entity had only been described four years previously [4]. A neurologist performed an electroencephalogram and ruled out petite mal epilepsy, and a hearing test was normal. T.G. was diagnosed with "minimal brain damage" at two years old. At two and a half years old, expressive and receptive language delays were noted (she was non-verbal), and she was enrolled in speech therapy. With intensive treatment and emphasis on turn-taking games, she began speaking at age three and a half and was fully verbal by age four. She maintained typical autistic behaviors, such as repetitively dribbling sand through her hands at the beach and tantrums after sudden loud noises. The associated issues included stuttering and challenges with auditory and sensory integration. In particular, touch aversion improved with the use of a squeezing machine during childhood, which T.G. designed and has adapted in order to improve ethical animal husbandry management [5]. She also reported difficulties with social interactions in childhood and adolescence. The diagnosis of autism came later in elementary school by a psychiatrist. At age nine years and at age twelve years, her IQ was tested with the Wechsler. Her full-scale IQ was 120 on the first test and 137 on the second test. The measurement of cognitive abilities ranged from disability to gifted; for example, cognitive skills that require visualization, including block design and puzzle completion, were superior, whereas auditory integration was reduced [5,6]. T.G. received a Ph.D. and has been highly successful in her career regarding animal behavior-informed agriculture design prior to her career lecturing on ASD. Despite the absence of formal testing for ASD (based on age), T.G. meets current diagnostic criteria for ASD, given her childhood clinical history. The family history is unremarkable for specific diagnoses of ASD or intellectual disability in other family members. However, T.G.'s maternal grandmother was medicated for anxiety, and T.G. suspects that her father had high-functioning ASD. Both of T.G.'s parents held college degrees, and her maternal grandfather had particular academic success as an MIT-trained engineer and co-inventor of the autopilot for airplanes. The remainder of the family history is non-contributory without a suggested inheritance pattern; further details on family history is withheld in order to maintain privacy.

T.G. reports a substantial, decades-long history of chronic myalgia, muscle rigidity, paresthesia, and hyperesthesia of the feet, as well as sudden episodes of feeling "boiling hot". She reports coordination difficulties and tires easily with exercise, reportedly since childhood. She experiences insomnia and requires physical exercise at bedtime ("100 sit-ups") to sleep. Severe anxiety and panic attacks have been major life-long issues that have moderately improved on desipramine (50 mg/day since 1980). Desipramine also alleviated colitis-like symptoms. Her diet is high in animal protein, and its reduction or elimination results in perceived irritability. A peculiar rash with eczema- and psoriasis-like features, which had been present since early childhood and diagnosed in adolescence as eczema, responds to topical steroids. She has a widow's peak (a V-shaped growth of hair in the center of the forehead). Cranial MRI identified the asymmetry of the ventricles and a reduction in cerebellum size [7].

T.G. has microdontia and hypodontia, including six missing adult teeth (two on bottom, four on top, bilaterally symmetrical) which are absent on x-ray; she did not lose the corresponding deciduous teeth until the third to sixth decades. Two dentists have commented that she has a high arched palate. Additional ectodermal dysplasia (ED) related manifestations include soft and very brittle nails, hyperhidrosis, and body hair loss since the fifth decade of life.

#### *2.2. Sample and Genetic Analysis*

We obtained written consent from T.G. to disclose her name and health information for this study and publication. We did not have access to parental samples. Genomic DNA was extracted from an oral swab (OC-100Dx, DNA Genotek, Kantana, ON, Canada) using the PureGene extraction kit (Qiagen, Inc., Valencia, CA, USA). DNA extraction and all the analyses were performed in CAP and CLIA certified laboratories. Table 1 presents details regarding the various genetic testing technologies utilized in this study.



Caption. <sup>1</sup> Affymetrix CytoScan-HD microarray plus 88,435 custom probes added to improve detection of copy number variants (CNVs) associated with neurodevelopmental disorders [8]. <sup>2</sup> Default settings used <sup>3</sup> >97% coverage of 22,000 genes in the genome at >30x.

#### **3. Results**

In our testing population on the aforementioned custom array, we have historically observed that 28% of patients with neurodevelopmental disorders have one or more abnormal or potentially abnormal copy number variants (CNVs) [11]. In the case of T.G., we did not identify any pathogenic or likely pathogenic CNVs on either custom CMA or WGS, as reported through our clinical pipeline and ACMG reporting criteria [12], and the results were consistent with a normal female chromosome complement. However, out of over 4000 structural variants of different types (including deletions, duplications, inversions, LOH, break points, and insertions of transposable elements; see Table S2 for complete list) some of the variants of unknown significance could be relevant to patient's phenotype given their relevance to brain pathology. Two of these variants are discussed in more detail below, both being located on the q arm of chromosome 9.

One of them, a heterozygous duplication of chr9q34.3q34.3x3(138,014,000–138,228,000) is about 200 kbp long and it includes several noncoding genes and exons 19 to 47 of a calcium channel gene CACNA1B. *CACNA1B* is associated with Neurodevelopmental

disorder with seizures and nonepileptic hyperkinetic movements, according to OMIM. The disease is autosomal recessive, and while the deletion has a 0.00035 allele frequency in general population (four cases out of 11,295 in DGV database), one cannot completely exclude mild phonotype in heterozygotes.

Another candidate structural variant is a 1656 bp heterozygous deletion of chr9q34.13q34.13 x1(131,153,102–131,154,758), which is not found in the general population. The deletion affects exon 3 of non-coding gene RP11-544A12.4. Interestingly, this gene overlaps *NUP214*, which is located on the opposite DNA strand and, according to OMIM, is associated with susceptibility to acute infection-induced encephalopathy-9. However, the facts that disease is recessive and for *NUP214* the deletion is entirely intronic suggest that this variant is a less feasible candidate to be causative, at least not by itself.

WGS interpretation also revealed that CGG repeats that correspond to fragile X syndrome are 30,30 (a frequent normal genotype).

Three sequence variants of interest were identified in suspected or known ASD risk genes SHANK2, ALX1, and RELN (Table 2). Nevertheless, as of November 2020, none of these variants met the ACMG guidelines for "pathogenic" or "likely pathogenic" designation and, thus, are clinically classified as variants of uncertain significance (VUS) [12]. A heterozygous missense variant in *SHANK2* (p.H64R) was identified. This missense variant is a change from histidine to arginine. The histidine at this location does not lie in any well-defined protein domains. However, histidine is present at this location in primates. Further, the substitution of histidine to arginine is predicted by SIFT to be deleterious. This suggests that the his to arg amino acid substitution may alter protein function. Additionally, the variant is only observed in seven of 184,874 reference alleles (allele frequency: 3.79 <sup>×</sup> <sup>10</sup>−<sup>5</sup> ) in the Genome Aggregation Database, a database of approximately 141,000 individuals without severe genetic conditions (gnomAD) [13,14]. Several studies suggest a role for *SHANK2* in ASD and/or intellectual disability (ID). In one publication, a patient with ASD harbored a *de novo* nonsense variant in *SHANK2*, while two additional, unrelated patients with ASD and mild-to-moderate intellectual disability had *de novo* deletions in *SHANK2* [15]. This study suggests that the haploinsufficiency of the SHANK2 gene may affect synaptic function and predispose to ASD and/or ID. In a subsequent study, a novel *de novo SHANK2* deletion was identified in another patient with ASD. Further, sequencing identified a significant enrichment of variants affecting conserved amino acids in *SHANK2* (3.4% of autism cases and 1.5% of controls, *P* = 0.004, OR = 2.37) [16]. In neuronal cell cultures, the variants that were identified in patients were associated with reduced synaptic density at dendrites when compared to variants that were only detected in controls. Interestingly, the three patients with *de novo* deletions identified in the two aforementioned studies also carry inherited CNVs at 15q11-q13 previously associated with neuropsychiatric disorders [17–19]. These data strengthen the role of synaptic gene dysfunction in ASD and support the "multiple hit model", suggesting that a better knowledge of these genetic interactions will be important in understanding the complex inheritance pattern of ASD [18,20].

A heterozygous missense variant (p.R64L) was identified in the ALX1 gene. *ALX1* encodes a transcription factor that plays a role in development, including proper neural crest migration in animal models [21]. Bi-allelic loss-of-function variants in *ALX1* cause frontonasal dysplasia, while gain-of-function variants are hypothesized to impact neurodevelopment [22,23]. T.G.'s craniofacial finding of a "widow's peak" could be related to this variant, as neural crest cells can be found in hair follicles [24]. This variant was found in 1203 of the 280,730 reference alleles (allele frequency: 0.004285) in the gnomAD database [14]. Additionally, the variant identified in T.G. was one of several potential ASD risk variants that were identified in two unrelated multiplex families [25]. In one family, this variant was shared by two siblings with ASD and it was inherited from their unaffected father. In the second family, the variant was found in an individual with ASD and it was not found in his two brothers with ASD, but was inherited from his unaffected father. Importantly, reportedly unaffected parents were not phenotyped in detail in that publication. The authors showed the *ALX1* variant was observed multiple times in their population study (27/1541 cases and 58/5785 controls), yielding an odds ratio of 1.75 (95% confidence interval 1.11 to 2.77; *p* = 0.022; on page 7 of Matsunami et al., 2014 [25]). Although this specific variant has been observed in a supposedly unaffected control population, its higher prevalence in individuals with ASD when compared to those without ASD supports it as a potential risk factor. Further research is needed in order to confirm the impact of this variant on gene function and the role of *ALX1* in ASD susceptibility.

A heterozygous missense variant in the RELN gene (p.T1002S) was also identified in T.G. *RELN* encodes the reelin protein, which is thought to control interactions between cells for cell positioning and neuronal migration. Although the serine for threonine substitution is conservative and does not lie within any known protein domain, this variant was not present in the gnomAD database and it is conserved across different vertebrate species, except lamprey (PhyloP 1.048) [14]. Variants in *RELN* are associated with autosomal recessive lissencephaly with cerebellar hypoplasia (OMIM). In addition, *de novo* variants in *RELN* have been observed in individuals with ASD in several studies [26–28].

*MEFV* is a fourth gene that harbored variants with clinical overlap for T.G. Two heterozygous variants, p.R408Q and p.P369S, were identified in T.G. and they have been reported to be disease-associated. Allele frequencies in the gnomAD database are 0.00001595 (four out of 250,794 reference alleles) and 0.01470 (4150 out of 282,228 reference alleles, respectively [14]. The MEFV gene encodes a protein, called pyrin, whose function is not fully understood, but appears to direct the migration of white blood cells to sites of inflammation and downregulate the inflammatory response following the improvement of infection or injury. Over 80 variants in *MEFV* have been associated with familial Mediterranean fever (FMF), a highly complex and variable condition that can exhibit either autosomal dominant or autosomal recessive inheritance [29]. Studies suggest these variants are in linkage disequilibrium and are, thus, in cis [30]; indeed, a review of the WGS read data that were utilized in this study (2 × 150 bp) was long enough to confirm cis phasing.]. Despite having ClinVar associations that range from "benign" to "pathogenic", these variants, when found together, have been published as associated with disease and they are often included in clinical gene panels that are designed to test for FMF [31,32]. Most of the patients with both variants are reported to have an atypical clinical presentation. Although T.G. does not strictly meet the Tel-Hashomer clinical criteria for FMF, she has symptoms that are consistent with the atypical presentation of the condition seen in those with the same genotype, including frequent intermittent hot spells, muscles that are stiff and sore, episodes of calor, and paresthesia in both feet, and lifelong skin rashes that are diagnosed as eczema [32]. FMF has not been reported to be associated with ASD; however, inflammation is one of many pathways implemented in ASD pathogenesis and, thus, we cannot exclude these variants as being risk factors for ASD in T.G.

Further, *WNT10A* is another gene harboring a variant with overlap to T.G.'s phenotype. A homozygous variant (p.F228I) has been previously reported as pathogenic. *WNT10A* is a member of the WNT gene family, which encodes proteins that are implicated in several developmental processes, including the regulation of cell fate and patterning during embryogenesis [33]. Although p.F228I is a conservative amino acid substitution, the amino acid at this position is conserved across different species (PhyloP 0.964) and the variant has been predicted to be deleterious to the protein structure or function by in silico prediction tools. The variant identified here has been reported previously, either in the homozygous state or in trans with a second pathogenic variant, in individuals with either isolated oligodontia, tooth agenesis, or with other features of ED [34–38]. The p.F228I variant that is identified in T.G. is observed at a relatively high frequency in the general population (heterozygous carrier frequency: 0.0137, homozygote frequency: 0.000153 in the gnomAD database) [14]. Oligodontia is observed in approximately 0.14% of the population and, in one study, it was shown that variants in *WNT10A* were present in more than half of the cases of isolated oligodontia [38,39]. The variant in WNT10A possibly explains the multiple ED-like manifestations in T.G. involving her teeth, nails, hair, and sweat glands.


 *WNT10A* 2:219755011 NM\_025216.3:c.682T>A p.Phe228Ile Deleterious Pathogenic rs121908120 0.0137 Homo ED Caption. VUS-variant of uncertain clinical significance; NA-not available; Het-Heterozygous; Homo- Homozygous; ASD-autism spectrum disorder; FMF-familial Mediterranean fever;

ED-ectodermal dysplasia; rows 1–3 strong evidence to support autism susceptibility; rows 4–6 lists published pathogenic variants with excellent clinical correlation. Based on review of WGS raw data, *MEFV* variants are in *cis*. Note: *VKORC1* and *CYP2C9* variants described in the text are polymorphisms and thus not represented in the rare variant table above.

Please see Supplementary Table S1 for additional variants, which were identified in both the whole exome and whole genome sequencing assays that were run independently, with additional clinical overlap that potentially contributes to T.G.'s ASD, sleep pathogenesis, anxiety, mitochondrial function, and more, including some with potential therapeutic targets.

From a proactive standpoint, sequencing also identified a *CYP2C9* genotype that was associated with slowed metabolism of many drugs, including the anticoagulant warfarin [40]. This information, coupled with T.G.'s heterozygous status of the *VKORC1* 1639G>A variant also revealed by this testing, provides specific dosing information if warfarin is prescribed to reduce the risk adverse events. Furthermore, this *CYP2C9* genotype is also correlated with slowed metabolism of the anti-epileptic drug phenytoin, which would impact the dosing recommendations; this is critical information, given that there is a high rate of comorbidity between epilepsy/seizures and ASD [41,42].

Given the presence of multiple genetic variants potentially contributing to the manifestation of ASD in T.G., we sought to explore her data through a currently available polygenic risk score (PRS) algorithm (impute.me). This algorithm showed that T.G.'s PRS for ASD is lower than 99% and greater than 1% of the general population by assessing 17 single nucleotide polymorphisms (SNPs) that were previously reported to be associated with ASD [43].

#### **4. Discussion**

We present this case of a female scientist, Prof. Temple Grandin, with high-functioning ASD and other clinical sequelae, who was referred for clinical diagnostic testing. Through various test methodologies, we identified variants of unknown significance in three ASD risk genes (*SHANK2*, *ALX1*, and *RELN*) and other variants that impact genes that are possibly relevant to ASD pathology. This supports the concept of a polygenic model in ASD. Surprisingly, the PRS model used for T.G. showed her risk for ASD in the 1st centile of the general population. The tool did provide a pie chart indicating that the genetic liability captured by this assessment is very small (~1%), which echoes the disclaimer supplied in other publications regarding the limitations of PRS for clinical application [44]. There are several potential reasons for the contradictory ASD PRS score in T.G., given her clinical diagnosis. The SNPs that were used to generate the score in the impute.me tool were derived from case cohort individuals diagnosed with ASD prior to 2014 [43]. Importantly, Asperger disorder was a diagnostic entity until 2013, according to the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM). Therefore, the case cohort whose data are represented in the tool is likely more representative of those individuals who are lower functioning, which is in stark contrast to the phenotype of T.G. Additionally, we, as a community, do not yet know all of the genes and variants contributing to ASD. Additionally, finally, some co-occurring traits in those with ASD are also clearly polygenic (IQ and anxiety for instance) and may skew the algorithmic outcomes. In summary, a low PRS score, at least for ASD, cannot be used to rule out the potential for a future clinical diagnosis.

While T.G. herself does not harbor any known, identifiable ASD-related variants that garner specific medical management changes, this is a possibility for others with ASD and genetic testing should be pursued as a standard of care in line with ACMG and other medical guidelines [45]. Importantly, genetic testing did identify actionable variants that contribute to T.G.'s clinical symptomatology and can be specifically addressed in order to improve her functional symptoms and prevent further medical issues. Of note, prior to testing, additional neurological and non-neurological symptoms were attributed to a broader diagnosis of ASD and not specifically addressed based on underlying genetic etiology. This is critical for medical action, as well as family understanding, coping, and improved quality of life.

For example, clinical features that disturb T.G.'s activities of daily living are longstanding calor and paresthesia in both feet. These correlate with the two identified *MEFV* variants that were reported in cases with atypical familial Mediterranean fever. FMF is a

treatable condition, and T.G. has since been referred to see a specialist. In addition, the homozygous pathogenic variant in *WNT10A* is likely to explain T.G.'s multiple ED-related symptoms. Dentists and orthodontists incorporate specific management decisions for individuals with ED-related disorders. Fortunately, T.G. herself chose not to have dental implants; the avoidance of dental implants would have been advised previously had a diagnosis of ED been known at the time. Moreover, genetic testing has provided actionable guidelines for the future prescription of certain pharmacologic treatments that were impacted by *CYP2C9* metabolism. There are a variety of other findings for T.G. with currently less well-supported medical literature that will advance with time. For example, will some combination of CNVs and SNVs that are currently classified as benign emerge as a risk susceptibility for ASD? In the future, will the reclassification of VUS to pathogenic or likely pathogenic variants trigger medical action that prevents medical morbidity or mortality? A reinterpretation of the raw data and integration with medical care can be pursued based on T.G.'s medical course and preferences in the future.

The power of clinically available genetic testing for those with ASD with or without co-occurring morbidities can be substantial for neurologic and non-neurologic precision medicine, as shown in this study. However, there are associated challenges that we must address as a medical community to make this process more impactful. One is a lack of trained clinical experts that are comfortable in reading vast amounts of genetic data and translating it in order to inform disease-associated factors and treatment options. Efforts should be made to train physicians and other healthcare providers in the practical use of genomics, as the future of health care depends on its understanding and application, including the limitations of using PRS models to predict future presence and severity for ASD. A second challenge is that the feasibility of a "genomics board" (akin to that of a multidisciplinary "tumor board", which is standard in oncologic care) is hindered by state medical licensing and telemedicine laws. Genetic counselors (GCs) are essential in this process and growing in number, but they typically have long wait lists or their own state licensure barriers that encumber integrated care. Peer-to-peer consultation between physicians and GCs is not precluded by such laws; however, including the family in the discussion constitutes the practice of medicine in most jurisdictions. Third, market forces have resulted in low-cost exome and genome testing. However, careful report generation and detailed discussion with the patient's attending providers takes substantial professional time and it is not practical under the current cost structures. Paradoxically, this results in healthcare providers being unaware of or unable to utilize the resulting complex genetic reports in order to improve clinical care and leads insurance companies to deny coverage for lack of clinical application. We propose that payer policies should be devised to commensurately compensate parties that are involved in sequencing and variant interpretation, as well as physicians and GCs for effective use of the data and the treatment of the patients.

To summarize, these data support the concept that the genetic etiology of high-functioning ASD in T.G. could result from a combination of multiple genetic factors interacting in order to yield the observed clinical features. We demonstrated that comprehensive clinical phenotype information and genomics-trained providers/laboratorians are critical in the interpretation of genomic variants that were identified through these high-throughput genomic technologies. The genomic analysis that was carried out in this study provides a basis for at least part of T.G.'s clinical features and delivers suggestions for effective management of some of the symptoms. While improvements can be made to the process and application of genetic testing for individuals with ASD, it is effective and critical, as it currently exists for optimal medical and improved personal and family quality of life.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2075-4 426/11/1/21/s1, Table S1: Detailed Sequence Variant Spreadsheet, Table S2: Detailed Structural Variant Spreadsheet.

**Author Contributions:** Conceptualization, R.J.V., C.H.H., E.R.W., T.G., R.G.B.; resources, R.J.V., L.S., E.R.W., T.G., R.G.B.; data curation, A.P., M.A.S., A.K.; writing—original draft preparation, R.J.V., A.P.; writing—review and editing, R.J.V., A.P., L.S., M.A.S., R.G.B.; visualization, R.J.V., E.R.W., T.G., A.K., R.G.B.; supervision, R.J.V., C.H.H., E.R.W., R.G.B.; project administration, R.J.V.; funding acquisition, R.J.V., C.H.H., E.R.W., T.G., R.G.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was funded by Lineagen, Inc.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki and approved by Western Institutional Review Board (protocol code 20162032 and date of approval 11/1/2018).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available in the supplemental materials of this manuscript submission.

**Acknowledgments:** We thank members of our CAP/CLIA laboratory partners (including Predictive Laboratories, formerly known as Taueret Laboratories, LLC and PerkinElmer Genomics) for their assistance in processing and analyzing this sample and the team at Variantyx, Inc. for providing access to its Genomic Intelligence®software for WGS analysis. We also thank Jehannine Austin for providing expert guidance regarding the impute.me website and utility of polygenic risk scores as well as the entire Lineagen team for upholding our vision and mission to improve the lives of all individuals with autism spectrum disorder.

**Conflicts of Interest:** A.P., C.H.H., E.R.W., and L.S. were employees of Lineagen. M.A.S. and R.J.V. are employees and shareholders of Bionano Genomics, Lineagen's owner. A.K. is an employee and shareholder of Variantyx, Inc. T.G. and R.G.B. have no conflicts to declare.

#### **References**

