Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9

National Institute of Health Study FDA Psychiatry Division 301-796-2260 IND # 131455; ClinicalTrials.gov ID: NCT02839915 Efficacy of Folinic Acid in the Improvement of Language Impairment in Children with Autism Spectrum Disorder

Protocol Version: 2.9 Draft Date: 09.07.17 Emory IRB # 00089662

Richard Frye, MD, PhD Arkansas Children’s Hospital (University of Arkansas)

Lawrence Scahill, MSN, PhD Marcus Autism Center (Emory University)

Christopher McDougle, MD Massachusetts General Hospital (Harvard)

History of Protocol Revisions Version 1.0 April 24, 2016 Version 1.1 May 2, 2016 Version 2 June 1, 2016 Version 2.1 June 2, 2016 Version 2.2 June 3, 2016 (exclude children on anticonvulsants for seizures) Version 2.3 June 4, 2016 (minor change to DSMP) Version 2.4 June 25, 2016 (corrections to reference list; added Emory IRB number) Version 2.5 July 13, 2016 (minor corrections to the content and legend of Table1 Schedule of measures; clarification in paragraph entitled: Visit Description and Schedule)  Date submitted to Emory IRB: June 7, 2016  Date submitted to FDA 6/27/16 Version 2.61 July 18, 2016 (clarified when allowed anticonvulsants would be monitored for blood levels and when a second or third attempt to collect research blood samples would be done in cases for whom screening blood sample was insufficient or not obtained); (Replaced mention of Yale in Data Safety and Monitoring Plan with Emory). Version 2.7 August 18, 2016 (edits in abstract, Aim 1a, statistical section regarding statement on the primary outcome measure; edits to Aim 1c regarding statement on the key secondary outcome; corrected inconsistency on entry criterion for the Core-CELF - now noted as < 80 throughout; added that score of 40 on Core-CELF is exclusionary; minor changes in Schedule of Measures; added comment on rationale for pre- and post- measurement of Core-CELF; noted on CGI-I, drop outs will be classified as non- responders). Version 2.8 August 31, 2016 (revised dose schedule on Table 2; dropped the 5 mg evening dose for subjects > 15 - < 20 kg; clarified the taper schedule.) Added a statement on disposition. Version 2.9 September 7, 2017 (added to analytic plan to include moderator analysis). Added clinicaltrials.gov ID #.

1

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9

Protocol Signature Page

The signature below constitutes the approval of this protocol and any attachments and provides necessary assurances that this trial will be conducted according to all stipulations of the protocol. This includes all updates from previous versions (where applicable), all statements regarding confidentiality and according to the local legal and regulatory requirements and applicable federal regulations and ICH guidelines. No changes to this protocol will be made without approval of the Emory IRB except when necessary to protect the safety, rights or welfare of study participants

Principal Investigatory name (print): Lawrence Scahill, MSN, PhD

Principal Investigator Electronic Signature:

E-signed Date: 05/02/16

2

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 ABSTRACT Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental disorder often with life-long consequences that affects young children during critical developmental periods. The Centers for Disease Control estimates that ASD affects as many as 14.7 per 1000 children (1 in 68).1 Despite the dramatic rise in the detected prevalence of ASD over the past two decades, no effective medical treatment has been developed to address core ASD symptoms (social communication and repetitive behavior), the closely associated problem of language impairment, or the underlying pathophysiology of ASD. Currently, the only accepted treatment for core ASD symptoms is behavior therapy, which may entail intensive one-on-one treatment over several years. Such therapy is inconsistently covered by medical insurance and may not be available in special education settings, thereby limiting its application. The de facto standard-of-care for treating ASD is a combination of educational and behavioral interventions. Although interventions such as speech and occupational therapy may help modify neural pathways, they may not address the pathophysiology that underlies ASD. Despite these interventions, most children with ASD have incomplete recovery and need for long-term supportive care. Thus, a safe, effective and well-tolerated medical treatment that improves core and closely associated ASD symptoms, addresses the underlying pathophysiology of ASD, and augments ongoing behavioral and educational interventions would have disease-modifying potential.

The primary aims of this study are to evaluate the efficacy and tolerability of high-dose folinic acid for improving language impairment in children with ASD. Improvement in delayed language may also benefit the core ASD problem of social communication. The study will also focus on identification of biomarkers in pre-specified subgroups of children with ASD that may moderate positive response to folinic acid. Our model is that high-dose folinic acid will improve language and set the stage for improved social communication in children with ASD and moderate language impairment. To test whether folinic acid is superior to placebo, 160 children (age 5 to 12 yrs 6 months, inclusive) with ASD and moderate language will be randomly assigned to folinic acid or placebo for 12 weeks under double-blind conditions. We will also test whether abnormalities in folate-dependent pathways, such as dysfunctional transport of folate across the blood-brain barrier, will moderate positive response to folinic acid treatment. This clinical trial is based on encouraging results from our pilot randomized, double-blind, placebo-controlled (DBPC) trial.

3

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 A. BACKGROUND and SIGNIFICANCE Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental disorder often with life-long consequences that affects young children during critical times in their development.2 The latest estimates from the Center for Disease Control estimates an ASD prevalence of 14.7 per 1000 (1 in 68) children in the United States.1 ASD is defined by impairments in social communication as well as the presence of restrictive interests and repetitive behaviors4. ASD is frequently associated with co- occurring problems such as gastrointestinal abnormalities5, seizures6, attention deficits7, anxiety8 and allergies9 - to name a few.

The detected prevalence of ASD has dramatically increased over the past two decades due to the expanded definition of the disorder. The increase in the detected prevalence notwithstanding, we do not have effective medical treatment for children with ASD. A recent multicenter study showed that optimal outcomes can be achieved in a subset of ASD children.10 Moreover, a study that analyzed the annual change in ASD symptoms in 6,975 children showed that some children with ASD can substantially improve.11 However, these studies did not address whether particular therapies were associated with better outcomes. In addition, the findings suggest that only a minority of children with ASD show substantial improvement over time, underscoring the need for more effective treatments.

Currently the only well-accepted treatment for core ASD symptoms is behavior therapy such as Applied Behavioral Analysis (ABA) and Early Intensive Behavioral Intervention. However, such therapy requires near full-time engagement with a one-on-one therapist over several years. The evidence for early intervention rests primarily on single-subject design studies with few randomized controlled studies and these interventions rarely result in complete recovery.12 Moreover, access to such intensive treatments is often limited and may only be available in highly specialized centers.13 Behavior therapy is inconsistently covered by medical insurance and not uniformly available in the education system.14,15 Thus, the de facto standard-of-care for treating ASD is a combination of educational and allied health therapies (speech and occupational therapies) in various settings.

Although these interventions may promote development in children with ASD, they may not address the underlying pathophysiological abnormalities in ASD. As noted, optimal outcomes may occur in a minority of individuals with ASD. However, most require life-long supportive care and do not achieve independent living in adulthood.16 Thus, a safe, effective and well-tolerated treatment that addresses underlying physiological abnormalities in ASD, augments available habilitative and educational interventions could accelerate achievement of optimal outcomes for a greater proportion of children with ASD.

Evidence-based medical treatments for ASD are limited. There is no United States (US) Food and Drug Administration (FDA) approved medical therapy that addresses core ASD symptoms or the pathophysiological processes that underlie ASD.17 The two FDA approved drugs, both atypical antipsychotics, are indicated for serious behavioral problems in children with ASD. Although these drugs, risperidone and aripiprazole, may improve social interactions as a secondary benefit,18 the detrimental effects on lipid and glucose metabolism and body weight are well established and may develop quickly (within 12 weeks).19-21 These metabolic changes increase the risk of cardiovascular disease and type-2 diabetes. Although the incidence is low, tardive dyskinesia, a potentially permanent movement disorder, is an additional risk.22 There is a pressing need for safe medications that affect target pathophysiological processes and treat core ASD symptoms.

Several medications are used off-label to treat associated, but not core, ASD symptoms. These medications do not address underlying pathophysiological abnormalities. A systematic review of attention-deficit hyperactivity disorder treatments for ASD highlights the complicated nature of medication treatment due to the diverse nature of children with ASD and associated adverse effects.23 A recent Cochrane review found no evidence for efficacy of selective serotonin reuptake inhibitors in 4

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v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Į )5Į 7KURXJKHQHUJ\GHSHQGHQWHQGRF\WRVLVIRODWH LVWUDQVSRUWHGDWWDFKHGWRWKH)5ĮIURPWKHDSLFDOWR WKHEDVRODWHUDOVLGHRIWKHFHOODJDLQVWDFKHPLFDO JUDGLHQW )LJXUH 7KLVDFWLYHWUDQVSRUWSURFHVVLV QHFHVVDU\EHFDXVHIRODWHFRQFHQWUDWLRQLQWKHFHQWUDO QHUYRXVV\VWHPLVVHYHUDOWLPHVKLJKHUWKDQVHUXP $ERXW\HDUVDJRLQDUHSRUWLQWKH1HZ(QJODQG -RXUQDORI0HGLFLQH'U4XDGURV &R, DQGFROOHDJXHV GHVFULEHGDQHZQHXURPHWDEROLFGLVRUGHUFDOOHG FHUHEUDOIRODWHGHILFLHQF\ &)' ZKLFKLVFKDUDFWHUL]HG E\DEQRUPDOO\ORZIRODWHOHYHOVLQWKHFHUHEURVSLQDOIOXLG GHVSLWHQRUPDOVHUXPIRODWHOHYHOV&)'LV DVVRFLDWHGZLWKWZRW\SHVRI)5ĮDXWRDQWLERGLHV EORFNLQJDQGELQGLQJZKLFKLPSDLULWVIXQFWLRQ6HUXP WLWHUVRI)5ĮDXWRDQWLERGLHV )5$$V KDYHEHHQ FRUUHODWHGZLWKFHUHEURVSLQDOIOXLGIRODWHFRQFHQWUDWLRQV Figure 1. Folate transport across LQLQGHSHQGHQWVWXGLHV0LWRFKRQGULDOGLVRUGHUVKDYH the blood brain barrier EHHQDVVRFLDWHGZLWK&)'GXHWRDODFNRIHQHUJ\IRU 5

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 the active transportation of folate. Early case-series of children with CFD described many with ASD features.52,53

Frye (Arkansas PI) and colleagues29, FRAAs were measured in 93 children with ASD using the assay development by Quadros (Co-I). Of these, 60% were positive for the blocking FRAAs and 44% for binding FRAAs. Overall, 29% were positive for both FRAAs, 46% were positive for only one FRAA and 75% were positive for at least one FRAA. These prevalence rates are clearly higher than the prevalence of blocking FRAAs in the general population United States (10-15%), Spain (7.2%), and Ireland (12.6%).54 In a study from Belgium, 47% of ASD children were positive for the blocking FRAA compared to 3.3% of developmentally delayed non-ASD controls.54

In summary, deficits in folate metabolism and/or interference with the transport of folate into the brain could play a causal role in the pathophysiology of some children with ASD. Folinic acid may be effective way to correct one or more of these abnormal folate pathways.

B. SPECIFIC AIMS 1. To evaluate the efficacy, safety and tolerability of a high-dose folinic acid treatment for improving language in 160 children with ASD and moderate language impairment (age 5-12 yrs 6 months, inclusive). The 12-week randomized, DBPC trial will be followed by a 12 week open- label extension. a. The primary outcome will be the change in language function from baseline to Week 12 in children randomly assigned to folinic acid compared to placebo as measured on age- appropriate Core subtests of the Clinical Evaluation of Language Function – 4. b. Tolerability of high-dose folinic acid will be systematically assessed and documented during the 12-week DBPC phase. c. The key secondary outcome will be the rate of positive response in the folinic acid group compared to placebo as measured on the Improvement item of the Clinical Global Impression scale at Week 12 of the Double-blind phase. d. Other secondary outcomes will include: change in repetitive behavior (CYBOCS-ASD), change in Irritability, Social Withdrawal and Hyperactivity subscales of the Aberrant Behavior Checklist between treatment groups at week 12 of the DBPC phase.

2. To determine if immune and/or genetic biomarkers can predict response to high-dose folinic acid. Abnormalities in folate and folate-related pathways could explain the need for high- dose folinic acid. Preliminary data suggest that the presence of abnormalities in the folate pathway predict positive response to folinic acid. Biomarkers of folate-related abnormalities include genetic polymorphisms in folate and folate-related pathway enzymes and autoantibodies to the folate receptor alpha. To determine if these biomarkers moderate treatment response, we will: a. Test if immune biomarkers (folate receptor alpha antibodies) predict clinical response to folinic acid b. Test if specific genetic polymorphisms (MTHFR 667C>T, MTHFR 1298A>C, DHFR 19 bp del, RFC 80A>G, TCN2 776C>G, COMT 472G>A, GSTM1 Null, MTRR 66A>G, MTR 2756A>G, and SHMT 1420C>T) predict clinical response to folinic acid. c. Combine the candidate immune and genetic biomarkers into a predicting equation that will evaluate the probability of positive response to high-dose folinic acid treatment.

C. RESEARCH PLAN Design 160 children with ASD and moderate language impairment will be randomized in 1:1 ratio to receive folinic acid or placebo for 12 weeks under double blind conditions. We hypothesize that folinic acid treatment will be superior to placebo on the Core tests of the Clinical Evaluation of Language Fundamentals (CELF). On the phone screen, we will identify children (age > 5 and < 12 years 6 6

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 months) with a diagnosis of ASD and language impairment. These children will be invited to participate in formal screening. The screening visit will be used to confirm the diagnosis of ASD and moderate language delay (score> 40 and < 80 on the Core CELF). Children who need a different or more intensive treatment (e.g., clinically significant self-injurious or aggressive behavior or preverbal) will be excluded (see below).

Protection of the blind, the two clinician model: Participants will be followed by two blinded clinicians: a treating clinician and an independent evaluator (IE). The treating clinician will monitor dose, AEs and compliance. The IE will see the child and family at Baseline, Weeks 4, 8 and 12 to evaluate therapeutic outcomes without any discussion of medication dose or AEs. Treating clinicians and IEs will conduct their assessments separately and will not confer about the case except in urgent situations. Ratings and progress notes for the treating clinician and IE will be kept in separate folders.

Inclusion criteria - Boys and girls ≥ 5 years of age < 12 years 6 months of age; - Weight ≥ 15 kg; - DSM-5 diagnosis of Autism Spectrum Disorder as established by clinical assessment, corroborated by the Social Communication Questionnaire and the Autism Diagnostic Observational Schedule. - A score < 80 on the Core Language score of the Clinical Evaluation of Language Fundamentals (CELF)– 4. - Current Clinical Global Impression Severity score ≥ 4 on ASD + communication delay. - IQ at least 40 as measured by the Leiter or mental age at least 18 months as measured on the Receptive Language Scale of the Mullen. - Stable educational plan (one month) with no planned changes in the intensity of treatment for 12 weeks. (Otherwise eligible subjects with anticipated changes in their school program in the near term will be invited to return when the transition has been accomplished. - Stable speech therapy program in the community (one month) with no planned changes for 12 weeks. - English is spoken in the home and at least one parent is able to read, write and speak English. - Stable medication (no changes in past 6 weeks and no planned changes for the next 6 months (duration of the study). - Able to swallow pills whole, as reported by parent or demonstrated by child – if parents cannot confirm.

Exclusion criteria - IQ below 40 as measured by the Leiter or below a mental age of 18 months on the Receptive Language Scale of Mullen. (N.B. subjects who test below 18 months of age, but are otherwise eligible, may be enrolled following a case review by the Steering Committee – e.g., child’s uncooperative behavior resulted in a likely underestimate of intellectual ability); - Score of 40 on the Core-CELF (indicates more severe language); - Current DSM-IV diagnosis requiring alternative pharmacotherapy, e.g., Major Depression, Bipolar Disorder, a psychotic disorder (based on clinical assessment assisted by the Child and Adolescent Symptom Inventory); - Presence of serious behavioral problems (tantrums, aggression, self-injury) for which another treatment is warranted. - Significant medical condition by history or by physical examination or lab tests that would be incompatible with the study drug. - Children taking anticonvulsant medication for seizures.

Procedures Randomization: Randomization will be carried out within site using permuted blocks of 4 subjects without additional stratification to drug or placebo in a 1:1 ratio. Statisticians will prepare the permuted 7

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 blocks for each site and will be sent to the Investigational Pharmacist at each site. Investigators and study staff will be blind to the allocation sequence. Our pilot data suggest that the presence or absence of FRAAs may have prognostic value. Although the prevalence of the autoantibody in children with ASD is not known with certainty, we are convinced that the prevalence is not rare and that in this sample of 160 subjects, we can rely on randomization to provide equal numbers of children with ASD that are positive and negative for FRAAs.55

Table 1. Schedule of Measures Double blind Phase Open Label Phase *2 4 *6 8 *10 12 *15 18 *21 24 Measure Screen Baseline wk wk wk wk wk wk wk wk wk wk Parent Ratings [1] Demographics X SCQ-Lifetime [2] X Vineland II Caregiver Form X X CASI-5 [3] X ABC [4] X X X X X X HSQ-ASD [5] X X X X X X CGSQ [6] X X Subject Assessments Core CELF [8] X X X ADOS [9] X Leiter IQ Test [10] X Safety Laboratory Testing [11] X X Biomarker samples X [X] [X] Pregnancy (B-HCG) [12] X X X Vital Signs X Height and Weight X X X X X X X Head Circumference X [X] Treating Clinician Medical/Psych History X Physical Exam X Health/ Behavior Review X X Adverse Effects Review X X X X X X X X X X Suicide Assessment [13] X X X X X X X Concomitant Treatments X X X X X X X X X X X X Drug Compliance X X X X X X X X X X Independent Evaluator CGI-Severity [14] X X X CGI-Improvement [15] X X X X X Parent Target Problems X X X X X X CYBOCS-ASD [16] X X X X X X * telephone checks; [1] Parent or primary caretaker; [2] SCQ Lifetime = Social Communications Questionnaire - Lifetime;[3] CASI = Child & Adolescent Symptom Inventory [4] ABC = Aberrant Behavior Checklist; [5] HSQ- M=Home Situations Questionnaire- Modified; [6] CGSQ = Caregiver Strain Questionnaire; [8] Core CELF (Clinical Evaluation of Language Fundamentals); [9] ADOS = Autism Diagnostic Observational Schedule; [10] or Mullen; [11] Routine blood tests and urinalysis for all subjects; [12] Urine (or blood) for B-HCG in girls of child bearing potential; For repeat B-HCG at Weeks 12 and 24 (or early termination) use a urine sample; [13] Included in Adverse Event Review; [14] CGI-S = Clinical Global Impression for Severity; [15] CGI-1 = Clinical Global Impression for Improvement; [16] CYBOCS-ASD = Children Yale Brown Obsessive Compulsive Scales Modified for ASD. [X] Biomarker samples (DNA or folate antibody) at Week 12 for subjects on monitored anticonvulsant, Week 24 or early termination only if missed at Screening;

Visit Description and Schedule Screening Assessment The RUPP Autism Network has established a systematic set of assessment methods. This project will also benefit from established 8

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 training and quality control procedures for cross-site reliability in clinical assessments. Prior to entry into the study all subjects will be evaluated for eligibility. This assessment will take place over 2 visits in most cases (total of 4-6 hours). The pretreatment evaluation will include measures listed in Table 1 and described below. The health assessment will include a physical examination and health history [including sleep, appetite, current treatment interventions]; laboratory tests (CBC, liver function tests, electrolytes, BUN, creatinine and urine for urinalysis will be collected during the screening visit. For girls of childbearing potential, blood or urine sample for Beta-HCG will be obtained to confirm a negative test prior to randomization. As noted, children taking an anticonvulsant for seizures will be excluded. For subjects taking an anticonvulsant for another purpose (e.g., as a mood stabilizer) that requires blood level tracking ( e.g., valproate), anticonvulsant level will be checked at screening, Week 12 of Double-blind phase and Week 12 of the Open-label phase. For research purposes a blood sample for DNA and for the putative blocking folate antibody will be collected at the screening visit. As noted in Table 1, if the research blood sample was not obtained at the screening, we will try to obtain the sample at Week 12 for subjects on pertinent anticonvulsant, at scheduled blood draw at endpoint (Week 24) or early termination.

Baseline Assessment In the baseline assessment, we will collect measures to complete the characterization of subjects and collect pre-treatment measures in a medication-free state. The Baseline visit will take about 3 hours. The purposes of the Screening and Baseline are to confirm eligibility, characterize the sample, and establish pretreatment measures that will be used to evaluate outcome or potential moderators of outcome. The Baseline visit will typically occur 7 to 14 days after Screening (except when taper of pre-study medication requires a longer duration). At baseline, the parent will pick up the study medication or exactly appearing placebo.

Follow up Contacts The primary caregiver will be contacted by telephone at Weeks 2, 6 and 10 to monitor adverse events. The parent and subject will return to the center at Weeks 4, 8 and 12 (endpoint in DBPC phase) (see Table 1). The blind will not be broken at the end of the DBPC phase. Following the DBPC phase, participants will be offered treatment with folinic acid in a 12-week open- label extension phase (Phase II). The medication dose will start over again (see Table 2 below). Follow up visits in Phase II will occur every 3 weeks. Clinic visits at Weeks 4 and 8 will take 45 to 60 minutes. Endpoint Visits (Week 12 of DBPC trail and Week 12 of the open trial) will 90 to 120 minutes.

Measures Subject Characterization The Autism Diagnostic Observation Schedule (ADOS) is a 40-minute investigator-based procedure that places the child in naturalistic social situations demanding specific social and communication reactions. Behaviors are coded in the areas of social communication, social relatedness, play and imagination, and restricted interests and/or repetitive behaviors. The ADOS provides a DSM-5 based algorithm to assist with the diagnosis of autism spectrum disorder.56 Social Communication Questionnaire (Lifetime) (SCQ) The SCQ is a 40-item caregiver survey that evaluates communication skills and social functioning in children who may have autism spectrum disorder. The Lifetime Form focuses on the child's entire developmental history. It provides a Total Score that can be compared to normative data using specific cutoff points. The SCQ takes 10 to 15 minutes and shows high agreement with the much longer interview called the Autism Diagnostic Interview – Revised.57 Pretreatment Health and Behavior Review This 34-item form was developed by the RUPP Autism Network to assist with the elicitation and assessment of health complaints, appetite, sleep, activity level and general health at baseline. It follows the format of the Adverse Effects review (described below) in order to establish current health status prior to randomization.25 Intellect: The Leiter-R, due to its non-verbal nature, is an appropriate measure of cognitive ability in children with ASD and language impairment.58 Children who are unable to do the Leiter will be tested with the Mullen. 9

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 Clinical Global Impression - Severity (CGI-S) This is a 7-item scale ranging from a score of 1 for Normal to 7 for Extreme.59 Although language impairment will be given particular weight in this study, independent evaluators will consider all aspects of the child’s condition to assign the CGI-S score. Adaptive Behavior: The Vineland Adaptive Behavior Scale II is a widely used standardized, well- validated assessment tool for children with developmental delays that measures functional abilities within several domains.60 The Vineland II relies on an informant (mother or primary caretaker) to complete the survey about the child’s adaptive behavior – i.e., what the child actually does in the course of daily living. We will use the parent checklist version. Child and Adolescent Symptom Inventory (CASI) is a 132-item DSM 5 based checklist that asks parents to rate the severity of symptoms in the major DSM 5 diagnostic categories on a four point scale where 0 = never; 1 = sometimes; 2 = often; and 3 = very often. Scores of 2 or higher are usually regarded a positive symptom. The CASI will assist with the identification of other DSM 5 conditions.61

Primary Outcome The primary outcome will be the change in the standard score on the Core Language battery of the Clinical Evaluation of Language Fundamentals (Core-CELF).62 The Core-CELF is comprised of 4 subtests intended to identify language problems and can be used to track progress over time. Administration of the CELF takes 30-45 minutes. The four subtests of the Core-CELF vary slightly by age. Across the entire age range of the study (> 5 to < 12 yrs 6 mos), three of the four subtests are the same: a) Concepts and Following Directions; b) Recalling Sentences; and c) Formulating Sentences. For children 5 to 8 years old, the fourth is the Word Structure subtest. For children 9 to 12 yrs6 months, the fourth is the Word Classes subtest. Despite this minor difference the Core-CELF score is a composite that includes receptive and expressive language. Using the tables in the manual, raw scores from the four subtests are compared to normative data by age. The population mean = 100 + 15. To be eligible, subjects have to have a standard score on the Core- CELF < 80 (slightly more than 1 std dev below score for age). Because we are focused on children with moderate language impairment, those who score at the floor (score of 40) will be excluded.

Standardized tests such as the Core-CELF are age and gender normed and tend to be stable over time. Typically developing children are expected to gain language skills over time. If a child keeps pace with development, the Core-CELF standard score will remain stable over time. Children with ASD, however, often show declines in the standard score over time because they do not keep pace with their typically developing age mates. By definition, subjects in this study will have moderate language delay at baseline. Gain in the standard score after twelve weeks would signal that the child is actually catching up to peers.

From a statistical point of view and as practical hedge against attrition, it could be an advantage to repeat the Core-CELF at mid-point (Week 6) and endpoint (Week 12) in the trial. However, test- retests on the Core-CELF over brief periods of time are not likely to show change. In addition, frequent re-administration of the Core-CELF would be vulnerable to practice effects. Given these issues, we decided to repeat the Core-CELF at endpoint only (see below for description on we will handle missing data).

Clinician-rated outcome assessments Clinical Global Impression for Improvement The CGI-I is a 7-point measure of overall symptomatic change compared to baseline59 that will be used as the key secondary outcome measure. Scores range from 1 (Very Much Improved) through 4 (Unchanged) to 7 (Very Much Worse). The CGI-I will be rated by a clinician who is blind to treatment assignment, adverse events and medication dose. Ratings of Much Improved or Very Much Improved on the CGI-I will be used to define positive response. In the RUPP Autism Network, we have developed a high degree of cross-site reliability with the CGI-I.25

10

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 Parent Target Problems. At baseline, parents will be asked to nominate the two most important core ASD problems for the child. Through brief discussion, the frequency (for episodic behaviors such as insists on daily routines) or constancy (for problems such as language delay or rigidity reflecting more enduring patterns), intensity and impact of the behavior on the family are established. Responses from this systematic inquiry are documented in a brief narrative. The narrative will be reviewed monthly and a new narrative documented. This review will assist with the scoring of the CGI-I in real time.25 At the end of the trial, the narratives will be reviewed by a panel blind to treatment assignment. This method has been shown to be reliable and valid in placebo-controlled studies.63

Children’s Yale-Brown Obsessive-Compulsive Scales-ASD (CYBOCS-ASD) The CYBOCS-ASD is a modified version of the CYBOCS developed for use in children with Obsessive-Compulsive Disorder.64 The modified version is a semi-structured clinician-rated scale designed to rate the current severity of repetitive behavior in children with ASD (i.e., Time Spent, Interference, Distress, Resistance, and Control). Each item is scored from 0 (least symptomatic) to 4 (most symptomatic), yielding a Total score from 0 to 20. It has established reliability and validity65 and is sensitive to change.66

Assessment of Suicide and Self-harm: Participants in this study will be young school-age children with ASD and language delay. It is likely that many will also have cognitive delay. As part of the screening procedure, the treating clinician will determine if the participant understands the concepts of death, suicide and self-harm. If the participant is deemed able to understand these concepts, the treating clinician will ask if the child has had any thoughts about wanting to die, self-harm, has plans or had actual episodes of self-harm. Any child deemed to be suicidal at screening will be excluded from the study. Assessments of suicidal thought, suicidal plan, self-harm plans or actual episodes will occur at Weeks 4, 8, 12, 15 18 and 24. The caregiver will be present during these assessments and encouraged to provide additional information.

Parent Questionnaires One parent or guardian will be identified at the screening visit and every reasonable effort will be used to ensure that this same informant completes ratings throughout the study. Ratings will be given to the parent at each visit and collected before the parent leaves. Aberrant Behavior Checklist (ABC): The ABC is a 58-item consisting five subscales: hyperactivity, irritability, social withdrawal, stereotypic behavior and inappropriate speech in children with developmental disabilities.67,68 It has been used in multiple autism clinical trials.24,25,69 Home Situations Questionnaire- Modified for ASD (HSQ-ASD): This is a parent-rated scale of child noncompliance. The parent reports on the child’s difficulties with compliance in 24 everyday situations. Questions answered affirmatively are then rated on a 1 to 9 Likert scale, with higher scores indicating more severe noncompliance. The scale yields a count of “yes” responses (0 to 27) and a severity score (total of 1 through 9 for all “yes” responses, for a range of 0 to 216). This total score is usually expressed as a per item mean (total score ÷ 24). The HSQ-ASD has established reliability, validity and sensitivity to change in children with ASD.70

Reliability The study start up meeting will be used to make certain that all sites have appropriately trained personnel to conduct the ADOS, Leiter IQ test, CELF, Vineland II, CGI-S, CGI-I, Parent Nominated Target Problems, CYBOCS-ASD and medical procedures. Drs. Scahill and McDougle have extensive experience training investigators for multisite trial in children with ASD. 24,25,69

CELF 4: The CELF 4 will be administered by trained personnel (e.g., psychologist, speech pathologist, psychometrician). Clinicians who are not speech pathologists will be trained by a speech pathologist on the administration of the CELF 4. The speech pathologist will review the first three

11

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 administrations of the CELF to confirm reliable administration and conduct quarterly conference calls to review cases and to promote a common approach to conducting assessments across sites.

This study includes 4 measures by the blinded Independent Evaluator: CGI-S, CGI-I, Parent Nominated Target Problems, CYBOCS-ASD. Training on these measures will be led by Dr. Scahill through a series of vignettes. For example, the CGI-I is a key secondary outcome measure. To be considered reliable, raters will achieve perfect agreement with gold standard on 3 of 4 training vignettes and within one unit on the fourth vignette. If IEs do not meet criterion, they will be given additional training and given additional vignettes to score until reliability is achieved. On the CYBOCS- ASD: Raters must achieve within 2 points of the gold standard (approximate margin of 15%) on the total CYBOCS-ASD score on 3 video-recorded interviews. Parent Nominated Target Problems: The procedure for eliciting and documenting the Parent Nominated Target Problems will be reviewed prior to study launch. IEs will participate in monthly conference calls to review cases and to promote a common approach to conducting assessments across sites

Early Termination Every reasonable effort will be applied to keep subjects in the trial. Parents who indicate that they are unable or uninterested in continuing the study treatment will be asked to continue coming in for assessments until Week 12 of the DBPC phase. The assessments obtained will be used in the primary data analysis. However, if it is clear that the family is unable or unwilling to return for regular visits, a premature endpoint assessment will be scheduled. Data from this premature endpoint will be carried forward to Week 12 or Week 24 (see Table 1Schedule of Measures).

Need for Subject Rescue For subjects who show deterioration during the DBPC phase, the child will be tapered off the study medication (25% every other day until discontinuation). Following the taper, the research team will confer with the child and family to discuss treatment options and referrals offered as necessary. We will maintain contact with the family for 30 days after study exit.

Genotyping: We will examine 10 polymorphisms to see if they can predict response to folinic acid treatment: MTHFR 667C>T, MTHFR 1298A>C, DHFR 19 bp del, RFC 80A>G, TCN2 776C>G, COMT 472G>A, GSTM1 Null, MTRR 66A>G, MTR 2756A>G, and SHMT 1420C>T

Medication Management Table 2. Twice daily dose schedule in mg by weight group The study medication will be obtained ≥ 20 - < 25 kg ≥ 25 kg from Lee Silsby Compounding ≥ 15 - < 20 kg Pharmacy (Cleveland Heights, OH), Study Day AM PM AM PM AM PM who will provide folinic acid and 1-13 10 - 10 10 10 10 placebo. Dye-free, milk-product-free 14 - 28 10 10 20 20 25 25 vegetarian capsules will be used. A certificate of analysis by an 29-84 20 10 20 20 25 25 independent laboratory will be obtained for each strength (10, 20 and 25 mg capsules). Our target dose is 2 mg/kg/day with a maximum of 50 mg/day, divided in two daily doses. To avoid the need for multiple capsules, parents will be instructed to give one capsule twice a day (Table 2) [Exception: children in the lowest weight group ( ≥ 15 - < 20 kg) will start with 10 mg capsule once a day for Days1-13.] Parents will be instructed to give the capsules at the same time each day. The research pharmacist will dispense the capsules in a single bottle during the DBPC and Open-label phases). [Exception: children in the lowest weight group ( ≥ 15 - < 20 kg) will receive two bottles clearly marked for the morning and evening dose from Day 29 to 84.] Study medication will be securely stored at the Investigational Pharmacy at each site in accordance with manufacturer’s instructions. This dose schedule serves as a guide and is not fixed. The treating clinician can delay a scheduled increase or reduce the dose to manage an adverse effect.

12

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 Compliance: A pill count will be conducted on returned pill bottles. Bottles have an extra 10 days of medication so they are not expected to be empty on return.

Concomitant Medications: Excluded are mineral or vitamin supplementation that exceeds the Recommended Daily Allowance limit set by the FDA and IOM and medications that interfere folate metabolism. Any other medications must be unchanged two months prior to the study and must be maintained at the current dose throughout the study.

Disposition: At the end of the double-blind phase, the treatment blind will not be broken. Subjects will be classified according to positive response or negative response (based on the CGI-I). All subjects will be offered Open-label treatment. Subjects who decline entry into the Open-label phase will be tapered off the study drug (folinic acid or placebo). The taper will reverse the does schedule in Table 2 over a week. For example, children on blinded medication at 20 mg twice a day will cut back to10 mg twice a day for 7 days and then stop. [Exception: children in the lowest weight group ( ≥ 15 - < 20 kg) will cut back to 10 mg twice a day for 4 days and then 10 mg once a day for 3 days and stop.] Children who continue in the Open-label will follow the schedule in Table 2 starting with Day 1. At the end of the Open-label phase, the child and family will meet with the treating clinician to consider treatment options. The blind will not be broken. All subjects will be tapered off the study medication as described above.

D. DATA MANGEMENT AND STATISTICS Storage and Security Study data will be managed using the UAMS Clinical Trial Management System (CRIS), a 21 CFR Part 11 compliant web-based data entry system supported by the UAMS Clinical and Translational Research Center (NCRR/NIH Grant 1 UL1 RR02988). This system was used by the NICHD-funded National Children’s Study and supports 500+ ongoing human clinical trials. We used this system in the pilot trial of folinic acid. Thus, data collection templates will only need minor modification. Study staff will be trained on CRIS by Christine Rodgers (UAMS Bioinformatics), an expert in database development and CRIS.

Statistical Analysis All data analysis for the primary study aims will be conducted by the statistics team in close collaboration with principal investigators, but will remain separate from the clinical team members. In our previous multisite trials, once analyses for the primary reports are completed, a copy of the data set will be circulated to all sites so they can pursue additional analyses.

For the inferential methods we will use general linear models (GLMs). The primary analysis will be intent-to-treat, defined as all randomized subjects. A secondary analysis will be the “per protocol” population defined as participants who adhered to protocol. Analyses will be performed with R 3.02 (http://www.r-project.org) or SAS 9.3 (SAS Institute, Cary, NC).

Sample Size Justification and Power Analysis: This study is designed to have sufficient power to detect a clinically meaningful difference between the treatment groups. The power calculation is based on change in the primary outcome (standard score on the Core-CELF). The scores on the Core-CELF will be treated as a continuous measure A sample size of 80 in each group will have 90% and 95% power to detect medium effect sizes of 0.40 and 0.45, respectively (with alpha set at 0.05 for a two-tailed test). In our pilot study, the placebo group showed a 1.7 point change. Assuming 0 to 2 points change in the placebo group and at least a 6.5 point change in the folinic acid group, we will accept a treatment difference in the range of 4.5 to 6 points on the standard score of the Core CELF as meaningful (effect size between .45 to .55). Sample size calculations were performed using nQuery Advisor software using a two-group t-test. For randomized trials, such sample size

13

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 calculations are adequate for estimating the sample size needed for a GLM given that treatment assignment tends to be uncorrelated with covariates that affect the outcome.71

Primary Hypothesis: After 12 weeks of treatment, folinic acid will be superior to placebo in improving the standard score on the Core-CELF in 160 children with ASD and moderate language delay. The primary statistical analysis will assess the effect of folinic acid compared to placebo group on our primary outcome variable: Core-CELF. GLM is equivalent to an analysis of covariance when the outcome is normally distributed. GLMs can accommodate variance heterogeneity and skewed distributions if present. The response parameter will be the change between the participant’s baseline and Week 12 score.

Secondary Hypothesis: At Week-12 the rate of overall improvement in the folinic acid group will be significantly greater than placebo as measured on the Improvement item of the Clinical Global Impression scale. The proportion of positive response on the CGI-I will be compared across treatment groups by Chi square. In a sample of 160 (80 per group), we have sufficient power to detect a 20% difference: 50% positive response rate in folinic acid group vs 30% in placebo: Chi square = 5.86, p =0.02. Treatment drop outs who return for assessments will be analyzed in their originally assigned groups. Subjects who drop out completely will be classified as negative responders.

Biomarkers of Predicting Treatment Response We will develop an equation to predict the probability of positive response to high-dose folinic acid treatment based on biomarkers of abnormalities in folate and folate-related metabolism. Participants will be classified as responder or non-responder (CGI-I of 2 or 1; “much improved” to “very much improved”).

- First, blocking and binding folate receptor autoantibodies (FRAA) will be dichotomized as negative or positive. The association between each FRAA and response will be estimated by the odds ratios (OR) of the FRAA-by-treatment using logistic regression. - Second, ten polymorphisms will be measured: MTHFR 667C>T, MTHFR 1298A>C, DHFR 19 bp del, RFC 80A>G, TCN2 776C>G, COMT 472G>A, GSTM1 Null, MTRR 66A>G, MTR 2756A>G, and SHMT 1420C>T. Polymorphism will be removed for high no call rate (>5%), low minor allele frequency (<5%) or deviation from Hardy-Weinberg equilibrium. The association between each genetic variant and the response will be estimated by computing ORs of the gene-by-treatment effect using logistic regression. - Third, a multivariable logistic regression will be developed that predicts the probability of positive response to the high-dose folinic acid treatment. Independent candidate predictors will include biomarkers selected from the two previous analyses as well as their interaction with treatment. Additional covariates such as age and baseline ASD severity and language ability will also be included. Model discrimination ability will be assessed using the C-statistic and by plotting the area under the receiver operating characteristic (ROC) curve. The ROC curve will be used to determine cutoff value with the highest sensitivity and specificity.

Exploratory Analyses: Other secondary outcomes will include: change in repetitive behavior (CYBOCS-ASD), change in Irritability, Social Withdrawal and Hyperactivity subscales of the Aberrant Behavior Checklist between treatment groups at week 12 of the DBPC phase. We will also conduct exploratory moderator analysis with candidate moderators: IQ (< 70 vs > 70), gender, racial category (e.g., white vs minority), CGI-S (Moderate vs > Moderate), baseline scores on indices of ASD severity, behavioral problems and anxiety. For ease of presentation and interpretation, continuous moderators will be split above and below the 75th percentile to demonstrate differences in magnitudes of treatment effects for each candidate moderator. We will also look at baseline Vineland, and CELF scores. On the Vineland and CELF, higher scores indicate better functioning – so we will explore other cut points to classify subjects into dichotomized groups (e.g., median split). 14

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 Safety analyses: Adverse events (AE) will be systematically monitored using a 34-item form that was developed by the RUPP Autism Network to assist with the elicitation and assessment of adverse effects. It includes specific queries about major body systems, activity level, sleep, appetite, and general health. New events, whether presumed related to treatment or not, are classified as AEs and rated as Mild, Moderate or Severe (see details below). The occurrence of an AE will be counted once at the highest level of severity (e.g., a report of mild nausea followed by a report of moderate nausea in the same child would counted as an occurrence of moderate nausea). The frequency of AEs by severity will be evaluated using a Fisher’s exact test. We will pursue exploratory analyses such as survival analyses using time to AE as the response variable.

Missing Data: In our recently completed six-month randomized trial of Parent Training in young children with ASD (NIMH grant MH081148; Scahill, PI) attrition was less than 7% (13 of 180).72 To minimize attrition and missing data in this study, we will apply similar procedures. First, during the consent process we will engage parents in a direct discussion about random assignment, the 12-week duration of the trial, and the opportunity to receive folinic acid in the open extension phase. This discussion allows expression of parental misgivings about the study and clarity about the study demands. Second, we distinguish between subjects who drop out completely and those who agree to return for assessments and remain on study through Week 12 – even if the subject is no longer taking the assigned study medication. In keeping with the intent-to-treat principle, we will encourage subjects and families who encounter barriers to remain on study for assessment. Finally, for parents who indicate intention to withdraw the child completely from the randomized trial, we will conduct an early termination visit to ensure a post-randomization assessment of the primary and key secondary outcomes.73

Missing data will be checked for data entry errors. All reasons for drop out and missing data will be documented. The pattern of missing data due to drop outs will be examined by comparing subjects with missing data to subjects with no missing data. If substantial differences in missingness occur across treatment groups that cannot be adequately explained by observed variables (data not missing at random), secondary sensitivity analyses employing the methods described by Little and Wang (1996)74 and Kenward and Carpenter (2007)75 will be employed. If necessary, we will use multiple imputation strategies to impute missing values. We will use the following steps: Five complete datasets will be created using PROC MI in SAS®. Each of these datasets will be analyzed using regular analytical methods. The inferences will be combined using PROC MIANALYZE in SAS®.

E. HUMAN SUBJECTS These inclusion and exclusion criteria described above reflect our best estimate for subject selection in this clinical trial. In our past RUPP Autism Network trials, we rarely encountered subjects who did not satisfy all selection criteria, but, in the judgment of the research team, appeared appropriate for the study. For example, a healthy, developmentally disabled child who was not toilet trained could not provide a urine sample. Therefore, we could not say with certainty that his urine test was “normal.” To exclude such a child would threaten the generalizability of the study findings and might systematically exclude lower-functioning children. In such cases, a cross-site Steering Committee will review each case, and the exception will be documented. This procedure worked well in the RUPP trials, and we will continue this procedure in this multisite trial. The Steering Committee will convene weekly by conference calls. Thus, subject-specific exceptions to the protocol specifications will not be determined unilaterally by a single site. In making such decisions, the Steering Committee will insure that the exception 1) does not decrease the benefit-risk ratio for the subject in question; 2) does not compromise the scientific integrity of the study; and 3) does not depart from good clinical practice. Such exceptions will be coded as approved protocol deviations and will be documented in the database and reported to the DSMB (below).

Sources of Materials 15

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 The sources of the research material will be data collected in the course of the clinical trial to determine the efficacy and also to ensure the safety of study participants. The principal types of data will be clinical rating scales and standardized tests (e.g., IQ test and the CELF-4). Routine laboratory measures (blood chemistry, blood cell counts, urinalysis) and research blood sample will also be obtained.

Potential Risks The risks associated with these procedures include exposure to the study drug or placebo, and the risk associated with venipuncture. The blood draws poses a possible risk of bruising at the site of the draw, light-headedness and fainting along with a remote chance of infection.

Folinic acid is closely related molecule to the B vitamin, folate. It is approved to help with the adverse of effects of certain medications such as methotrexate, which depletes folate in the body.76 Folinic acid is also approved to be used in combination with chemotherapeutic agents in certain cancers.77 The identification of side effects of folinic acid is somewhat confounded by the combined use of potent chemotherapeutic drugs. Adverse effects of folinic acid may include: skin rashes, itching, facial flushing, nausea or vomiting. Acute allergic reactions have been reported – but considered rare.

Although highly unlikely, there are potential privacy risks associated with participating in such a study. There is minimal risk associated with the completion of rating scales and assessment measures.

Protection from Risks & Discomforts Blood draws are common procedures and will be performed by a certified phlebotomist. We will offer a numbing cream prior to the blood drawing procedure. We will systematically review adverse at follow up phone calls and clinic visits (see below). Parents will be given phone numbers of the principal investigator and study physician – who wil be available after hours. The PI and study physician have access to the pharmacy after hours – if it is necessary to break the blind.

The proposed dose of folinic acid in this study is higher than when it is used in combination with chemotherapeutic drugs. At the end of the double-blind phase, subjects on folinic acid (known to pharmacist) will have the drug dose reduced by in half for day one of the open-label extension. The subjects on placebo will start with the same dose (Table 2 above). At the end of the open-label extension, the dose will be reduced by half immediately and then by half again after 4 days and finally stop after 4 more days. There is potential harm to a participant on placebo as a result of not receiving active medicine for ASD. However, there is no approved treatment for core features of ASD. Thus, no effective treatment is being withheld. This point will be made clear in the consent process.

Confidentiality will be maintained throughout the project period. With the exception of tests that require date of birth, research records will not include names or contact information. Subjects will be identified only by their code numbers. A file with personal health information (e.g., names and addresses) will be kept separately in a locked file cabinet.

Recruitment All recruitment materials such as flyers, letters and social media posts will be IRB approved. Each site will use site-specific means for recruitment including but not limited to posting to their local website, contacting clinical populations, using social media and contacting local organizations, physicians and schools.

Compensation Subjects will be compensated for study participation according to usual practice at each site. Incentive will be offered for the following visits: screening, baseline and at Weeks 4, 8, 12, 18 and 24. This 16

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 incentive will be $50 at the Emory and Arkansas Site a payment of $50 and $25 at the Harvard site per the local standard incentive rate.

Consent Process Interested families will be screened on the telephone by the study coordinator. Presumably eligible and willing families will be invited to the site for a formal screening visit. Parental consent will be obtained at the screening visit prior to the collection of any study data. The consent procedure will include the following: (a) provide the consent forms to parents and guardians together with a verbal summary of the contents of all paragraphs, and (b) a description of alternatives to participating in the study (not to take part, use of other treatments, and obtaining similar treatment through other clinical services). Parents (guardians) will also be told that declining participation will not prejudice their right to pursue other treatments at the local site or any of its affiliates.

Parental consent will be obtained by trained investigators as documented on each sites delegation of authority log. Children whose parents indicate notable barriers to meeting the demands of the study or whose parents lack capacity to understand the purpose and expectations of the study will not be randomized in the study. These judgments will be made by the investigator conducting the consent discussion in consultation with the research team as needed. Parents will be given the opportunity to ask questions. Once all questions have been answered, parent(s) will be asked to sign and date the consent form. Parents will be provided with a copy of the singed consent form.

Child Assent The children in this study (ages 5 to12 yrs 6 months) will be diagnosed with ASD and language delay. Some will also be cognitively delayed. The site PI or designated Co-investigator conducting the consent procedure will evaluate the child’s capacity to provide written or verbal assent. The investigator’s judgment will be documented on the assent form.

Consent for Video Recording Consent for video recording for research purposes will be obtained from parents. If parents are willing, video recordings will be used for training purposes (e.g., presentations at professional meetings). Parents can indicate agreement or disagreement with this additional use of the videos directly on the video consent form. Agreement to use the videos for these additional training purposes is not required for study participation.

Study Withdrawal Parents are free to withdraw from the study at any time. This will be clearly stated during the consent process. Subjects may need to discontinue participation in the study before completion for various reasons (e.g., family move, withdrawal of consent, adverse events, need for a different treatment). Parents who indicate intention to withdraw the child from the study will be offered to have a case conference with the study team to discuss the matter. If the discussion indicates that the child needs another treatment, we will assist the family to locate that treatment. Parents who drop out of treatment will be invited to return for scheduled assessments or an early termination visit. Data collected from children who drop out of treatment, but return for assessments, will be analyzed in their originally randomized group.

CONFIDENTIALITY Case report forms (CRFs) contain the subject’s unique ID number – but no identifying information. UAMS Clinical Trial Management System (CRIS) is a secure, password protected data base. The folders containing CRFs data will be kept in locked files and access to these files is only granted to members of the research team. Standardized tests (e.g. IQ and language tests) record the child’s date of birth in order to calculate the child’s age for comparison to normative data. No other personal health information will be documented on research forms. Contact information for the family is kept in a separate file that is in a locked cabinet. Certain measures will be recorded on video. The digital files are stored on a secure, password- protected site. Recordings may also be stored electronically on an encrypted external hard drive that 17

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 is kept in a locked closet or cabinet. The digital files will be marked with a Subject ID instead of names of children or their parents. It is possible that names may be mentioned on the recordings. Video recordings from this research study will be retained for up to 5 years after the study is over. Recordings may be used for other purposes (e.g. presentations at professional meetings) if separate consent is obtained from the parent (or guardian).

Access to study records (case report forms) will be restricted to study staff. Others that have access to the record include individuals with regulatory responsibility at Emory, Children’s Health Care of Atlanta (CHOA). All clinical trials are subject to routine audits by offices at Emory or CHOA (or similar offices at the University of Arkansas or Harvard University). Others that may review research records include study monitors, representatives from National Institute of Child Health and Development. We are also required to submit de-identified study data to the National Database for Autism Research (NDAR).

Potential Benefits of the Proposed Research to the Subjects and Others Subjects may directly benefit by the receipt of folinic acid. Subjects initially randomized to placebo will be offered treatment with folinic acid in the 12-week, open-label extension trial. Overall, the risks to the subjects in this research are similar to those that exist in standard clinical practice and the proposed research project offers considerable promise in improving treatment of moderate language delay in children with ASD. Indeed, this research may lead to a new treatment option for children with ASD and language delay. Thus, the potential benefits of the research outweigh the risks involved.

Importance of the Knowledge To Be Gained Children with ASD with language delay are impaired and, untreated, their long-term prognosis is guarded. Although current treatments are helpful in some cases, significant gaps in our treatment options remain. There are no large, controlled clinical trials of folinic acid in children with ASD and no medication treatments for core features of ASD or language delay. This study may provide new information about language delay in children with ASD and may identify subgroups of affected children who could benefit from this treatment.

Data Safety and Monitoring Plan The following is a general description of the monitoring plan and our overall framework for data and safety monitoring. We will empanel an external DSMB to monitor this multi-site clinical trial. If the NIMH DSMB does not provide one, the Data Coordinating Center at Emory University will convene a DSMB for this study. This Board includes members with appropriate medical and statistical expertise who are not involved in this study. We will invite a pediatric neurologist, a child psychiatrist, and a parent from Autism Speaks to serve on the DSMB. The unblinded statistician will serve as an ex officio member. This DSMB will meet by teleconference 3 times in Year 1; 4 times a year in Years 2 through 4 and 3 times a year in Year 5. Reports on recruitment, recruitment by site, adverse events, attrition, etc will be generated by the unblinded statistician prior to each review (see below).

Monitoring for Adverse Events (AEs): At each visit, including baseline and interim contacts, the treating clinician will systematically review for adverse effects and concomitant medications using an Adverse Effects Review form developed for this study. We have used versions of this form, in several multi-site clinical trials in children with autism (RUPP Autism Network, 2002; Autism Network, 2005b; Scahill et al., 2006; Aman et al., 2009; Scahill et al., 2015). It contains a general inquiry, drug-specific queries, and several questions about daily activities (e.g., sleep, appetite, energy level, bowel and bladder functions). The general inquiry includes an open-ended question about any problems or complaints, as well as questions regarding the need for other medications and doctor or health care encounters since the last study visit. The next section includes drug-specific queries, probing for adverse events that have been reported in the scientific literature. These items are rated as not present, mild, moderate or severe. The ratings: mild, moderate or severe are defined as follows: mild= present, but no intervention required; 18

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 moderate=present, may be bothersome or may require intervention; severe=present, bothersome and requires intervention. The last section includes specific questions about daily activities. All new adverse events (mild, moderate or severe) will be documented on the adverse event log. The status of previously-reported adverse events will be monitored as well. In the prior RUPP Autism Network studies, the form was easily modified to make it relevant for risperidone (RUPP Autism Network, 2002), methylphenidate (RUPP Autism Network, 2005b) and guanfacine (Scahill wet al., 2015). We will include queries about self-harm or expressions of suicidal thoughts or plan.

When the treating clinician elicits an adverse event (AE), it will be documented regardless of suspected relationship to the study drug. The adverse event log requires the treating clinician to label the AE (using a list of preferred terms and a numeric code) and to document the severity, onset, course, outcome, and attribution (study drug related or not).

Attribution is classified as follows:  Definite: AE is clearly related to the study drug.  Probable: AE is likely to be related to the study drug.  Possible: AE may be related to the study drug.  Unlikely: AE is doubtfully related to the study drug.  Unrelated: AE is clearly not related to the study drug. Severity of AEs is classified as follows:  mild= present, but no intervention required;  moderate = present, may be bothersome or may require intervention;  severe = present, bothersome, requires intervention and may produce sequelae.

Decisions on the appropriate care of the subject will be made by the treating clinician (who will remain blind to treatment assignment), the local PI, and in collaboration with the family’s primary care provider whenever possible. Abnormal findings on laboratory tests that exceed established alarm values will be investigated to determine the nature of the adverse events.

Routine reporting of AEs Adverse events will be documented on the Adverse Event Log as described above. All AE data will be captured in the electronic database and reviewed at the DSMB meetings. The DSMB will report to the investigators any AEs occurring at a greater than expected frequency. Investigators will provide this information to local IRBs at the time of annual renewal. The multi-site steering committee will decide if modifications to the protocol or consent form are required.

AEs resulting in treatment discontinuation: Adverse events that are the primary reason for study discontinuation will be tracked separately. The Data Coordinating Center will provide a table with all such side effects to the DSMB, four times a year. High rates of such events in the folinic acid group compared to the placebo group will prompt evaluation of potential significant safety concerns in the trial. Rates of AEs will also be monitored during the open-label extension trial.

Serious adverse events: A serious adverse event (SAE) is defined as an event that poses a threat to the participant’s life or functioning. We note that “severe” is not necessarily equivalent to “serious.” Thus, a severe rash is not a serious adverse event, whereas a heart attack of any severity is likely to be a serious adverse event. A serious adverse event is defined as any event that entails one of the following: • Death; • Threat to the individual’s life; • Inpatient hospitalization or prolongation of existing hospitalization; • Persistent or significant disability or incapacity; • Intentional drug overdose; • Any other significant event that jeopardizes the participant. 19

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 Assessment of Serious Adverse Events The PI at the local site will send preliminary notice of SAEs to the Coordinating Center (CC) and other PIs within 24 hours. With help from the local PI, the data (CC) will distribute a narrative description of the SAE to all PIs within 48 hours of receiving notice from the site PI where the SAE occurred.

The Steering Committee will convene a teleconference to review the SAE. The SAE will be classified as: a) serious AND anticipated; b) serious AND anticipated, BUT occurring with a greater frequency than expected; c) serious AND unanticipated. The Steering Committee will deliberate on whether the SAE is unrelated, possibly, probably or definitely related to the study treatment. SAEs that are unanticipated or anticipated SAEs that occur at a greater frequency than expected and deemed at least possibly related to the study treatment will be reported with seven days following this determination by the Steering Committee. The report will be sent to study IRBs, NICHD, DSMB. The report will be sent to company and the FDA within 14 days.

SAEs that do not meet either of these criteria will be included in DSMB reports and annual reports to the IRBs.

REFERENCES 1. Developmental Disabilities Monitoring Network Surveillance Year Principal I, Centers for Disease C, Prevention. Prevalence of autism spectrum disorder among children aged 8 years - autism and developmental disabilities monitoring network, 11 sites, United States, 2010. Morbidity and mortality weekly report. Surveillance summaries. Mar 28 2014;63(2):1-21. PMCID: 2. Szatmari P. The causes of autism spectrum disorders. BMJ. Jan 25 2003;326(7382):173-174. PMCID: 1125046 3. Elsabbagh M, Divan G, Yun-Joo Koh YJ, Kim YS, Kauchali S, Marcín C, Montiel-Nava C, Patel V, Paula CS, Wang C, Yasamy MT, Fombonne E: Global Prevalence of Autism and Other Pervasive Developmental Disorders. Autism Res 2012; 5(3):160–179 4. American Psychiatric Association . Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994. 5. Chaidez V, Hansen RL, Hertz-Picciotto I. Gastrointestinal problems in children with autism, developmental delays or typical development. Journal of autism and developmental disorders. May 2014;44(5):1117-1127. PMCID: 3981895 6. Frye RE, Rossignol D, Casanova MF, Brown GL, Martin V, Edelson S, Coben R, Lewine J, Slattery JC, Lau C, Hardy P, Fatemi SH, Folsom TD, Macfabe D, Adams JB. A review of traditional and novel treatments for seizures in autism spectrum disorder: findings from a systematic review and expert panel. Frontiers in public health. 2013;1:31. PMCID: 3859980 7. Taurines R, Schwenck C, Westerwald E, Sachse M, Siniatchkin M, Freitag C. ADHD and autism: differential diagnosis or overlapping traits? A selective review. Attention deficit and hyperactivity disorders. Sep 2012;4(3):115-139. PMCID: 8. Hallett, V., Lecavalier,L., Sukhodolsky, D.G., Cipriano,N., Aman, M.G., McCracken, J.T., McDougle,C.J., Tierney,E., King,B.H.,Hollander, E.,Sikich,L., Bregman,J., Anagnostou,E. Donnelly,C.,Katsovich, L.,Dukes, K.,Vitiello,B., Gadow, K., & Scahill, L. (2013). Exploring the manifestations of anxiety in children with Autism Spectrum Disorders. Journal of Autism and Developmental Disorders, 43 (10), 2341-2352 9. Angelidou A, Alysandratos KD, Asadi S, Zhang B, Francis K, Vasiadi M, Kalogeromitros D, Theoharides TC. Brief report: "allergic symptoms" in children with Autism Spectrum Disorders. More than meets the eye? Journal of Autism and Developmental Disorders. Nov 2011;41(11):1579-1585. PMCID: 10. Fein D, Barton M, Eigsti IM, Kelley E, Naigles L, Schultz RT, Stevens M, Helt M, Orinstein A, Rosenthal M, Troyb E, Tyson K. Optimal outcome in individuals with a history of autism. 20

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 Journal of child psychology and psychiatry, and allied disciplines. Feb 2013;54(2):195-205. PMCID: 3547539 11. Fountain C, Winter AS, Bearman PS. Six developmental trajectories characterize children with autism. Pediatrics. May 2012;129(5):e1112-1120. PMCID: 3340586 12. Tonge BJ, Bull K, Brereton A, Wilson R. A review of evidence-based early intervention for behavioural problems in children with autism spectrum disorder: the core components of effective programs, child-focused interventions and comprehensive treatment models. Current opinion in psychiatry. Mar 2014;27(2):158-165. PMCID: 13. Reichow B, Barton EE, Boyd BA, Hume K. Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). The Cochrane database of systematic reviews. 2012;10:CD009260. PMCID: 14. Karst JS, Van Hecke AV. Parent and family impact of autism spectrum disorders: a review and proposed model for intervention evaluation. Clinical child and family psychology review. Sep 2012;15(3):247-277. PMCID: 15. Broder-Fingert S, Shui A, Pulcini CD, Kurowski D, Perrin JM. Racial and ethnic differences in subspecialty service use by children with autism. Pediatrics. Jul 2013;132(1):94-100. PMCID: 16. Magiati I, Tay XW, Howlin P. Cognitive, language, social and behavioural outcomes in adults with autism spectrum disorders: a systematic review of longitudinal follow-up studies in adulthood. Clin Psychol Rev. Feb 2014;34(1):73-86. PMCID: 17. Ghosh A, Michalon A, Lindemann L, Fontoura P, Santarelli L. Drug discovery for autism spectrum disorder: challenges and opportunities. Nature reviews. Drug discovery. Oct 2013;12(10):777-790. PMCID: 18. Scahill L, Hallett V, Aman MG, McDougle CJ, Eugene Arnold L, McCracken JT, Tierney E, Deng Y, Dziura J, Vitiello B, Research Units on Pediatric Psychopharmacology Autism N. Brief Report: social disability in autism spectrum disorder: results from Research Units on Pediatric Psychopharmacology (RUPP) Autism Network trials. Journal of autism and developmental disorders. Mar 2013;43(3):739-746. PMCID: 3886122 19. Correll CU, Manu P, Olshanskiy V, Napolitano B, Kane JM, Malhotra AK. Cardiometabolic risk of second-generation antipsychotic medications during first-time use in children and adolescents. JAMA : the journal of the American Medical Association. Oct 28 2009;302(16):1765-1773. PMCID: 3055794 20. Scahill, L., Jeon, S., Boorin, S.G., McDougle, C.J., Aman, M.G., Dziura, J., McCracken, J.T., Caprio, S., Arnold, L.E., Nicol, G., Deng, Y., Challa, S.A., and Vitiello, B. (in press). Weight and metabolic consequences following 6 months exposure to risperidone in young children with autism spectrum disorder and serious behavioral problems. Journal of the American Academy of Child and Adolescent Psychiatry 21. Bobo WV, Cooper WO, Stein CM, Olfson M, Graham D, Daugherty J, Fuchs DC, Ray WA. Antipsychotics and the risk of type 2 diabetes mellitus in children and youth. JAMA psychiatry. Oct 2013;70(10):1067-1075. PMCID: 22. Correll CU, Kane JM. One-year incidence rates of tardive dyskinesia in children and adolescents treated with second-generation antipsychotics: a systematic review. Journal of child and adolescent psychopharmacology. Oct 2007;17(5):647-656. PMCID: 23. Mahajan R, Bernal MP, Panzer R, Whitaker A, Roberts W, Handen B, Hardan A, Anagnostou E, Veenstra-VanderWeele J, Autism Speaks Autism Treatment Network Psychopharmacology C. Clinical practice pathways for evaluation and medication choice for attention- deficit/hyperactivity disorder symptoms in autism spectrum disorders. Pediatrics. Nov 2012;130 Suppl 2:S125-138. PMCID: 24. Research Units on Pediatic Psychopharmacology (RUPP) Autism Network Randomized, controlled, crossover trial of methylphenidate in pervasive developmental disorders with hyperactivity. Arch Gen Psychiatry. 2005;62(11):1266-1274. PMCID: 25. Scahill L, McCracken JT, King BH, Rockhill C, Shah B, Politte L, Sanders R, Minjarez M, Cowen J, Mullett J, Page C, Ward D, Deng Y, Loo S, Dziura J, McDougle CJ, Research Units 21

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 on Pediatric Psychopharmacology Autism N. Extended-Release Guanfacine for Hyperactivity in Children With Autism Spectrum Disorder. Am J Psychiatry. Aug 28 2015:appiajp201515010055. PMCID: 26. Webb S. Drugmakers dance with autism. Nature biotechnology. Aug 2010;28(8):772-774. PMCID: 27. Frye RE. Clinical potential, safety, and tolerability of arbaclofen in the treatment of autism spectrum disorder. Drug, healthcare and patient safety. 2014;6:69-76. PMCID: 4025936 28. Frye RE, Rossignol DA. Treatments for biomedical abnormalities associated with autism spectrum disorder. Frontiers in pediatrics. 2014;2:66. PMCID: 4073259 29. Frye RE, Sequeira JM, Quadros EV, James SJ, Rossignol DA. Cerebral folate receptor autoantibodies in autism spectrum disorder. Molecular psychiatry. Mar 2013;18(3):369-381. PMCID: 3578948 30. Frye RE, Melnyk S, Fuchs G, Reid T, Jernigan S, Pavliv O, Hubanks A, Gaylor DW, Walters L, James SJ. Effectiveness of Methylcobalamin and Folinic Acid Treatment on Adaptive Behavior in Children with Autistic Disorder Is Related to Glutathione Redox Status. Autism Research and Treatment. 2013;2013:609705. PMCID: 31. Mukaddes NM, Tutkunkardas MD, Sari O, Aydin A, Kozanoglu P. Characteristics of children who lost the diagnosis of autism: a sample from Istanbul, Turkey. Autism Res Treat. 2014;2014:472120. PMCID: 4022169 32. Tager-Flusberg H, Rogers S, Cooper J, Landa R, Lord C, Paul R, Rice M, Stoel-Gammon C, Wetherby A, Yoder P. Defining spoken language benchmarks and selecting measures of expressive language development for young children with autism spectrum disorders. Journal of speech, language, and hearing research : JSLHR. Jun 2009;52(3):643-652. PMCID: 2819321 33. Luyster R, Qiu S, Lopez K, Lord C. Predicting outcomes of children referred for autism using the MacArthur-Bates Communicative Development Inventory. Journal of speech, language, and hearing research : JSLHR. Jun 2007;50(3):667-681. PMCID: 34. Szatmari P, Bryson SE, Boyle MH, Streiner DL, Duku E. Predictors of outcome among high functioning children with autism and Asperger syndrome. Journal of child psychology and psychiatry, and allied disciplines. May 2003;44(4):520-528. PMCID: 35. Anderson DK, Lord C, Risi S, DiLavore PS, Shulman C, Thurm A, Welch K, Pickles A. Patterns of growth in verbal abilities among children with autism spectrum disorder. J Consult Clin Psychol. Aug 2007;75(4):594-604. PMCID: 36. Tilford JM, Payakachat N, Kovacs E, Pyne JM, Brouwer W, Nick TG, Bellando J, Kuhlthau KA. Preference-based health-related quality-of-life outcomes in children with autism spectrum disorders: a comparison of generic instruments. PharmacoEconomics. Aug 1 2012;30(8):661- 679. PMCID: 3423960 37. Lai MC, Lombardo MV, Ecker C, Chakrabarti B, Suckling J, Bullmore ET, Happe F, Consortium MA, Murphy DG, Baron-Cohen S. Neuroanatomy of Individual Differences in Language in Adult Males with Autism. Cerebral cortex. Sep 23 2014. PMCID: 38. Greenblatt JM, Huffman LC, Reiss AL. Folic acid in neurodevelopment and child psychiatry. Progress in neuro-psychopharmacology & biological psychiatry. Jul 1994;18(4):647-660. PMCID: 39. Black MM. Effects of vitamin B12 and folate deficiency on brain development in children. Food and nutrition bulletin. Jun 2008;29(2 Suppl):S126-131. PMCID: 3137939 40. Neale BM, Kou Y, Liu L, Ma'ayan A, Samocha KE, Sabo A, Lin CF, Stevens C, Wang LS, Makarov V, Polak P, Yoon S, Maguire J, Crawford EL, Campbell NG, Geller ET, Valladares O, Schafer C, Liu H, Zhao T, Cai G, Lihm J, Dannenfelser R, Jabado O, Peralta Z, Nagaswamy U, Muzny D, Reid JG, Newsham I, Wu Y, Lewis L, Han Y, Voight BF, Lim E, Rossin E, Kirby A, Flannick J, Fromer M, Shakir K, Fennell T, Garimella K, Banks E, Poplin R, Gabriel S, DePristo M, Wimbish JR, Boone BE, Levy SE, Betancur C, Sunyaev S, Boerwinkle E, Buxbaum JD, Cook EH, Jr., Devlin B, Gibbs RA, Roeder K, Schellenberg GD, Sutcliffe JS, 22

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 Daly MJ. Patterns and rates of exonic de novo mutations in autism spectrum disorders. Nature. May 10 2012;485(7397):242-245. PMCID: 3613847 41. Sanders SJ, Murtha MT, Gupta AR, Murdoch JD, Raubeson MJ, Willsey AJ, Ercan-Sencicek AG, DiLullo NM, Parikshak NN, Stein JL, Walker MF, Ober GT, Teran NA, Song Y, El-Fishawy P, Murtha RC, Choi M, Overton JD, Bjornson RD, Carriero NJ, Meyer KA, Bilguvar K, Mane SM, Sestan N, Lifton RP, Gunel M, Roeder K, Geschwind DH, Devlin B, State MW. De novo mutations revealed by whole-exome sequencing are strongly associated with autism. Nature. May 10 2012;485(7397):237-241. PMCID: 3667984 42. Adams M, Lucock M, Stuart J, Fardell S, Baker K, Ng X. Preliminary evidence for involvement of the folate gene polymorphism 19bp deletion-DHFR in occurrence of autism. Neuroscience letters. Jul 5 2007;422(1):24-29. PMCID: 43. James SJ, Melnyk S, Jernigan S, Cleves MA, Halsted CH, Wong DH, Cutler P, Bock K, Boris M, Bradstreet JJ, Baker SM, Gaylor DW. Metabolic endophenotype and related genotypes are associated with oxidative stress in children with autism. American journal of medical genetics. Part B, Neuropsychiatric genetics : the official publication of the International Society of Psychiatric Genetics. Dec 5 2006;141B(8):947-956. PMCID: 2610366 44. Boris M, Goldblatt A, Galanko J, James SJ. Association of MTHFR gene variants with autism. Journal of American Physicians and Surgeons. 2004;9(4):106-108. PMCID: 45. Mohammad NS, Jain JM, Chintakindi KP, Singh RP, Naik U, Akella RR. Aberrations in folate metabolic pathway and altered susceptibility to autism. Psychiatric genetics. Aug 2009;19(4):171-176. PMCID: 46. Guo T, Chen H, Liu B, Ji W, Yang C. Methylenetetrahydrofolate reductase polymorphisms C677T and risk of autism in the Chinese Han population. Genetic testing and molecular biomarkers. Aug 2012;16(8):968-973. PMCID: 47. Schmidt RJ, Tancredi DJ, Ozonoff S, Hansen RL, Hartiala J, Allayee H, Schmidt LC, Tassone F, Hertz-Picciotto I. Maternal periconceptional folic acid intake and risk of autism spectrum disorders and developmental delay in the CHARGE (CHildhood Autism Risks from Genetics and Environment) case-control study. The American journal of clinical nutrition. Jul 2012;96(1):80-89. PMCID: 3374734 48. Liu X, Solehdin F, Cohen IL, Gonzalez MG, Jenkins EC, Lewis ME, Holden JJ. Population- and family-based studies associate the MTHFR gene with idiopathic autism in simplex families. Journal of autism and developmental disorders. Jul 2011;41(7):938-944. PMCID: 49. Goin-Kochel RP, Porter AE, Peters SU, Shinawi M, Sahoo T, Beaudet AL. The MTHFR 677C-- >T polymorphism and behaviors in children with autism: exploratory genotype-phenotype correlations. Autism research : official journal of the International Society for Autism Research. Apr 2009;2(2):98-108. PMCID: 50. Pasca SP, Dronca E, Kaucsar T, Craciun EC, Endreffy E, Ferencz BK, Iftene F, Benga I, Cornean R, Banerjee R, Dronca M. One carbon metabolism disturbances and the C677T MTHFR gene polymorphism in children with autism spectrum disorders. Journal of cellular and molecular medicine. Oct 2009;13(10):4229-4238. PMCID: 51. Pu D, Shen Y, Wu J. Association between MTHFR gene polymorphisms and the risk of autism spectrum disorders: a meta-analysis. Autism research : official journal of the International Society for Autism Research. Oct 2013;6(5):384-392. PMCID: 52 Ramaekers VT, Hausler M, Opladen T, Heimann G, Blau N. Psychomotor retardation, spastic paraplegia, cerebellar ataxia and dyskinesia associated with low 5-methyltetrahydrofolate in cerebrospinal fluid: a novel neurometabolic condition responding to folinic acid substitution. Neuropediatrics. Dec 2002;33(6):301-308. PMCID: 53. Ramaekers VT, Rothenberg SP, Sequeira JM, Opladen T, Blau N, Quadros EV, Selhub J. Autoantibodies to folate receptors in the cerebral folate deficiency syndrome. The New England journal of medicine. May 12 2005;352(19):1985-1991. PMCID: 54. Ramaekers VT, Quadros EV, Sequeira JM. Role of folate receptor autoantibodies in infantile autism. Molecular psychiatry. Mar 2013;18(3):270-271. PMCID: 23

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 55. Kernan WN, Viscoli CM, Makuch RW, Brass LM, Horwitz RI. Stratified randomization for clinical trials. J Clin Epidemiol. Jan 1999;52(1):19-26. PMCID: 56. Gotham K, Pickles A,Lord C. Standardizing ADOS Scores for a Measure of Severity in Autism Spectrum Disorders. J Autism Dev Disord, 2009; 39:693–705 57. Corsello C, Hus V, Pickles A, Risi S, Cook EH Jr, Leventhal BL, Lord C: Between a ROC and a hard place: decision making and making decisions about using the SCQ. J Child Psychol Psychiatry 2007; 48(9):932-40 58. Roid GH, Miller LJ. Leiter International Performance Scale – Revised. Wood Dale, IL: Stoelting Co; 1997. 59. Guy W (1976): ECDEU Assessment Manual for Psychopharmacology: Publication ADM 76- 338. Washington, DC, US Department of Health, Education, and Welfare, pp 218–222 60. Sparrow S, Cicchetti D, Balla D. Vineland Adaptive Behavior Scales (2nd ed.). Minneapolis, MN2005. 61. Gadow KD, De Vincent C, Pomeroy J, Azizian A: Comparison of DSM-IV symptoms in elementary school-aged ADHD symptoms in children with PDD versus clinic and community samples. Autism 2005; 9(4): 392–415 62. Semel E, Wiig EH, Secord WA. Clinical evaluation of language fundamentals, fourth edition (CELF-4). Toronto, Canada: The Psychological Corporation/A Harcourt Assessment Company; 2003. 63. Arnold LE, Vitiello B, McDougle C, Scahill L, Shah B, Gonzalez NM, Chuang S, Davies M, Hollway J, Aman MG, Cronin P, Koenig K, Kohn AE, McMahon DJ, Tierney E. Parent-defined target symptoms respond to risperidone in RUPP autism study: customer approach to clinical trials. J Am Acad Child Adolesc Psychiatry. Dec 2003;42(12):1443-1450. PMCID: 64. Scahill L, Riddle MA, McSwiggin-Hardin M, Ort SI, King RA, Goodman WK, Cicchetti D, Leckman JF. Children's Yale-Brown Obsessive Compulsive Scale: reliability and validity. J Am Acad Child Adolesc Psychiatry. Jun 1997;36(6):844-852. PMCID: 65. Scahill L, McDougle CJ, Williams SK, Dimitropoulos A, Aman MG, McCracken JT, Tierney E, Arnold LE, Cronin P, Grados M, Ghuman J, Koenig K, Lam KS, McGough J, Posey DJ, Ritz L, Swiezy NB, Vitiello B, Research Units on Pediatric Psychopharmacology Autism N. Children's Yale-Brown Obsessive Compulsive Scale modified for pervasive developmental disorders. J Am Acad Child Adolesc Psychiatry. Sep 2006;45(9):1114-1123. PMCID: 66. Scahill, L., Sukhodolsky, D.G., Anderberg, E., Dimitropoulos, A., Dziura, J., Aman, M.G., McCracken, J.T., Tierney, E., Hallett, V., Katz, K., Vitiello, B., McDougle, C.J. (2015). Sensitivity of the modified Children’s Yale-Brown Obsessive Compulsive Scale to detect change: results from two multi-site trials. , Autism: The International Journal of Research and Practice, Vol. 19(1) 38–52, DOI: 10.1177. 67. Aman MG, Singh NN, Stewart AW, Field CJ. The aberrant behavior checklist: a behavior rating scale for the assessment of treatment effects. Am J Mental Deficiency. Mar 1985;89(5):485-491. PMCID: 68. Kaat AJ, Lecavalier L, Aman MG. Validity of the Aberrant Behavior Checklist in children with autism spectrum disorder. J of Autism Dev disorders. May 2014;44(5):1103-1116. PMCID: 69. Research Units on Pediatric Psychopharmacology (RUPP)Autism Network (2002). Risperidone in children with autism for serious behavioral problems. New England Journal of Medicine, 347(5), 314-321. 70. Chowdhury M, Aman MG, Lecavalier L, Smith T, Johnson C, Swiezy N, McCracken JT, King B, McDougle CJ, Bearss K, Deng Y, Scahill L. Factor structure and psychometric properties of the revised Home Situations Questionnaire for autism spectrum disorder: The Home Situations Questionnaire-Autism Spectrum Disorder. Autism : the international journal of research and practice. Jul 17 2015. PMCID: 71. Aickin M, Ritenbaugh C. A criterion for the adequacy of a simple design when a complex model will be used for analysis. Control Clin Trials. 1991;12(5):560-565. PMCID:

24

Frye (Arkansas), McDougle(Harvard), Scahill (Emory) Folinic Acid for Language Impairment in Children with Autism Spectrum Disorder v2.9 72. Bearss K, Johnson C, Smith T, Lecavalier L, Swiezy N, Aman M, McAdam DB, Butter E, Stillitano C, Minshawi N, Sukhodolsky DG, Mruzek DW, Turner K, Neal T, Hallett V, Mulick JA, Green B, Handen B, Deng Y, Dziura J, Scahill L. Effect of parent training vs parent education on behavioral problems in children with autism spectrum disorder: a randomized clinical trial. JAMA : 2015;313(15):1524-1533. PMCID: 73. Leon AC, Mallinckrodt CH, Chuang-Stein C, Archibald DG, Archer GE, Chartier K. Attrition in randomized controlled clinical trials: methodological issues in psychopharmacology. Biol Psychiatry. Jun 1 2006;59(11):1001-1005. PMCID: 74. Little RJ, Wang Y. Pattern-mixture models for multivariate incomplete data with covariates. Biometrics. Mar 1996;52(1):98-111. PMCID: 75. Kenward MG, Carpenter J. Multiple imputation: current perspectives. Stat Methods Med Res. Jun 2007;16(3):199-218. PMCID: 76. Boarman DM, Baram J, Allegra CJ. Mechanism of leucovorin reversal of methotrexate cytotoxicity in human MCF-7 breast cancer cells. Biochemical pharmacology, 1990;40(12):2651-2660. PMCID: 77. Marsh Rde W; Talamonti MS; Katz MH; Herman JM. Pancreatic cancer and FOLFIRINOX: a new era and new questions. Cancer Medicine, 2015; 4(6):853-63

25