New evidence suggests that autistic disorder (AD) may be associated with abnormalities in folate metabolism, which is a process that affects genetic expression by facilitating the formation of methyl donors for DNA methylation. Limited data show that some children with AD show behavioral improvements with folic acid (FA) therapy, while others show a worsening effect. If behavioral worsening is linked with abnormalities in folate metabolism, then nutritional modifications could normalize these processes and result in clinical improvements. To address this premise, we propose a randomized, placebo-controlled crossover pilot study with two phases. The first phase will focus on the behavioral and biochemical responses of children with AD to high-dose folic acid supplementation. Because FA is an inactive folate that requires biochemical conversion to become active, and select genotypes impede this conversion, our general hypothesis is that FA will yield behavioral improvements in some children but exacerbate problem behaviors in others. During the second phase, children who had a worsened behavioral response to FA during phase 1 will participate in an open-label trial of high-dose Metafolin® supplementation. The focus here would similarly be on the behavioral and biochemical outcomes of participating children following treatment with the study supplement. Because Metafolin® is an active folate metabolite that should not be affected by genotypes in the folate pathway, our general hypothesis for phase 2 is that Metafolin® would yield behavioral improvements without the risk for behavioral worsening. Results from this project may provide support for continued study of the potential relationship between folate metabolism and problem behaviors among children with AD, potentially justifying the need to examine effects of folate supplementation among a larger sample of affected children.

There are limited data on the use of folic acid as a treatment for children with autistic disorder (AD). Most experimentation thus far has focused on the effects of folic acid supplementation in small samples of males with fragile X syndrome (FRAX), and many of these participants did not meet criteria for AD but were said to exhibit autistic-like features. Initial case reports suggested folic acid might be an effective treatment for FRAX, some of which noted a reduced expression of the fragile site Xq27 in blood cultures. Subsequent work found that high doses of folic acid supplementation reduced the percentage of FRAX positive cells and increased plasma-folate concentrations among 9 males with FRAX. Autistic-like behaviors were said to improve for the 5 youngest participants (age range 3 years, 5 months to 9 years, 10 months); however, one adult, aged 49, developed an increase in motor activity and aggressiveness during treatment. His behavior returned to pretreatment levels following discontinuation of folic acid. In similar work, researchers employed a crossover design using 4 boys (ages 6 to 14) with comorbid autism and FRAX to assess efficacy of folic acid supplementation. Favorable behavioral changes were noted for 3 boys; however, for 2 of them, carryover effects of folic acid into the placebo period may have clouded the magnitude of potential differences in folic acid versus placebo effects. No changes throughout the trial were evidenced for the fourth child, who was also the oldest and the only one to have commenced puberty.

Overall, investigations on folic acid supplementation with small groups of FRAX participants have yielded inconsistent findings, with favorable behavioral and/or biochemical effects for some but not others. Moreover, the findings cannot be generalized to the broader population of children with AD who do not have FRAX. Our search of the literature uncovered only one report of folic acid treatment prescribed for 2 children without FRAX, one of whom was described as psychotic with mental retardation and the other, as having a diagnosis of autism. For both children, behaviors were notably improved with treatment and returned to problematic levels when folic acid therapy stopped.

In addition to these few empirical studies, there are numerous anecdotal reports of families using folate treatments with their children, most of which are promulgated through Internet websites and the Autism Research Institute's (ARI) publication, Autism Research Review International. For the most part, these reports are favorable. Additionally, the ARI recently published findings from their web-based investigation on the parent-reported use and efficacy of a wide range of treatments for children with AD. Of the 1437 families who indicated use of folic acid as a therapeutic supplement, 42% said their children got better, 54% said their children did not exhibit noticeable changes, and 3% said their children got worse. Taken together, these limited data support the possibility for folic acid and other folates to generate favorable outcomes among children with AD; this may hold true particularly for pre-adolescent children.

The lack of empirically sound information about folic-acid efficacy among children with AD, however, also leaves open the possibility that increased supplementation may not be helpful and could, in fact, be harmful. There is some evidence to suggest that high doses of folic acid supplementation produce negative physiological and/or behavioral responses, such as gastrointestinal disturbances, sleep difficulties, malaise/irritability, and excitability/overactivity in a sample of typical adults. Many of these symptoms are similar if not identical to those described more generally in children with AD. Additionally, just as there are favorable anecdotal reports of folic acid treatment, there are negative ones, as well. For example, parents of one teenager with AD told us that, when their son participated in a stepwise-administered supplement program and folic acid was administered as the final supplement, he became extremely agitated, self-injurious, aggressive toward parents, prone to frantic gesturing, and developed sleep difficulties. When the folic acid was withdrawn, his behaviors gradually subsided to a pre-supplementation level (personal communication, September 27, 2003).

As to whether folic acid supplementation might be "good" or "bad," an alternative explanation is that folic acid may be beneficial in some circumstances and harmful in others. This explanation is potentially attractive if one hypothesizes that the conversion of homocysteine to methionine (for which folate is necessary) is beneficial, but that high doses of folic acid can lead to anti-folate effects by accumulation of "free folic acid." For individuals with an imbalanced metabolic profile but adequate levels of B12, it seems that folic acid supplementation would yield favorable effects by normalizing this profile. However, folic acid supplementation could lead to problematic behaviors through any one or combination of biological scenarios, including accumulation of unmetabolized pteroylglutamic acid (PGA) in serum, B12 deficiency, and/or genotypes known to impede the biochemical conversion of folic acid. Given that negative responses to folic acid supplementation have typically been reported among adolescents and adults, it is further possible that this phenomenon is regulated by maturational processes.

One possibility for achieving maximum folate benefit without the risk of anti-folate effects is through supplementation with a form of folate, Metafolin, that can serve directly as a methyl donor. Metafolin is the L-form isomer of 5- methyltetrahydrofolate (5-MTHF) and, presumably, a more active compound than folic acid or folinic acid (leucovorin). Metafolin has been approved for use in the U.S. as a dietary supplement since 2001. While studies around the globe have tested the safety of 5-MTHF at doses up to 17mg, there have not yet been any studies to examine its efficacy as a therapeutic agent in specific neurodevelopmental disorders. However, folinic acid (5,10- MTHF), which immediately precedes 5-MTHF in the folate-metabolism cycle, has been investigated repeatedly as a potential therapeutic agent for this population. In one such study, James et al. examined the biochemical effects of folinic acid supplementation in conjunction with betaine and vitamin B12 on the biochemical outcomes of children with imbalanced metabolic profiles. This team initially discovered that, compared with a control group, children with AD had significantly lower baseline concentrations of methionine, S-adenosylmethionine (SAM), and homocysteine and significantly higher concentrations of S-adenosylhomocysteine (SAH) and adenosine--a metabolic profile that is consistent with impaired capacity for methylation. Following supplementation with the study treatments, the metabolic imbalance within the AD group was normalized. This same research team recently published results from an extension of this work in which they assessed levels of folate-related metabolites and genotypes for select enzymes. Approximately 50% of children with AD had severely abnormal metabolic profiles; additionally, "...several susceptibility alleles that perturb a common metabolic pathway were increased among the autistic children. The hypothesis that a genetic component of autism could involve multiple susceptibility alleles that interact to create a fragile, environmentally sensitive metabolic imbalance is worthy of further pursuit," (electronic copy; no page number available).

Inclusion Criteria:

1. Child participant has a confirmed diagnosis of AD by Autism Diagnostic Observation Schedule and Autism Diagnostic Interview--Revised criteria
2. Child participant is in stable condition with relatively good control of seizures and no other significant medical problems, including liver, kidney, or heart problems, at the time of entrance to the study. If the child participant is taking medication for a seizure disorder, the investigators will assess his/her eligibility with particular regard to type of seizure medication and other health-related information gleaned during the medical examination
3. Child participant and parents are willing to comply with the proposed treatments
4. Child participant is able to take oral medication
5. Family is fluent in the English language
6. Parent/caregiver agrees to provide behavioral data on participating children at the requested time points
7. Family agrees to be contacted weekly by study personnel during the treatment phases

Exclusion Criteria:

1. Child participant has co-morbid medical and/or genetic disorders, including celiac disease
2. Child participant has a history of liver or renal disease
3. Child participant is currently being treated for a serious acute illness
4. Child participant has a known allergy to any of the proposed supplements
5. Child participant has uncontrolled seizures
6. Child participant meets criteria for Asperger's syndrome, PDD-NOS, or does not meet strict criteria for AD
7. Family is not proficient in the English language