Broccoli Sprout Extract Improves Symptoms of Autism Spectrum Disorder

Autism Spectrum Disorder (ASD) affects 1 in 68 children in the United States by the age of 8 years. There is no cure and the available pharmaceutical treatments carry the risk of severe side effects. Sulforaphane (SF; 4-methyl sulfinyl butyl isothiocyanate) is an anti-inflammatory and chemoprotective constituent of broccoli (Brassica oleracea var. Italica, Brassicaceae) seed sprouts. SF has been evaluated as a treatment for a variety of medical conditions, including ASD. From 2011 to 2013, the authors conducted a small (n = 44), randomized, double-blind, placebo-controlled study where men with ASD aged 13-27 years received a weight-adjusted daily dose of SF (n = 26) or placebo (n = 14) for 18 weeks.1Four validated instruments were used to evaluate behavior; 2 were completed by parents or caregivers (P/C), and 2 were completed by physicians. While there were no significant changes in the placebo group after 18 weeks of treatment, 65% of the SF group had improved significantly compared with baseline (P values not reported). However, 4 weeks after treatment discontinuation at the end of this randomized controlled trial (RCT), most of the SF responders had regressed to baseline scores and many of the P/C requested a continuation of SF treatment.

There is a myriad of dietary supplements containing broccoli sprout extract (BSE) on the retail market. Some are SF-enriched, some are rich in glucoraphanin (GR; the more stable but inactive precursor to SF), and some contain GR plus myrosinase (MYR; the enzyme that catalyzes the conversion of GR to SF). BSE rich in GR has the lowest average bioavailability (~10%), while the bioavailability of BSE containing GR + MYR is approximately 35%, and the bioavailability of SF-rich BSE is ~70% (bioavailability varies significantly among individuals). The authors analyzed the content of a number of commercial BSE supplements and found that many products did not contain the amount of SF or GR claimed on the label (brand names were deliberately withheld, and no further information was provided). Ultimately, the authors recommended a small number of retail products for which they had verified the veracity of the label claims to the P/C.

Of the 26 males randomly assigned to the SF group for the RCT, P/C of 16 patients (14 SF responders and 2 non-responders) agreed to respond to follow-up questions after the clinical trial was completed. This case series covering the period from 2013 to 2016 summarizes the subjective responses of the 16 P/C and the post-hoc analysis of that data.

Ex post facto (after the fact) analysis of the responses made by the 16 P/C indicated that they fell into 4 categories. (1) Three patients had no major improvements during the study. P/C were not able to tell if the patient was on SF or placebo. [Note: One of these patients was classified as an SF responder based on assessment scores, but their P/C did not perceive any significant improvement.] (2) One patient began to improve within a few days of SF treatment initiation (for most patients, it took 3-4 weeks before significant changes were observed). SF was discontinued after his study participation ended in 2012; however, the dramatic improvements were sustained throughout the 3-year follow-up period. (3) Two patients continued SF supplementation after the study but later were discontinued due to other health issues. Subsequently, the P/C saw a regression in the improvement achieved while on SF and indicated they wanted the patient to restart SF therapy. (4) Ten patients were either still taking SF (n = 9) or wanted to take a different SF supplement (n = 1). Of the 9 patients who continued on BSE, P/C reported the following improvements: discontinuation of long-term fingernail picking, more verbal, calmer, happier, less stressed, improved focus, stopped biting and ripping shirt (behavior re-occurred when SF was discontinued), more responsive, increased socialization, showing interest in others, engaged in casual conversation, and doing well. Two P/C reported that the patient regressed when SF was discontinued. Refer to the original article for anecdotal quotes from the P/C.

In summary, of the 14 SF responders in the RCT whose P/C agreed to participate in the follow-up study, 13 P/C thought that SF significantly improved the behavior of their sons. The authors point out that there is a broad spectrum of BSE dietary supplements on the market, and there is significant variation in both the claimed content of GR or SF and the manufacturer’s recommended dose; consequently, the dose of SF that consumers take may vary up to 10-fold. They also stress that while the content of the BSE they recommended to the P/C participating in this follow-up study was verified, many of the products they tested did not meet label claims and there are “a great many more” supplements which they did not test. Thus, the P/C of patients with ASD who try commercial BSE supplements may not observe any benefit because a sub-therapeutic dose was taken. While the evidence from this study and others provides some guidance regarding dosage, the optimum dose of SF has not been established. The issue of the optimum therapeutic dose of SF is further complicated by the fact that there are significant inter-individual differences in the bioavailability of SF.

The authors acknowledge that the anecdotal evidence reported in this case series is much weaker than that obtained from RCTs. However, this evidence has helped stimulate the initiation of at least 5 studies of SF in patients with ASD and 2 related studies on symptoms of schizophrenia (see for descriptions). The pending research is of high interest as it will provide further evidence regarding the safety, efficacy, and optimum dose of SF in the treatment of ASD.

Two of the authors (Zimmerman and Talalay) are named inventors on a patent pertaining to the treatment of ASD with SF, the rights to which they have assigned to The Johns Hopkins University School of Medicine; Baltimore, Maryland. This study was funded in part by The Brassica Foundation for Chemoprotection Research (Baltimore, Maryland).


1Singh K, Connors SL, Macklin EA, et al. Sulforaphane treatment of autism spectrum disorder (ASD). Proc Natl Acad Sci U S A. 2014;111(43):15550-15555. 

Lynch R, Diggins EL, Connors SL, et al. Sulforaphane from broccoli reduces symptoms of autism: a follow-up case series from a randomized double-blind study. Glob Adv Health Med. October 26, 2017;6. doi: 10.1177/2164957X17735826.