No Solid Evidence to Support SSRIs in Autism

Pauline Anderson

October 22, 2019

The selective serotonin reuptake inhibitor (SSRI) fluoxetine (Prozac, Eli Lilly) appears to provide no benefit in treating obsessive-compulsive symptoms in children with autism spectrum disorders (ASDs).

Results of a randomized trial showed that although the SSRI fluoxetine reduced obsessive-compulsive behaviors compared with placebo, the difference was statistically nonsignificant when sex, verbal ability, and baseline imbalances were considered.

"Clinicians and families should be cautious about the use of these medications where, to-date, there has been no published evidence of their effectiveness in children and young people with ASDs," lead author Dinah S. Reddihough, MD, a pediatrician at the Royal Children's Hospital, and Research Fellow at the Murdoch Children's Research Institute, Melbourne, Australia, told Medscape Medical News.

The findings were published online today in the Journal of the American Medical Association.

Commonly Prescribed

Autism is characterized by restricted and repetitive interests and behaviors that can include stereotypic motor movements, ritualistic behaviors, and difficulty coping with change. This often manifests as anxiety, irritability, aggression, and self-injury, which frequently interferes with everyday functioning.

Despite inconclusive evidence that they are effective, antidepressants are prescribed for up to one third of children and adolescents with ASDs.

The rationale for prescribing these medications in this population is based on evidence that serotonin plays a contributory role in the pathophysiology of ASDs.

There is support from genetic, biological, and neuroimaging studies indicating that individuals with ASDs have higher serotonin levels than those without such disorders, the authors note.

To examine the efficacy of SSRIs, the investigators conducted the Fluoxetine for Autism Behaviors (FAB) study. The trial included 146 children and adolescents aged 7.5 to 18 years (mean age 11.2 years; 85% male) at three sites.

The original minimum age for the trial was 8 years but was lowered because recruitment proved difficult. It took 7 years to complete the trial.

"Many families did not want to risk being randomized to the placebo group," said Reddihough. "Fluoxetine was available off-label from community pediatricians, making it difficult for families to understand why they should participate in a trial."

Study participants had "significant problems" with obsessive-compulsive behaviors, with a total score of six or more on the Children's Yale-Brown Obsessive Compulsive Scale–Modified for pervasive developmental disorders (CYBOCS-PDD), said Reddihough.

Participants were randomly assigned to receive fluoxetine or placebo. The once daily dose of the active drug was tied to weight, starting at 4 mg/day (for participants under 40 kg [88 lb]) or 8 mg/day (for those 40 or more kg) for the first week and was then titrated weekly over 3 weeks. The maximum dose was 20 mg/day to 30 mg/day. After a follow-up assessment at 16 weeks, the medication was tapered over 4 weeks.

High Dropout Rate

The primary outcome was the difference in total score on the CYBOCS-PDD. This scale includes a checklist of possible obsessions and compulsions, rated from 0 to 4 across five items, with a focus on stereotypic and complex behaviors typically associated with ASDs. The total score can range from 0 to 20, with higher scores indicating higher levels of maladaptive behaviors.

The CYBOCS-PDD was selected as the primary outcome based on findings of a previous trial of fluvoxamine (Luvox, ANI Pharmaceuticals), an SSRI used primarily for treatment of obsessive–compulsive disorder, in children with ASDs. That study showed that the CYBOCS was sensitive to change across the treatment period, said Reddihough.

Secondary outcomes included differences in scores on a number of rating scales, among them the Repetitive Behavior Scale-Revised (RBS) and the Aberrant Behavior Checklist-Community Version (ABC), which measures inappropriate behaviors in individuals with developmental or intellectual impairments.

The placebo group had worse baseline scores on the RBS and the ABC lethargy scale than the fluoxetine groups.

Many participants — 41% in the fluoxetine group and 30% in the placebo group — did not complete the trial. Reddihough noted that families of children with autism have high levels of maternal depression and family stress. 

"Adhering to a treatment regimen over 16 weeks would likely have been difficult for many families, particularly if they felt that the medication was not of benefit," she said.

Results showed that from baseline to study end, CYBOCS-PDD scores were lowered more in the fluoxetine than in the placebo groups (from 12.80 to 9.02 compared to 13.13 to 10.89; mean difference of −2.01; 95% confidence interval [CI], −3.77 to −0.25; P = .03).

However, this difference became nonsignificant after controlling for sex, verbal ability, baseline measure of the CYBOCS-PDD, and baseline differences in the RBS and ABC lethargy subscale (mean difference −1.17; 95% CI, −3.01 to 0.67; P = .21).

An additional analysis using multiple imputation for missing data was also not statistically significant (mean difference −1.82; 95% CI, −3.71 to 0.06; P = .06).

In total, 45% of the fluoxetine group and 42% of the placebo group experienced adverse events. The most common of these were mood disturbance, particularly irritability, gastrointestinal problems such as nausea and diarrhea, and sleep disorders. Two participants in the placebo group and none in the fluoxetine group experienced serious adverse events.

Although the evidence "is not strong enough" to recommend fluoxetine as a treatment for this study population, "we can't exclude that it is helpful for some individual children. More research is needed," said Reddihough."

While conducting drug studies in children and adolescents with autism is "extremely difficult," they still need to be done, she added.

No Added Value

In an accompanying editorial, Bryan H. King, MD, from the Department of Psychiatry at the University of California San Francisco's Weill Institute for Neurosciences, said the study outcome "contributes new evidence that SSRIs do not add any value over placebo for repetitive behaviors in children and adolescents with ASD."

King noted that the baseline mean score on the CYBOCS-PDD was only 12.8 in the fluoxetine-treated group. With 6 being the minimum entry score, the study population may not have been "enriched for children for whom repetitive behaviors were specific or significant problems," he notes.

Using entry criteria that did not produce a study cohort of children at greatest risk "may have made it more difficult to find a therapeutic signal," he added.

King also noted that the lack of between-group differences in adverse events could suggest that the dosing of the treatment may have been too low.

Additional rigorous studies are needed, King writes, both to identify other potential treatments for core symptoms and to determine whether clinical indications other than repetitive behaviors might account for the persistent widespread use of SSRIs in ASD.

Don't Close the Door

Commenting for Medscape Medical News, Gabrielle L. Shapiro, MD, clinical professor of psychiatry at the Icahn School of Medicine at Mount Sinai in New York City, and chair of the American Psychiatric Association (APA) council on children, adolescents and families, described the study as "somewhat flawed."

In addition to the CYBOCS, the inclusion of other outcome scales "more specific" to autistic children in the primary analysis may have produced different results, said Shapiro, who was not involved with the current study.

Although the study was "pretty long," researchers may want to look at longer-term effects, said Shapiro. She noted that when treating children not on the autism spectrum with an SSRI for anxiety, OCD, or other symptoms, "we usually do it for more than 16 weeks; we usually do it for 6 months to a year or even longer."

Clinicians continue to use SSRIs in children with autism, even without solid evidence, because they simply "don't have many tricks in their bag," said Shapiro. "We don't have many proven useful treatments."

She added that in many cases, other medications have too many unacceptable side effects.

Shapiro stressed that as children grow, they pass through developmental stages and experience changes in hormones, so their symptoms "tend to wax and wane," and their behaviors change, said Shapiro.

"You may have to change medications at different times to get good results," she added.

While Shapiro said she is pleased to see studies like the current one, "the door shouldn't be closed."

"We need more funding for studies to look at different treatment strategies for children with autism," she added.

The study was supported by Australia's National Health and Medical Research Council, the Royal Children's Hospital Foundation in Melbourne, the Victorian government's Operational Infrastructure Support Program, and the Murdoch Children’s Research Institute. Reddihough, King, and Shapiro have disclosed no relevant financial relationships. 

JAMA. Published online October 22, 2019. Full text, Editorial

For more Medscape Psychiatry news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: