Fluoxetine appeared to lower scores for obsessive-compulsive behaviors among a group of children with autism spectrum disorders (ASDs), but the positive finding fell apart during multiple secondary analyses, Dinah S. Reddihough, MD, and colleagues have reported.

At 16 weeks, children and adolescents randomized to receive the SSRI had about a 2-point improvement on the Children’s Yale-Brown Obsessive Compulsive Scale, modified for pervasive developmental disorder (CYBOCS-PDD), compared with those taking placebo. But the finding lost significance in a multivariate analysis that accounted for a between-group difference in baseline scores – an uncontrollable variable that occurred during randomization, wrote Dr. Reddihough, of the Royal Children’s Hospital in Victoria, Australia, and coauthors.

“Moreover, repeating the analyses with multiple imputation to handle the missing data, arguably a preferable analysis, also failed to show evidence of benefit of fluoxetine compared with placebo irrespective of adjustment for the baseline imbalance,” the team noted. The study was published in JAMA.

Despite the null findings of the additional adjusted analyses, the authors held out hope for fluoxetine.

“Although cautious interpretation of the results from the primary analysis is warranted, all analyses of the primary outcome yielded 95% confidence intervals that extended well above the minimum clinically important difference of 2 points, indicating that fluoxetine may reduce the frequency and severity of obsessive-compulsive behaviors in children and adolescents with ASDs. Given the large amount of missing data, the study may have been underpowered to detect the minimum clinically important difference of 2 points.”

The study comprised 146 children (mean age, 11 years) recruited through three large practices in Australia. Children were randomized to fluoxetine or placebo for 16 weeks. Fluoxetine was weight-dosed and then titrated every week for the first month to a maximum of 20 mg/day.

The primary outcome was the difference between groups in the total score on the CYBOCS-PDD at 16 weeks. Secondary endpoints included changes on the Repetitive Behavior Scale–Revised, the Spence Children’s Anxiety Scale Aberrant Behavior Checklist–Community Version, the Clinical Global Impression Scale–Global Improvement and Efficacy Index, and a Disruptiveness Assessment.

Of the cohort, 85% were male, and 30% had an intellectual disability. The placebo group had higher scores on the Repetitive Behavior Scale–Revised and the Aberrant Behavior Checklist lethargy scale than did the fluoxetine groups.

There was a very high rate of nonadherence to study protocol, with 41% of those in the active group and 30% in the placebo group not completing the treatment regimen. The most often cited reasons for treatment discontinuation included parent decision to drop out (20 fluoxetine, 12 placebo), adverse events (5 fluoxetine, 4 placebo), and clinician decision (2 fluoxetine, 2 placebo).

The primary analysis found that scores on the CYBOCS-PDD were significantly lower in the fluoxetine group at 16 weeks; the fluoxetine group had decreased its score from 12.80 to 9.02, while the placebo group went from 13.13 to 10.89. This mean 2-point difference was statistically significant and, the authors wrote, met the minimum threshold for a clinically significant difference.

But the mean between-group difference decreased to a nonsignificant 1.17 points in the sensitivity analysis that controlled for sex, verbal ability, baseline CYBOCS-PDD, and imbalances found at baseline in some of the measures.

“Moreover, repeating the analyses with multiple imputation to handle the missing data, arguably a preferable analysis, also failed to show evidence of benefit of fluoxetine compared with placebo irrespective of adjustment for the baseline imbalance,” the team said.

There were no significant differences on any of the secondary measures.

Adverse events were similar in the active and placebo groups (45% and 42%, respectively). These included mood disturbances particularly irritability (9 fluoxetine, 12 placebo), gastrointestinal problems such as nausea and diarrhea (10 fluoxetine, 7 placebo), and sleep disorders (13 fluoxetine, 16 placebo). Two patients in the placebo group and none in the active group experienced suicidality.

Dr. Reddihough and coauthors cited the study’s high dropout rate as one of its limitations.

The study was supported by a federal grant from the Australian government. Dr. Reddihough had no financial disclosures.

SOURCE: Reddihough DS et al. JAMA. 2019;322(16):1561-9.