Randomized, Controlled, Phase 2 Trial of STX209 for Social Function in ASD

Background:   STX209 (arbaclofen) is a selective GABA-B agonist associated with behavioral improvements and disease-modifying effects in animal models of fragile X syndrome (FXS).  It is hypothesized to modulate mGluR5 receptor signaling, and to augment inhibitory neurotransmission.

In a previous 8-week, open-label study of 32 subjects with Autism Spectrum Disorder (ASD), STX209 was associated with significant improvement on the ABC-Lethargy/Social Withdrawal (ABC-LSW) subscale, the Social Responsiveness Scale, and the Vineland Adaptive Behavior Scales, Second Edition (VABS) Communication domain.  In a placebo-controlled trial in 63 subjects with FXS, STX209 was associated with significant improvement on the ABC-Social Avoidance subscale, which is specifically validated for FXS.  VABS Socialization scores also improved in the subgroup of 27 subjects with more severe social impairments.

Objectives:   To examine the efficacy, safety, and tolerability of STX209 in patients with ASD, age 5-21 years.

Methods:  A 12-week, double-blind, placebo-controlled trial was conducted at 24 sites in the USA.  Subjects met DSM-IV criteria for Autistic Disorder, Asperger’s Disorder, or PDD-NOS and had a minimum score of 8 on the ABC-LSW subscale.  Up to 2 concomitant psychoactive medications were permitted, excluding antipsychotics and medications with anxiolytic effects.

Subjects were randomized to either STX209 or placebo, with stratification by age and concomitant use of psychoactive medication.  Study drug was titrated over 4 weeks, to a maximum of 10 mg TID (age 5-11 years) or 15 mg TID (age 12-21 years), with fixed dosing for the subsequent 8 weeks.  Efficacy assessments included the ABC subscales, CGI-I, CGI-S, and VABS Social and Communication domain scores.

Results:  150 subjects (124 male; 130 DSM-IV Autistic Disorder; 120 ADOS-Autism) were randomized in the study, with 26 receiving at least one concomitant psychoactive medication.  130 subjects completed the study, with 10 (8 on STX209, 2 on placebo) discontinuing due to adverse events, which were generally behavioral (e.g., aggression, sleep disturbances).  There were 2 serious adverse events (suicidal ideation on STX209; anaphylaxis on placebo).  Suicidal ideation also occurred in 1 subject on placebo.  Overall, STX209 was well-tolerated, with a 9% incidence of somnolence.

On the primary endpoint, the ABC-LSW subscale, subjects on STX209 and placebo showed very similar improvements (change from baseline -5.4±0.78 vs. -6.0±0.75, LS mean±SEM, p=0.518).  On the CGI-S, subjects improved significantly more on STX209 (-0.6±0.10 vs. -0.2±0.10, p=0.006).  On all other secondary endpoints, results favored STX209 numerically, but did not reach statistical significance (e.g., VABS Socialization standard score: 3.8±1.27 vs. 2.1±1.22, p=0.362).  In a post-hoc analysis among subjects whose Vinelands were completed by the same clinician and caregiver (n=96), as the study protocol had required, those receiving STX209 showed greater improvement on the VABS Socialization scale (7.2±1.40 vs. 1.8±1.27, p=0.006).  Subgroup analyses indicated that improvement on VABS Socialization scores was notably larger in subjects with IQ≥70.

Conclusions:   STX209 was well-tolerated and shows potential for clinically-meaningful improvements in social function.  Drug effects were more evident in subjects with higher IQs. Further prospective trials of STX209 are needed.