Setback For Autism Pioneer Seaside

Phase II study testing arbaclofen in patients with idiopathic autism showed no difference from placebo on a measure of social withdrawal but demonstrated significant improvement on disease severity and social function. That’s a hint the drug may yet be the first to affect the course of the disease and not just its symptoms – if a trial can be designed to test the right endpoint.

Seaside Therapeutics LLC’s most advanced candidate to treat autism spectrum disorders missed its primary endpoint in a Phase II trial, a setback evocative of the Alzheimer’s field in that there’s hope the miss may relate more to a mismatch of patient population and measurement tool than the efficacy of the drug.

Data from an exploratory Phase IIb trial testing arbaclofen (STX209) in patients with autism spectrum disorder were reported May 1 at the International Society for Autism Research annual meeting, in Donostia/San Sebastian, Spain.

In the trial, which involved investigators at 22 sites, 150 patients age 5 to 21, most of them with classic autistic disorder, were randomized to arbaclofen, a GABA-B agonist, or placebo for 12 weeks with the goal of showing improvement on core impairments of social function. But in the end there was no difference in scores on the defined primary endpoint, the Aberrant Behavior Checklist Social Withdrawal Subscale.

Arbaclofen, however, demonstrated a significant improvement that represents “a clinically relevant level of change” on the Clinical Global Impression of Severity scale, a clinician rating system, said Jeremy Veenstra-Vander Weele, Vanderbilt University, the lead author on the study. Those who showed the most improvement were those with a higher IQ, he said.

A secondary analysis of the data also revealed significant improvement on the Vineland-11 Socialization Scale, a psycho-educational measure of social function in higher functioning patients.

Arbaclofen was generally well-tolerated, with roughly 10% of patients having difficulty with emotional lability, or mood changes, while 9% had difficulty with sedation, Veenstra-Vander Weele said. One patient in each group had suicidal ideation.

Lessons Learned And Next Steps

“One of the exciting things about being involved in autism treatment research right now is trying to figure out how we may measure some of these outcomes that, really, we haven’t been able to see significant change in before,” Veenstra-Vander Weele said. “As we see new treatments come forward, we hope we’ll be able to identify scales that better measure some of these things … like how a child processes social information that’s coming in from the environment.”

A lot of behavioral studies have led the way here, have shown change, but “a lot of that has been measured by IQ and not necessarily autism-specific measures,” Veenstra-Vander Weele said. In fact, the Aberrant Behavior Checklist social withdrawal/lethargy subscale was chosen “because there was precedent for the FDA considering subscales of the [ABC] checklist when they make decisions about whether a medication could be said to be effective in a particular situation.”

The study results also suffered from the fact both the placebo and medication arms had a “substantial response,” Veenstra-Vander Weele said. The message there may be that similarities or overlap between “lethargy” and “social withdrawal” cause them to be conflated, and “when you look at this checklist, that removes your power to focus in on specific social deficits in autism.”

Another point to be considered is that social withdrawal may not be the defining characteristic in autism at all. Autism could well be defined by impairment in social ability/social engagement, not avoidance, he said. And that particular ABC checklist hones in on avoidance or withdrawal instead of capturing social function across different settings.

In the post-hoc analysis, “what seemed to move was day-to-day social function,” measured by the Vineland adaptive behavior scale, Veenstra-Vander Weele said. “This is a scale that a lot of us as investigators didn’t think could move over this period of time, so this is potentially a very exciting, novel outcome,” he said. “This would be really an unprecedented treatment effect for a medication.”

Seaside is working with FDA to design a trial to replicate the signals that came from this trial and determine specifically who benefits and to what degree, Veenstra-Vander Weele said.

The company confirmed that the new primary endpoint will likely be the Socialization subset of the Vineland-11 Adaptive Behavior Scale. Further it will be a pivotal trial that could be part of a regulatory submission if it is successful, likely a Phase II/III. The patient number is not finalized, but it will be large, around 250-300 patients.

About Arbaclofen

Privately held Seaside, in Cambridge, Mass., is a pioneer in the development of drugs that work to modulate glutamate, the major excitatory neurotransmitter in the brain, with the idea that once the synaptic junctions in that pathway are in balance, patients with autism spectrum disorders will have a better chance to learn and succeed in their environment.

The company’s main focus has been fragile X syndrome, caused by a single-gene defect or deletion, but the hope is that a successful therapy for fragile X will be effective in other gene-related and sporadic syndromes on the autism spectrum (Also see "Developing Novel Drugs To Treat Autism" - Pink Sheet, 18 Dec, 2012.).

Seaside is approaching the problem from two directions, through two pathways that in normal neural function provide balance between overstimulation and understimulation at crucial synaptic junctions. Glutamate is the major excitatory neurotransmitter in the brain, while the major inhibitory neurotransmitter is gamma-amino butyric acid type B (GABA-B).

One way to stop that runaway car is to put your foot on the brake by reducing the receptor response to glutamate with an mGluR5 inhibitor, but another is to cut back on the release of glutamate in the first place, in effect taking your foot off the gas by releasing more GABA-B.

The different mechanisms will probably turn out to have different effects because of the distribution of receptors in the brain, but the company’s GABA-B compound may increase inhibitory tone generally, which the company hopes will be good for both fragile X and idiopathic autism.

Two Phase III trials testing arbaclofen in fragile X, one in adolescents and adults ages 12 to 50 and one in children ages 5 to 11, are fully enrolled with results expected in 2013.

The company’s lead mGluR5 antagonist candidate STX107 is in a Phase II trial against fragile X. The mGluR5 program is partnered with Roche ([See Deal]).