By Julie Steenhuysen
CHICAGO (Reuters) - The first-ever drug designed to treat social impairments associated with autism failed to show a benefit in a midstage trial, representing a blow to families and to privately held drugmaker Seaside Therapeutics.
Results of the study, presented on Wednesday at the International Meeting for Autism Research in Spain, showed the drug known as STX209 failed to improve symptoms of social withdrawal in a 12-month study of 150 individuals aged 5 to 21, most with classic autistic disorder.
The drug is the first of a handful of treatments in development that are expressly designed to correct the genetically induced signaling problems in the brain that result in autism.
Seaside's drug STX209, or arbaclofen, works to control an overabundance of signaling or "noise" at brain synapses by reducing the amount of the neurotransmitter glutamate that is available in the brain.
Other companies pursuing drugs for autism include Swiss drugmakers Novartis AG and Roche Holding AG , which last year licensed patents from Seaside for the use of drugs known as mGluR5 antagonists that attack the signaling problem in a different way.
Based on the results, however, Roche has decided to pass on its option to develop arbaclofen, said Dr. Randall Carpenter, president and chief executive officer of Cambridge, Massachusetts-based company. Roche did not return calls seeking comment.
Carpenter said Seaside will seek new partners and has already met with the U.S. Food and Drug Administration to devise a new study based on the current trial's findings that he believes may yet lead to the drug's approval.
"We believe our drug works at a molecular level that corrects signaling pathway abnormalities, circuit abnormalities and even anatomical abnormalities," Carpenter said.
"So the real question for us is, if you have a disease-modifying therapeutic, how would you demonstrate that in a short-term clinical trial?" Carpenter said.
In general, individuals with autism struggle with difficulties in communication, behavior and social interaction. U.S. government researchers said in March that as many as one in 50 American school-age children have a diagnosis of autism, which can range from highly functioning individuals to those with severe speech and intellectual disabilities.
While intensive behavioral therapy has been shown to help, there are no approved drugs that can reverse or improve core symptoms of autism.
In the Seaside trial, the primary goal of improvements in social withdrawal was assessed based on parent observations, and that may have contributed to high response rates among those who got a placebo, likely reflecting parents' eagerness to see an improvement.
Dr. Bryan King, director of the Seattle Children's Autism Center at the University of Washington and Seattle Children's Hospital, said the issue of a strong placebo effect has been a problem in prior trials. "It's not uncommon at all to see a 25 to 30 percent placebo response rate in these studies," he said.
But researchers saw other signs that the Seaside drug was having an effect. On a secondary measure of severity, which is scored by clinicians, there was a significant difference between those in the study on the drug and those who were on a placebo.
"This is a clinically relevant level of change," Dr. Jeremy Veenstra-Vanderweele Of Vanderbilt University, one of the study's key investigators, said in a webcast from the meeting.
Veenstra-Vanderweele said an analysis of the results showed improvements in day-to-day social function, as measured on a scale known as the Vineland Adaptive Behavior Scale.
"This is a scale that a lot of us as investigators didn't think could move over this period of time. So this is a potentially very exciting, novel outcome," he said.
He said the next step is to test to see if this same finding can be observed in a larger clinical trial that specifically targets changes in this behavior.
Carpenter said the company has collected genetic samples from more than 400 subjects and plans to study the specific genetic differences to get a better idea of which patients will likely respond to the drug.
(This version of the story corrects the name of behavior scale in paragraph 16.)
(Editing by Doina Chiacu)