Automated Speech Analysis to Predict Development of Psychosis: A Novel Endpoint in a Randomized, Phase II Trial of BI 409306 in Attenuated Psychosis Syndrome
Poster B108, Tuesday, October%209, 11:30%20am%20-%201:00%20pm, Essex%20Ballroom
Natalia Mota1, Mauro Copelli2, Sidarta Ribeiro1, Visar Berisha3, Diego Fernandez Slezak4,5, Kristen Daniels6, Michael Sand6; 1Brain Institute, Federal University of Rio Grande do Norte, 2Physics Department, Federal University of Pernambuco, 3Department of Speech and Hearing Science and School of Electrical Computer and Energy Engineering, Arizona State University, 4Universidad de Buenos Aires, Facultad de Ciencias Exactas y Naturales, Departamento de Computación, 5CONICET-Universidad de Buenos Aires, Instituto de Investigación en Ciencias de la Computación (ICC),, 6Boehringer Ingelheim Pharmaceuticals Inc.
Although individuals with attenuated psychosis syndrome (APS) can be identified, reliable prediction of frank psychosis onset among them remains elusive. Improving the capacity to predict those at highest risk for conversion to psychosis would have important ramifications for early identification and preventive intervention, potentially critically altering the long-term life trajectory of people with emergent psychotic disorders. Subtle disorganization in speech has consistently been identified as predictive of psychosis (i.e., with classification accuracy of ~60%) among young people identified as clinically high-risk. BI 409306, a potent and selective phosphodiesterase-9 inhibitor that improves NMDA signaling, is under development for early intervention in APS. An ongoing multinational, double-blind, parallel-group study will assess the efficacy, safety, and tolerability of BI 409306 in patients with APS (16–30 years of age). Patients will be randomized (300 planned, 1:1) to BI 409306 or placebo for 52 weeks. The primary endpoint is time to first episode psychosis. Secondary endpoints will assess functional capacity, cognition, and disease symptoms. Novel automated speech analysis, an exploratory biomarker of brain plasticity, will also be assessed. Patients will participate in audio-recorded interviews (baseline, Weeks 18, 42, 52). Audio recordings and transcripts will be analyzed to extract acoustic parameters and generate speech graphs and semantic features. Classifiers will be applied to determine the optimal combination of acoustic, graph-theoretical, and semantic features to predict psychosis. This novel speech analysis, as part of a clinical trial, may provide valuable data to aid early detection of APS. Recruitment is ongoing. Funding: Boehringer Ingelheim (NCT03230097).
Topic Area: Ultra High Risk / Prodromal Research