Understanding Antipsychotic Drug Treatment Effects: A Novel Method to Reduce Pseudospecificity of the Positive and Negative Syndrome Scale (PANSS) Factors

| December 1, 2017 | 0 Comments

by Seth C. Hopkins, PhD; Ajay Ogirala, PhD, Antony Loebel, MD; and Kenneth S. Koblan, PhD 

Drs. Hopkins, Ogirala, Loebel, and Koblan are with Sunovion Pharmaceuticals Inc, Marlborough, Massachusetts.

Funding: Funding was provided by Sunovion Pharmaceuticals Inc.

Disclosures: The authors are employees of Sunovion Pharmaceuticals Inc.

Abtract: The Positive and Negative Syndrome Scale (PANSS) is the most widely used efficacy measure in acute treatment studies of schizophrenia. However, interpretation of the efficacy of antipsychotics in improving specific symptom domains is confounded by moderate-to-high correlations among standard (Marder) PANSS factors. The authors review the results of an uncorrelated PANSS score matrix (UPSM) transform designed to reduce pseudospecificity in assessment of symptom change in patients with schizophrenia. Based on a factor analysis of five pooled, placebo-controlled lurasidone clinical trials (N=1,710 patients), a UPSM transform was identified that generated PANSS factors with high face validity (good correlation with standard Marder PANSS factors), and high specificity/ orthogonality (low levels of between-factor correlation measuring change during treatment). Between-factor correlations were low at baseline for both standard (Marder) PANSS factors and transformed PANSS factors. However, when measured change in symptom severity was measured during treatment (in a pooled 5-study analysis), there was a notable difference for standard PANSS factors, where changes across factors were found to be highly correlated (factors exhibited pseudospecificity), compared to transformed PANSS factors, where factor change scores exhibited the same low levels of between-factor correlation observed at baseline. At Week 6-endpoint, correlations among PANSS factor severity scores were moderate-to-high for standard factors (0.34–0.68), but continued to be low for the transformed factors (-0.22–0.20). As an additional validity check, we analyzed data from one of the original five pooled clinical trials that included other well-validated assessment scales (MADRS, Negative Symptom Assessment scale [NSA]). In this baseline analysis, UPSM-transformed PANSS factor severity scores (negative and depression factors) were found to correlate well with the MADRS and NSA. The availability of transformed PANSS factors with a high degree of orthogonality/specificity, but which retain a high degree of concurrent and face validity, can reduce pseudospecificity as a measurement confound, and should facilitate the drug development process, permitting a more accurate characterization of the efficacy of putative new agents in targeting specific symptom domains in patients with psychotic illness.

Keywords: Schizophrenia, antipsychotic agents, factor analysis, efficacy, clinical, clinical trials

Innov Clin Neurosci. 2017;14(11–12):54–58


The content you are trying to view is restricted to registered members only.
Membership to Innovations in Clinical Neuroscience is free, so please consider becoming a member today.
Thank you
ICNS Online Editor
Please Login or Register for access.

Tags: , , , , ,

Category: Assessment Tools, Drug Development, Past Articles, Patient Assessment, Psychiatry, Psychopharmacology, Review, Scales, Schizophrenia, Trial Methodology

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.