Anne Chain (1), Maria Luisa Sardu (2), Chao Xu (1), Juub Jan Kleijn (2), Li Qin (2), Ferdous Gheyas (1), Sreeraj Macha (1), Eugene Cox (2)
(1) MSD, New Jersey, USA, (2) Certara Strategic Consulting, The Netherlands
Objectives: This analysis aimed to develop a Model Based Meta-Analysis (MBMA)comparator model to provide a quantitative framework for the comparison and benchmarking of second generation antipsychotics (SGA) used for the treatment of schizophrenia under clinical development. Specific objectives include development of a network MBMA model describing the change from baseline of the total positive and negative symptom score (PANSS) with SGAs. As secondary objectives, MBMA will also be developed for only describing the positive and negative subscale scores.
Methods: A systemic literature review was conducted from the Quantify Schizophrenia Database, a schizophrenia literature database that consists of publicly available summary-level safety and efficacy data from trials investigating first- and second-generation antipsychotics. The database included 267 randomized controlled trials with approximately 60,000 patients. The database comprised of a source database that maintained the sources of information as well as a clinical outcomes database that contained information on trial, treatment, patient characteristics, efficacy and safety results of the trials.
Pre-specified list of inclusion/exclusion criteria, i.e. double-blind, sponsor- or investigator-initiated registration trials in adult patients with a typical disease duration of 1 years after the diagnosis of a first psychotic break up to ~20 years with a trial duration of 4-12 weeks, was used to select the final analysis dataset for the MBMA of sGAs. Model development, evaluation and simulations were performed using R 3.4.2. The models were developed using the generalized nonlinear least squares (gnls) function in R. Placebo response was described using non-parametric method to adjust for trial-to-trial variability. Drug effect was estimated using parametric method and dose-response was assessed where possible using an Emax model. Model appropriateness was assessed using diagnostic plots and covariates were graphically explored. Treatment effect estimates with associated 95% confidence intervals for each drug were derived from 10000 simulations with parameter values sampled from the multivariate normal variance-covariance matrix of the estimates.
Results: The selected analysis dataset contained monotherapy data from 33 trials of 10 oral SGAs as well as haloperidol for acute schizophrenia that either has forced dose titration or no dose titration. Long-acting or injectable formulations of SGAs were excluded. The models describing total PANSS score, positive and negative subscales were able to capture the data well and treatment effect was identified for each drug and dose-response relationship was obtained for lurasidone, risperidone and paliperidone. Age and black race were identified as significant covariates in the model. Treatment effects based on simulations using the MBMA model were obtained.
Conclusions: MBMA models for total PANSS, positive subscale and negative subscale scores of sGAs were successfully developed. Model estimated treatment effects provide a quantitative framework for benchmarking new investigational compounds in schizophrenia under clinical development.
Reference: PAGE 28 (2019) Abstr 9137 [www.page-meeting.org/?abstract=9137]
Poster: Drug/Disease Modelling - CNS