No abstract

BackgroundA phase II randomized, placebo-controlled, double-blind study and subsequent open-label extension study evaluated the efficacy, safety, and tolerability of mavoglurant (AFQ056), a selective metabotropic glutamate receptor subtype-5 antagonist, in treating behavioral symptoms in adolescent patients with fragile X syndrome (FXS). A novel method was applied to analyze changes in symptom domains in patients with FXS using the narratives associated with the clinician-rated Clinical Global Impression-Improvement (CGI-I) scale.MethodsIn the core study, patients were randomized to receive mavoglurant (25, 50, or 100 mg BID) or placebo over 12 weeks. In the extension, patients received 100 mg BID mavoglurant (or the highest tolerated dose) for up to 32 months. Global improvement, as a measure of treatment response, was assessed using the CGI-I scale. Investigators assigning CGI-I scores of 1 (very much improved), 2 (much improved), 6 (much worse), or 7 (very much worse) were provided a standard narrative template to collect further information about the changes observed in patients. Investigator feedback was coded and clustered into categories of improvement or worsening to identify potential areas of improvement with mavoglurant. Treatment effect in each category was characterized using the Cochran–Mantel–Haenszel test.ResultsA total of 134 and 103 patients had reached 2 weeks or more of core and extension study treatment, respectively, by the pre-assigned cutoff date for investigator feedback. In the core study, 34 CGI-I scores of 1 or 2 were reported in 28 patients; one patient scored 6. Analysis of the CGI-I narratives did not indicate greater treatment response in patients receiving mavoglurant compared with placebo in any specific improvement domain. There were 54 CGI-I scores of 1 or 2 in 47 patients in the extension study. The most frequently reported categories of improvement were behavior and mood (79.3 and 76.6 % in core and extension studies, respectively), engagement (75.9 and 78.7 %), and communication (69.0 and 61.7 %).ConclusionsA method was established to capture and categorize FXS symptoms using CGI-I narratives. Although this method did not show benefit of drug over placebo, narratives from investigators were mostly based on parental report and thus do not represent a completely objective alternative assessment.Trial registrationThe studies described are registered at ClinicalTrials.gov with clinical trial identifier numbers NCT01357239 and NCT01433354.

The proof-of-concept (PoC) decision is a key milestone in the clinical development of an experimental treatment. A decision is taken on whether the experimental treatment is further developed (GO), whether its development is stopped (NO-GO), or whether further information is needed to make a decision. The PoC decision is typically based on a PoC clinical trial in patients comparing the experimental treatment with a control treatment. It is important that the PoC trial be designed such that a GO/NO-GO decision can be made. The present work develops a generic, Bayesian framework for defining quantitative PoC criteria, against which the PoC trial results can be assessed. It is argued that PoC criteria based solely on significance testing versus the control are not appropriate in this decision context. A dual PoC criterion is proposed that includes assessment of superiority over the control and relevance of the effect size and hence better matches clinical decision making. The approach is illustrated for 2 PoC trials in cystic fibrosis and psoriasis.

Identifying subgroups of treatment responders through the different phases of clinical trials has the potential to increase success in drug development. Recent developments in subgroup analysis consider subgroups that are defined in terms of the predicted individual treatment effect, i.e. the difference between the predicted outcome under treatment and the predicted outcome under control for each individual, which in turn may depend on multiple biomarkers. In this work, we study the properties of different modelling strategies to estimate the predicted individual treatment effect. We explore linear models and compare different estimation methods, such as maximum likelihood and the Lasso with and without randomized response. For the latter, we implement confidence intervals based on the selective inference framework to account for the model selection stage. We illustrate the methods in a dataset of a treatment for Alzheimer disease (normal response) and in a dataset of a treatment for prostate cancer (survival outcome). We also evaluate via simulations the performance of using the predicted individual treatment effect to identify subgroups where a novel treatment leads to better outcomes compared to a control treatment.