KNOWLEDGE / Publication / POST
August 7, 2017

Scientific Poster Publication made on June 2017 at the Promoting Statistical Insight Conference, London (UK).

In analgesia randomized clinical trials (RCTs), the magnitude and the variability of the placebo response have a negative influence when testing the statistically significant superiority of active compounds compared to placebo. Furthermore, the magnitude of this effect has tended to increase over time, including in peripheral neuropathic pain (PNP) trials. The main objective of this study was to investigate parameters influencing the placebo response as a way to control this major confounding factor. Eighty-seven PNP patients were enrolled and blindly given a placebo during 4 weeks. The placebo response was estimated as the difference in pain between baseline and end of the treatment. In addition, patients filled a psychological questionnaire at baseline assessing several components of their personality.

We modeled the placebo response from patient’s characteristics using a Bayesian machine learning
approach: Gaussian processes with a linear kernel. The covariates used in the model were selected
using a multivariate recursive feature elimination (RFE). The advantage of this Bayesian modeling is to
predict the placebo response and to give confidence intervals on the predictions. The predictive performances of this model were estimated in a repeated random sub-sampling scheme (or Monte Carlo cross-validation). The model explained almost 30% of the variance in new patients
(pvalue< 0.001).

Using the model predictions as a covariate could thus reduce the placebo variance by 30% in
subsequent PNP studies. This reduction of variance could in turns lead to an increased effect size and
study power. Such a tool to characterize and predict this important source of variance would thus be of great value in analgesia randomized clinical trials.

Scientific Poster
Samuel Branders, PhD; Alvaro Pereira, PhD; Frederic Clermont, PhD; Chantal Gossuin; Dominique Demolle, PhD
June 1, 2017
Promoting Statistical Insight Conference


Senior Project Leader

Related content


Correcting For The Individual Patient Regression To The Mean Effect

Often, the primary endpoint of RCTs is defined as a change from baseline of a continuous outcome. In…

Type: Scientific Poster
Authors: Samuel Branders, PhD; Guillaume Bernard, PhD; Alvaro Pereira, PhD
Conference: American Society for Clinical Pharmacology and Therapeutics
Read More

Do Environmental Parameters Influence The Prediction Of The Placebo Response?

This proof-of-concept study on peripheral neuropathic pain patients investigates the potential influence of the investigator on the placebo…

Read More

Leveraging Historical Data For High-dimensional Covariate-adaptive Randomization, A Machine Learning Approach.

There is a continuous growth in data collected in clinical trials. Many of those patient's characteristics are potential…

Type: Scientific Presentation
Authors: Samuel Branders, PhD; Guillaume Bernard, PhD; Alvaro Pereira, PhD
Conference: Promoting Statistical Insight Conference
Read More

The next frontier in clinical research & patient management

We’re proud to be leading the charge into the next era of drug development.
Cognivia helps clinical trials reduce data variability, empower decision-making, and accelerate the launch of new therapies.
Tell us about your clinical trial below and we’ll be in touch.

"*" indicates required fields

This field is for validation purposes and should be left unchanged.