Bayesian Pharmacometrics Analysis of Baclofen for Alcohol Use Disorder

Abstract

Alcohol use disorder (AUD) also called alcohol dependence is a major public health problem, which affects almost 10% of the world's population. Baclofen as a selective GABA_B receptor agonist has emerged as a promising drug for the treatment of AUD, however, its optimal dosage varies according to individuals, and its exposure-response relationship has not been well established yet. In this study, we use a principled Bayesian workflow to estimate the parameters of a pharmacokinetic (PK) population model from Baclofen administration to patients with AUD. By monitoring various convergence diagnostics, the probabilistic methodology is first validated on synthetic longitudinal datasets and then, applied to infer the PK model parameters based on the clinical data that were retrospectively collected from outpatients treated with oral Baclofen. We show that state-of-the-art advances in automatic Bayesian inference using self-tuning Hamiltonian Monte Carlo (HMC) algorithms with a leveraged level of information in priors provide accurate predictions on Baclofen plasma concentration in individuals. This approach may pave the way to render non-parametric HMC sampling methods sufficiently easy and reliable to use in clinical schedules for personalized treatment of AUD.

Probabilistic Inference on Virtual Brain Models of Disorders

“…The posterior distribution encompasses all possible parameter combinations that produce a simulation output that explains the data with quantifying the uncertainty in estimation, which is critical for model comparison, selection and averaging, hypothesis testing, and decision-making process. Bayesian inference also allows us to integrate the prior knowledge in inference and prediction (Gelman et al, 2014) e.g., the physiological, anatomical, or clinical knowledge to maximize the model predictive power against measurements (Hashemi et al, 2021; Baldy et al, 2023).…”

“…Although, we have shown that this non-parametric approach is able to accurately estimate the spatial map of epileptogenicity across whole-brain areas, it required a reparameterization over model configuration space to facilitate efficient exploration of the posterior distribution in terms of computational time and convergence diagnostics (Jha et al, 2022). In the presence of metastability in the state space (see Fig 1, Fig S1), MCMC methods require either more computational cost or intricately designed sampling strategies (Gabrié et al, 2022; Jha et al, 2022; Baldy et al, 2023), whereas SBI allows for efficient Bayesian estimation with-out the access to the full knowledge on the state space representation of a system (Baldy et al, 2024). Note that we used the data features derived from only firing rates, while the information related to membrane potential activities was treated as missing data (i.e., no access to the full state space behavior).…”

“…The Marko-vian (sequential) structure of MCMC methods interacts poorly with the highly multimodal, non-convex posteriors of generative VBMs, where nonlinear latent dynamics generally imply multistability in a high dimensional state space. Hence, MCMC methods require restrictive reparametrizations to deal with geometrical issues (Betancourt et al, 2014; Betancourt, 2016), immense computational cost (Baldy et al, 2024), or intricately designed sampling strategies (Hashemi et al, 2020; Jha et al, 2022) to efficiently converge (Gabrié et al, 2022; Baldy et al, 2023).…”

Efficient Inference on a Network of Spiking Neurons using Deep Learning

“…The selection of the objective function plays a crucial role in determining the estimation through optimization methods (Svensson et al, 2012; Hashemi et al, 2018). The error explanation with distance metrics such as RMSE is limited in accurately capturing the underlying data generation process (Baldy et al, 2023). This is because when generative parameters remain unchanged but when dynamic noise is introduced, the time-series can show large fluctuations, resulting in deviations from the observed data.…”

“…In particular, the NUTS calibrates the number of steps and step-size of the leapfrog integrator (in solving the Hamiltonian equations of motion) during a warm-up phase to achieve a target Metropolis acceptance rate. For more details see Betancourt (2013); Baldy et al (2023). Moreover, Stan offers alternative methods such as MAP estimation using L-BFGS optimization, automatic differentiation for efficient gradient computation, and various diagnostics to assess the convergence of the inference process (see https://mc-stan.org).…”