Pharmacometrics (PM) and machine learning (ML) are both valuable for drug development to characterize pharmacokinetics (PK) and pharmacodynamics (PD). Pharmacokinetic/pharmacodynamic (PKPD) analysis using PM provides mechanistic insight into biological processes but is time- and labor-intensive. In contrast, ML models are much quicker trained, but offer less mechanistic insights. The opportunity of using ML predictions of drug PK as input for a PKPD model could strongly accelerate analysis efforts. Here exemplified by rifampicin, a widely used antibiotic, we explore the ability of different ML algorithms to predict drug PK. Based on simulated data, we trained linear regressions (LASSO), Gradient Boosting Machines, XGBoost and Random Forest to predict the plasma concentration-time series and rifampicin area under the concentration-versus-time curve from 0–24 h (AUC0–24h) after repeated dosing. XGBoost performed best for prediction of the entire PK series (R2: 0.84, root mean square error (RMSE): 6.9 mg/L, mean absolute error (MAE): 4.0 mg/L) for the scenario with the largest data size. For AUC0–24h prediction, LASSO showed the highest performance (R2: 0.97, RMSE: 29.1 h·mg/L, MAE: 18.8 h·mg/L). Increasing the number of plasma concentrations per patient (0, 2 or 6 concentrations per occasion) improved model performance. For example, for AUC0–24h prediction using LASSO, the R2 was 0.41, 0.69 and 0.97 when using predictors only (no plasma concentrations), 2 or 6 plasma concentrations per occasion as input, respectively. Run times for the ML models ranged from 1.0 s to 8 min, while the run time for the PM model was more than 3 h. Furthermore, building a PM model is more time- and labor-intensive compared with ML. ML predictions of drug PK could thus be used as input into a PKPD model, enabling time-efficient analysis.

Pharmaceutical sciences experts and regulators acknowledge that pharmaceutical development as well as drug usage requires more than scientific advancements to cope with current attrition rates/therapeutic failures. Drug disease modeling and simulation (DDM&S) creates a paradigm to enable an integrated and higher-level understanding of drugs, (diseased)systems, and their interactions (systems pharmacology) through mathematical/statistical models (pharmacometrics)1—hence facilitating decision making during drug development and therapeutic usage of medicines. To identify gaps and challenges in DDM&S, an inventory of skills and competencies currently available in academia, industry, and clinical practice was obtained through survey. The survey outcomes revealed benefits, weaknesses, and hurdles for the implementation of DDM&S. In addition, the survey indicated that no consensus exists about the knowledge, skills, and attributes required to perform DDM&S activities effectively. Hence, a landscape of technical and conceptual requirements for DDM&S was identified and serves as a basis for developing a framework of competencies to guide future education and training in DDM&S.

Context GLP-1 receptor agonists are an established therapy in patients with type 2 diabetes; however, their role in type 1 diabetes remains to be determined. Objective Determine efficacy and safety of once-weekly albiglutide 30 mg (up-titration to 50 mg at week 6) versus placebo together with insulin in patients with new-onset type 1 diabetes and residual insulin production. Design 52-week, randomized, phase 2 study (NCT02284009). Methods A prespecified Bayesian approach, incorporating placebo data from a prior study, allowed for 3:1 (albiglutide:placebo) randomization. The primary endpoint was 52-week change from baseline in mixed meal tolerance test (MMTT) stimulated 2-h plasma C-peptide area under the curve (AUC). Secondary endpoints included metabolic measures and pharmacokinetics of albiglutide. Results 12/17 (70.6%, placebo) and 40/50 (80.0%, albiglutide) patients completed the study. Within our study, mean (standard deviation) change from baseline to week 52 in MMTT-stimulated 2-h plasma C-peptide AUC was −0.16 nmol/L (0.366) with placebo and −0.13 nmol/L (0.244) with albiglutide. For the primary Bayesian analysis (including prior study data) the posterior treatment difference (95% credible interval) was estimated at 0.12 nmol/L (0–0.24); the probability of a difference ≥0.2 nmol/L between treatments was low (0.097). A transient significant difference in maximum C-peptide was seen at week 28. Otherwise, no significant secondary endpoint differences were noted. On-therapy adverse events were reported in 82.0% (albiglutide) and 76.5% (placebo) of patients. Conclusion In newly diagnosed patients with type 1 diabetes, albiglutide 30 to 50 mg weekly for 1 year had no appreciable effect on preserving residual β-cell function versus placebo.

Background Buruli ulcer (BU) is a neglected tropical disease caused by Mycobacterium ulcerans that affects skin, soft tissues, and bones, causing long-term morbidity, stigma, and disability. The recommended treatment for BU requires 8 weeks of daily rifampicin and clarithromycin together with wound care, physiotherapy, and sometimes tissue grafting and surgery. Recovery can take up to 1 year, and it may pose an unbearable financial burden to the household. Recent in vitro studies demonstrated that beta-lactams combined with rifampicin and clarithromycin are synergistic against M. ulcerans. Consequently, inclusion of amoxicillin/clavulanate in a triple oral therapy may potentially improve and shorten the healing process. The BLMs4BU trial aims to assess whether co-administration of amoxicillin/clavulanate with rifampicin and clarithromycin could reduce BU treatment from 8 to 4 weeks. Methods We propose a randomized, controlled, open-label, parallel-group, non-inferiority phase II, multi-centre trial in Benin with participants stratified according to BU category lesions and randomized to two oral regimens: (i) Standard: rifampicin plus clarithromycin therapy for 8 weeks; and (ii) Investigational: standard plus amoxicillin/clavulanate for 4 weeks. The primary efficacy outcome will be lesion healing without recurrence and without excision surgery 12 months after start of treatment (i.e. cure rate). Seventy clinically diagnosed BU patients will be recruited per arm. Patients will be followed up over 12 months and managed according to standard clinical care procedures. Decision for excision surgery will be delayed to 14 weeks after start of treatment. Two sub-studies will also be performed: a pharmacokinetic and a microbiology study. Discussion If successful, this study will create a new paradigm for BU treatment, which could inform World Health Organization policy and practice. A shortened, highly effective, all-oral regimen will improve care of BU patients and will lead to a decrease in hospitalization-related expenses and indirect and social costs and improve treatment adherence. This trial may also provide information on treatment shortening strategies for other mycobacterial infections (tuberculosis, leprosy, or non-tuberculous mycobacteria infections). Trial registration ClinicalTrials.gov NCT05169554. Registered on 27 December 2021.

There is an evidence gap to ensuring safe/effective use of medicines in older adults. Generating clinical data in these patients poses ethical and operational challenges, yielding results that may not be generalizable to the overall population. Modeling and simulation (M&S) is proposed as a basis for assessing the impact of age-related changes and their clinical implications. M&S can be used in conjunction with bridging and extrapolation to ensure the selection of appropriate dose(s)/regimen(s) in this population.