Scripts to generate the figures and tables of Towards Developing Alternative Opioid Antagonists for Treating Community Overdose – A model-based evaluation of important pharmacological attributes

Anik Chaturbedi, John Mann, Zhihua Li

Division of Applied Regulatory Science, Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, Food & Drug Administration, Silver Spring, Maryland, USA

1. Brief description of models and parameters

1a. Models

Model for simulating intravenous opioid overdose and intranasal naloxone administration: models/modelIN_2Tr1C

Model for simulating intravenous opioid overdose and intranasal nalmefene administration: models/modelIN_2Tr1C1P_RepeatedDosing

Model for simulating transmucosal opioid overdose and intranasal naloxone administration: models/model_TM_2Tr1C1P_IN_2Tr1C

Model for simulating transmucosal opioid overdose and intranasal nalmefene administration: models/model_TM_2Tr1C1P_IN_2Tr1C1P_RepeatedDosing

In each of these folders delaymymod files contain the model equations described in detail in Mann et al. and other files serve as auxillary files.

1b. Parameters

The parameters for a "typical" subject can be found in parameters/optimalParameters . The opioid (fentanyl and carfentanil) pharmacokinetic and receptor binding parameters are available in parameters/optimalParameters/opioid . Pharmacokinetic parameters and receptor binding parameters for IN naloxone and IN nalmefene are provided in parameters/optimalParameters/antagonist . The parameters representing various physiological variables which are used in the physiological and ventilatory component of the model are provided in parameters/optimalParameters/physiological/physiologicalParameters . The pharmacodynamic parameters for a chronic opioid user is given in parameters/optimalParameters/subject/chronic .

The corresponding parameters for all of the 2000 subjects in the virtual population are available in parameters/populationParameters . The opioid pharmacokinetic and receptor binding parameters are available in parameters/populationParameters/opioid . Similar to the "typical" subject, fentanyl pharmacokinetic parameters are modified to match a longer half-life for carfentanil. The pharmacokinetic and receptor binding parameters for IN naloxone and IN nalmefene are available in parameters/populationParameters/antagonist .

Additional simulation related parameters used both for simulating the "typical" subject as well as the virtual population are defined in input/simulationParameters .

1c. Clinical data

The data used for plotting are in data. Note: data/nalmefeneCIndividualSubjectData.csv and data/naloxoneIndividualSubjectData.csv were digitized from Figure 2.

2. Workflow

2a. Pharmacokinetics

2a1. Figure 2 Pharmacokinetics of the various µ-receptor antagonist formulations studied in this work.

1. run SimulateVirtualPopulationsAntagonistPKOnly.sh (OR simulateVirtualSubject.R)
2. run SimulateVirtualSubjectsAntagonistPKOnly.sh (OR simulateVirtualSubject.R)
3. run PlottingAntagonistPKTimeCourseAndPKParameters.sh (or plottingAntagonistPKTimecourseAndPKParameters.R)

2b. IV opioid overdose simulations

2b1. Figure 3 Simulated antagonist induced reversal for various scenarios of opioid overdose with intravenous fentanyl and carfentanil.

1. run SimulateVirtualSubjects.sh (or simulateVirtualSubject.R)
2. run PlotVirtualSubjects.sh

2b2. Figure 4 Immediate antagonist induced recovery in the aftermath of various levels of opioid overdose with intravenous fentanyl and carfentanil in a virtual population.

1. run SimulateVirtualPopulations.sh (or simulateVirtualSubject.R)
2. run CalculatePopulationLevelCAMetrics.sh (or calculatePopulationLevelCAMetric.R)
3. run multipleDoses.R

2c. TM opioid overdose simulations

2c1. Figure 5 Simulated prevention of renarcotization induced respiratory depression in a typical subject.

1. run SimulateVirtualSubjectsRenarcotization.sh (OR simulateVirtualSubject.R)
2. run PlotVirtualSubjectsRenarcotization.sh (OR plottingRenarcotization4Cases.R)

2c2. Figure S2 Plasma concentration of long exposure opioid formulation used here to simulate renarcotization.

1. run MatchOpioidPlasmaConcentration.sh (or simulateVirtualSubject.R followed by plottingRenarcotizationPlasmaConcentration.R)

2c3. Figure S3 Plasma concentration of the various antagonist formulations studied in this work in a “typical” subject.

1. run SimulateVirtualSubjectsRenarcotization.sh (OR simulateVirtualSubject.R)
2. run PlotVirtualSubjectsRenarcotization.sh (OR plottingRenarcotization4CasesAPC.R)

2c4. Table 1 Minute ventilation with various antagonist formulations (and without any antagonist) at different times after exposure to an opioid with slower absorption.

1. run SimulateVirtualPopulationsRenarcotization.sh (or simulateVirtualSubject.R)
2. run CalculatePopulationLevelMVMetricsRenarcotization.sh (or calculatePopulationLevelMVMetric.R)

2c5. Table S5 Plasma concentration for various antagonist formulations at different times, for the case of long-exposure opioid administration.

1. run SimulateVirtualPopulationsRenarcotization.sh (or simulateVirtualSubject.R)
2. run CalculatePopulationLevelAPCMetricsRenarcotization.sh (or calculatePopulationLevelAPCMetric.R)

3. Requirements

These codes were developed in R-4.4.1 and C.