Treatment response prediction: Is model selection unreliable?

Augustin, David; Wang, Ken; Walz, Antje-Christine; Lambert, Ben; Clerx, Michael; Robinson, Michael F.; Gavaghan, David J.

doi:10.1101/2022.03.19.483454

Cited by 2 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This indicates that the PKPD model oversimplifies crucial elements of the treatment response mechanisms, resulting in a tendency to underestimate the treatment response of individuals. The risk for model misspecification is a generic limitation of PKPD modelling, which needs to be mitigated prior to clinical applications, for example by quantifying the structural uncertainty of PKPD models using model selection criteria or probabilistic model averaging ( Uster et al, 2021 ; Augustin et al, 2022 ).…”

Section: Resultsmentioning

confidence: 99%

“…This indicates that across applications PKPD models are the most promising approach for precision dosing. However, PKPD models are more susceptible to model misspecifications than the other two approaches (see Section 3.5.3 ), necessitating a careful evaluation of the predictive uncertainty of the model prior to MIPD, for example by means of model selection criteria or probabilistic model averaging ( Augustin et al, 2022 ).…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Simulating clinical trials for model-informed precision dosing: using warfarin treatment as a use case

Augustin,

Lambert,

Robinson

et al. 2023

Front. Pharmacol.

Self Cite

View full text Add to dashboard Cite

Treatment response variability across patients is a common phenomenon in clinical practice. For many drugs this inter-individual variability does not require much (if any) individualisation of dosing strategies. However, for some drugs, including chemotherapies and some monoclonal antibody treatments, individualisation of dosages are needed to avoid harmful adverse events. Model-informed precision dosing (MIPD) is an emerging approach to guide the individualisation of dosing regimens of otherwise difficult-to-administer drugs. Several MIPD approaches have been suggested to predict dosing strategies, including regression, reinforcement learning (RL) and pharmacokinetic and pharmacodynamic (PKPD) modelling. A unified framework to study the strengths and limitations of these approaches is missing. We develop a framework to simulate clinical MIPD trials, providing a cost and time efficient way to test different MIPD approaches. Central for our framework is a clinical trial model that emulates the complexities in clinical practice that challenge successful treatment individualisation. We demonstrate this framework using warfarin treatment as a use case and investigate three popular MIPD methods: 1. Neural network regression; 2. Deep RL; and 3. PKPD modelling. We find that the PKPD model individualises warfarin dosing regimens with the highest success rate and the highest efficiency: 75.1% of the individuals display INRs inside the therapeutic range at the end of the simulated trial; and the median time in the therapeutic range (TTR) is 74%. In comparison, the regression model and the deep RL model have success rates of 47.0% and 65.8%, and median TTRs of 45% and 68%. We also find that the MIPD models can attain different degrees of individualisation: the Regression model individualises dosing regimens up to variability explained by covariates; the Deep RL model and the PKPD model individualise dosing regimens accounting also for additional variation using monitoring data. However, the Deep RL model focusses on control of the treatment response, while the PKPD model uses the data also to further the individualisation of predictions.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Simulating clinical trials for model-informed precision dosing: using warfarin treatment as a use case

Augustin,

Lambert,

Robinson

et al. 2023

Front. Pharmacol.

Self Cite

View full text Add to dashboard Cite

show abstract

Section: The Pkpd Modelmentioning

confidence: 99%

Simulating clinical trials for model-informed precision dosing: Using warfarin treatment as a use case

Augustin

Lambert

Robinson

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Treatment response variability across patients is a common phenomenon in clinical practice. For many drugs this inter-individual variability does not require much (if any) individualisation of dosing strategies. However, for some drugs, including chemotherapies and some monoclonal antibody treatments, individualisation of dosages are needed to avoid harmful adverse events. Model-informed precision dosing (MIPD) is an emerging approach to guide the individualisation of dosing regimens of otherwise difficult-to-administer drugs. Several MIPD approaches have been suggested to predict dosing strategies, including regression, reinforcement learning (RL) and pharmacokinetic and pharmacodynamic (PKPD) modelling. A unified framework to study the strengths and limitations of these approaches is missing. We develop a framework to simulate clinical MIPD trials, providing a cost and time efficient way to test different MIPD approaches. Central for our framework is a clinical trial model that emulates the complexities in clinical practice that challenge successful treatment individualisation. We demonstrate this framework using warfarin treatment as a use case and investigate three popular MIPD methods: 1. neural network regression; 2. deep RL; and 3. PKPD modelling. We find that the PKPD model individualises warfarin dosing regimens with the highest success rate and the highest efficiency: 75.1% of the individuals display INRs inside the therapeutic range at the end of the simulated trial; and the median time in the therapeutic range (TTR) is 74 %. In comparison, the regression model and the deep RL model have success rates of 47.9% and 65.8 %, and median TTRs of 45 % and 68 %. We also find that the MIPD models can attain different degrees of individualisation: the Regression model individualises dosing regimens up to variability explained by covariates; the Deep RL model and the PKPD model individualise dosing regimens accounting also for additional variation using monitoring data. However, the Deep RL model focusses on control of the treatment response, while the PKPD model uses the data also to further the individualisation of predictions.

show abstract

Treatment response prediction: Is model selection unreliable?

Cited by 2 publications

References 42 publications

Simulating clinical trials for model-informed precision dosing: using warfarin treatment as a use case

Simulating clinical trials for model-informed precision dosing: using warfarin treatment as a use case

Simulating clinical trials for model-informed precision dosing: Using warfarin treatment as a use case

Contact Info

Product

Resources

About