PFIM 7.0 – What’s New

<style> body { max-width: 960px; margin: 2em auto; font-family: “Lato”, “Helvetica Neue”, Helvetica, Arial, sans-serif; line-height: 1.7; color: #DDD; background-color: #121212; -webkit-font-smoothing: antialiased; -moz-osx-font-smoothing: grayscale; }

h1, h2, h3 { margin-top: 1.5em; margin-bottom: 0.75em; font-weight: 700; color: #EEE; }

pre, code { font-family: Consolas, Monaco, “Courier New”, monospace; font-size: 90%; background-color: #1E1E1E; color: #DDD; padding: 0.3em 0.5em; border-radius: 4px; overflow-x: auto; }

a { color: #58A6FF; text-decoration: none; }

a:hover { text-decoration: underline; }

img { border: none; } </style>

This version fully replaces the version of PFIM 6.1

Complete Package Redesign

• Functional Programming Architecture: Version 7.0 is entirely rebuilt using functional programming principles for improved performance and maintainability.

• Breaking Changes: This version introduces significant user script changes that require script updates from previous versions

• Enhanced Features: New modeling capabilities and improved Bayesian FIM estimation

Key New Features

• Flexible PK/PD Modeling: Support for multiple PK responses with PD components (ode pk pk pk .. pd)

• Expanded User Defined Model: Additional models available (see BJCP test in https://github.com/packagePFIM)

• Bayesian FIM Estimation: Enhanced estimation for single sampling time scenarios

• Multi-dose Bolus Models: New ODE-based multi-dose bolus modeling capabilities

Migration Guide from PFIM 6.1 to PFIM 7.0

1. Equations of a user defined model

• PFIM 6.1:

{r, echo = TRUE, comment='',eval=FALSE} modelEquations` = `list( outcomes = list( "RespPK", "RespPD" ), equations = list( "RespPK" = "dose_RespPK/V * ka/( ka - Cl/V ) * ( exp( -Cl/V *t ) - exp(-ka * t ) )", "RespPD" = "S0 * ( 1 - Imax * RespPK/( RespPK + C50 ) )" ) ) - PFIM 7.0: {r, echo = TRUE, comment='',eval=FALSE} modelEquations = list( "RespPK" = "dose_RespPK/V * ka/( ka - Cl/V )*( exp( -Cl/V * t ) - exp( -ka * t ) )", "RespPD" = "S0 * ( 1 - Imax * RespPK/( RespPK + C50 ) )" )

2. Model equations from the library of models.

PFIM 7.0 introduces modelFromLibrary for predefined models:

{r, echo = TRUE, comment='',eval=FALSE} modelFromLibrary = list( "PKModel" = "MichaelisMenten2InfusionSingleDose_VmKmk12k21V1V2", "PDModel" = "TurnoverRinEmax_RinEmaxCC50koutE" )

and the evaluation writes:

{r, echo = TRUE, comment='',eval=FALSE} evaluation = Evaluation( modelFromLibrary = modelFromLibrary, ... )

without the slots modelEquations.

3. Model error

The outcome od an error model is modified.

• PFIM 6.1: outcomes
• PFIM 7.0: outputs

4. Arm

The specification of the dose in the initial conditions for an ode model is modified.

• PFIM 6.1: initialCondition = list( "C1" = "dose" )
• PFIM 7.0: initialCondition = list( "C1" = "dose_C1" )

5. Administration constraints

The constraints for the administration for the design optimization is changed.

• PFIM 6.1 (vector): doses = c( 0.2, 0.64, 2, 6.24, 11.24, 20 )
• PFIM 7.0 (list): doses = list( 0.2, 0.64, 2, 6.24, 11.24, 20 )

6. Evaluation

The evaluation outcomes are changed.

• PFIM 6.1: outcomes = list( "RespPK" )
• PFIM 7.0: outputs = list( "RespPK" )

The specification of the Fim type is changed.

• PFIM 6.1: fim = " ... "\
• PFIM 7.0 fimType = " ... "

7. Output methods

The plot method for the sensitivity indices is changed.

• PFIM 6.1: plotSensitivityIndice = plotSensitivityIndice( ..., ... )
• PFIM 7.0: plotSensitivityIndices = plotSensitivityIndices( ..., ... )
• Correlation matrix removed from report in this version
• SE and RSE extraction simplified with direct access without [[designName]] indexing