Accumulation and Other Comparisons

AutoPilot Toolkit can run analyses to compare output from previously run NCA models. The Accumu­lation, Absolute Bioavailability, and Renal Clearance Comparison objects combine output from previ­ously created NCA models to create additional report output.

The AP Accumulation, Absolute Bioavailability, and Renal Clearance Comparison objects compare output from two studies. Comparison projects use NCA output or study data that is stored locally or in Certara PKS scenarios. The NCA projects can be imported into Phoenix from a disk or loaded from PKS. If PKS is used, then all NCA output must reside in the same PKS study.

Note:There are a number of situations where comparing treatment values of differing levels between two studies can create output that may not be useful. There are also cases where such compari­sons may prove valuable. The AutoPilot Toolkit does not make any assessment of when such comparisons should be done or not. For this reason, the user should be careful in the selection of the study data to be processed by the AutoPilot Toolkit’s Comparison tool, particularly when there are treatments in one study that do not appear in the other.

Use one of the following to add a Comparison object to a Workflow:

Right-click menu for a Workflow object: New > AutoPilot > AP Accumulation/Absolute Bio­availability/Renal Clearance Comparison.
Or Main menu: Insert > AutoPilot > AP Accumulation/Absolute Bioavailability/Renal Clear­ance Comparison.
Or right-click menu for a worksheet: Send To > AutoPilot > AP Accumulation/Absolute Bio­availability/Renal Clearance Comparison.

An AP Comparison object must have an assigned source of input data for the Reference run and at least one other Test run as a first step. If stacked data is being used, then only one input data source is needed. When the input sources are selected, AutoPilot Toolkit then creates the AP Comparison interface based on the input data.

 

When connected to an NCA object, the AP Comparison object detects any changes to the NCA model and generates an alert. The AP Comparison object does not correct the problem. It only alerts users that changes were detected. Users must revert their NCA changes or make the necessary changes to the AP Comparison object.

The Send To command cannot be used to map data to the Test inputs. The Select source button or the Diagram tab must be used to map data to the Test inputs. The Input checkboxes in the Setup tab are also a quick way to map/unmap sources to the reference and various tests.

To change a source of input data

The source of the input data for an Comparison object can be changed by simply remapping the input to the new source. AutoPilot Toolkit will check the compatibility of the new source’s study variables with the variables in the Comparison object’s original dataset.

This section contains the following topics:

Comparison types

List of output types

Input panel

Table panels

Graph panels

General tab

Stratification/Normalization tab

Display tabs

Ordering tab

Comparison results

See also:

“PK Comparison tables” for a listing of tables available for each combination of study design, dosing, regimen, and matrix.

“PK Comparison graphs” for a listing of graphs available for each combination of study design, dosing, and matrix.

“PK Comparison appendix output” for a listing of appendices available for each study design.

Comparison types

The differences between the three Accumulation, Absolute Bioavailability, and Renal Clearance com­parison types are outlined below.

Comparison Type

Input Data Requirements

Output

Accumulation

Both datasets must be either stacked or unstacked.

Same analyte must be compared between the two datasets.

NCA Models 200–202a 

Runs must use same matrix and route.

Reference input data must be from a single-dose NCA model.

Test input data must be from a multiple-dose NCA model.

Overlaid time-concentration graphs

Calculation of additional parameters such as Accumu­lation (RA) and Linearity (LI). Comparisons are between the single-dose day and a multi-dose day, but not between two multi-dose days.

Absolute Bio­availability

Both datasets must be either stacked or unstacked.

Same analyte must be compared between the two datasets.

Study design type must be Crossover.

NCA Models 200–202a 

Runs must use same matrix and regimen.

Reference input data must be from a non-extravascular route containing one Treat­ment, which is either IV Bolus (model 201) or IV Infusion (model 202).

Test input data must be from an extravascu­lar route (model 200) and can have multiple Treatments.

Overlaid time-concentration graphs

Calculation of additional parameters such as Absolute Bioavailability (F).

Calculation of ratios for PK parameters.

If route information is unavail­able from the reference and test studies, the text strings “IV” and “Ext” will be used in legends or X-axis levels as appropriate.

Renal Clear­ance

Both datasets must be either stacked or unstacked.

Same analyte must be compared between the two datasets.

NCA Models 200–212a 

Runs must use the same route and regimen.

Reference input data must be from a non-urine-based matrix (model 200–202).

Test input data must be from a urine-based matrix (model 210–212).

For renal clearance calculation, the urine and blood draw collection intervals must be the same.

Calculation of additional parameters such as Renal Clearance (CLr=Ae/AUClast).

Calculation of ratios for PK parameters.

aEach NCA model whose output is being compared must be of the same study design type. For PKS studies, all NCA output must reside in the same PKS study. Additionally, only one NCA model is required for any comparison if stacked data is used. For non-stacked comparisons, two NCA models are required.

Note:Phoenix does not allow users to select steady state dosing for urine models. AutoPilot Toolkit uses the steady state flag to distinguish between SD (single dose) and MD (multiple dose) regimens. Therefore, AutoPilot Toolkit considers a urine study to be MD or steady state if Day is used as a Sort Variable and SD if Day is not used as a Sort Variable.

List of output types

The Setup tab consists of two areas, a hierarchical listing consisting primarily of output types available for the AP Analyte Comparison object selected in the Object Browser, and a panel area for displaying options specific to an item selected in the hierarchical list.

accabsbiocomp01277.png 

Setup tab for Accumulation, Absolute Bioavailability, and Renal Clearance Comparison objects

The Input list will vary depending on whether a stacked or unstacked data source was mapped to the Reference item. Test data inputs are not available if a stacked dataset is used as the reference.

To identify types of output:

Note:Selecting Tables in the hierarchical list is only possible when stratifications or exclusions are set, or when the input data is stacked by analyte.

Input panel

When an AP Comparison object is inserted into a project, the input source(s) must be assigned before the object can be used or modifications to object settings can be made. The input source can be mapped to NCA Final Parameters worksheets or observations datasets.

Note:The reference and test data sets used in a comparison must contain matching units for the vari­ables being compared.

OR

In the Diagram tab, right-click an AP Comparison object and select View Setup.

Comp_Input_panel_no_source.png 

Setup tab with no input source defined

If no input sources have been defined, the options available are restricted to selecting the source and specifying an alternative source for configuration settings (refer to “General tab” for more information on configuration settings).

Table panels

Users can set the variables, statistics, or precision for each table type. The options available depend on the type of table selected.

The following sections describe the table options available:

“Main Tables panel”

“Variables and Statistics tabs”

“Standard/Normalize tab”

“Precision tab”

Main Tables panel

The main Tables panel shows options that can be applied when generating the tables. The options vary depending on whether stratifications are defined.

Main_Table_Comparison.png 

Tables panel for unstacked data

AbsBio_Comp_Tables_panel.png 

Tables panel for stacked data

Note:At least one stratified table must be selected if stratifications are specified or the Comparison object will not pass verification.

Variables and Statistics tabs

The Variables and Statistics tabs are formatted the same for most tables.

By default, AP Comparison output can include all PK parameters from the PK_Parameter (A – F) tables or the Intext tables and the additional parameters listed under “PK comparison parameters”.

For Absolute Bioavailability and Accumulation comparisons, the PK Parameters that are common in all NCA models are listed in the Selected list and all other parameters are listed in the Unselected list.

For Renal Clearance comparisons, since there are no common Plasma and Urine parameters, all parameters from either run are available in the comparison Unselected list.

See “PK Parameters” for a full list and descriptions of supported PK parameter study variables. See “Summary Statistics” for a full list and descriptions of supported statistics.

Note:The Variables tab may/may not be available, depending on the table type.

Variables and statistics that are in the Selected column will be included in the output and will be reported in the order that they appear in the column.

Comparison_Tables_panel_1.png 

Statistics tab

The following instructions apply to both the Variables tab and the Statistics tab.

Intext Table Statistics tab

The Statistics tab for Intext PK Parameter tables contains different options than Statistics tabs for other tables.

Intext_Table_Statistics_tab_2.png 

Statistics tab for intext PK parameters

Standard/Normalize tab

This tab becomes available for PK Parameter tables when normalization schemes are defined (see “Stratification/Normalization tab”).

Precision tab

For each variable and statistic, the precision can be set by the number of significant digits or decimal places. Selection of the type and value of numerical precision is also done through this tab.

Precision_1_2.png 

Precision_2_2.png 

Precision tab

Graph panels

AutoPilot Toolkit allows the user to apply different attributes to each graph. These attributes include Y-axis scaling, summary value display, error bar display, and regression line options. Selection of PK parameters to include in the graphs is also available.

The following sections describe the graph options available for each graph type:

“Main Graphs panel”

“Time Concentration panel”

“Comparison Categorical Standard panel”

“Comparison Categorical Box & Whisker panel”

“Continuous Demographic panel”

Main Graphs panel

Parameters that are in the Selected column will be included in the output.

PK_parameter_Graphs_menu_2.png 

Graphs panel

Time Concentration panel

Time Concentration graphs are available for Absolute Bioavailability and Accumulation Comparisons, but not for Renal Clearance Comparisons.

AbsBio_Comp_time_and_conc_graph_options_1.png 

AbsBio_Comp_time_and_conc_graph_options_2.png 

Time Concentration panel

There are two types of Table Concentration graphs available:

The panel displays a table of options for the Time Concentration graphs, grouped into several catego­ries and sub-categories:

Output 

Time_Concen_Comparison_Graph_Strat_Options_1.png 

Output section of Time Concentration panel

Display 

In the Graph section, check the box to use the Y-axis scaling method. Unselect to not scale the Y-axis. Selecting this checkbox under both the Linear and Log sections will generate two graphs, one using each method.

For Summary by Treatment type, select the Summary Value to use for plotting the sum­mary line: Mean, Median, Geometric Mean, Harmonic Mean.

Specify the Value (SD, SE, Variance, Min and Max, None, 68% Range) and Direction   (Both, Down, Up) of error bars to display on Summary by Treatment graphs.

Comparison Categorical Standard panel

Comparison_Categorical_Std_Graph_Options_1.png 

Categorical Standard panel

The PK Parameters available in the study are listed as rows in the table.

The panel displays a table of options for the Comparison Categorical Standard graphs, grouped into several categories and sub-categories:

Output 

If normalization schemes have also been defined (see “Stratification/Normalization tab”), check/uncheck the Normalize subcategory boxes to normalize/not normalize the graphs.

accabsbiocomp01278.png 

Output section of Categorical Standard panel

If normalization schemes have also been defined, check/uncheck the Normalize subcategory boxes to normalize/not normalize the graphs.

Display 

Select the statistic to use as the Summary Value when plotting the summary line: Mean, Median, Geometric Mean, Harmonic Mean.

Specify the Value (Min and Max, Pseudo SD, SD, SE, Variance, 68% Range) of the Error bars. The only option available for Direction is Both.

For Treatment Display, check the All in same graph box to generate a graph that includes data from all treatments. Select the Separate per graph checkbox to generate a separate graph for each treatment.

Comparison Categorical Box & Whisker panel

Note:There must be at least three subjects in the study to create Box & Whisker graphs.

Ana_Comp_Categorical_Box_Graph_Options_1.png 

Categorical Box and Whisker panel

The PK Parameters available in the study are listed as rows in the table.

Output 

If normalization schemes are also defined (see “Stratification/Normalization tab”), check/uncheck the Normalize subcategory boxes to normalize/not normalize the graphs.

accabsbiocomp01280.png 

Output section of Categorical Box and Whisker panel

If normalization schemes have also been defined, the stratified graphs can be normalized using the Normalize subcategory checkboxes.

Use the Selected checkboxes to produce stratified graphs that are not normalized.

Treatment Display 

Continuous Demographic panel

Comp_Continuous_Demographic_Graph_Options_1.png 

Setup tab for Continuous Demographic graphs

The panel displays a section for selecting up to two demographic(s) to use for the X-axis (Demo­graphic 1 and Demographic 2).

When a second demographic type is selected, a second tab is created in the Comparison Contin­uous Demographic panel.

Continuous_Demographic_Graph_2tabs_2.png 

Each demographic type has its own tab in the Continuous Demographic panel

The PK Parameters available in the study are listed as rows in the table. Each parameter will have two sub-rows:

The lower part of the panel contains a table of options for the Comparison Continuous Demographic graphs, grouped into several categories and sub-categories:

Output 

If normalization schemes are also defined (see “Stratification/Normalization tab”), check/uncheck the Normalize subcategory boxes to normalize/not normalize the graphs.

accabsbiocomp01282.png 

Output section of Continuous Demographic panel

If normalization schemes have also been defined, check/uncheck the Normalize subcate­gory boxes to normalize/not normalize the graphs.

Regression 

General tab

The General tab allows users to select the study design type, configuration settings, and whether or not to try to complete a comparison run if an error occurs.

General_tab_2.png 

General tab in the Properties panel

The study design options in the General tab depend on the configuration settings. If the settings are changed, then the options could be different from the options listed below.

Note:The configuration settings must be specified before a dataset is mapped to the AP Comparison object.

Study Design 

Configuration Settings 

Stratification/Normalization tab

The Stratification and Normalization options allow users to create additional table and/or graph out­put.

Stratification-Normalization_tab_2.png 

Stratification/Normalization tab in the Properties panel

Stratification 

Results can be stratified (i.e., layered) using discrete demographic variables. Each stratification level can use one or two discrete demographic variables. If two variables are specified, they are associated using the logical operator AND.

Note:At least one stratified output type must be selected if stratification is enabled or the Comparison project will not pass verification.

If stratifications are selected, the comparison run creates one table per stratum for the time and con­centration, PK parameter, and intext PK parameter tables, using the stratification scheme as an addi­tional group variable.

If graphs include stratification, the stratification schemes are used either as X-axis variables (sorted by treatment) or new sort variables (sorted by demographics), depending on the AutoPilot Toolkit Admin settings.

Normalization 

Use the Normalization section to define normalization schemes to apply to the results. Each normal­ization scheme must use a different continuous demographic variable.

AutoPilot Toolkit calculates the normalized PK parameters and includes them in the results. Users can select the PK Parameter and Intext PK Parameter tables in the hierarchical list and choose which normalized parameters to display in each table. This allows PK Parameter automation tables to include both normalized and non-normalized values.

For more on using the table panels, see “Table panels”.

Column headers for the normalized variables include a normalization variable and its units. For exam­ple, oral clearance normalized by weight: CL/F/Weight (L/hr/kg). If graph output is selected that includes normalization, each normalized PK parameter is displayed in a separate graph. The Y-axis labels display the normalization in the same manner as tables.

PK parameters that are excluded from normalization are listed in “PK automation parameters”.

Display tabs

The Display tab contains four tabs that allow users to set output and display options, table and graph orientation, and the X- and Y-axes scaling for graphs.

Output Options tab

The Output Options tab allows users to define exclusions, the LOQ value, and the AUC percent extrapolation threshold value.

Comparison_Display_tab_1.png 

Output Options — Display tab in the Properties panel

The options for LOQ vary depending on the type of data used and the system configuration settings.

To set the LOQ value using the input data, click Use setting from data.

To enter a value, click Use value and type a value in the corresponding field.

Click No limit of quantification to not set an LOQ limit.

If the input dataset contains stacked data, a different LOQ can be set for each analyte.

Comparison_Display_-_Output_Options_tab_for_Stacked_data_1.png 

Output options for stacked data

Turn on the Use LOQ values found in study data checkbox to set the values for LOQ using the input data.

Enter an LOQ value for each analyte in the Value column. (The concentration units are taken from the input dataset.)

To not use LOQ values, turn off the Use LOQ values found in study data checkbox and leave the Value column entries blank.

Note:Setting the LOQ value for all analyts can significantly extend the execution time.

For more information, see “LOQ replacement”.

The following option is applicable to both stacked or unstacked input data:

See “PK parameter percent-extrapolated threshold” for details.

Display Options tab

The Display Options tab allows users to set table and graph output display options, select the time and concentration variables in the input dataset, and choose whether or not to include data source information.

Display_-_Display_Options_tab_2.png 

Display Options — Display tab in the Properties panel

Orientation tab

Through the Orientation tab, the orientation of each output item is set. A few additional settings regarding the appearance of graphs in a Word document are available for appendices involving indi­vidual time-concentration graphs.

Comparison_Display_-_Orientation_tab_1.png 

Orientation — Display tab in the Properties panel

Note:Only certain tables and graphs can be changed to Landscape. If Landscape is not supported, the Orientation setting for that table or graph defaults to Portrait and the pull-down menu is disabled.

For the Individual Time Concentration appendix output type, the following specifications can also be made:

All linear then all log: Display the linear graphs (sorted by subject) before the log graphs (sorted by subject).

All log then all linear: Display the log graphs (sorted by subject) before the linear graphs (sorted by subject).

Per profile, linear then log: Graphs are grouped by subject and then by analyte, with the lin­ear graph presented before the log graph. In the output, the graphs are displayed in subject order.

Per profile, log then linear: Graphs are grouped by subject and then by analyte, with the log graph presented before the linear graph. In the output, the graphs are displayed in subject order.

Time Scale Algorithm tab

The Time Scale Algorithm tab is used to specify the scaling options for the axes, the lower and upper bounds for time scale ticks, the tick frequency, the tick units, and the maximum time scale on the X-axis. See “Time scale algorithm” for more information.

Display_-_Time_Scale_Algorithm_2.png 

Time Scale Algorithm-Display tab in the Properties panel

The Major Ticks area contains a table where each row represents a separate time scale.

Use the Tick frequency and Tick units columns together to define the frequency with which tick marks are displayed along the X-axis.

A new row is added to the table below the row that was selected or modified last.

A minimum of two defined time scales is required.

Ordering tab

The Ordering tab is used to specify how the treatment descriptions and demographic study variables are ordered in the output.

Ordering_tab.png 

Treatments — Ordering tab in the Properties panel

Discrete_Variables_tab_1.png 

Discrete Variables — Ordering tab in the Properties panel

Analytes tab

This tab is present only if an Accumulation, Absolute Bioavailability, or Renal Comparison project uses stacked datasets for reference and test inputs.

When a stacked data PK Automation project is chosen for comparison, and an analyte compari­son type is selected, users can select the analyte to be used as reference, as well as change the display order of the remaining test analytes.

Comparison results

Caution:Do not perform any operations on the computer while the comparison run is in progress. Doing so could cause unpredictable results; keyboard and mouse input during a comparison run might affect automated AutoPilot Toolkit operations.

After the project is run, all output is arranged in groups in the Results tab.

Not all output can be viewed in Phoenix. In such cases, the right side of the Results tab will display a message with suggestions on how to view the results. One suggestion is to open an external program and load the results by clicking View in External Viewer.

AutoPilot Toolkit output can be individually exported to disk or copied to Phoenix’s Data folder. All results can be exported using AutoPilot File Explorer, which is located in the Reporting tab. For more using AutoPilot File Explorer, see “AutoPilot File Explorer”.

 


Last modified date:6/26/19
Certara USA, Inc.
Legal Notice | Contact Certara
© 2019 Certara USA, Inc. All rights reserved.