One Arm Time to Event Simulator
The One Arm Time to Event Simulator is a user friendly implementation of the One-Arm
Time-to-Event trial design.
The statistical method is based on the paper “Monitoring event times in early phase clinical trials: some
practical issues” by Thall, Wooten and Tannir.
NOTE: Information about specific versions is below.
The software was developed by John Aven, Clift Norris, and John Venier.
The software tools were: Microsoft Visual Studio 2015 (Update 3); C#; .Net Framework 4.6.1.
System Requirements
- Windows 10 (but may work on earlier versions of Windows as well, although this has not been tested)
- Microsoft .NET Framework version 4.6.1 (x86 and x64)
- Windows Installer 4.5
- Minimum Screen Resolution 1235x850
If any required software component is absent from your system, the installation process will install it.
About Version 3.0.9
Note: Version 3.0.9 fixes a minor bug in simulations which may affect the reported operating
characteristics. We recommend re-simulating with the latest version (v3.0.9).
About One Arm Time to Event Simulator
-
Model Parameters Tab Page:
Model Parameters Tab Page
-
Prior Distributions:
The method requires the specification of prior distributions for the Standard Treatment Arm and an
Experimental Treatment Arm. The prior distributions are Inverse Gamma distributions.
The prior distributions may be defined by specifying the Mean and Standard Deviation, the shape and
scale, or two quantiles.
-
Mean or Median:
The software supports parameterizing the Exponential Time-to-Event distributions in terms of the
Mean TTE or the Median TTE.
-
Stopping Rules:
Trials can be designed to Stop Early due either to superiority or inferiority (or both if desired).
Also, the design will specify whether maximizing or minimizing the time-to-event is desirable.
-
Trial Monitoring:
You can specify that the trial be monitored at periodic times (Periodic Monitoring) and
after a Post-Accrual Follow-Up Period. These monitoring times are in addition to the monitoring
which (optionally) occurs immediately before each patient is accrued.
Specifying the minimum number of patients to be greater than one postpones any monitoring of the
trial from occurring until at least the minimum number of patients has been accrued.
-
Time Units:
Many of the parameters in the software have a time unit associated with them. These units can be
set to be Days, Weeks or Months.
-
Help:
Clicking on the blue or firebrick text in
the form will open context-specific help in the window on the right hand side of the form.
-
Simulation Run Tab Page:
Simulation Run Tab Page
-
Scenarios:
Scenarios are specified by providing a Name, True Mean/Median Time-to-Event, Accrual Rate, and Seed
for the random number generator.
The specified time units for True Mean/Median Time-to-Event and Accrual Rate will be used for all
scenarios.
Scenarios can be added, deleted, duplicated, and moved up and down the list.
-
Credible Interval Percentage:
Some Operating Characteristics are reported with computed Sample Quantiles. These quantiles are
computed using a Credible Interval Percentage as descriptive of a centralized Sample Quantiles
range.
For example, a Credible Interval Percentage of 95% will correspond to a Sample Quantiles range of
( 0.025 , 0.975 ).
-
Simulation Output:
After the simulations have completed, or have been aborted, an HTML formatted report will be
output to a window on the right hand side of the form. This will contain a many expandable regions:
Model Parameters (Input), Summary Table, and an expandable region for each
Scenario.
-
Help:
Clicking on the blue or firebrick text in
the form will open context-specific help in the window on the right hand side of the form.
References
-
Peter F. Thall,
Leiko H. Wooten, and
Nizar M. Tannir
(2005). Monitoring event times in early phase clinical trials: some practical issues.
Clinical Trials, 2(6), 467-478.