By David Underwood
An ISS is an integrated summary of safety, and an ISE is an integrated summary of effectiveness. An ISS combines the safety results from different studies conducted in a compound, while the ISE combines efficacy results. The regulatory authorities state both the ISS and ISE are critical components of a submission.
Why Are An ISS And An ISE Considered Necessary?
The combining of data in integrated summaries helps to address safety and efficacy concerns that are difficult to address using the data from individual trials. However, the regulators make it clear that a statistically significant result in an ISE is not sufficient to replace positive results in individual trials.
Combining the results from a number of different studies provides considerably more power for these important comparisons of safety and efficacy, while not taking away from a study’s primary endpoint. The increased power in an ISS enables identification of rarer adverse events (AEs) that may not be evident from a single study.
Planning The Analyses And Summaries Required In The ISS/ISE
Early planning of the ISS and ISE and the use of a statistician in the planning can help to identify and resolve potential problems at an early stage. This will make the process as efficient and cost-effective as possible. Planning an ISS and ISE prior to starting your pivotal studies enables you to introduce efficiencies, allowing data to be collected to answer specific questions. Producing a submission that is complete, consistent, and easy to follow will make the review process for the regulatory authorities easier and therefore quicker.
For the statisticians and programmers working on the integrated summaries, one of the most time-consuming tasks is the production of a database containing the combined study data. Differences in how the data is collected may determine if it is sensible to combine the data or what the results of the combined analyses actually mean.
As each submission is different, there are key messages you will need to address or specific statistical methodologies that are required and are unique to your submission. Early identification of these enables the statistician to address them prior to combining the results.
Combining results from independently designed studies that often address slightly different objectives is always going to be difficult. For example, study A may dose patients for three weeks, compared to study B that doses patients for six weeks. A combined summary of the number of patients who reported an AE may not be appropriate as one group of patients was ‘at risk’ and followed up for a greater time period. The statistician can help identify an appropriate methodology for addressing such issues. In this example, a solution may be to present AEs based using a denominator that adjusts for time at risk.
Reporting Of The Study
Once planning of the integrated analyses is complete, the statisticians and programmers will be involved in the production of the results based on the planned analyses. At this stage, close collaboration between the biometric team and the rest of the study team is still vital to ensuring a successful submission.
It is surprising how difficult it can sometimes be to locate the validated datasets and full supporting documentation. Production of combined derived datasets can be a lengthy task, and you should take into consideration the standards of older studies compared to newer.
In a well-planned submission, once the datasets are final, you can use macros to reduce the complexity of the output production for the ISS and ISE. If the biometrics of the ISS/ISE and pivotal studies have been centralized, then these macros can be used to produce the results from the pivotal studies too. As well as significantly reducing the time to reporting once the pivotal studies unblind, this can help to ensure consistency in the presentation of the results.
The integrated database is also essential for allowing rapid turnaround of questions from the regulatory authorities, as it simplifies production of outputs from the central database.
David Underwood is CEO and chairman of Quanticate. He has been in the pharmaceutical industry for more than 30 years.