US Food and Drug Administration (FDA) has issued specifications to prepare and submit summary level clinical site dataset in electronic form for new drug applications (NDAs), biologics licensing applications (BLAs), and NDA or BLA supplemental applications submitted to the Centre for Drug Evaluation and Research (CDER).

In this regard the regulatory authority has called to submit a single summary level clinical site dataset that contains data from all major studies used to support safety and efficacy in the application, including studies with different treatment indications.

For each major study used to support safety and efficacy, there is need to  submit data by clinical site and treatment arm for the intent-to-treat (ITT) population.

For clinical investigator sites involved in multiple studies, there is need to provide the data independently for each study within the dataset.

For each study and investigator site, the regulatory authority has also variables associated with efficacy. These include Treatment Efficacy Endpoint (TRTEFFE) which is the summary statistic for each primary efficacy endpoint, by treatment arm. Under Treatment Efficacy Endpoint Standard Deviation (TRTEFFS), the standard deviation of the summary statistic (TRTEFFE) for each primary endpoint, by treatment arm needs to be included.

The section on Site-specific Treatment Effect (SITEEFFE), the treatment effect should be reported using the same representation as reported for the primary efficacy analysis. For the Site-specific Treatment Effect Standard Deviation (SITEEFFS), the standard deviation of the site-specific treatment effect (SITEEFFE) needs to be provided.

Under Endpoint, a plain text label that describes the primary endpoint as described in the data definition file data dictionary included with each application. In the section on Treatment Arm (ARM), a plain text label for the treatment arm that is used in the Clinical Study Report will need to be provided.

In addition, for studies whose primary endpoint is a time-to-event endpoint, there is need to include the Censored Observations (CENSOR) which cover the number of censored observations for the given site and treatment. If a study does not contain a time-to-event endpoint, record this data element as a missing value said the regulatory authority.

To accommodate the variety of endpoint types that can be used in analyses need to have information on discrete endpoints which are based on efficacy observations that can take on a discrete number of values like for instance  binary and categorical).

The Continuous Endpoints are based on efficacy observations that can take on an infinite number of values.

Time-to-Event Endpoints are those where the time to occurrence of an event is the primary efficacy measurement.

If the primary efficacy endpoint cannot be summarised in terms of the previous guidelines, a single or multiple values with precisely defined variable interpretations should be submitted as part of the dataset.

In all cases, the endpoint description provided in the ‘endpoint’ plain text label should be expressed clearly to interpret the value provided in the (TRTEFFE) variable, said the regulatory author.

The site-specific treatment effect (SITEEFFE) should be summarized in terms of the primary efficacy analysis which covers the difference of means, difference of proportions, odds ratio, hazard ratio and should be defined identically for all records in the dataset regardless of treatment. The summary level clinical site dataset should be accompanied by a data definition file. There is also need to create a data file template and structure for Bioresearch Monitoring (BIMO) Clinical Data to submit in electronic form, according to the regulatory authority.