Source Data Agreement

Before the trial begins, all technical aspects, such as E-CRF, fax machines, printers, etc., need to be clarified and problems resolved. In case of difficulties during the study, sponsor must be informed and backup plans must be agreed until the issue is resolved. When original lab data sets or test data sets are sent to the central site for evaluation, a process must be put in place to ensure that a duplicated or certified compliant copy is available in the site data series. Sponsors generally do not have sufficient internal knowledge or resources to develop and/or manage electronic systems used in clinical trials, such as. B as systems used for randomization and distribution management (IMP) and/or for the collection of clinical trials (eCRF and ePRO). As a result, sponsors very often delegate related tasks to third parties. In these cases, sponsors remain responsible for conducting the study in accordance with the protocol and principles of good clinical practice (Clinical Trials Regulation Art 47, ICH E6 (R2) Section 5.2.1). The EMA has issued a notice to clinical trial proponents that highlights the qualification and validation requirements for computerized clinical trial data management systems. This is based on the results of reflection and taking into account the impact on reliability, robustness and acceptance of data in applications for authorization: According to ICH E6 (R2), sections 5.2.1 and, “the ultimate responsibility for the quality and integrity of test data is always the responsibility of the sponsor” and “the sponsor should ensure and document that the electronic data processing system or systems meet the requirements of completeness, accuracy, reliability and consistent validation set by sponsors.” EU GCP inspectors do not prefer a specific solution, such as a third-party supplier. B, data printed before it is transferred to the database or the recording of a simultaneous copy on the examiner`s local hard drive; the key point is that the choice of an electronic solution should not compromise the credibility of the data and should not lead to a lower quality than a paper FRC.