Validation & Data Integrity in eClinical Platforms
1 Introduction
It is well understood that each clinical trial is conducted and managed as an independent project even if several trials investigate the same Investigational Medicinal Product (IMP). Each clinical trial is different as each addresses different parts of the development cycle or varying product indications or endpoints. Trial projects, especially across the various phases, vary greatly in terms of duration, number of patients to be recruited, the pace of enrollment, and the spread of geographic location(s) involved.
To address the trial specific setups, eClinical solutions have to be built from various integrated technologies and tools designed to be utilized in clinical trials, working together sharing data, eliminating duplication of activities, and streamlining the use of multiple technologies for the end user (Figure 1.1).
Figure 1.1: Mapping of Sample Systems to the Clinical Process
The need to ensure data integrity through the life cycle of a clinical trial and across all the systems involved is of paramount importance as inconsistent, incorrect or corrupted data could endanger the safety of patients, adversely affect the outcome of the trial and increase the risk of a failure during the submission procedure. Therefore, this aspect has increasingly become the focus of regulatory oversight. One of the main drivers for this has been that the industry has embraced individual or strategic outsourcing of clinical trial activities to Contract Research Organizations (CROs) and sponsors as well as CROs also adopting Software as a Service (SaaS) offerings especially in the area of Electronic Data Capture (EDC) or Interactive Voice Response Systems/Interactive Web Response Systems (IVRS/IWRS). Often this leads to a chain of partners with an increasing risk of losing direct control for the sponsor. Even when strategically partnered with a CRO, the responsibility to address these risks resides with the Sponsor and cannot be delegated. This requires extensive and increasing efforts for oversight, which must be considered when addressing the risks with regard to data integrity. While the application of the GAMP® 5 principles [1] to the validation of GCP relevant systems has already been discussed in an ISPE Concept Paper, “The Application of GAMP® 5 to the Implementation and Operation of a GxP Compliant Clinical System,” [2] the challenges in the setup and maintenance of an eClinical Platform are largely not addressed.
2 eClinical Platforms
The introduction of the GAMP® Good Practice Guide: IT Infrastructure Control and Compliance, has also seen the term “platform” officially associated with the IT infrastructure of GxP regulated environments for the first time. A platform provides the technological environment (hardware and software) required for an application to fulfill its intended use. The efficient and quality-assured operation of IT infrastructure is facilitated by the use of reusable building blocks, which consist of logical groupings of standardized system components.
The introduction of GAMP® 5 extended the GAMP® software Category 1 to include infrastructure software (infrastructure software tools and layered software), thereby effectively achieving an interface to IT infrastructure. The so called “layered software” includes software such as operating systems, table calculation applications or statistical programming tools, which constitutes platforms for the development of applications.
The main difference between this and an eClinical Platform is the fact that the “layered software” focuses on individual software products (instances) in their condition at delivery, while the eClinical Platform constitutes a preconfigured application portfolio as an intermediate layer between the clinical trial process and IT infrastructure. For the purpose of this paper, an eClinical Platform is defined as a pre-existing environment of integrated computerized systems that can be adapted to support the conduct of a clinical trial by utilizing existing, validated functionality and processes.
Typical platforms include Electronic Data Capture (EDC) system, Clinical Trial Management System (CTMS), electronic Trial Master File (eTFM), statistical systems as well as safety systems and others. Individual components of the eClinical Platform may require set-up or configuration to meet the requirement of the individual clinical trial.
Figure 2.1: Flow of Information through the Different System Layers
To support the collection, analysis and processing of data collected in a clinical trial, highly specialized tools like an EDC system, CTMS or IVRS/IWRS have been developed and have been in use for years. However, today these tools are no longer stand-alone systems as they have been in the past. These systems have become the building blocks of an integrated eClinical Platform that supports the efficient conduct of clinical trials (Figure 2.1).
Not all clinical trials will need all systems being part of such an eClinical Platform (e.g., an open-label trial does not require systems that support blinding of trials). For instance, a Phase I trial may require different systems than a Phase IV trial. A Phase I trial may not require an expensive, complex, multi-lingual and web-based EDC system, as Phase I trials are often conducted in only one location with very few users and subjects. Other examples include a subject recruitment database or a barcode reader that may not be necessary in a Phase IV trial.
Additionally, some systems (e.g., a CTMS) will collect and process data from all clinical trials conducted by the organization without further customization while others (e.g., EDC system) may need to be setup and configured for each individual trial based on the protocol requirements.
A further aspect that needs to be considered is potential outsourcing of activities and the usage of SaaS offerings. The resulting eClinical Platform might span across multiple organizations (e.g., the sponsor of the trial), one or more CROs and SaaS vendors (e.g., for an EDC system) and could even include Electronic Health Records (EHR) systems at the investigator sites.
As mentioned in the introduction, a practical and efficient approach for the validation of GCP relevant systems has already been provided in the ISPE Concept Paper [2].
A generic example of an eClinical Platform is provided in Figure 2.2.
Figure 2.2: Example of a Generic eClinical Platform
Definition Data integrity can be defined as the validity of data and their relationships. For electronic records collected and processed as part of a clinical trial to be trustworthy and reliable, the links between raw data, metadata, and results must not be compromised or broken. Without data integrity, it is not possible to regenerate a previous result of a clinical trial reliably.
Obviously, maintaining data integrity is an important aspect not only for clinical trials and eClinical Platforms. This needs to be addressed throughout the product lifecycle spreading across GMP, GLP, GCP and other GxP areas.
In considering all of these aspects, it becomes obvious that data integrity cannot be ensured by the validation of the individual systems and their point-to-point interfaces alone. A more holistic approach toward validation, including relevant processes, data and quality management is necessary because those systems are acting together across corporate borders and controlled by different quality systems. Similar to validating individual systems following a risk based approach, the risks of using the eClinical Platform must be identified, assessed and adequately addressed. In addition to the system and study life cycles, this holistic validation approach needs to support the complete data life cycle from the first generation of the data till the end of the retention periods and should include the relevant metadata. To limit the scope of this paper, only the validation aspects of the systems and platforms are investigated. Other techniques and controls need to be in place to assure data integrity along the complete data life cycle.
Read more by downloading Validation and Data Integrity in eClinical Platforms (Published: June 2014).