Collaborative Strategies for Technology Integration in Clinical Studies

Image Credit: © putilov_denis - stock.adobe.com

Digital technologies have significantly transformed the landscape of clinical trials, revolutionizing the way they are designed, conducted, monitored, and analyzed. These advancements have brought numerous benefits, making trials more efficient, cost-effective, patient-centric, and data-rich. Technology, however, is outpacing regulation. Regulations, standards, and guidance documents governing clinical trials have been created and revised. In this dynamic environment where compliance reference systems are still evolving, biopharma companies, in particular quality and auditing functions, are challenged with fulfilling their oversight role of ensuring quality, data integrity, and compliance.

Clinical trials of yesterday, today, and tomorrow

In today’s clinical trials, compared to those of the past, there is increasing complexity, greater data collection, and a growing reliance on digital technologies to facilitate data gathering. Companies are now adopting a patient-focused approach and opting for decentralized clinical trial (DCT) models, bringing the trial directly to the patient. Consequently, the use of digital tools has become inevitable, serving as the central hub for remote data collection from various technologies such as wearables, mobile electrocardiogram (ECG) devices, smartphones and tablet computers. However, it comes with challenges related to measurement science, regulatory acceptance, error reduction, handling missing data, and ensuring data integrity. The new concept of bring your own device (BYOD) is evolving and has the following potential benefits:

• It is more convenient for the study participant.
• It simplifies trial logistics.
• It comes at a reduced cost.
• Patient centric:
  • The majority of trial participants own a mobile smart device.
  • Smartphone users often routinely check devices for messages and applications throughout the day.
  • The requirement to use/carry and maintain a separate device for electronic patient reported outcomes (ePRO) may be inconvenient.
  • BYOD may lead to better diary compliance and increased convenience for participants.

Generally, digital technologies used in clinical trials can be categorized into various groups. These categories encompass a wide array of tools and technologies employed to collect diverse types of data at different stages of a clinical trial. The information gathered via these tools is valuable for analysis, assessment, and decision-making.

Here are several categories of data collection tools:

• Direct data capture (DDC)—Data collection from sources directly from the patient
  • eConsent: Informing study participants and obtain documented informed consent remotely.
  • Televisit: Remote study visits.
  • Electronic clinical outcome assessment (eCOA): Measures and records how a patient is feeling or functioning. It is used as part of a clinical trial to measure the efficacy of a health intervention.
    • ePRO: Allow patients to answer questions and report on their health through an electronic device, such as a smartphone or tablet.
    • eDiary: An electronic patient diary registers the data in a storage device and allows for automatically monitoring the time the entry was made.
    • Clinician reported outcome (ClinRO): A measurement of any aspect of a patient’s health status that comes directly from a clinician or other member of the investigational team with the patients` responses to the clinician (investigator) or qualified staff members.
    • Observational reported outcome (ObsRO): A measurement of any aspect of a patient’s health status that comes from a family member or caretaker.
    • Performance outcome (PerfO): A measurement based on standardized task(s) actively undertaken by a patient according to a set of instructions. A PerfO assessment may be administered by an appropriately trained individual or completed by the patient independently.
  • Mobile devices & technology: Includes applications, software, or platforms accessible via mobile devices such as smartphones or tablets, allowing patients or participants to directly enter data related to their health, symptoms, or other trial-related information.
  • Sensors & wearables: Utilizes various sensors and wearable devices to collect real-time data, tracking biometric information, movement, vital signs, or specific health-related metrics.
  • Digital biomarkers: A digital biomarker refers to objective and quantifiable physiological and behavioral data collected and measured through digital devices. These devices include wearables, portables, implantables, and digestibles.
• Indirect data capture—Data collection from sources other than directly from the patient
  • ECGs: Manage, interpret, and store ECG data.
  • Imaging: Medical image data management, access and review
  • Laboratory information management systems (LIMS): Centralized and integrated lab data for visibility, context and insight across stakeholders
  • EDC: Electronic data capture
• Management of clinical trials:
  • Study data tabulation model (SDTM): Defines a standard structure for human clinical trial data tabulations and for nonclinical study data tabulations that are to be submitted as part of a product application to a regulatory authority.
  • Analysis data model (ADaM): Specifies principles for analysis datasets and standards for a subject-level analysis file and for a basic data structure, which can be used for a wide variety of analysis methods. Note: While SDTM is used to create and map collected data from raw sources, ADaM is all about creating data that is ready for analysis. SDTM is ALWAYS the source of the ADaM data. It could be the domain or a supplemental qualifier, but the source is always the SDTM datasets.
• Supply management: Interactive response technology (IRT): Typical functionalities are:
  • Patient randomization
  • Kit assignment
  • Kit traceability
  • Supply management

Another category is the electronic case report form (eCRF). The eCRF is not explicitly listed among the main categories of data collection tools because the eCRFs primarily serve as a means to collect structured data and does not fall under the same technological umbrella as the other categories mentioned earlier.

In order to collect individual study participant's data from the investigational sites, CRFs were designed to collect the data, required by the clinical trial protocol, in a structured manner per study participant. Initially the paper CRFs were manually collected from sites and required manual data entry in the clinical database. As time progressed eCRFs were designed, based on standard data collection forms, which were built to enter data directly into the clinical database on an EDC platform.

During clinical trials, a vast amount of data is collected, with some data being more critical than others due to its role as primary or secondary efficacy or safety endpoints. Regardless of the data’s significance, adherence to data integrity principles is crucial at all stages of the data flow. Typically, this extensive data is gathered using central capturing tools such as EDC platforms and derived SDTM platforms. Additionally, the IRT system plays a vital role, offering functions such as randomization, kit allocation and clinical supply management, among others.

Data, cloud computing, and software as a service

All data, information and records are collected, processed and structured using technology solutions that are designed to do so and developed by specialized technology providers. Their services are offered as cloud-based technology solutions which offer a comprehensive suite of tools and infrastructure to manage the complexities of data collection, processing, and structuring in clinical trials. These solutions are pivotal in ensuring data security, accessibility, compliance, data integrity and efficiency throughout the clinical trial lifecycle, thereby contributing significantly to the success of clinical trials. The cloud services can be divided in three main categories:

• Platform as a service (PaaS): Offers a platform for developers to build, deploy, and manage applications without worrying about underlying infrastructure.
• Infrastructure as a service (IaaS): Provides virtualized computing resources over the internet, including servers, storage, and networking components.
• Software as a service (SaaS): Delivers software applications via the internet, allowing users to access and use them without local installation.

These cloud-based services are essential to the success conduct of a clinical trial. The leading PaaS and IaaS providers, such as AWS (Amazon Web Services), Microsoft Azure, and Google Cloud, set industry standards for cloud computing. They offer robust, scalable, and secure infrastructure for hosting various applications and services, including those critical to clinical trials. In summary, these providers not only offer robust infrastructure but also prioritize:

• Security: They adhere to rigorous standards and undergo frequent audits to ensure data security, including access security to their data farms
• Compliance: They publish compliance certifications on their websites, including ISO certifications, SOC compliance, HIPAA compliance for healthcare data, GDPR compliance for data protection in the European Union, and others.
• Transparency: Their commitment to transparency builds trust by providing clear information about their compliance practices.
• Scalability: Their infrastructure can seamlessly handle growing demands.

As a result, these providers are the go-to for many organizations involved in clinical trials and healthcare-related activities.

This article mainly focuses on SaaS solutions. In the pharmaceutical industry, there has been a significant shift from on-premise to cloud-based solutions. Previously, technology companies developed computerized systems as products that were either provided as-is, configurable, or customizable. The pharmaceutical companies were responsible for installation, configuration, validation, and maintenance activities. Today, the same technology companies are offering cloud-based solutions that include additional services such as hosting, maintenance, and specific configuration. They can deliver a complete validation package as part of their service.

In the past, during execution of clinical trials, there used to be paper based CRFs and COAs. IRT systems were updated/managed via telephone calls, whereas today all these paper-based processes have been replaced by digital technologies. Technology companies designing and developing these digital tools now provide cloud-based solutions, which include additional services such as hosting, maintenance, validation of the core platform, study-specific configuration, and the opportunity to perform user acceptance testing (UAT) for the study-specific setup.

As a result, sponsors and contract research organizations (CROs) are increasingly outsourcing activities to technology companies. These providers not only offer digital technology but also additional services as part of their evolving business model. Historically, sponsors, CROs, and sites have faced scrutiny from regulators, e.g. during sponsor, CRO and site inspections. However, digital technology companies assuming greater responsibilities in trial conduct are not directly prioritized by regulators. As a result, these technology companies may not fully appreciate the importance of complying with all relevant regulations and good clinical practice (GCP) standards.

Sponsor oversight and quality auditing

During the setup of clinical trials, sponsors often outsource various activities to CROs, including the completion of the sponsor’s electronic trial master file (eTMF). However, many sponsors and CROs may overlook the importance of keeping the actual data and metadata together. It’s crucial to maintain both the raw data and its accompanying metadata in unison to ensure a comprehensive and compliant record. For example, if documentation of UATs for digital records are uploaded without the necessary approvals, or if audit trails that detail the sequence of activities in the system are missing, merely finalizing the record in the eTMF will not meet compliance standards. Even if the record includes all required metadata, it could still be non-compliant without the inclusion of essential supporting data and approvals. It is important to understand that during a sponsor inspection, the eTMF serves as the primary source for documentation review, making it the ‘source of truth.’

For eCOA services (e.g. system development, validation, go live and maintenance), there are a handful of vendors that have the knowhow. Typically, a consortium establishes standards and offers sponsors the option to acquire copyrighted versions of COA`s as part of their study protocols. These versions underwent meticulous validation to confirm their clinical relevance, accuracy, and suitability for data collection in specific medical conditions, ensuring they meet the required standards and serve their intended purpose effectively. The copyright versions are then purchased by the sponsor and the digital technology company is developing, configuring, or customizing and testing the CoA on their platform.

For rare diseases, there might not always be a suitable validated questionnaire or other assessment tool, so it becomes the responsibility of the sponsor to perform the clinical validation and collect the clinical data supporting the claim that the data collected via the COA is a relevant parameter to collect primary, secondary or exploratory efficacy/safety endpoints. In view of the complexity, it is highly recommended to only collect data that is necessary and has added-value and not to collect unnecessary data. An existing trend is that sponsors are making the clinical trial protocols too complex without added-value of specific data being collected. Another aspect is the intricacy of the process itself including:

• Questionnaire validation
• Linguistic validation from original language
• Migration to electronic platform
• Questionnaire licensing (as outlined in previous section)
• Traditional setup: At home (unsupervised) versus at site (supervised)
• DCT setup: flexible with home/site setup (mostly unsupervised)
• BYOD vs provisioned device (PD)

It is the accountability of the sponsor to maintain quality oversight throughout this whole process. Generally, technology companies are up-to-date with the latest advancements, are technically sound and are innovative, incorporate the latest developments and know their domain very well. However, there have been quite some updates in legislation/regulations complemented by guidance documents reflecting the current thinking of the regulators. The following example is cited from the EMA Guidance document `Guideline on computerized systems and electronic data in clinical trials` in `Annex 1 – Agreements:` ‘The responsible party should ensure that service providers (including vendors of computerized systems) have the knowledge and the processes to ensure that they can perform their tasks in accordance with ICH E6, as appropriate to their tasks. Standards to be followed, e.g. clinical trial legislation and guidance should be specified in the agreement, where relevant. A number of tasks involve accessing, reviewing, collecting and/or analyzing data, much of which is personal/pseudonymized data. In addition, in specific cases involving contact with (potential) trial participants, data protection legislation needs to be followed, in addition to the clinical trial legislation and guidance.’

Common challenges in data quality

Main concerns or risks as a result of a lack of sponsor` oversight with these technology companies are following deficiencies, shortcomings or considerations:

1. Data security and privacy: Ensuring patient data security and privacy is crucial. Compliance with regulations like HIPAA (Health Insurance Portability and Accountability Act) or GDPR (General Data Protection Regulation) is vital to safeguard sensitive subject information collected through digital technologies.
   1. The location of data collection and processing matters. It is important to ensure that data is not sent across continents.
   2. Data should always be encrypted, especially since it might pass through several systems before reaching the statistical team. Understanding the data flows is crucial.
   3. Access rights should be correctly set and adequately tested.
   4. Hosting companies should physically segregate servers to prevent data loss.
2. Technological accessibility: Access to technology and digital devices might limit patient participation, particularly among elderly or underprivileged populations. Ensuring inclusivity and providing support for individuals with limited technology access is essential.
   1. If mobile devices are provided by the sites through the sponsor or CRO sites instead of using personal devices, users might forget passwords or get confused, leading to data not being entered at required timepoints or not at all or passwords created by or shared with site staff.
   2. User-friendly interfaces are crucial for patient engagement. Complex interfaces or difficult navigation can lead to non-compliance or inaccurate reporting.
   3. The stability and ongoing support provided by technology solution vendors throughout the duration of a clinical trial are critical. Vendors should demonstrate their ability to provide consistent support, maintenance, and timely updates of their systems, minimizing disruptions and ensuring seamless operation throughout the trial period.
   4. Depending on the disease profile, subjects might need support from caregivers during and after an epileptic seizure, for example.
   5. Depending on the patient population (elderly people, caregivers, pediatric population, type of disease, complexity of the questionnaires, and data to collect), it is important to ensure that the needs and information to be collected are clear.
   6. Issues such as insufficient training or translation errors need to be resolved rapidly.
3. Data quality and reliability: Ensuring the accuracy and reliability of data collected electronically is important. Factors such as incomplete surveys, data entry errors, or misinterpretation of questions might affect the quality of reported outcomes.
   1. Missing data and data entered retrospectively can compromise data integrity.
   2. Backups in segregated locations should always be available to prevent data loss.
   3. The reliability and validity of the technology platforms provided by vendors are essential factors influencing the outcomes of clinical trials. Issues such as system downtime, technical glitches, or inaccuracies in data collection may compromise the integrity of the clinical trial. Thorough vetting of vendors, including their track record, system validation processes, and adherence to industry standards, is necessary to mitigate these risks.
   4. An inadequate data management plan can result in missing data due to queries not being inclusive.
   5. Enrollment strategy issues can arise when the target population is elderly study participants in a clinical trial set-up with ePROs or wearables to the completed by the study participants at home, which may be beyond their capabilities.
   6. It is important to ensure that data is standardized across different systems and platforms. This can facilitate data integration and analysis and improve the accuracy and reliability of the results.
4. Integration with clinical workflow: Seamless integration of ePRO into existing clinical workflows is necessary for efficient data collection and analysis. Compatibility with other clinical databases is crucial for effective utilization of the collected data.
   1. For randomization systems, it is important to have an interface with IRT and other systems for kit allocation. These interfaces should be adequately validated.
   2. More comprehensive testing for eCOA that determine efficacy or safety endpoints is necessary.
   3. Standardization and Interoperability: The lack of standardization and interoperability among different eCOA systems might hinder data exchange and comparability across different studies or healthcare settings.

One of the trends observed when performing audits and compliance oversight activities of these technology companies is inadequate validation documentation, non-compliance with data integrity principles and incomplete validation coverage e.g. not covering variations, different scenarios, testing against reference points etc. Typically, the technology company is providing the core platform, setting up study specific configuration, performing validation activities and maintenance activities. The UAT (user acceptance testing) exercise is executed either by the sponsor and/or the CRO. This exercise should not commence until the technology company provides written confirmation that they have completed their role in the validation. Therefore, it should be noted that it is not the intention of the UAT to provide a complete revalidation but rather the effort to determine whether the technology solution meets expectations, complies with the system requirements and ultimately reflects the needs of the study protocol. Hence, it is paramount that the validation process and the resulting documentation provided by the technology company are comprehensive, accurate, compliant, and capable of reconstructing the validation exercise while adhering to data integrity principles both at the platform and study-specific configuration levels.

This means that both the validation process conducted by the technology company and the UAT performed by the sponsor and/or the CRO play essential roles in ensuring the generation and processing of reliable, accurate, and complete data. These efforts are essential for maintaining compliance with data integrity principles.

Transforming challenges into opportunities: The way forward

To ensure that technology companies build, customize, configure, and validate solutions appropriately, thorough assessments are essential as part of quality oversight. Additionally, the validation conducted by the technology company is not repeated during the UAT exercise; instead, it contributes to the overall validation both at the platform and study-specific configuration levels.

During validation, merely verifying the presence of forms, fields, or availability is insufficient. Validation testing should encompass more extensive methodologies, including stress testing, negative testing, volume/load testing, performance testing, repeatability testing, structural/path testing, smoke testing, and regression testing. It is also crucial to test different scenarios, including comparisons against controlled reference points.

Therefore, sponsors should establish an effective vendor qualification and vendor oversight program. This ensures that any shortcomings in validation and produced documentation are mitigated to an acceptable level. Ultimately, the sponsor remains accountable for overseeing and conducting the clinical trial.

Typically, the following activities are part of the standard qualification process:

• Request for information/request for proposal (RFI/RFP)
• Vendor questionnaire
• Qualification audit

During vendor qualification, it is important to assess the technology, the software development lifecycle process and how the technology company is validating, including the documentation they are generating. The technology company might be asked to build a proof of concept or a prototype, do a pilot project or have a collaborative workshop upfront of awarding work and signing contracts. This all can be enabled through a non-disclosure agreement and will allow the sponsor to thoroughly assess and audit how the technology company is performing all their activities, including the documentation they are producing.

Prior to starting any engagement, it is important to clearly define the expectations related to the quality of deliverables and system validation in contractual agreements between the technology company and sponsor. If contracted out to a CRO, the sponsor should oversee the execution of the contract between the CRO and the technology company and ensure the same set of clear expectations are included in the contract between the sponsor and CRO. Any contract should explicitly define data security protocols, system reliability, compliance measures, support services, and provisions for resolution of disagreements, ensuring the technology vendor meets the requirements of the clinical trial.

After the contract is signed, continuous monitoring and quality assurance measures should be implemented throughout the collaboration with the technology company. Regular audits, performance assessments, and feedback mechanisms can help identify and address issues promptly, ensuring that vendors uphold their commitments and deliver as per agreed standards.

The above examples primarily focus on eCOA providers, but they are equally applicable to EDC, IRT, and other technology providers in clinical studies. Furthermore, it’s highly recommended to maintain an up-to-date data flow diagram. This diagram should illustrate how data flows between different databases, where data alterations or calculations occur, and what quality controls or queries are implemented. This comprehensive overview can serve as a foundation for performing a data management risk assessment, helping to identify areas that require the most effort and attention.

Conclusion

In conclusion, technology companies play a pivotal role in the success of clinical trials, but concerns regarding data integrity, reliability, compliance, and ongoing support remain. Mitigating these concerns requires a vigilant approach that involves rigorous vendor selection, clear communication and clear expectations and standards in contractual agreements, adherence to regulatory standards, auditing and continuous sponsor oversight throughout the collaboration. By addressing these concerns proactively, pharmaceutical companies can leverage technology solutions effectively, ensuring the integrity and efficiency of clinical trials while prioritizing patient safety, data integrity and data confidentiality. Pharmaceutical companies should not take on this challenge on their own. Instead, they should collaborate through industry associations such as TransCelerate, ISPE, EFPIA in order to align on current good practices to provide guidance for technology companies on robust and defendable methods and standards.

Bart Dewolf, Head of Technology Auditing & Compliance, UCB Pharma