Vice President, Vault CDMS, Veeva Systems

Richard Young

Each year, the global health community celebrates International Clinical Trials Day on May 20, the anniversary of James Lind’s famous scurvy experiment in 1747 on H.M.S.

. Lind’s discovery—that sufferers of the disease recovered faster if they included citrus fruits in their diet–is considered one of the earliest examples of a successful controlled clinical trial.

Today’s clinical studies benefit from access to advanced technology, vast volumes of data, and expertise that spans therapy areas and regions. Yet, the overall objective is consistent with Lind’s study 275 years ago: getting innovative, tested, and safe treatments to those who need them most.

Given the scale of the ambition, it’s easy to overlook just how quickly change is happening in the clinical research sector. The advent of paperless trials is a good example. The first study to be conducted using electronic data capture (EDC) took place in the 1970s; by 2012, clinical trials had come to rely on EDC, and paper-based trials had gone from default to exception. Moving to EDC has significantly improved the oversight and accuracy of clinical data management. Here are a few other areas where advancements are simplifying studies and shaping the future of clinical trials.

An industry united to accelerate innovation

The unprecedented speed of the COVID-19 vaccine trials has also reset expectations of study lifecycles. It took just 11 months for the first vaccines to be granted emergency authorization, beating the previous record from 1967, when the mumps vaccine was developed within four years.

Outside of COVID research, other concepts such as platform, bucket, and umbrella studies continue to drive new collaborations across our industry, including with regulators. The shared goal is to reduce the time and cost of providing life-saving treatments to patients. One great example is the 

 attempting to re-engineer Phase II breast cancer trials by aligning personalized medicine with adaptive trials.

Another area of significant progress is in executing digital trials where in just ten years, technology has made studies more patient-centric. 

The first, widely discussed virtual trial took place in 2011

, allowing volunteers to participate regardless of their proximity to sites. This has evolved to enable decentralized trials, helping patients gain access to treatments and making it easier to participate in a study while allowing study teams to capture data remotely.

Looking ahead, we hope to see an ambitious standard adopted industry-wide for decentralized trials in the coming years. Our aspiration is to execute trials according to the patient’s wishes, from one day to the next. Data collection will become distributed, but trial data will continue to be aggregated and harmonized on a central platform for effective oversight and management.

For patients, studies should start to feel like part of everyday life so that data is collected and shared seamlessly from wearables or patient-facing applications. To make this a reality, the seamless integration of technology is becoming a priority across the industry. The FDA made progress in this area in 2018 when it 

, allowing patient data to be linked to traditional clinical trials, real-world studies, observation studies, and registries.

The COVID-era further underscored that patients need ways to contribute clinical trial data quickly, and the industry needs to streamline processes for accelerated drug development. We will soon see the focus for sponsors and CROs shift from integrating technology to empowering patients to easily share outcomes during a study without navigating through a myriad of applications and site portals.

Laying the foundation for the future of connected digital trials

There are many reasons to be optimistic about the next era of clinical trials and the milestones we will achieve in the years ahead. Now is the time to lay the 

 for complete and concurrent trial data to speed innovations. As clinical trials evolve and become more complex and dynamic, they will require flexible protocol design, modern data management, and patient-centric technologies that can support decisions in close to real-time.

Despite Lind’s discovery, it took 42 years for the Admiralty to make lemon juice compulsory on ships. The industry is now making progress much faster, supported by technology. Companies are committing to a digital trials future for seamless information sharing across sponsors, CROs, research sites, and patients. This accelerates trial execution, improves information sharing, and speeds the delivery of innovative medicines and treatments to patients in need.

We’ve seen what a united industry can do to accelerate new treatments in the face of great obstacles. With digital solutions, patients can be assured of trials that deliver safe medicines to improve health outcomes at a pace we could not have imagined in past decades.

, vice president, Vault CDMS, Veeva Systems

Articles

From the first trial on H.M.S. Salisbury to advancing towards a digital trials future.

Clinical Trials Day: Celebrating 275 Years Since the First Controlled Trial

Innovation and adaptation around COVID-19 restrictions have brought a “can do” spirit to the industry, particularly in clinical trials. This led to an increase in adoption of decentralized models that allow patients to participate in trials without repeated site visits, offering sponsors a way to increase enrollment and prevent attrition.

But the buzz surrounding patient centricity and decentralized trials has drowned out an inconvenient truth. Sponsors have no clear-cut way to aggregate and review the huge volume of patient data being gathered from disparate sources. This problem is not new - the industry has been working with decentralized trials, in some form, for over a decade. And looking ahead, to truly run a connected and digital clinical trial that enables decentralized approaches, companies will need complete and concurrent clinical data.

Sponsors may be able to capture patient data remotely and in real-time, but they cannot verify and reconcile it in anything close to real-time. Instead, most organizations use tedious, manual methods to aggregate and clean each silo individually.

We’ve heard from one Top 20 sponsor that, for a single trial, such efforts required 27 people working in shifts, 24 hours a day, for six weeks. Considering that most big pharma companies run hundreds of trials each year, the time and costs required to do this for every trial are prohibitive.

An Explosion in third-party sources and non-static formats

The roots of this problem go back to the earliest days of traditional electronic data capture (EDC), where we saw data sources fragmented and isolated as we opted for speed of collection and sacrificed speed of analysis. With novel trial designs creating a drive for faster decisions, the ongoing explosion of data required for both traditional and decentralized clinical trials is changing our thinking. Today, a typical Phase III trial uses close to 10 data sources and generates an average of 3.6 million data points, three times the level that was seen 10 years ago.

 One study found the cost to support data transfers between systems or companies in life sciences is $156 million annually.

Increased use of third-party data sources and non-static patient data have added complexity. Where, 10 years ago, most key clinical data came from physicians and was stored in the EDC, only about one-quarter of that data is now stored in the EDC. The remaining 75%—third-party data from smartphones and other sources—is managed independently and must be reconciled against the EDC. More and more frequently we see primary efficacy and safety data coming from outside of the eCRF, creating increased pressure on data integration strategies.

Further complications are introduced by non-static data formats, such as readings from wearable devices and monitors. Data formats differ between different systems, even for the same type of data—patient heart rate, for example.If measured via a stethoscope in the clinic, it will be recorded as a single data point, but a wearable device will produce continuous, high-frequency data. In the end, huge volumes of data collected at different frequencies will need to be managed, reconciled, and interpreted.

The added complexity of digital and decentralized trials

With decentralized trials, data that was once solely collected at the site may also be collected during telehealth visits, in-home visits, through ePRO apps, and more, each creating its own silo. Frequently, these scenarios require additional systems to collect the data and one size rarely fits all.

To get a clear view of the patient, this data must be aggregated and harmonized. For example, to see a patient’s heart rate, there are multiple places to look—the EDC, the in-home visit log, or the iWatch reading. The source for each patient can also vary by visit—a trip to the clinic one day and an iWatch reading the next.

The same data is collected by different systems, at different times, and structured in different formats for decentralized trials, which makes synchronization significantly harder. This makes traditional data management capabilities such as querying important, yet it is not included in newer data collection tools.

One sponsor shared the challenges it faced after installing a system designed to handle patient data from remote nursing visits. The application did not include a querying tool. As a result, when data discrepancies emerged, the company faced three ugly choices: querying the data via email outside of the system; re-keying the data into its EDC; or paying for an expensive one-off integration. As this example illustrates, sponsors need an infrastructure that brings data together in an effective and scalable way.

When trials are fully digital with no paper, sites must use eSource, and an incremental challenge emerges. In these cases, data are not anchored to an EDC, which has traditionally served as the backbone for trial data, providing a reference point for other sources to be checked against. In a world of eSource that is managed by sites and where practices vary by site, working without the EDC means losing a fixed anchor against which to reconcile data. As a result, the data is much more difficult to clean.

When eSource and EDC co-exist in separate systems, sponsors must bear the enormous effort and cost required to reconcile the two. What will happen when we create truly patient-centric processes, whereby the patient will decide which visit they attend in person, versus perform remotely?

Mixing and matching at the data point level will become the norm, and current linear solutions are not designed with that in mind. Add to that the impact of protocol amendments and adaptive designs, and the conclusion is simple: companies can make it work for their important trials, but costs are prohibitive for the average study.

An integrated platform that connects the patients and sites with the sponsor’s infrastructure for one point of cleaning and review would eliminate many of these challenges. Unfortunately, such an infrastructure doesn’t yet exist, although many technology providers are working to deliver such solutions.

Time lag prevents data-driven decision making

In decentralized trials, the time lag between data collection and the availability of clean data makes it more difficult to make informed decisions during the trial. Many new data-collection instruments are stand-alone tools that lack data review capabilities. When data needs to be reviewed, it must be transferred to the sponsor or CRO and imported into a separate system. When discrepancies are found, data managers must resort to disconnected email exchanges to issue queries, adding further delays and more manual work.

In 2020 and 2021, after COVID-19 restrictions took effect, some pharma companies invested considerably in decentralized trial technologies, only to find themselves with data that could not be connected or verified. They’ve waited months to extract, clean, and reconcile it with their EDC data, finding unexpected anomalies, such as different dates for a patient’s adverse event. There has been no easy way to query data sources and ensure validity.

Delays viewing the data prevent trial practitioners from making data-driven decisions in a timely manner or being able to assure regulators that the data represents a completely accurate account of each patient’s experience. For example, investigators may need to determine why a sensor reading appears out of range, such as when one patient’s blood pressure suddenly spikes. Currently, there is no rapid way to verify these type of root causes.

Consider a rare disease trial where each patient’s outcomes are potentially meaningful to the others. If one patient’s diagnostic readings trigger a change in the treatment plan, sponsors should be able to make that change instantly. If there is a delay before the data can be cleaned and checked, it leaves them liable for failing to stop potentially harmful treatments.

Establishing end-to-end data flows will be crucial if decentralized trials—and not just data collection—are to run in real-time. Not only will it be key to clinical trial agility and ensuring the validity of results, but it will also be a pre-requisite for adaptive trials.

Ironically, the industry’s push for clinical innovation has only compounded the data management challenge. New technologies are being overlaid on a data management foundation that hasn’t changed in decades. Designing a trial that is adaptive, digital and connected, and allows for decentralized execution (all in one protocol) with systems and solutions that support it must be the long-term goal.

Standards help but will never fully solve the challenge

There is a clear and understandable desire for greater development and use of standards to address challenges. The industry has spent over a decade working to define data standards, yet still struggles with overall diversity and complexity.

“Standardization is very important if you want to simplify trials. Right now, you have 10 to 15 different external data providers, and everyone uses a different way to ingest data. Imagine what happens when you move to 20 or 30 different data sources. How will you standardize that data?” asks Mayank Anand, vice president and global head of data strategy and management at GSK.

But there is no easy way to standardize. The needs for standards are diverse, and there are different standards for how data should be collected, moved, analyzed, and submitted. In addition, the clinical environment is dynamic: data changes; needs change; the understanding of science, and the human body, changes. Considering the pace of innovation in life sciences and the speed at which new data sources are introduced, it is not realistic to rely solely on standards.

The path toward complete and concurrent data

Company strategies for digital and decentralized trials must incorporate plans to connect the myriad sources of patient data into a single clinical data management system. With decentralized trials, the same data for different patients will be collected in different ways. Aggregating data in a central clinical data management system (CDMS) is crucial to achieving the visibility and timeliness that we know contribute to more effective trials.

Technology providers are exploring different ways to achieve this connection, but some options scale better than others. One approach is to use a clinical database or data workbench that stores clinical data in one place, allowing it to be cleaned and harmonized.

Veeva is working on such an approach, Veeva CDB, with several of the top 20 pharma companies. Veeva CDB includes a data lake that holds disparate datasets in their native structure.

These datasets are mapped to the study backbone using five metadata fields as a simple “key,” rather than requiring a complete mapping, transformation, or adherence to a standard. This auto-mapping helps make data for the same point from different sources more equivalent (e.g., a blood pressure reading from a site visit vs. one taken from a device at home). Data can be ingested, aggregated, cleaned, and made readily accessible to other stakeholders in the organization.

One thing is clear, sponsors seeking decentralized data collection on the front end need centralized data management at the back end to prevent fragmented, heterogeneous data from slowing trials down. Automating the ingestion and harmonization steps will eliminate the time lag between data collection and access to clean data. Aggregating and cleaning sources simultaneously also address the need for patients to be treated consistently. For example, patient data from digital sources do not receive preferential treatment, which could otherwise result in questions of bias.

An additional approach to this problem will be to unify EDC and eSource solutions so that they capture data into the same system, which would allow data queries to be handled via EDC tools. Sponsors should also consider the benefit of having a single platform and data model extending from the patient to the sponsor. These requirements leave data experts with a strategic question: should they extend patient-facing data collection tools into the sponsor data environment, or move in the opposite direction?

Whatever approaches are used in the future, the industry clearly needs to make it easier to work with external data. “Pharma is already struggling to manage the volume of data we have for trials today,” said Anand. “Over the next few years, as the speed of data ingestion increases, the industry will expect clean data output to be faster too.”

, Vice President, Vault CDMS, Veeva Systems

Rising protocol design complexity is driving rapid growth in clinical trial data volume. Volume 23

Managing and Integrating Clinical Trials Data: A Challenge for Pharma and their CRO Partners

Without a strategy for managing clinical data centrally, sponsors lack the visibility needed to optimize trial outcomes.

Data First: Building a Foundation for Digital and Decentralized Clinical Trials

When data sources are stored and managed separately, it is difficult to see patterns and reconcile differences. Each source creates and stores data in a different format, with a different schema. Mapping data across sources is labor intensive and introduces traceability challenges. Data managers use data science to ensure data from diverse sources are gathered, harmonized, and formatted for the research scientists to analyze. Organizations that want to analyze big data should equip data managers with the technology and training necessary to ensure the new data is usable.

	Quality issues that slip through data management impact downstream users such as medical monitors, potentially the most expensive resource in our organizations. These are precious resources, and the financial impact of their decisions is substantial. Poor data quality increases the likelihood of inaccurate decisions, and both false positives and false negatives can cost a pharmaceutical company millions of dollars.

	Data management is largely manual today, but that is beginning to change. At the 2019 Society for Clinical Data Management (SCDM) Annual Conference, leading organizations discussed technologies to improve efficiency, such as machine learning (ML) to help automate mapping into the study data tabulation model (SDTM) format required by FDA and robotic process automation to help code incoming data. Equipping data managers with tools that assist in cleaning and harmonizing data is the most cost-effective means to supply downstream stakeholders with critical information faster.

	Currently, data managers spend a significant portion of their day on relatively menial and manual tasks, with a heavy focus on managing electronic data capture (EDC) data. Technologies can streamline, automate, and even obviate many of those tasks and help data managers provide the organization with greater visibility and insights into the data.

	Lotus Clinical Research, a CRO and research site network specializing in pain management, recently invested in training for one of its data managers and quickly reaped the benefit. “I highly recommend investing in a reporting power-user. We have a new EDC platform and our reporting guru created reports and dashboards that automated hours of work previously spent on manual tracking and reporting,” said Jeanne Strain, vice president of data services, Lotus Clinical Research. “Sharing those reports with colleagues on other trials has saved time across the organization and gives sponsors faster visibility into their data.”

	There are numerous examples of how real-time visibility into patient data can help an organization run more effective and efficient clinical trials.

		Identifying compliant patients and decreasing loss-to-follow-up reduces the number of patients needed and speeds database lock. 

		Enrollment dashboards showing site performance help surface what is or is not working earlier so other sites can course correct.

		Data visualizations of patient data such as blood pressure readings can surface outliers or potential adverse events.

	Clean, trustworthy data is a prerequisite to extracting those insights. Getting clean data to downstream decision-makers faster results in higher-quality outcomes.

	Artificial Intelligence (AI), ML, sensors, and wearables each hold significant promise to help design more effective trials, improve patient safety, and identify which patients benefit from a specific investigational product.

	However, these technologies have thus far failed to live up to their hype. A major impediment to the success of any AI or ML project is the heterogeneity and discord in their data.

	When Vas Narasimhan, CEO of Novartis, was asked about AI and ML, he spoke first about the challenge of getting clean data to work with. “The first thing we’ve learned is the importance of having outstanding data to actually base your machine learning on,” said Narasimhan. “In our own shop, we’ve been working on a few big projects, and we’ve had to spend most of the time just cleaning the data sets before you can even run the algorithm. That’s taken us years just to clean the datasets. I think people underestimate how little clean data there is out there, and how hard it is to clean and link the data.”

	The promise of AI will remain out of reach until organizations find a sustainable and repeatable means to manage and harmonize data from disparate sources. A critical component will be data managers applying the practice of data science to master the idiosyncrasies of biologic and real-world data. Clean data allows greater confidence in patterns and ability to predict outcomes. Once achieved, organizations can affordably and productively apply AI to produce the much-anticipated benefits.

	Case report forms (CRFs) in EDC systems, which for years have been the primary source and store of clinical data, can no longer be the center of a data manager’s universe. Industry leaders at the SCDM 2018 Leadership Forum estimated that less than 30% of the data volume comes from the EDC. In most organizations, the EDC is the sole system designed to manage clinical data. According to a 2017 report by Tufts Center for the Study of Drug Development, most organizations (77%) reported difficulty loading external data sources into their EDC and 66% attributed those difficulties to EDC system limitations and integration issues. 

	Organizations ready to invest in their clinical data infrastructure should consider three main technologies:

 Data managers will work more efficiently with a purpose-built platform to aggregate, clean, and normalize data. While a dedicated platform for managing all data sources was historically a luxury, they could now be considered core. Two of the most important steps for data management are selecting the appropriate data aggregation platform and learning to use the query and reporting capabilities provided. Capabilities that automate or eliminate manual data cleaning tasks will free the data manager’s time, while advanced reporting equips them to deliver high-value insights.

 Visualization tools help you look at data and extract meaning from what the data is showing. Usability is key to enabling data managers to move quickly and explore different aspects of a problem. Visualizations help you explore different ideas and formulate the right questions to ask.

Augmented intelligence applications such as ML, natural language processing, and robotic process automation are valuable tools for a data scientist and should be considered in partnership with the data management team, not as a replacement of the team. Data scientists play an important role in training the learning algorithms and are the natural partners enabling these systems to deliver value.

	Pragmatically, what does this mean for data managers today? The following are four specific recommendations for how data managers can educate and position themselves to help their organization leverage today’s data, and enable data managers to contribute as trusted advisors.

		Researching the specific ALCOA (attributable, legible, contemporaneous, original, accurate) risks and limitations associated with each real-world data (RWD) source. For example, patient registries provide extensive observational data on patients that is of good quality and relatively inexpensive. However, confounding (i.e., the inability to isolate the impact of relationships between dependent and independent variables) is a significant problem and makes it difficult to attribute trends in registry data to particular therapies.

		Familiarizing oneself with FDA guidance on using electronic health record (EHR) data in clinical investigations, which includes recommended practices for common situations such as handling data modifications. EHR-EDC integrations reduce the need for data entry by sites and associated source data verification (SDV), but numerous complications exist. Data managers should advise the organization on integration decisions that help preserve data lineage, protect personal identifiable data, and preserve identity masking.  

		Advising study teams on data source selections. Selecting the appropriate data source requires making informed decisions and balancing trade-offs -i.e., which sources contain the desired information with the greatest reliability, fewest gaps, and in high enough volume to balance across treatment arms? Educational courses from SCDM, such as mastering mobile and digital technologies, provide a framework for evaluating new sources in clinical trials.

		Advising organizations on data management technology selections. Many established systems were architected before the recent expansion in data sources. Data managers should research the capabilities of modern tools and platforms to identify those designed with diverse, heterogeneous data sources in mind. Industry conferences such as the DIA and SCDM annual meetings are most valuable during times of change. For anyone dealing with requests for new data sources, it is worth attending an industry conference to speak first-hand with the early adopters and supporting-technology providers.

	Challenges with data quality were the third most-cited barrier to completing clinical trials in 2018

 and data management is becoming even more difficult with each new source. And yet, innovation in the tools and training to support data managers has been stagnant. To truly capitalize on AI and RWD, biopharmaceutical companies must invest in the data science skills and resources of their data management teams. 

		The Burden of the “False-Negatives” in Clinical Development: Analyses of Current and Alternative Scenarios and Corrective Measures. 

		Novartis CEO Who Wanted To Bring Tech Into Pharma Now Explains Why It’s So Hard.

		Global Top Health Industry Issues: Defining the Healthcare of the Future. PwC Health Research Institute. 2019.

, Vice President of Strategy, Veeva Vault CDMS; email: 

The need for biopharma companies to equip data managers with the training and resources necessary to capitalize on new digital health tools. 

Richard Young

Articles

Clinical Trials Day: Celebrating 275 Years Since the First Controlled Trial

Data First: Building a Foundation for Digital and Decentralized Clinical Trials

Investing in Data Science to Unlock Clinical Data Value