Feb 03 2023

How the Cloud Can Improve Data Sets for Real-World Evidence in Clinical Trials

Real-world data can help researchers make important decisions about clinical trials. However, that data must be organized and accessible.

The volume of data being collected in the healthcare industry is increasing rapidly. If researchers and healthcare organizations can access quality data sets, they could unlock new insights to drive improved health outcomes.

This is especially useful in clinical trials, where pharmaceutical companies and researchers are actively working toward new treatments. Real-world data and real-world evidence can help scientists to make important decisions for clinical trials. Making effective use of RWD also goes beyond clinical trials. It can help payers, regulatory bodies and policymakers.

However, to access large patient data sets, organizations often need to rely on the cloud for its scalability and flexibility.

To use RWD and RWE, it’s important for healthcare stakeholders to understand what they are, how they relate to clinical trials and how to cloud can support data initiatives to improve patient outcomes.

Click the banner for access to exclusive HealthTech content and a customized experience.

What Is Real-World Evidence?

Real-world data is a collection of information about a person’s journey through healthcare collected at various points of time and amalgamated, according to Guardant Health Chief Medical Officer Dr. Craig Eagle. The data is collected from sources such as electronic health records, disease registries, claims and billing activity, patient-generated data, and even data gathered from personal mobile devices and wearables. This data is de-identified when building real-world databases.

“This type of data collection has the potential to enhance understanding of unmet patient needs, identify their commonalities and differences with other patients, and how to fulfill them — how to treat the patient based on their own needs versus treating symptoms of an ailment or disease,” says Chris McCurdy, senior manager of global life sciences solutions architecture at Amazon Web Services.

Real-world evidence involves analyzing data sets and interpreting results to determine what’s happening in the real world. Analysis of real-world data produces real-world evidence.

“We’re sampling real-world data to try to understand the real world,” says Eagle.

One example of using RWD would be to build a better understanding of how cancer diagnoses impact patient outcomes, how patients respond to specific cancer treatments, why cancer is breaking through if a patient is not responding to a treatment, and what the appropriate next steps are in the patient’s care.

Eagle explains that Guardant researchers can identify a cancer DNA mutation in the blood, for example, that affects a protein called the epidermal growth factor receptor (EGFR) in some patients with advanced lung cancer. The EGFR mutation allows the lung cancer to grow quickly. A treatment has been developed to target that specific mutation.

However, this treatment can become less beneficial over time. Using real-world evidence on cancer outcomes and aligning it with Guardant’s cancer genomic liquid biopsy data, the organization can start to determine if there’s a specific mutation that causes resistance to the therapy. This can enable future work to discover a treatment to target that mutation.

READ MORE: Why should healthcare organizations hire a data quality manager?

What Role Does the Cloud Play in Real-World Data Platforms?

Data platforms are required to realize the power of real-world evidence and could range in complexity for a simple Excel spreadsheet to complex data analysis platforms.

“Formatting data to be on these platforms is a key step in the process of creating an effective data platform and eventually a real-world database,” says Eagle.

However, a simple list of formatted RWD is inefficient for analysis if it isn’t organized, linked and easily searchable. Eagle says that a classic example of a real-world data platform is the one created by the Surveillance, Epidemiology, and End Results Program (SEER), which provides information on cancer statistics in the U.S.

In addition to public databases, researchers or healthcare organizations can create their own in-house databases or aggregate data from several public databases and apply an artificial intelligence (AI) algorithm to enhance the analysis of real-world data with fewer mistakes compared with a human.

Many organizations rely on the cloud to bring real-world data together. McCurdy explains that real-world data platforms leverage cloud computing technologies such as cloud storage, flexible computing, geographic reach and AI with machine learning (ML) to bring varied real-world data sources together, which is especially important when it comes to clinical trials.

“For example, to help customers generate maximum evidence from trials, Precision Digital Health developed a cloud-based platform capable of integrating and harmonizing disparate data assets in the research and development and life sciences industry,” he says. “PDH’s SUMMA platform enables compliant evidence generation powered by cloud, with a configurable nature that keeps innovation at pace with the rapidly evolving healthcare and life sciences landscape. This approach enables data management for all aspects of the trials process, with self-service analytics and visualization tools to support a wide range of research needs.”

Using the cloud to host data can enable healthcare organizations to access important patient information, and healthcare providers to optimize clinical decision-making and focus on preventative strategies and personalized treatment. Eagle points out that a benefit of using the cloud is the ability for multiple people to access the data at once from different locations and different interfaces.

Chris McCurdy
In the next decade, making sense of all this data will provide the biggest opportunity to transform care.”

Chris McCurdy Senior Manager of Global Life Sciences Solutions Architecture, AWS

According to McCurdy, another benefit of using cloud technology is the ability to analyze data and break down silos, making it easier to use AI and ML to address interoperability challenges.

Common challenges include the local storage of large amounts of data from research equipment such as microscopes and spectrometers, which creates a barrier for archiving, processing and sharing data securely, says McCurdy. In addition to that data, there’s been an influx of data from sensors, mobile devices and medical devices in recent years. Overcoming these challenges has the potential to advance precision medicine and improve patient outcomes.

“Most healthcare data is frequently incomplete and inconsistent. It’s often unstructured and stored in disparate formats such as clinical notes, lab reports, insurance claims, medical images, recorded conversations and time-series data. This makes it incredibly challenging for organizations to process, extract information and analyze at scale,” he says. “In the next decade, making sense of all this data will provide the biggest opportunity to transform care. However, this transformation will primarily depend on data flowing where it needs to at the right time, all while being processed in a way that is secure and protects patients’ private information.”

Cloud computing and ML models can help healthcare organizations break down data silos and digest information to be accurate, relevant and actionable. This allows organizations to focus on patient care while the cloud technology automatically normalizes, indexes, structures and analyzes the data for them, says McCurdy.

“Today, we are seeing a wave of health care organizations moving to the cloud, which is enabling researchers to aggregate and harmonize research and development data with information from across the value chain while benefitting from compute and storage options that are more cost-effective than on-premises infrastructure,” he says. “Cloud-based hyperscale computing and ML enable organizations to collaborate across data sets, create and leverage global infrastructures to maintain data integrity, and more easily perform ML-based analysis to accelerate discoveries and de-risk candidates faster.”

EXPLORE: How CDW can help health systems improve data-driven decision making with analytics.

How Does Real-World Evidence Speed Up Clinical Trials?

According to the U.S. Food and Drug Administration, medical product developers use real-world data and evidence to design clinical trials and observational studies with the goal of creating new treatment approaches. RWE can also be used to monitor post-market safety and adverse events.

RWD can add value throughout the clinical trial process by increasing efficiencies and reducing time to market for potentially lifesaving treatments.

“Those customers are seeing gains across all phases of the clinical trial process, including reduced drug development timelines, simplified regulatory complexity and more holistic patient views for better insights,” says McCurdy. “In drug discovery, the analysis of RWD uncovers vital information into drug performance and safety, so it provides an important feedback loop into the drug development R&D teams. The same is true in post-approval, where RWD is used to ensure industry compliance and identify any adverse events.”

Eagle says clinical trials are a sample of data from the real world that control for many differences to assess the impact of one variable. RWD can’t replace clinical trials, but it can speed up researchers' ability to ask the right questions in a trial. They can turn to the real-world evidence to determine what is happening and take the steps to determine why.

This potential research approach can only enable these determinations when the data sets have the right kind of information.

“My biggest tip for building and launching into real-world evidence is to find the database that covers the parameters or variables that are to be analyzed rather than building a database and then realizing later that the data sets are missing important data,” says Eagle.

MORE FROM HEALTHTECH: How does hybrid cloud unlock the power of clinical data management?

What Is the Future of Real-World Data in Healthcare?

With investments in AI and ML, the healthcare and life sciences industries are now witnessing the democratization of genomics, and “multi-omics” is becoming the new norm for better understanding the body, how it reacts and how to best treat it versus treating a disease or population, says McCurdy.

He adds that by using cloud technologies and natural language processing, healthcare organizations can pair ML with data interoperability to help uncover new ways to enhance patient care, improve outcomes and save lives while also driving operational efficiency to help lower the overall cost of care.

As the field of data analytics and AI evolves in healthcare, Eagle says the ability to access real-time or nearly real-time data and analyses will increase, making the industry more agile.

gorodenkoff/Getty Images

Become an Insider

Unlock white papers, personalized recommendations and other premium content for an in-depth look at evolving IT