Clinical research is evolving in a new direction – a belief echoed by Saama Technologies, an analytics platform organization that enables the life sciences industry to conduct faster and safer clinical development and regulatory programs.
Saama’s Sagar Anisingaraju, Chief Strategy Officer, and Malaikannan Sankarasubbu, VP of Artificial Intelligence (AI) Research, recently shared their insights on rethinking clinical research at the inaugural CORE West (Clinical Operations Retreat for Executives hosted by the Halloran Consulting Group) in Sonoma, California, in a presentation called “Leveraging AI During COVID to Accelerate Key Product Development.”
Saama began its journey in 1997 when company executives identified a critical need for business intelligence through data. In 2015, the company focused its attention on life sciences, attempting to help clinical trial sponsors connect, contextualize, and converse with data for better clinical trial outcomes. In this article, we’re going to expand on lessons learned along the way and share insights discussed during CORE West.
Disrupting Clinical Trial Operations
Linear and sequential clinical trials remain the accepted way to ensure the efficacy and safety of new medicines. This lengthy, yet tried and tested process of discrete and fixed phases of randomized controlled trials (RCTs) was designed principally for testing mass-market drugs. It has changed little in recent decades.
Historically, the life sciences industry has been limited to manual, inefficient data review processes to validate data from clinical trials. These historic models often lack the analytical power, flexibility, and speed required to develop complex new therapies that target niche populations. As a result, the healthcare industry has long struggled with being data rich and insight poor.
While dozens of systems and platforms collect information on patients and populations during clinical trials, extracting value from the resulting data to improve patient outcomes and operational efficiency is still an uphill battle. Additionally, the life sciences industry has observed that suboptimal patient selection, recruitment and retention, and patient monitoring are contributing to high clinical trial failure rates, resulting in higher research and development costs and delayed treatments to patients.
So, let’s switch our attention to AI-powered tools that enable organizations to significantly improve their processes and data accuracy and propel their clinical development into the next phase. Fueled by rapidly increasing amounts of medical data from multiple sources, including apps and wearable devices, we anticipate a major disruption in the way clinical trials are designed and conducted. Machine learning and other AI technologies will be an exciting part of the way forward.
How Pfizer Hyper Accelerated its Vaccine Development
Saama was invited to work with Pfizer to develop and deploy an AI powered analytical tool geared toward clearing many of the obstacles faced by study data managers and monitors. Saama was selected after showing its competencies during a Pfizer-sponsored hackathon.
Pfizer’s willingness to embrace the AI solution, known as Smart Data Query (SDQ), helped deliver higher quality work in less time during the company’s COVID-19 vaccine development.
During the trial, Pfizer provided the clinical data and domain knowledge to train Saama’s models. In turn, SDQ automated the process of identifying data discrepancies, generating queries, and resolving queries in a predictive way. With a human always in the loop for review, the application’s predictions could be approved or rejected, as needed. This type of feedback made the AI smarter every time, improving predictions and requiring less human intervention.
If done manually, this would have been a time-consuming effort. SDQ helped ensure data quality throughout the trial and cut the data cleaning time down from 30+ days to just 22 hours.
SDQ helped expedite the Pfizer COVID-19 vaccine trial in these other ways, too:
In a critical time, Pfizer was able to simplify and rapidly accelerate its vaccine development through prescriptive analytics.
Data Managers Help Set Up SDQ for Success
From the beginning, data managers who work with SDQ need to establish ground truth: labeled examples of the data that needs to be analyzed and metrics that need to be achieved.
And, as the model gets trained, data managers take on the critical role of mapping workflows and implementing processes for generating prescriptive analytics and insights.
Looking Ahead and Building the Business Case
As data becomes more voluminous and complex, mitigating data quality risk becomes increasingly difficult. Saama makes it easier for data managers to keep data clean and up to date, resolve queries faster, coordinate activities from start-up to database lock more efficiently, and make strategic contributions to risk-based monitoring (RBM) initiatives through their Smart Applications.
What if the first pass of the schedule of assessments of a clinical trial protocol was drafted by an AI-powered machine? What a future that would be. But that’s a long journey ahead.
Often, new methods of conducting clinical trials are met with resistance. A sponsor’s challenge will be to show that AI technology does, in fact, improve clinical trials enough to warrant the investment and change in operations. Take Pfizer, for example. Concrete time (and money) savings opened the executive team to a new way of doing things. The success of the vaccine trial enabled others in the organization to embrace SDQ, and it is fast becoming standard operating procedure throughout the organization.
About Saama Technologies, Inc.
Saama is the #1 clinical analytics platform company, enabling the life sciences industry to conduct faster and safer clinical development and regulatory programs. Saama’s Life Science Analytics Cloud (LSAC) is a single, unified data aggregation and analytics platform that helps drug development teams—including clinical operations, medical review, data management, biostatistics, pharmacovigilance, and translational research—make faster, better decisions.