Time to take trial data analysis to the next level: ConcertAI

By Jenni Spinner

- Last updated on GMT

(alengo/iStock via Getty Images Plus)
(alengo/iStock via Getty Images Plus)
A leader from the advanced data and artificial intelligence specialist suggests ways to advance data collection through new tools, techniques, and thinking.

Clinical trial professionals have at their disposal more advanced data collection and analytical tools than ever before. Still, they might not be making the best, most efficient use of all that information they have access to.

Outsourcing-Pharma recently spoke with Jeff Elton, CEO of ConcertAI, to gain some perspective on how the use of real-world data (RWD) has evolved and possible ways to build upon the recent advances in data collection and analysis.

OSP: Could you share your perspective on how the research industry’s understanding of and attitudes toward RWD has evolved over the years?

JE: RWD has enormously transformed over the years. Before the broad use of electronic medical records, RWD was referred to as ‘Chart Reviews’ wherein a design for a study would be advanced and then patients manually accessed through their paper records, with forms being completed for the specific study design. There were no data to query with analytic tools, nothing could be scaled beyond the study designs, studies were small, and we had no idea if the results were representative of larger populations. For an array of reasons, this was seen as substantially inferior to Randomized Controlled Trials (RCT) studies.

With the passing of the Health Information Technology for Economic and Clinical Health (HITECH) Act in February of 2009, there was a focus on ubiquitous health information technologies. It was a heavily subsidized implementation that later came with reimbursement penalties for non-compliance. HITECH was intended to improve individual and population outcomes, increase transparency and efficiency, and improve the ability to study and enhance care delivery. This was the beginning of RWD as it is known today, and it is also why RWD and real-world evidence (RWE) generation is more advanced in the US than in any other region of the world.

Today we have hundreds of publications that support the use of RWD across an array of questions. These also create a foundation for what is considered highly reliable research methodologies, e.g., use of specific rules for processing unstructured data such as physician notes, radiologist interpretation reports, and other diagnostic or lab reports.

The data from companies like ConcertAI, frankly, are now actively used by biopharma innovators, academic researchers, and regulators as the basis for their analyses and to generate ‘evidence.’ This use of the term ‘evidence’ is itself significant in that it indicates that changes in decisions, treatments, priorities can be made based on RWD analyses.

As of September 30 of this year, and then again in November, the US Food and Drug Administration (FDA) advanced preliminary guidance for comment on the use of RWD in support of regulatory decisions and the creation of RWD-based registries to further support regulatory surveillance of long-term safety and outcomes.

So, if 2009 was the beginning of modern RWD, 2021 is the watershed point where RWE analyses evolved to be considered alongside RCT derived analyses and insights. With this has come new and higher standards. For example, we need to establish data provenance, set data abstraction rules with the study design, use peer-assessed methods that have established their equivalence across multiple datasets and analyses, to name only a few of the changes to come into place as part of the new ‘RWE Standard.”

OSP: As you point out, use of RWD in clinical trials has increased since COVID-19 plopped down upon us. Please share some insights as to why the surge has transpired, and some of the benefits/advantages RWD offers to trials, especially with the obstacles the virus has put up.

OSP_ConcertAIrwd_JE
Jeff Elton, CEO, ConcertAI

JE: As part of COVID-19 vaccines trial design, COVID-19 direct therapeutics, and more, there were initiatives to confederate data from multiple sources; assess who was most vulnerable; understand testing approaches that identified infected patients, and determine response and durability of that response. This was done internationally. It was done at ‘super scale’ with publications and insights coming at an unprecedented rate. It integrated RWD and RCT analyses into the same set of workflows, publications, and regulatory considerations.

It also quickly became clear that populations were suffering as a byproduct of the pandemic – e.g., cancer diagnoses rates were declining owing to lower clinic access during the height of the pandemic. It also initiated specific initiatives within the FDA for bringing more novel trial designs to bear for relaunching studies or to reconsider those that were stalled. Altogether these have collapsed what likely would have five to seven years of initiatives into 18 to 24 months – not lightly, but with rigor and urgency.

So, it is not a surprise, nor unrelated, that the FDA guidance on RWD and RWE came out with timing that indicated that the drafting occurred during the height of the pandemic. For oncology, this is particularly significant, as 80 to 90% of the programs achieving breakthrough designation are cancer drugs. Therefore, RWE standard of care analyses often inform the RCT study design and the underlying deficiencies in the standard of care that inform the importance of the novel approach and medicines being advanced.

RWE analyses, in the form of External Controls, also can accompany a single-arm study to provide confidence in the analysis and prospective approval decision. Finally, with the approval of a new drug application that achieved breakthrough designation, it will be important to complete a range of post-approval studies as confirmation of the benefit, safety, and durability of that response.

OSP: You also suggested that the industry isn’t likely to reap all the potential benefits of RWD until the mindset switches from thinking of research being limited to labs and clinics, to that of being a “continuum.” Could you please expand upon that?

JE: When we approve medicines, biopharma enters into something that has been termed ‘Life Cycle Management (LCM).’ There is an equivalent concept for research – though I’d maintain it is likely more important than the traditional aspect of commercial LCM. That is the Research Continuum. We need to frame early phase trials in terms of endpoints of interest, but we also need to begin capturing critical aspects of safety, based on concepts that apply to the current standard of care and those that may relate specifically to the unique biological mechanism of a novel therapeutic.

Also, oncology therapeutics are addressing very complex aspects of the human immune system or targeting a specific genetic mutation in a specific organ system tumor. As such, targeting, interrupting, correcting, and enhancing the function of biological pathways and the body’s own immune capacity requires working within very complex systems.

Ongoing RCT research and post-approval research can assure improvement in outcomes and safety. This already happens today to a degree – we see that most therapeutics targeting hematological malignancies for adults and pediatric patients improve in their outcomes over time based on investigator-initiated and company-sponsored studies.

What we are suggesting now is that a Research Continuum strategy would accelerate this further, be a plan, designate resources, and create a body of knowledge in a highly structured process that can deliver those benefits more rapidly and with a greater translation to clinical practice – both for safety and efficacy.

OSP: From your point of view, how good of a job do trial teams do regarding collection of data? And, how well do they do using insights from that data to help improve their studies?

JE: Trial teams do a good job and are trying to do a good job – but it is a tough one to do well. Look at this from the perspective of the healthcare provider. A provider may have multiple studies, from multiple sponsors, supported by multiple CROs, using multiple technologies for the collection of these data. That’s complex, involves a lot of time, and is filled with significant ‘non-value adding costs’ – what I define as supporting the research but not impacting the direct research process.

COVID-19 limited the ability for sponsors and CROs to access trial sites and ushered in new solutions – solutions that use decentralized trial technologies, source document upload capabilities, etc. Some of ConcertAI’s AI-enabled eScreening solutions and newer Digital Clinical Trial technologies are now part of that non-linear inflection that occurred during the pandemic.

So now we need to be thinking about evidence-generation as a system wherein I can use retrospective RWE, define prospective use of RWD, and generate RCT data with an ‘enter once’ and low burden for patient and provider mindset. This is going to be the future, no doubt at all about that, and we need to commit and accelerate this transformation as patient benefit and lives are at stake.

OSP: You suggested thinking bigger about trial data to yield additional benefits—incorporating patient data to improve trials before they kick off, using the data to enhance future trials, etc. Could you please share your vision, and how this stands to improve standard of care?

OSP_ConcertAIrwd_pic
(alengo/iStock via Getty Images Plus)

JE: Again, very important considerations. We have rich data at scales previously unimaginable available to design clinical trials and post-approval studies, define the sites and settings where these are best run, and optimize them for the endpoints of interest and assurance of lowest burden. Trials need to be reflective of the patients who will ultimately receive the medicines. They need to be run within and modeled for the community – that is where 85% of patients receive and want to receive their care.

Today we work to assure that the insights informing trial designs can be derived in a fraction of the time of what this required only three to four years ago. We can model a trial and optimize it with AI-enabled solutions in a few days or at most weeks. We can find the clinical sites most suited to the studies in a few hours. And we can also deploy AI solutions for identifying patients eligible for these studies in a few hours.

Post-approval pragmatic and Phase IV studies will advance standard-of-care insights. Registrational studies underpinned by RWD analyses and conducted in the community will assure outcomes and decisions that are more immediately generalizable to the 85%. That should be our goal, and it is achievable now.

OSP: How can the experts at ConcertAI and other companies help sites and sponsors bring their RWD use to the next level?

JE: There are a number of ways this can be accelerated and advanced. Health technology firms, especially the Electronic Medical Record companies, need to recognize that the healthcare provider and the patient are the owners of their data and provide tools and open Application Program Interfaces (APIs) that actively support healthcare providers to use their data, in an array of third-party applications, and with an array of partners. Today they are slow to meet interoperability standards and even slower to take a customer-, provider-, and patient-centric view on data access and use. This is ‘pre-competitive’ in that data access is a requirement for quality of care and new evidence generation approaches.

With that foundation, biopharma sponsors can work collaboratively, and again pre-competitively, to establish some common standards for data reliability, veracity, and use. With those common and higher standards, it is easier for regulatory bodies to integrate these data into their early consultative conversations, interim assessment, and final approvals. The same is true of government and private payers who afford patients access and financial support for specific therapeutics based on their effectiveness absolutely and relative to the current standards of care.

OSP: Anything to add?

JE: I am personally highly optimistic about the promise of RWE to make substantial contributions in evidence generation as a complement to RCT studies. It is an important field where our focused efforts and maintenance of the highest standards can assure that patients can be beneficiaries immediately and for the long term. 

Related news

Show more

Related products

show more

Saama accelerates data review processes

Saama accelerates data review processes

Content provided by Saama | 25-Mar-2024 | Infographic

In this new infographic, learn how Saama accelerates data review processes. Only Saama has AI/ML models trained for life sciences on over 300 million data...

More Data, More Insights, More Progress

More Data, More Insights, More Progress

Content provided by Saama | 04-Mar-2024 | Case Study

The sponsor’s clinical development team needed a flexible solution to quickly visualize patient and site data in a single location

Using Define-XML to build more efficient studies

Using Define-XML to build more efficient studies

Content provided by Formedix | 14-Nov-2023 | White Paper

It is commonly thought that Define-XML is simply a dataset descriptor: a way to document what datasets look like, including the names and labels of datasets...

Related suppliers

Follow us

Products

View more

Webinars