Pharma firms should harness synthetic data: Phesi report
Clinical development analytics specialist Phesi has released the first in a series of reports, generated in partnership with consulting firm Accenture, on the value of synthetic data in clinical development. The document outlines a number of reported advantages of synthetic data (streamlined patient recruitment, eased patient burden, reduction in cost, and accelerated time to market) and offers advice to that end.
The Phesi-Accenture report includes two case studies where the use of synthetic data made notable improvements to the trial process: oncology and gastroenterology.
Oncology trials can be costly and time-intensive—a recent study has shown that without a synthetic control arm, the median R&D cost for oncology drugs is $2.77b, driven in part by a high failure rate (up to 80%). Timely patient recruitment also presents a challenge, leading to a delay in getting cancer treatments to market. Synthetic data can help reduce patient numbers and eliminate placebos, in turn helping speed the process.
In ulcerative colitis, synthetic patient data can be harnessed to synthesize baseline patient characteristics. This can help optimize the protocol design and to reduce patient numbers and exposure to a comparator.
Gen Li, Phesi’s president and founder, explained the inaugural report to Outsourcing-Pharma and detailed the advantages the use of synthetic data can offer sites and sponsors.
OSP: Could you please share how understanding and use of synthetic data in trials has evolved in recent years?
GL: Utilizing synthetic data in clinical development is not a new concept. For a number of years, various patient data has been gathered outside of randomized controlled clinical trials, and then used in clinical development, ranging from projects supporting drug life cycle management to going as far as supporting new drug registration, although sporadically.
What has changed in recent years, particularly in the past two years, is that firstly, we now have access to a massive amount of patient data in historical clinical trial settings, and in healthcare and other settings, which has become readily available, thanks to various technology progresses in big data, machine learning and artificial intelligence; and secondly, there has been some conceptual breakthroughs enabling us to better interpret and analyze the patient data from various settings.
OSP: Please share your perspective on the key advantages of harnessing synthetic data, and any reasons why sites/sponsors might be hesitant or slow to take advantage.
GL: Synthetic data, first and foremost, is a patient-centric concept. There will always be ethical concerns subjecting patients to a placebo, or an inferior active comparator, when we consciously know that there is a better alternative.
We have been routinely using synthesized patient profiles to better understand various trial designs and their impact on the wider patient population, without conducting clinical trials. This exercise enables us to predict which clinical trials will have a probability of failure, and prevent them from even starting.
The COVID-19 pandemic has exacerbated a number of existing challenges in clinical trial planning and execution and shone a spotlight on the urgency for companies to starting systematically utilizing synthetic data in clinical development, including using synthetic clinical trial arms.
Yet, there is still widespread hesitation among sponsors to use synthetic data more broadly, for two main reasons:
- There is a lack of understanding in how much synthetic data utilization has progressed
- The conservative nature of our industry, and any marginal risks associated with the failure of a usually innovative clinical development candidate when using new methods.
OSP: What in your opinion are the key takeaways from the report you’ve issued with Accenture?
GL: The key takeaway from this report is to help demonstrate the “art of the possible” in clinical development, brought to us by available patient data, and the enhanced ability to gain and utilize patient data without subjecting actual patients to the comparator arm part of the clinical trial process, as well as reducing costs and accelerating the delivery of innovative new drugs to patients.
The full report, Faster and Cheaper Clinical Trials: The Benefit of Synthetic Data, can be accessed and downloaded here: https://bit.ly/faster-cheaper-clinical-trials
