One of the world’s leading cancer research centres is considering radical changes to the way it tests new treatments.
Scientists at the University of Texas MD Anderson Cancer Center, lead by Donald Berry, the professor of Biostatistics and Applied Mathematics, believe the time may have come to ditch traditional clinical trial design.
They are testing whether the common “frequentist” or “Neyman-Pearson” method, in which researchers ignore prior results, should be replaced by a new approach. In this, the “Bayesian” method, prior discoveries and findings that emerge during a trial are allowed to influence how the study is subsequently conducted.
Prof Berry said the Bayesian method was more flexible and would allow clinical trial participants to benefit from medical knowledge as it emerged.
Already more than 100 cancer-related Phase I and II clinical trials are being planned or carried out at his centre using the new approach, which is more commonly used in non-clinical sciences such as physics and geology.
“We need to rethink how we design and conduct clinical trials in the United States,” said Prof Berry. “Our current system has served us well for the past 50 years, but the demands of 21st century medicine are beginning to put a strain on the current system, and we believe we have something to relieve that strain.”
In Nature Reviews Drug Discovery Prof Berry said that the frequentist method, used nearly exclusively to design and monitor clinical trials today, limited innovation and learning.
In contrast, the Bayesian method would allow doctors much greater flexibility the ability to respond to individual patients. “At the end of the day, when they enroll the last patient in the study doctors want to be able to treat that patient optimally depending on the patient’s disease characteristics,” he said.
“Using a Bayesian approach, the trial design exploits the results as the trial is ongoing and adapts based on these interim results.” He admitted this was “anathema” to the traditional approach.
Stephen Porter, professor of oral medicine and an expert in clinical trial design at University College London, said the Bayesian approach was not without drawbacks: “The problem is if you’re going to change the trial as you go along it can be difficult to get the statistical data to show that ‘A did B’ as you set out to.
“However, sometimes in trials it’s possible to spot trends in sub-groups of patients, and a more pragmatic approach which enables doctors to respond to increased side-effects or worsening outcome in these subgroups could be a good thing.”
Prof Berry said that with the Bayesian method it might be possible to reduce the number of patients required for a trial by as much as 30%, thereby reducing the numbers of people put at risk – and the cost and time required to develop new treatments.
He added that efforts were being made both in persuading both pharmaceutical companies and the US Food and Drug Administration of the benefits of Bayesian methods.
“Our biggest challenge is to convince the regulators that we are not throwing the baby out with the bathwater by using a Bayesian approach. It is rigorous and we are not losing science by using it,” he said.
