Califf Seeks Clinical Trials That Inform Product Labels

Newly confirmed FDA Commissioner Robert Califf has long advocated for a “learning healthcare system” that supports clinical research by tapping electronic health records to facilitate clinical trial enrollment and provide ongoing information on the effects and side-effects of therapies in real-world use. Now at the helm of FDA, Califf has a ready platform to discuss such strategies for improving clinical trial methodology, execution and evaluation as part of the larger goal of developing more innovative treatments for patients. 

Such issues were addressed last week at an Institute of Medicine (IOM) workshop on “Neuroscience Trials of the Future.”  In a discussion on regulatory pathways in the U.S. and other countries in this field, Califf addressed some of challenges facing FDA and sponsors in achieving more effective clinical trials for nervous system disorders.

Sponsors now take different approaches for “regulatory” trials to gain market registration, and “real world” studies to determine how a product should be used in practice, Califf observed. But these two worlds should not be separate, he said, as FDA’s job is to approve a label that informs how to use a product in practice, not in theory. Even though doctors don’t read drug labels, these documents are important, Califf observed, because they form the basis for practice guidelines and compendia. Sponsors thus should start by envisioning the product label and corresponding clinical practice guidelines.

In the “ideal world,” said Califf, sponsors and investigators would conduct group concept studies that lead to informed clinical trials, and those results then would be used to write practice guidelines and to guide more “real-world” studies; these, in turn, would provide more evidence and refine practice. “This would just be routine” in a learning health-care system, he commented, where research influences practice, and practice influences research based on constant tracking, adjusting, evaluating and measuring impact.

Complexity drives up costs

Unfortunately, Califf sees the research community opting for larger, more complex clinical trials, with the result that costs are “going off the scale.” Investigators enroll fewer patients per site because “we’re making things more and more complex,” something that he hopes FDA can address. He suggested that trials stop running multiple blood tests and collecting rare, non-serious adverse events from all patients, which “costs a ton of money.” Sponsors could schedule patient visits less frequently, and possibly improve tracking off site through the use of new technology.

It’s not necessarily FDA that seeks more data from larger studies, Califf observed. He recalled from previous work managing clinical trials for biopharma companies that in FDA-sponsor meetings, the agency might offer a strategy for simplifying a study, but the company refused the proposal. Any uncertainty for a sponsor “feels intolerable,” he noted, so the challenge for FDA is to develop study models that all parties “feel good about.” FDA can’t promise it will approve a new product if the researchers do things in a certain way, as there are “always surprises with medical products,” he commented. But we can assure, Califf said, that we won’t come back later and say we don’t like that approach.

More informative, streamlined clinical trials are particularly needed in neuroscience research, where the pace of developing new treatments for Alzheimer’s, Parkinson’s and psychiatric disorders has been stymied by high failure rates in early studies, poor assay sensitivity, low treatment adherence, and weak patient engagement and retention, according to scientists at the IOM workshop. Steven Romano, senior vice president and chief science officer at Mallinckrodt Pharmaceuticals, noted that industry investment in new drug development has shifted away from central nervous system (CNS) disorders to other areas with clearer R&D pathways. CNS studies are challenging, he commented, because the biology is unclear, genetics complex, the brain is difficult to examine, and there are few validated molecular targets or compelling biomarkers.

A main topic of discussion was whether expanded collection of real-world, real-time data through “patient-centric technologies,” such as smartphones and wearable devices, could improve the efficiency of traditional CNS trials. The hope is that such methods can boost patient retention and adherence to assigned treatment, which are particular problems for researchers in the neuroscience field.

Patient privacy and data sharing also are important concerns in conducting trials for depression and neurological disorders. Researchers are looking to engage more with patients to improve study screening, adherence and retention.