Diagnostics & Imaging Week

CHICAGO – The FDA had some words of advice for attendees at this week's annual meeting of the American Association for Clinical Chemistry (AACC; Washington) on the challenges of biomarker and drug co-development and regulation – commonly known as the path to "personalized" medicine.

While saying that the "diagnostic lifecycle is by now well-defined," Steven Gutman, MD, of the FDA's Center for Devices and Radiological Health, advised companies that if they have an assay that they "think is good," then certainly pursue studies that, with any luck, will have clinical significance and will ultimately be reimbursed.

"What companies like more than an FDA [approvable] marketing letter is to get paid for the tests," Gutman said.

He noted the importance of co-development in which biomarkers are used both for diagnosis and drug development, with diagnostics a key vehicle for improving a drug discovery pipelines.

However, he warned, the drug can become "hostage" to the device in the co-development process, meaning that there needs to be a balance between drug and device, and "safety and effectiveness have to be [determined] in both."

He also said that the "whole notion of personalized medicine" is perhaps oversimplified in the way that it currently is discussed, because often omitting the complications of the required regulatory processes involved.

There are two ways of credentialing diagnostics and drugs in co-development: first identify the patients affected or recruit and treat" according to the biomarker of interest. "In my view, the ideal is a matrix of all patients," he said.

Secondly, studies for a diagnostic could be done in a way in which a drug is working in a subset of patients, but Gutman said he is "not aware of any submissions on this predicate."

"What we see at the FDA most commonly is co-development based on enrichment studies," he said, noting that one of the drawbacks to these studies is companies only having "predictive value of positive results only."

To determine the timing of a diagnostic study, Gutman said it is "optimal" if the diagnostic study and the drug study are coordinated. If the studies cannot be conducted simultaneously, then Gutman suggested saving the samples from one study for use in the other.

He acknowledged that there is "little regulatory or scientific history" on which to base questions of biology when co-development is approached, which ultimately means that "for companies, there are probably financial risks, as well."

However, the scientific "literature is growing," he said. For instance, the FDA is developing an updated concept paper on co-development that should help with a roadmap for companies wishing to pursue this avenue. And, Gutman said, the CDRH and the FDA's Center for Drug Evaluation and Research (CDER) "work together well."

In a comment obviously responding to persistent criticisms of the agency, he said: "Our purpose is to protect the public's health . . . you can firebomb the FDA, and you can get rid of us, but those . . . pesky questions will stay on the table."

Representing the CDER, Felix Frueh, MD, associate director of genomics at the FDA, told the audience: "Pharmacogenomic-based drug therapy is only possible if we have two things: a drug and a diagnostic" – and with information about both. "It is valuable in the marketplace also only if we know these two things," he said.

Frueh said the FDA is in the process of developing a web site on the genomics side of the equation. It will list biomarkers and drugs, so companies can conduct more complete searches as they ramp up development.

He pointed also to a proposal on biomarker validation, to be published soon, that will "discuss the validity of preclinical genomic markers" for drug safety. The goal of this and other initiatives is to create "regulatory buy-in."

As to the why of drug/test co-development, Frueh said the primary goal of this approach is to move therapy from a trial-and-error approach to a scientifically based foundation for drug efficacy.