He began by stating that the report outlines clearly “all of the factors and potential pitfalls” involved in determining what he called a “rock star biomarker.” Calling the three-step framework comprehensive, Dr. Johnson agreed that biomarker use demands analytical rigor, and he assessed the qualification phase as feasible based on prospective epidemiological studies. However, he added, “the rub comes in determining the clinical outcome of an intervention where you change just the biomarker and nothing else, and then wait to see what happens in a group of healthy people.” The report conveys the message that there are no shortcuts to a surrogate endpoint, he said.

Nevertheless, Dr. Johnson emphasized that biomarkers offer an important means to communicate useful information to companies and other organizations and to inform research. He expressed hope that the report can be used to “provide some structure around how that information can be communicated.” Such communication could happen during the utilization step, by distinguishing between predictive biomarkers, for which there is evidence of an association with a clinical endpoint (for example, C-reactive protein [CRP]) that could be used to inform less rigorous claims, from probable biomarkers for which validity is clearly established. “It is possible that there could be some kind of regulatory language that would allow information on those less-than-rock-star biomarkers to be communicated,” he suggested. “I don’t think the committee’s intent was to tell clinicians they can’t measure CRP because it hasn’t been fully established as a biomarker,” he said; rather, there need to be guidelines for how to make use of information from biomarkers that are not surrogate endpoints. “That’s where I think the real opportunity of this report is,” he said.

Having the same standards of scientific rigor for biomarkers used in foods and drugs makes perfect sense, Dr. Johnson said. However, there are inherent differences between the process one might use to determine the effect of a food on a biomarker and the effect of a drug on the same biomarker. For example, he continued, “it’s tough to do a randomized double-blind placebo-controlled trial with pomegranate juice [so that] … people don’t know that they are eating it.”

The report presents an opportunity to communicate information to consumers, as do biomarkers themselves, Dr. Johnson said. “Frankly, there’s a danger in not communicating information to consumers,” he said. On the other hand, he hoped that the report was not viewed as the “kiss of death” for a variety of biomarkers, such as low-density lipoprotein cholesterol (LDL-C).

Dr. Johnson also asked, in light of the report’s recommendations, whether the Food and Drug Administration (FDA) would need to convene a panel to examine the blood pressure/sodium biomarker. “There are no clinical trials that show that reducing sodium, and therefore blood



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement