WASHINGTON—Measurable biomarkers are a key tool of molecularly based precision medicine, but their success rate has been very poor. Now, a new alliance has come together in an attempt to improve that and create an evidence-based, “end-to-end” development process. The nonprofit group, the National Biomarker Development Alliance (NBDA), brings together academia, the private sector, insurers, patients, and advocates to improve biomarker discovery.
Speaking here at the National Press Club last month to launch the alliance, the group's President, Director, and cofounder, Anna D. Barker, PhD, Co-Director of Complex Adaptive Systems and Professor at Arizona State University's School of Life Sciences, called the existing success rate of biomarkers in clinical practice “dismal.”
Fewer than 100 biomarkers are routinely used in the clinic, she said. “The explosion of genomics-based assays and other non-regulated laboratory-developed tests discourages companies from pursuing more rigorous, uncertain, and expensive FDA biomarker/diagnostic approval pathways.”
Barker, a former Deputy Director of the National Cancer Institute, said the new alliance is needed to create standards to enhance the use of biomarkers in precision medicine. “It's a solvable problem, and we can do it. We believe this is the path to precision or personalized medicine.”
The Alliance is “hosted” by the Arizona State University (ASU) Foundation, and the early planning for NBDA was collaboratively undertaken by leadership from ASU and the Critical Path Institute (C-Path) International 2 Genome Consortium (IGC) and the Translational Genomics Research Institute (TGen). Information from NBDA notes that the organization has spent the past 18 months “engaging experts in workshops and think tanks to develop an in-depth understanding of the critical barriers at each stage of biomarker development.”
The aim now is to “focus on solutions, be agnostic in terms of disease, and inclusive versus exclusive of the affected stakeholders. The NBDA will not re-invent good work—but rather will evaluate and incorporate existing strengths into solutions as appropriate. All of the NBDA's results from demonstration projects, databases, workflows, etc. will be public and available on its website” (which is not yet operational).
Demonstration Projects in Four Areas
The NBDA is already setting up demonstration projects to develop standards for four classes of biomarkers: genomics, proteomics, imaging, and complex biomarkers (such as biosignatures).
Asked by OT if the alliance intends to work with specialty societies such as the American Society of Clinical Oncology, Barker said yes: “We hope to work with all the constituency societies, such as ASCO. If we do a good job of putting these standards together, they will be accepted.” She said that when new standards for biomarkers have been accepted, industry representatives developing one or more biomarkers that did not use those standards would be “somewhat delusional that they're ever going to get them approved.”
Asked at the news briefing for her opinion, Margaret Foti, PhD, CEO of the American Association for Cancer Research, said, “Certainly the AACR would be very excited to participate in this,” and that she would be happy to leverage the power of the AACR to help the NBDA succeed.
She cited the biomarker consensus report of the AACR, Food and Drug Administration, and National Cancer Institute published in 2010 in the AACR journal Clinical Cancer Research on advancing the use of biomarkers in cancer drug development” (OT 10/10/10 issue), which had said, “Biomarkers are central to accelerating the identification and adoption of new therapies, but currently, many barriers impede their use in drug development and clinical practice.”
The NBDA has been in the works for several years, Barker said, noting that earlier this month the FDA published final draft guidance on qualifying drug-development tools, including biomarkers.
FDA's Janet Woodcock
Also speaking at the news briefing, Janet Woodcock, MD, Director of FDA's Center for Drug Evaluation and Research, said about the new alliance, “We've wanted something like this for some time. Every day I hear from desperate and frustrated patients and relatives. They say, ‘Why can't you tell me earlier that I will not respond to a drug?'”
What are needed are reliable new tools to determine patient response to a given therapy early in the course of disease, she continued. “It is not our job to develop standards. We're stretched thin as it is. The FDA will help in this effort in any way we can,” noting that correct diagnosis is the very foundation of medicine.
“This is an unprecedented, exciting time in biomarker science,” said Steven D. Averbuch, MD, Vice President for Translational Clinical Development and Pharmacodiagnostics at Bristol-Myers Squibb. He said the NBDA will provide a needed platform for a consensus approach to defining and implementing standards for the biomarker-development process. As a specific example of a biomarker success story, he cited the KRAS mutation, which shows which colorectal cancer patients are likely to respond to cetuximab and which are not.
Other speakers noted that the disappointing success rate of biomarker in clinical practice is not due to a lack of scientific studies. On the contrary, some 150,000 scientific papers have been published documenting thousands of biomarker discoveries, said George Poste, DVM, PhD, the NBDA's Interim Chief Science Officer, and Arizona State's Co-Director of Complex Adaptive Systems; Regents' Professor; and Del E. Webb Chair in Health Innovation.
The rate of successful biomarker development is dismal because of what Poste called a lot of “s” problems:
- Sloppy science—“publish and vanish,” leaving behind unsolved issues of biomarker reproducibility and replication);
- Statistical issues—for example, biomarker studies that lack statistical power;
- Standards vacuum;
- Sharing gap (lack of data disclosure); and
- Silo science, which prevents scientists in different disciplines from communicating with each other about their findings.
Poste stressed the need for high research standards that include using biospecimens of reliable quality: “The key is to work with robust specimens, not specimens of convenience. “It really is a question of discipline.”
Another speaker, Carolyn Compton, MD, PhD, the NBDA's Chief Medical Officer and Professor at Arizona State's School of Life Sciences, said, “The pre-handling of the specimen is a black box.” As an example, she pointed out that the test results for HER2 in breast cancer will turn negative if the biospecimen is left out too long even if the biospecimen is from a woman who is HER2 positive; this specimen mishandling could prevent HER2-positive patients from receiving life-saving trastuzumab therapy. In 2010 ASCO issued a statement on “exemplary attributes” for research sites in biospecimen collection and use in clinical trials (Journal of Oncology Practice 2010; 6:206-209).
Using biomarkers in clinical trials is about using the ones that meet rigorous standards and “building a community that uses the same standards,” said another speaker, Laura J. Esserman, MD, MBA, Director of the Carol Franc Buck Breast Care Center and Professor of Surgery and Radiology at the University of California, San Francisco, and principal investigator for the I-SPY Trial program, a multisite neoadjuvant Phase II breast cancer adaptive clinical trial that uses biomarkers to select the most precisely targeted therapy for each trial participant.
In a seminar that followed the news briefing, Joshua LaBaer, MD, PhD, Professor and Director of the Biodesign Institute's Virginia G. Piper Center for Personalized Diagnostics at Arizona State, emphasized that “developing biomarkers is a team science.” He also noted that even though biomarker validation is vitally important, it is not easy to get funding to do biomarker validation studies, because “it is not a sexy field.”
Indeed, he said, “If you don't find a biomarker that succeeds, no one wants to hear about it.” Esserman agreed: “The journalists and the public only want to hear about things that work.” LaBaer said it costs about $60 million to $100 million to go through the discovery process on a given biomarker. In cancer, he noted, this expense can pay off if a biomarker becomes a “value added” diagnostic tool—for example, if it can be used to supplement mammography.
NBDA's Planning Team includes Anna Barker (Chair), Karen Anderson, Michael E. Berens, Kenneth Buetow, Martha Brumfield, Carolyn Compton, Raymond DuBois, Joshua LaBaer, Randy W. Nelson, Robert Penny, George H. Poste, and Robert Mittman.