BETHESDA, Md. — The beleaguered state of biomarker research was once again a topic of discussion at this week’s annual conference of the US Human Proteome Organization, held here this week.
With many proteomics experiments now largely fishing for biomarkers, the field is increasingly looking inward to reflect on where the research is and concluding that for all the work that has been done, it has fallen short of expectations.
This week, Leigh Anderson, founder and CEO of the Plasma Proteome Institute, kicked off the conference, the fourth such conference held by the US arm of HUPO, by saying that the state of clinical biomarker research is nowhere near where people thought it would be by now, and the field is still in the very early stages of the diagnostics-development pipeline.
A similar sentiment was expressed during a discussion with representatives from federal funding agencies.
While biomarker research has resulted in thousands of candidate biomarkers, so far not a single clinical biomarker has been discovered through proteomics methods, Anderson said. Furthermore, he added, since 1994, the number of protein-based diagnostic tests has been falling.
The diagnostics pipeline is increasingly resembling the pharmaceutical drug pipeline, “which means it’s going to be unglamorous and a lot of difficult things are going to have to be done” to get to the point where proteomics-based tests will have any clinical and marketplace significance, he told the audience.
His evaluation mirrors the views of many other proteomics insiders. In recent years, similar views have been repeated at several other forums, including HUPO’s annual conference held last October, where panel discussion participants said that bringing proteomics to the clinical setting remains a distant goal [See PM 10/11/07
In his keynote speech, Anderson said that while the term “proteomics” may just be in its teenager phase, the study of clinical proteins is approaching retirement age. But clinical proteomics is only now beginning to get to where he and other protein researchers thought it would be “in five to 10 years back in 1980,” he said.
“We’re still very much in the discovery stage. We are years away from getting anything of clinical [value].”
The reason for the dearth of protein biomarkers is the lack of technology and the complexity of the biology, Anderson said. Technologically, the tools and instruments are not good enough to find low-abundance proteins. While progress has been made in developing technology and methods for candidate biomarker discovery and assay development, less progress has been made in verifying, qualifying, and validating those biomarkers.
The two main technologies that exist, he said, are immunoassays and hybrid mass spec-based assays. While the former are sensitive, they are also expensive and have specificity issues and multiplexing limits, Anderson said.
Hybrid MS-based assays, meanwhile, are not sensitive enough for folding complicated applications, although they can yield absolute analyte specificity and analyze samples on a high multiplex level.
Biologically, in addition to the inherent complexity of biomarkers, researchers may be looking in the wrong places. Variance in population is also great, contributing to the muddled picture, he said.
Indeed, the whole field of biomarker research is based on a hypothesis that is presumed to be true, but one that no one knows for sure is a fact: For every disease state, there is an “accessible” protein biomarker, Anderson said.
“It is a pure hypothesis and there is no way of genuinely determining whether this is true,” he said, adding there is no “guiding theory” when it comes to biomarkers: The molecular weight of specific proteins, how they interact, and their biovariability are fundamental things that still perplex researchers.
Some at the conference agreed with Anderson. At a roundtable discussion held at the meeting, representatives from various US government funding agencies said the clinical biomarker field is still in the crawling stage.
“We’re still very much in the discovery stage,” Maureen Beanan, program officer at the National Institute of Allergy and Infectious Disease’s Division of Microbiology and Infectious Diseases, said about the state of biomarker work at her agency. “We are years away from getting anything of clinical [value].”
Susan Old, acting deputy director of the National Heart, Lung, and Blood Institute’s Division of Cardiovascular Diseases, said that even if new biomarkers are discovered, they aren’t necessarily of any value.
“We need to assess what is the added value in [providing the literature with] another marker or another panel,” she said. And if it turns out to have value, there are other issues, including whether insurers will pay for a protein-based biomarker.
The difficulty in biomarker work was also emphasized by Danilo Tagle, program director of the Neurogenetics Cluster at the National Institute of Neurological Disorders and Stroke. According to Tagle, his agency is still dealing with issues of tissue and fluid samples.
“Nobody wants to donate their brains” while they’re still alive, he said.
On funding, within the proteomics community, there is a pervasive sense that government funding in the US is tilted toward genomics. Asked to comment on that, the government representatives were mum.
Each agency representative gave an overview of the proteomic programs and projects currently underway at their respective agencies. An audience member said that while it was “wonderful” that such programs existed, he questioned whether the funding was getting to outside researchers.
Several panelists replied that their offices were especially trying to reach out to new investigators. Roundtable moderator Gil Omenn, who is a researcher at the University of Michigan, said that because many of the government-run proteomics programs were still relatively new, it was not yet known if funding would eventually reach more researchers and result in “a flood” of papers tied to government projects in proteomics.
“I hope so,” Old said.