SAN DIEGO – The familiar questions of reproducibility and throughput loomed large at the Association for Mass Spectrometry's fourth annual Applications to the Clinical Lab meeting, held here this week.
But while such technical hurdles still limit the utility of mass spec-based proteomics, several vendor announcements and research presentations indicated that the field is moving ever closer to fulfilling its clinical promise.
Given the struggles of discovery proteomics, such a move may be not only desirable but necessary, observed Swiss Federal Institute of Technology Zurich researcher Ruedi Aebersold. In a lecture at the meeting, he said that future biomarker discovery would increasingly rely on data from genomics research rather than comparative proteomics experiments, suggesting that proteomics will ultimately prove most useful as a tool for validation and clinical implementation of biomarkers.
"Tens of thousands of papers have been published trying to find [protein biomarkers via discovery proteomics], and the output has been rather meager to date," Aebersold said, noting that this was in large part due to the technical difficulty of analyzing the low-abundance proteins thought to be the most promising biomarkers.
"It has been enormously difficult for many laboratories to collect hundreds of samples of cohorts from healthy and disease-affected individuals and to do very thorough plasma proteomic analyses," he said. "So, I think that it's reasonable to assume that increasingly biomarker candidates will come from genomics and bioinformatics rather than from comparative proteomics ... [and that] the objective of proteomics in biomarker research will shift from discovery to validation."
This shift, Aebersold said, will necessitate technologies that "can reproducibly and quantitatively measure essentially any protein in a sample" and do so in large numbers of samples. He highlighted as one such technology the SRMAtlas developed jointly by ETH Zurich and the Institute of Systems Biology, which offers researchers a storehouse of selected-reaction monitoring or multiple-reaction monitoring mass spec assays they can use for protein quantitation (PM 9/24/2010). The atlas, Aebersold said, currently has assays for 170,000 proteotypic peptides, or roughly 99 percent of the proteins in the human proteome.
Due to its high sensitivity, specificity, and multiplexing potential, SRM/MRM-MS has emerged as the predominant technique in mass spec-based clinical proteomics. A number of other scientists besides Aebersold presented on the method's potential clinical applications, most prominently proteomics researcher Leigh Anderson, whose stable isotope standards and capture by anti-peptide antibodies, or SISCAPA, protein quantification technique, typically uses an MRM-MS workflow.
At last year's MSACL, Anderson detailed an automated SISCAPA workflow he had developed in collaboration with Agilent using an Agilent Bravo liquid handling system for sample prep attached to an Agilent 1200 series LC system and an Agilent 6490 triple quadrupole mass spec (PM 2/11/2011).
Using that system, Anderson said, he and Agilent had been able to bring the method's cycle time down from about 40 minutes to around 10 minutes per sample. At this year's meeting he noted that they had managed to bring the cycle time down even further, to around three minutes per sample.
"Russell Grant [director of mass spectrometry at Laboratory Corporation of America] once told me, 'Don't talk to me if [a clinical assay] takes longer than five minutes,'" Anderson said. "So this is less than five minutes."
He added that he hoped to further improve throughput and, ultimately, sensitivity, by eliminating the method's chromatography step. He is pursuing this goal via two different routes – on the one hand by adapting SISCAPA to MALDI-MS, a technology with a rising profile as a potential clinical platform (see story, this issue); and, on the other, by incorporating Agilent's RapidFire high-throughput separations platform, which allows for injections of samples into a mass spec every seven seconds.
Using the RapidFire in SISCAPA-based peptide quantitation "we can get reasonably decent linearity with a range of different peptides," Anderson said this week, adding that while "the [coefficients of variation] are not as low as we want them to be, they are reasonably low." The device, he added, looks particularly promising for measuring medium-abundance proteins, making it "possible to imagine doing very high-throughput quantitative analysis by this approach."
Through this work with Anderson and deals like the partnership agreement with Integrated Diagnostics that it announced last week (PM 1/13/2012), Agilent has clearly demonstrated an interest in moving its mass specs – and particularly its triple-quad instruments – into the clinic. The company further signaled this aim this week with the announcement that it had registered certain of its 1200 series LC instruments and 6000 series triple quads with the US Food and Drug Administration as Class I exempt medical devices (see story, this issue).
The move, said Gustavo Salem, vice president and general manager of Agilent's Biological Systems Division, stemmed from an acknowledgment that these systems are currently being used in clinical settings and a recognition of mass spec's growing usefulness as a clinical tool.
Registration is an "acknowledgment that we're seeing more and more of our business focused on the clinical laboratory," Salem told ProteoMonitor. "It's an important market. It's a strategically critical opportunity for us to continue to participate with the customers in these [clinical] laboratories, and we just think that there will continue to be more and more adoption of mass spectrometry in the clinical setting."
Waters, the only other vendor to have registered some of its mass spec instruments with FDA, also made a move with clinical implications this week, announcing an agreement with automation firm Tecan to combine Tecan's Freedom EVO liquid handling platform with its Acquity TQD LC/MS-MS system for running clinical assays.
Clinical customers' needs for higher throughput and better reproducibility in their mass spec work was a major factor driving the deal, Pat Martell, Waters clinical marketing director, told ProteoMonitor. "Many of our customers, when they look at LC/MS-MS solutions, they struggle with sample prep," he said. "It's a hurdle for them to get through – the time element, as well as the consistency. So, [the Tecan agreement] will really take away some of those hurdles that our customers have."
Currently the collaboration is aimed at providing a solution for existing clinical applications like immunosuppresant testing, Martell said, but an automated LC-MS/MS workflow could also prove useful for protein quantitation purposes down the line.
Significant support from large vendors like Agilent and Waters is key to making mass spec-based proteomics a clinical reality, Randall Nelson, a researcher at Arizona State University's Biodesign Institute, told ProteoMonitor, adding that he thought the big vendors were at last lining up behind the technology in a serious way. Among the signs of this, he said, was Thermo Fisher Scientific's purchase last year of his protein biomarker firm Intrinsic Bioprobes (PM 9/2/2011).
"If you really start to think about clinical and diagnostic medicine, it's not a small company endeavor," Nelson said. "You need to address not just a couple of people in an academic lab; you need to be able to address everybody in the US, every day. That's a full-on industrial push. It's big business. And so hopefully with some of the steps we've taken with Thermo – and I see some other companies doing some similar stuff – that will start to resolve some of these hindrances" that have kept mass spec-based proteomics out of the clinic.
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.