By Tony Fong
SAN DIEGO — With the use of mass spectrometers gaining traction in the clinical laboratory, a new organization formed to promote mass spectrometry and share expertise in the area specifically for clinical applications held its first conference here this week.
The conference, organized by the Association for Mass Spectrometry: Applications to the Clinical Laboratory, also took place in 2008 when it was held as part of a continuing education workshop by the University of California, San Diego, according to David Herold, chief of clinical chemistry at the VA Medical Center in San Diego, and conference chair and founder.
But based on the response to the 2008 event and in order to have greater autonomy, organizers of that event formed their own organization, MSACL, in May 2009 to have purview over the conference, Herold told ProteoMonitor.
Other organizations and conferences, such as the American Society for Mass Spectrometry and Pittcon, have as part of their programs courses and sessions on clinical mass spectrometry. And most recently, the American Association of Clinical Chemistry made proteomics a full division within the organization in recognition of the growing importance of the field [See PM 11/06/09]
But according to Herold, clinical mass spectrometry can get overlooked in such forums, and “we wanted to focus this on people who use mass specs in a clinical lab.”
A total of 440 people attended the conference, and what they experienced was a smorgasbord of mass spec-based research. This year’s program, spread over five days, covered the gamut in clinical mass spectrometry, from traditional applications such as small-molecule analysis and newborn screening, to emerging uses of the instruments such as for the investigation of microorganisms, and what Herold called “frontier” applications, including proteomics and metabolomics.
The conference kicked off with a plenary address by Lee Hood, co-founder of the Institute for Systems Biology, recapping what he calls a new paradigm in healthcare dubbed P4 that is based on the theory that it must be predictive, preventive, personalized, and participatory.
As part of the effort to move to a P4 approach, Hood and others started a new firm, Integrated Diagnostics, to leverage proteomics and other technologies to translate biomarkers to tests for a broad range of diseases [See PM 10/16/09].
While the use of mass specs for proteomics in a clinical environment is still in a nascent stage, its potential could have wide-ranging implications, according to speakers here, and proteomics was well represented at the conference. In one track, presenters took a retrospective look at where clinical mass spec-based proteomics was, where it is, and where it needs to go.
Keith Baggerly, associate professor of bioinformatics and computational biology at the University of Texas MD Anderson Cancer Center, started the session by revisiting the Lance Liotta-Emanuel Petricoin controversy from 2002 when the two researchers, who were then with the US Food and Drug Administration/National Cancer Institute Clinical Proteomics Program, announced they had identified protein biomarkers for ovarian cancer using the SELDI platform.
The finding was later questioned by Baggerly and eventually rebuked by the scientific community.
That talk was followed by another from David Ransohoff, professor of medicine at the University of North Carolina School of Medicine, who highlighted the importance of bias in study design, particularly sample bias, which is too often overlooked in proteomics studies. Included in the examples he pointed out was one that investigated prostate cancer in which disease samples came from older men but control samples included serum from women.
Finally, Steve Skates, associate professor of medicine at the Harvard Medical School, spoke about the use of selective-reaction monitoring for developing assays for use in longitudinal biomarker studies, and the importance of achieving single-digit coefficients of variation in developing mass spec-based assays.
[ pagebreak ]
Another track focused on discovery proteomics. In it, Steve Carr, director of the Proteomics Platform at the Broad Institute, presented work on developing a discovery-to-development pipeline based on a multiple-reaction monitoring method. The work builds on research that he and his collaborators have done as part of the NCI’s Clinical Proteomic Technology Assessment for Cancer Initiative [See PM 07/09/09].
To date, his lab has built about 200 assays, he said, and aims to increase that to about 400 within the next few years. In addition, they also plan to explore the use of the assays for the detection and verification of post-translationally modified proteins and splice variants.
During the same track, Leigh Anderson, CEO of the Plasma Proteome Institute, discussed regulatory, financial, and scientific roadblocks that have resulted in a paltry number of proteins used for diagnostic purposes — 205 — that have ever been passed by the FDA or been CLIA-approved.
And Andy Hoofnagle, director of clinical mass spectrometry at the University of Washington Medical Center, presented work he is conducting using proteomics methods into the role obesity and insulin resistance may have in the levels of clusterin in high-density lipoproteins and non-HDL particles.
While one of the major goals of proteomic mass spectrometry is to move the instruments into the clinical environment, it may still be some time before that happens, according to speakers this week. The consensus view is that hospitals are with increasing frequency purchasing triple-quadrupole instruments for screening purposes.
But movement of the instruments for proteomics-driven purposes, even into the clinical lab, is “still a ways away,” Christopher Kinsinger, proteomics program manager at the NCI, told the audience during his talk.
MALDI vs. ESI
Much of the mass spec work, especially in proteomics, uses electrospray ionization platforms, but during the conference Marvin Vestal discussed how his company, Virgin Instruments, where he is CEO, is developing a MALDI platform that could have widespread applications in proteomics.
While ESI instruments have high performance and are widely available, they suffer from low flow rates, limited capacity, and long LC runs, Vestal told the audience in his talk. MALDI-TOF, by contrast, is “very fast” but has inferior resolution and mass accuracy.
To address the shortfalls of MALDI technology, Virgin Instruments has designed platforms with optimized elements of TOF mass spectrometry, including one- and two-stage MALDI ion stages, two-stage ion mirrors, and pulsed and static ion accelerators.
His company has developed four instruments with a focus on optimizing resolving power, mass accuracy sensitivity, and dynamic range, Vestal said. And one platform being tested has a resolving power of 60,000 with a mass accuracy of 2 ppm.
If MALDI is to become a common tool for disease diagnosis, Vestal said, the technology will have to be simple, rugged, precise, and accurate with automated calibration. In addition, the platform will need to be high-throughput and capable of being used to develop multiplex assays. Cost per sample would need to be low, he added.
MALDI can meet these criteria, Vestal said, “but we’re not there yet. We’ve made a good start … but we still have a lot work” to do yet.
Another highlight of the conference was a presentation by Larry Bowers, chief science officer with the US Anti-Doping Agency on the evolution of the use of mass specs in the area of detection of performance-enhancing drugs.
While GC-MS instruments were used at least as early as the 1980s for detecting steroids in urine, most recently blood has replaced urine as the matrix of choice for detecting both hormones and their isoforms, as well as metabolic biomarkers.
Protein biomarkers, while still an emerging source for anti-doping detection, could become an important one, and the use of stable-isotope labeled internal standards could result in the development of better methods with increased accuracy, Bowers said.