By Tony Fong
The year 2009 was a watershed for the National Cancer Institute's proteomics division as projects begun at the inception of the program three years ago began bearing fruit, according to a report released last month by the division.
In its second annual report, the NCI's Clinical Proteomic Technologies for Cancer initiative cited further development of a workflow for verifying disease biomarkers, the creation of a yeast proteome reference, and the identification of standard operating procedures and performance metrics as high points during the past year in its mission to accelerate "the translation of new cancer [protein] biomarkers into diagnostic tests."
Created in 2006 with $104 million in funding over five years, CPTC is the NCI's main initiative directed at applying and improving proteomic technologies and methods for diagnosing and treating cancer.
CPTC is composed of three components: Advanced Proteomic Platforms and Computational Sciences, focused on the development of new tools, reagents and "enabling technologies for protein/peptide measurement"; the Proteomic Reagents and Resources Core, which develops tools, reagents, enabling technologies, "and other critical resources to support protein/peptide measurement and analysis efforts"; and Clinical Proteomic Technology Assessment for Cancer, for the development of standardized technologies and methods that would allow for inter-laboratory proteomics research.
While the NCI-CPTC reported progress in all three areas, the work by CPTAC made the biggest news. This past year, the CPTAC network of five teams — each comprising researchers from multiple universities and organizations — reported the completion of critical phases of multi-year projects aimed at improving the quality of proteomics experiments and the resulting data.
The CPTAC teams were named in 2006 and share a five-year, $35.5 million grant.
In its annual report, the NCI-CPTC said that CPTAC's goal is to "enable all researchers conducting cancer protein biomarker research at different laboratories to use proteomic technologies and methodologies," and that ultimately, its work "should lead … to improved diagnostics, therapies, and even prevention of cancer."
In one project focused on developing a multiple-reaction monitoring method, researchers reported they had designed and implemented an MRM protocol to measure absolute amounts of proteins that had been spiked into human plasma, "providing a foundation for the proteomics community," according to the NCI-CPTC annual report. In addition, they demonstrated the reproducibility of the MRM method across different laboratories and with different LC-MS platforms [See PM 07/09/09].
The work is being spearheaded by the CPTAC team led by the Broad Institute, whose "overarching goal" is to develop methods for verifying large numbers of candidate protein biomarkers. Specifically it is seeking to develop a "sensitive, specific, and quantitative technology based on [MRM-MS] that is capable of measuring hundreds of candidate cancer biomarker proteins in large sets of clinical plasma samples," according to the annual report.
The team is trying to couple MRM-based techniques with stable isotope standards and capture by anti-peptide antibodies, or SISCAPA, "which employs peptide-specific antibodies to improve sensitivity, speed, and robustness of the assays through enrichment," the annual report said.
It added that researchers have increased the sensitivity of MRM multiplexed assays for proteins in plasma by more than 500-fold. More than 20 SISCAPA-MRM assays have been constructed, including the first nine- and 10-plex assays.
In another project, members of CPTAC developed a yeast proteome as a performance standard [See PM 11/06/09]. According to the NCI-CPTC, the reference standard "will make it easier to standardize both current and emerging proteomic technologies and help ensure that only the highest quality data are being generated."
Another study resulted in the creation by researchers at the National Institute of Standards and Technology of more than 40 metrics for evaluating LC-MS system components. In addition to being a troubleshooting manual, the metrics also provide a comprehensive quality control profiler for LC-MS systems [See PM 11/06/09].
[ pagebreak ]
Lastly, CPTAC researchers published a study in November [See PM 11/20/09] on variability in LC-MS systems that suggested that "adhering to SOP-driven methods could effectively standardize LC-MS/MS proteomics across laboratories, optimizing biomarker discovery," the NCI-CPTC said.
In the annual report, the five CPTAC network teams also provided an update on their efforts. In addition to developing the MRM-SISCAPA assays, the annual report said the Broad Institute-led team had developed an optimized antibody reagent production and quality-control pipeline "to support creation of SISCAPA assays for candidate cancer biomarkers." Reagents have been generated for more than 100 CPTAC candidate proteins, and 16 other proteins from four separately funded projects.
Another CPTAC team, led by Memorial Sloan-Kettering Cancer Center in New York, is "evaluating and documenting whether serum peptide patterns, and/or the protease activities producing them" can be measured reproducibly, and what use, if any, they have for detecting cancer, making a prognosis, and differentiating clinically significant from and insignificant cancers.
The researchers also said that they have developed a test that can measure global exopeptidase activities "within individual proteomes of two or more groups of biological fluids." Using semi-automated MALDI-TOF mass spectrometry, the test offers advantages over standard peptidome measures in robustness, reproducibility, and quantitation, according to the NCI-CPTC.
The MSKCC team also is working on improving reproducibility and repeatability on the MALDI-TOF platform, and while "significant improvements" have been realized, "practical limitations on the selection of diagnostic peak patterns" exist. An assessment of the assay output patterns is under way, the annual report said.
A third CPTAC team, headed by researchers at Purdue University, is tackling reproducibility, and has developed robust protocols and standards for both electrospray ionization and MALDI mass spectrometry, "which include new or improved separation and/or enrichment systems."
A clinical proteomics data model for managing metadata from proteomics experiments was created, which "can be used to capture and manage experimental designs, experimental protocols, data storage parameters, search software, and annotation of result descriptions," the NCI-CPTC said. The system has been implemented as a customized application of the metadata framework in the Computational Proteomics Analysis System, it added.
In addition, the team has designed the Healthy Human Individual's Integrated Plasma Proteome Database as a way for researchers to search for plasma proteins gathered from different mass spec platforms. The database is to support "future clinical proteomics research, especially the discovery of biomarkers through plasma proteomics profiling," the annual report said.
At the University of California, San Francisco, CPTAC researchers are developing and implementing affinity-based workflows for the detection of proteins that "carry various cancer-related [post-translational modifications]" such as glycoproteins, phosphoproteins, and modifications indicative of oxidative damage, the NCI-CPTC said, adding the team is set to embark on a set of interlaboratory studies targeting glycopeptides.
The group also is researching protein biomarkers specific for metastasis-prone subtypes, which may have the largest impact on breast cancer survival, according to the annual report. The researchers have identified about 1,700 genes "that show strong evidence of alternative splicing across several breast cancer cell lines," and have shown that "basal subtype cell lines and breast cancer tissues express a glycosylated protein variant that may be involved in the metastatic process." Such biomarkers may eventually be used for the early detection of the disease, the NCI-CPTC said.
[ pagebreak ]
A fifth CPTAC team, led by Vanderbilt University researchers, has developed a statistical model for distinguishing disease-associated phenotype differences from shotgun proteomics datasets. Applying this new analytical platform to colon cancer samples, the group has performed successful proof-of-concept studies and is now using it to study tissues retrospectively from patients with stage 2 colon cancer. The goal is to characterize distinct biomarkers for recurrence of the cancer. "If successful, their work could help pave the way for the next generation of cancer diagnostic testing," according to the NCI-CPTC.
In addition, the researchers have successfully evaluated shotgun proteomic analysis of formalin-fixed paraffin embedded tissue. They have developed an integrated, open-source data analysis pipeline for shotgun proteomics "that exceeds the performance of other available tools," and developed a cost-effective method for targeted quantitation of hundreds of proteins, "thereby providing an effective means to configure specific, targeted assays for biomarker candidates rapidly."
In the Advanced Proteomic Platforms and Computational Sciences component, the NCI supports more than a dozen projects in which "investigators have made substantial achievements that are advancing" capabilities in proteomics, which allow the research community to better understand the differences between disease and healthy proteomes, the NCI-CPTC said.
Among the work being done is the development of a platform combining fast liquid chromatography separations, ion mobility spectrometry, and time-of-flight mass spectrometry. Richard Smith's lab at Battelle Pacific Northwest Laboratory is doing the work. Calling it a "next-generation cancer biomarker discovery and validation platform," the NCI-CPTC said it would allow measurements that are "more robust, more sensitive, have high throughput, and have improved quantitative utility" than current systems.
Also, Nathan Edwards at Georgetown University is developing informatics tools to "improve the characterization of alternative splicing, coding single nucleotide polymorphisms, [and] novel protein isoforms." The peptide sequence databases developed by Edwards and his collaborators have allowed for the detection of such splices and novel protein isoforms that otherwise would have been missed, according to the NCI-CPTC.
And Stephen Walton at Michigan State University is developing an aptamer-based strategy that provides better sensitivity and dynamic range than current array-based technologies. The strategy "ultimately provides the option of using oligonucleotide microarrays to quantify proteomic signatures," the annual report said. Walton and his team is developing methods in which aptamers that target specific proteins are each labeled with a "unique molecular barcode sequence serving as unique identifiers of specific aptamers."
They have developed assays for thrombin and PDGF-BB, and "according to measurements from scintillation counting and PCR, each pair of aptamers has a lower detection limit of 50 nM," NCI-CPTC said.
Lastly, as part of the Proteomic Reagents and Resources Core, the NCI-CPTC said that the Reagents Data Portal — a web-based service provide by NCI-Frederic to make reagents, standard operating procedures, and characterization data produced by the CPTC available to the research community — is expanding "as the program makes way for a great number [of] reagents in the pipeline that are needed for effective proteomic analysis."
Currently more than 25 antigens and 75 monoclonal antibodies have been generated against cancer-associated proteins.
Also, the NCI Clinical Proteomic Measurement Assessment Materials Program at NIST is developing and documenting the characterization of complex biological mixtures for use by the CPTAC network to evaluate proteomic analyses. The goal is to develop proteomic standard reference materials to provide them to the research community. So far, a yeast lysate and a peptide mixture have been approved for standard reference material development, the NCI-CPTC said.
Working with the NCI Office of Biorepositories and Biospecimen Research, CPTC also developed a biospecimen collection protocol, which the annual report said is the "first multicenter SOP generated by the program."