Skip to main content
Premium Trial:

Request an Annual Quote

Taking Aim at Clinical Oncology Market, Proteome Sciences Developing SRM-MS Assay for ER Status


By Adam Bonislawski

Snagging two new research collaborations, London-based proteomics firm Proteome Sciences is moving to apply its selected-reaction monitoring mass spectrometry platform to the cancer biomarker space and plans to have at least one such assay commercially available for research use by the end of the year.

The company has signed agreements in recent weeks with the Moffitt Cancer Center of Tampa, Fla. (PM 07/08/2011), and the Buck Institute for Research on Aging in Novato, Calif., with "the intention of building a portfolio of oncology-related assays primarily using selected-reaction monitoring as the detection platform," CEO Ian Pike told ProteoMonitor.

Proteome Sciences currently offers two SRM-MS assays for Alzheimer's research (PM 07/16/2010) – its AD-TMT-SRM assay, which measures levels of nine Alzheimer's biomarkers in serum, and an assay for measuring levels of phosphorylated tau protein in brain tissue – but it has yet to launch any assays aimed at cancer research.

The Moffitt agreement is focused on developing mass spec-based biomarker assays measuring signaling and repair pathway proteins for clinical tumor biopsy analysis. The company will be working primarily with John Koomen, scientific director of the center's proteomics core facility, whose research includes investigations into protein biomarkers for predicting patient outcomes in colon cancer as well as for directing therapy for multiple myeloma.

The Buck collaboration will focus on breast cancer – in particular on the development of mass spec assays measuring post-translational modifications in the protein biomarker estrogen receptor alpha.

While ERα – the target of anti-estrogen agents like tamoxifen and aromatase inhibitors – is a key breast cancer biomarker, its presence has proven only somewhat predictive of patient response to such therapies, noted Christopher Benz, a Buck Institute professor and leader of the collaboration with Proteome Sciences.

"It's long been known that [anti-estrogen therapies] are only effective in about maybe half the [breast cancer] cases that overexpress the estrogen receptor," he told ProteoMonitor. "And it's been the subject of intense investigation, but still not really fully understood, why many ER-positive [breast] cancers don't respond to these endocrine therapies."

Research into this question has indicated that crosstalk between ERα and other protein signaling pathways underlies patient response to anti-estrogen agents, Benz said. This, he noted, suggests that its post-translationally modified forms – its phosphorylated forms in particular – could be key to its effectiveness as a biomarker.

"The standard clinical assays that are done by pathology departments right now are antibody-based immunohistochemical assays" that simply measure total ERα expression, Benz said. The aim of the Proteome Sciences collaboration is to develop SRM-MS assays to measure levels of phosphorylated variants of the protein, which, he said, could be better predictors of how patients will respond to therapy.

Because the phosphorylation of different ERα phosphosites is done by different kinases from different signaling pathways, measuring levels of the protein's various phosphorylation patterns could let scientists determine which pathways are activated in ERα-positive tumors. This, Benz said, has led to considerable research aimed at unraveling which kinases are responsible for phosphorylating which ERα phosphosites, a question that he noted "still hasn't been fully resolved."

Thus far, Benz said, his team has identified one pattern of ERα phosphorylation that indicates the receptor is being activated by estrogen alone, another pattern that indicates it's being activated by other growth factors, and another pattern indicating it's being activated by both estrogen and other growth factors. Last year his lab received a five-year grant from the National Cancer Institute to continue research into decoding these phosphorylation patterns.

Antibodies exist for several ERα phosphorylations, but, Benz said, their sensitivity is relatively poor. This lack of good antibodies has led his team to take a mass-spec based approach to their research, through which they've recently achieved full coverage of the protein's phosphosites.

In a 2009 study published in Molecular & Cellular Proteomics, the Buck researchers used a combination of MALDI-TOF on an Applied Biosystems Voyager-DE STR Plus, MALDI-MS/MS on a Thermo Scientific vMALDI-LTQ ion trap, and SRM-MS on an AB Sciex 4000 QTRAP to identify a total of nine phosphorylated serine residues on ERα, bringing them, Benz said, to 95 percent coverage of the protein. More recently, he said, they achieved coverage of the remaining 5 percent of the protein.

"Putting all this together is going to allow us to decode" the relationship of ERα phosphorylation status to patient response to endocrine therapies, Benz said. "Proteome Sciences wants to leverage our basic approach and methodology and ultimately apply it to routine clinical samples."

Benz stressed, however, that there are considerable challenges standing between the work his group has done on ERα phosphorylation and applying it clinically. For instance, while the researchers have used SRM-MS to successfully measure ERα phosphorylation levels in MCF-7 cell lines, it's unclear how well the technique will work in actual tumor samples, which, Benz noted, typically have lower expression levels of the protein than do the cell lines.

Perhaps more significant, he said, are the sample collection and contamination challenges inherent in trying to measure labile modifications like phosphorylation.

"When you work with cell lines, they're relatively pure," Benz said. "When you take out human tumors you're already dealing with 50-50 mixes of normal cells of various origins mixed in with the tumor epithelium, so you have a lot of processing steps that can impact things, and when you're talking about post-translational modifications like phosphorylation you have to create conditions to preserve [the protein's] phosphorylation status."

Indeed, he said, studies using reverse-phase protein microarrays to examine phosphorylation in cancer signaling networks have found that even a 20-minute delay in preserving extracted tumor samples can cause the loss of some phosphorylation species.

"I don't think [mass spec] sensitivity is going to be the bottleneck of the future. Not with the technology improving as it is," he said. "But there's all this upfront stuff that needs to be worked out to optimize the entire approach."

According to Pike, Proteome Sciences hopes to have an initial assay for measuring ERα phosphorylation on the market by the end of this year. That will be a research-only product, but, he said, the ultimate goal is to develop clinical assays to predict likelihood of response to anti-estrogen therapies as well as ongoing monitoring of ERα phosphorylation status.

The company is also interested in possibly expanding beyond ERα and adding ErbB2/HER2 and progesterone receptor phosphorylation status to make a multiplex assay for breast cancer diagnosis, Pike said.

Last year Pike told ProteoMonitor that using SRM-MS to measure protein phosphorylation was of considerable interest to the firm, adding that it was working on assays to measure the effects of different kinase inhibitors on tumor cells (PM 11/19/2010).

This week, he noted that the company's interest in this area stemmed from his attendance in 2008 at the NCI's Clinical Proteomic Technologies for Cancer Initiative annual meeting where he "saw what was going on in the area of SRM-MS development at a very early stage" and realized that "it overlapped with what Proteome Sciences was doing at that time" in the neurological area.

The goal is to "develop some products that will transform the way tumor profiling is carried out in the laboratory," he said. "For us as a business opportunity, that's very important."

Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.