The Terry Fox Research Institute and the Canadian Partnership Against Cancer are providing C$5 million (US$4.9 million) in funding for a five-year effort to identify biomarkers for ovarian cancer.
The money will fund the compilation of roughly 2,000 samples from research centers across Canada. The samples will be used in validation studies of genomic and proteomic biomarkers, with an emphasis on markers useful in stratifying disease subtypes by their molecular pathology and predicting patient response to therapy.
"What we're really putting together is a validation platform," Anne-Marie Mes-Masson, an oncologist at the University of Montreal Hospital Research Centre and one of the leaders of the project, told ProteoMonitor. "We have throughout the world discovery labs that are busy identifying potential biomarkers for any number of clinical applications. We don't lack [potential] biomarkers. What we lack is a good platform to test them rapidly and rigorously and to [determine] which are the best ones to be pushed forward into the clinic."
Called the Ovarian Cancer Pan-Canadian Program, or COEUR, the initiative brings together 35 investigators from across the country, each of whom, Mes-Masson said, will contribute in various ways, including providing samples, proposing biomarkers for testing, and running assays.
The program also includes a study committee tasked with prioritizing which biomarkers should get access to the samples compiled by the group. The committee's primary criteria, Mes-Masson said, will be the amount and quality of the preliminary data suggesting a given marker's utility. Additionally, she noted, the committee will aim to avoid redundancy, "so if groups come together with the same biomarkers, we'll try to work with them together to make sure we do the assay only once."
Proteomics-based ovarian cancer tests like Vermillion's OVA1, Fujirebio's ROMA, and Correlogic's OvaCheck have been aimed at early diagnosis of the disease, but, said Mes-Masson, the COEUR project plans to concentrate more on biomarkers for companion diagnostic purposes rather than prediction or early detection.
"The focus will really be on stratifying the disease according to molecular pathology so that women can be better stratified in terms of clinical care," she said. "We know, for instance, that some [ovarian cancer] subtypes don't respond to first-line therapies. The other [goal] is to look at therapeutic resistance."
"We have [samples for] quite a wide range of all the steps of the disease," Diane Provencher, a gynecological oncologist at the University of Montreal Hospital Research Centre and leader on the project, told ProteoMonitor. "So the beauty is this is a large number of clinical samples that are annotated with a lot of clinical information that we'll [use to] try to differentiate all the heterogeneity of the disease."
The COEUR collection will be a retrospective cohort comprising fresh frozen tumor tissue and formalin-fixed paraffin-embedded tissue. The researchers also plan to build tissue microarrays "that will allow people through immunofluorescence or immunohistochemistry to test their biomarkers more rapidly and correlate them with the clinical parameters we have for the cohort," Mes-Masson said.
The researchers are currently in the process of putting together the samples, an undertaking that Mes-Masson predicted would take roughly two years. They plan to roll out the samples by disease subtype, meaning that some forms will be available before others. For instance, the platform for investigating biomarkers for high-grade serous carcinoma – one of the major forms of ovarian cancer – will be ready within the year. "Other, rarer, subtypes could take a bit longer to put together," she said.
In the meantime, the program is setting up workshops for discussing what biomarkers should be first in line for validation on the platform. Both Canadian and non-Canadian researchers can apply to test their markers on the samples, and, Mes-Masson said, while the cohort has not been put together to serve industry, "should industry have an important marker that needs to be tested, they would be considered by the study group like any other investigator that approached the group."
All results from the researchers will be combined in a database that will allow the COEUR study group and participants to do metanalysis work looking for combinations of biomarkers that might work well together, Mes-Masson said. "It's very likely that a single biomarker won't be sufficient and that a combination of biomarkers will give higher specificity or sensitivity for any kind of clinical testing we would want to do."
Recently, sample collection procedures have emerged as a significant concern for proteomics researchers, particularly with regard to highly labile post-translational modifications like phosphorylation, which may be lost or altered if a tissue sample isn't frozen immediately after being taken from a patient (PM 9/2/2011).
Mes-Masson acknowledged these concerns, noting that the COEUR samples might not be good for investigating phosphoproteomic markers. Aside from the COEUR project, she suggested that phosphorylation will be difficult to implement as a clinical biomarker generally.
"Phosphorylation could be a great biomarker, but it would be very difficult to implement in a hospital setting because getting the tissue out of the patient and frozen in 30 seconds is just not going to happen in that kind of setting," she said. "So we're looking for things that are robust enough that can be used in a clinical setting."
Phosphoproteomics aside, the COEUR project will benefit from sample-collection standards put into place in 2004 by the Canadian Institutes of Health Research, Mes-Masson said.
"The biobanks that are participating in the project have largely adhered to those standard operating procedures," she said. "So although they're not completely prescriptive, things like time from clamp[ing] to freezing are things that are at least documented" for the samples being used.
Mes-Masson said the program was aiming for samples that had been frozen within 30 minutes of collection but would accept longer lag times if they were documented and the samples passed a quality-control test.
According to Mes-Masson, COEUR aims to move any validated markers into the clinic as quickly as possible.
By the end of the five-year project, she said she hopes to see validated markers, if not in the clinic, at least "in prospective clinical trials and further validating what turn out to be the best markers we find in the first five years of the program."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.