Researchers at Arizona State University's Biodesign Institute are collaborating with Pfizer on a four-year, $5 million National Institutes of Health-funded project to identify and verify protein biomarkers linked to cardiovascular disease in diabetes patients.
The project will be led by Randall Nelson, director of the institute's Molecular Biosignatures Analysis Unit, and builds on a previous $3.1 million NIH award he received last year for heart disease and diabetes marker research (PM 5/7/2010).
"It's a natural segue" between the two initiatives, Nelson told ProteoMonitor this week. The initial grant, he said, focused on the development and verification of cardiovascular biomarkers that the current project aims to move into applied areas like drug development.
Specifically, Nelson's team is working with Pfizer to integrate the biomarkers into the company's diabetes research program. In 2007 Cleveland Clinic cardiologist Steve Nissen published a study in the New England Journal of Medicine linking the use of GlaxoSmithKline's diabetes drug Avandia to an increased risk of CVD. That link has led the US Food & Drug Administration to require pharma firms to monitor the cardiovascular effects of their diabetes therapies, raising the need for CVD biomarkers in diabetes drug development.
"FDA decided that all new diabetes drug approvals need to have long-term studies to monitor adverse cardiovascular effects," Jim Pearson, Personalized Medicine Directed-Proteomics lead at Pfizer, told ProteoMonitor. "So essentially they were telling drug companies, 'You can't show us data from just 12 weeks. You have to show us data from two or three years, and you have to include some sick patients in those studies.'"
However, "the problem is, there's no given panel to measure adverse cardiovascular risk," he said. "So how do you do these long-term cardiovascular studies? How do you know what to look for, when there's really no accepted panel out there? [Long-term trials] are going to be very expensive, and before we start those trials we want to have a good idea of what to put in the trial."
Nelson noted that the traditional marker of efficacy for diabetes drugs – lowering blood glucose – "doesn't really tell you much about the cardiovascular outcome." In light of that, "you've got to build panels of markers that tell you that you're getting the effect of the glucose-lowering drug while not increasing cardiovascular disease."
Nelson is investigating the CVD biomarkers using the mass spectrometry immunoassay, or MSIA, platform developed in his lab. MSIA relies on a patented pipette immunoenrichment technology that uses a high-throughput, high-binding-capacity microcolumn activated with antibodies to isolate low-abundance proteins in complex samples. The isolated proteins are then analyzed via selected-reaction monitoring mass spec, which enables quantitation of prognostically and diagnostically useful protein variants.
The ASU-Pfizer collaboration stems from a meeting between Nelson and Pearson several years ago at an American Society of Mass Spectrometry annual meeting.
"[Nelson] was telling me about all the applications [of the MSIA technology], and I told him that if you want to get into pharma you have to pick one application," Pearson said. "I thought the best application would be in diabetes, so we had [Nelson] run a certain assay in our diabetes program [using the MSIA technology], and what I was impressed with was that it took minimal sample volume for multiple assays; it was high-throughput; and he ran 400 samples and we got data on every sample – so there was a zero assay failure rate."
While ASU brings the MSIA technology to the arrangement, Pfizer is able to supply the large, well-characterized clinical sample sets needed for biomarker validation, including longitudinal sets following patients over a period of years, Nelson said, noting that the researchers will be testing their panel in samples from roughly 1,000 patients.
"Pfizer has a lot of clinical samples," Pearson said. "All drug companies have a lot of clinical samples. But the problem is getting the academics and the drug companies working together."
NIH, he said, served as a "matchmaker" for the two.
The Biodesign Institute's CVD panel is currently based on roughly 140 different protein species arising from 11 different gene products, Nelson said. He noted that his team will continue to add markers as they refine the panel.
"The products from those 11 genes turn into about 140 different molecular species because of post-translational modifications," he said. "So as we apply those 11 assays, we get about 140 pieces of data, and those 140 pieces of data tell you where you are in this merger of cardiovascular disease and diabetes. That's the original panel we're taking [into the new initiative]. A year from now it might be 12 or 15 targets for assays."
Provided their efforts at integrating the panel into drug development are successful, the researchers then hope to use it for "more general population screening." Nelson said, although, he added, such an application is still probably ten years off.
In terms of the panel's use in diabetes drug development, Pearson said it will likely be used in conjunction with markers from other studies like the Framingham Heart Study.
"To really get a good ROC area under the curve for a biomarker panel, I feel it's going to take [data] from several –omics and different modalities," he said. "One of the things we don't want to do is work in a silo and say all biomarkers have to be from a certain discipline, whether it's DNA, proteomics, metabolomics."
According to Nelson, ASU has patents on the protein markers and signatures in the current panel. He said that if Pfizer "wanted to implement [the panel] we'd certainly work with them to do that," but added that while the drug firm is an obvious partner right now, he would be open to collaborations with other companies in the future.
Nelson commercialized the MSIA technology being used in the project through the firm Intrinsic Bioprobes, which was acquired in September by Thermo Fisher Scientific (PM 9/2/2011). Thermo Fisher isn't a collaborator on the ASU-Pfizer project, but, Pearson said, from pharma's perspective, the acquisition increases the technology's appeal.
"From a pharmaceutical point of view I think it's good because Thermo Fisher is going to have more resources to put into their applications," he said. "If we go into Phase III clinical trials, we'll have thousands of samples, and I would rather give samples to Thermo Fisher than a little shop."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.