Protein biomarker firm Applied Proteomics has decided to use multiple-reaction mass spectrometry as the platform for its clinical proteomic tests, John Blume, the company's chief science officer, told ProteoMonitor this week.
The decision marks a switch from the firm's original plan to convert its tests to an immunoassay platform for commercialization and serves as another example of growing industry confidence in mass spec as a viable clinical technology.
As part of the new development plan, API inked an agreement earlier this month with Agilent to collaborate on clinical MRM mass spec assays and workflows. According to Blume, API aims this year to begin large-scale validation of its lead product – a protein panel for identifying patients with colon polyps or adenomas – with the goal of launching it commercially within the next few years.
Launched in 2007 by David Agus, director of the University of Southern California's Center for Applied Molecular Medicine, and Danny Hillis, chairman and co-founder of technology firm Applied Minds, API has been committed to mass spec for biomarker discovery but had until recently planned to offer its tests clinically in immunoassay format (PM 2/10/2012).
"We always used to say in our pitch decks that we would use this ... [mass spec-based] discovery engine to go find content, and then there would be a slide at the end that would basically say, 'And then we're going to turn it into a Luminex assay,'" Blume said.
"But it turns out that mass spec is actually a very good clinical test platform, particularly if you want to multiplex higher numbers of proteins," he said. "So we threw the Luminex slide out of the deck, essentially, and we are [now] developing laboratory-developed tests in a CLIA lab on mass spec." The company is in the process of leasing a space in which to construct its CLIA facility, he added.
API's move fits within a larger shift in clinical proteomics toward mass spectrometry, one that has picked up momentum in recent months as a number of academic and industry players – ranging from vendors to diagnostic developers – have published studies, released products, and entered collaborations suggesting that mass spec workflows have at last suitability as a clinical platform (PM 2/15/2013).
Among mass spec companies, Agilent has been at the forefront of this shift, inking deals with proteomics firms including Integrated Diagnostics, Sera Prognostics, and now API to develop and implement MRM-MS workflows suitable for clinical testing.
During Agilent's analyst day presentation this month, Nick Roelofs, president of its Life Sciences Group, noted that the company believes "there is a real opportunity for mass spectrometry technologies to penetrate the clinic," adding that the firm was positioning itself to take advantage of what it expected would be significant activity in the realm of laboratory-developed proteomic tests like those under development at InDi, Sera, and API.
Under the API-Agilent collaboration, the two companies will develop a commercial version of API's colon polyp/adenoma panel using Agilent's 6490 triple quadrupole, 1290 Infinity LC, RapidFire 360 MS, and Bravo liquid handling systems.
A primary area of focus for the collaboration will be upping the multiplexing capabilities of the RapidFire system, a high-throughput chromatography system that enables processing and injection of samples into mass spectrometers at rates as fast as one per every seven seconds.
The RapidFire is key to obtaining the sort of throughput needed for clinical implementation, Blume said, but, he noted, it is currently unable to process the large number of proteins per injection that API requires for its colon panel, which will likely consist of between 10 and 20 proteins.
"Right now we're working directly with [Agilent] on getting that workflow up to higher numbers," he said.
Agilent researchers have managed to perform as many as 20 MRM assays in a single RapidFire injection and believe they could theoretically perform close to 30 MRMs in one injection, Can Ozbal, director of Agilent's RapidFire business, told ProteoMonitor.
"So 30 MRMs could mean 30 peptides; it could mean 15 peptides and 15 internal standards," he said. "Its kind of how you divvy up those MRM transitions per experiment."
Given that API's colon panel could consist of up to 20 proteins, however, Ozbal said it was likely that running it would require multiple RapidFire injections per sample.
The company is "trying to multiplex 20 proteins into a single assay, and I can tell you right now that's not going to work in a single injection," he said. "The question is, will it take two injections, or three, or four?"
At a certain point, Ozbal noted, it could make more sense to simply run the panel using HPLC, but making it work on the RapidFire system would result in significant throughput improvements, bringing sample cycle times down from a matter of minutes to a matter of seconds.
"We know that this will work on an HPLC, and that will be kind of the fallback position," he said. "So then the other leg of the collaboration is to see what sort of throughput enhancement, cost decrease, economic impact, et cetera can the RapidFire have. If we can do it in one injection, it will be tremendous. If we can do it in two or three, it will be good. And at some point [if higher numbers of injections are required] maybe it doesn't make a lot of sense."
Ozbal noted that API was somewhat unique among its clinical proteomics collaborators in wanting to run such large panels of proteins. "For most of our collaborations, a single [RapidFire] injection will suffice just fine," he said.
API is positioning its colon panel as a test for driving patients to follow guidelines for colon cancer screening, Blume said. The test identifies patients likely to have colon polyps or adenomas with the idea being that such molecular evidence will encourage them to undergo recommended colonoscopies.
"Roughly half the people in the US who should have colonoscopies don't," he said. "So we decided that if we could provide a molecular tool to physicians to help them manage their patients to get colonoscopies, that could be quite useful."
The aim of the API panel, Blume said, is that "as part of your standard physical, you have a simple blood test, and now your doctor is able to say, 'Well, you have a four-fold chance over background that you actually have a polyp or adenoma – you really should have a colonoscopy.'"
He estimated the potential US market for the test to be around 14 million people per year.
In a 100-patient 10-fold cross-validation study presented at last month's Mass Spectrometry's Applications to the Clinical Lab annual meeting, the API panel identified patients with colon polyps or adenomas with an area under the curve of 0.92.
Blume said API now plans to validate those results in larger cohorts and has collected more than 1,000 samples for that purpose. He said the company aims to start large-scale validation efforts by the end of this year or early next year.
While API plans to launch the colon panel in the next few years, Blume said that he envisions the sample collection and validation process is one that will continue indefinitely.
"I think that many of the companies in this space underestimate the number of samples they need to collect," he said. "We just assume that we are going to collect samples basically forever. We see that a natural part of launching a test in this space is that you have to continually support it by demonstrating all the places it does work and guarding against the places it doesn't."
Blume said that the company might eventually go down the in vitro diagnostic route with the test, taking it through the US Food and Drug Administration, but, he added, API is confident it can make the panel commercially successful operating it as an LDT.
API currently has 25 employees, a number that Blume said will roughly double as it continues development of its CLIA lab and commercialization of the colon test.
Last February the company raised $22.5 million in a Series B financing round, adding to $4 million raised in a 2007 round.