NEW YORK (360Dx) – MRM Proteomics is aiming to push its immuno-MALDI (iMALDI) mass spec assay into the clinic through its collaboration with Canadian research network Exactis Innovation.
Under the terms of the collaboration, announced last week, Exactis will provide Montreal-based MRM with 100 well-characterized clinical samples from breast and colorectal cancer patients, and MRM will measure levels of protein expression and signaling to test whether such measurements could help guide treatment with immunotherapies and other targeted treatments.
The effort builds on work led by Christoph Borchers, MRM's chief science officer and head of the University of Victoria-Genome BC Proteomics Centre, to develop iMALDI assays to measure proteins and phosphoproteins involved in the PI3K signaling pathway, which has been linked to a variety of cancers.
Borchers and his colleagues have developed iMALDI assays to measure the expression and phosphorylation levels of the PI3K, PTEN, AKT1, and AKT2 proteins. In collaboration with Exactis, they now plan to develop iMALDI assays to a number of other proteins in the pathway.
Exactis is a nonprofit focused on recruiting cancer patients to clinical trials for targeted therapies. Patients recruited through the Personalize My Treatment registry agree to provide detailed molecular information about their cancer for storage in a digital biobank and to be contacted for future clinical trials that are a match for their characteristics.
"Hopefully [the 100 samples] will be enough to get some reasonable indications as to how useful this [protein and phosphoprotein] information about the expression and phosphorylation levels is and how it can be used to stratify patients [for treatment]," he said. Assuming the measurements do provide useful information for guiding patient treatment, MRM and Exactis might then expand their work to a larger validation study, Borchers added.
While precision medicine efforts have largely focused on using genomic information for targeting treatments, proteomics researchers have argued that protein-level data could ultimately prove more useful given that they are the molecules responsible for carrying out cellular processes and are typically the actual targets of drugs.
Research has also indicated that not just protein expression but also protein activation, as measured by phosphorylation levels, can be important in determining whether a patient is a good candidate for a particular therapy. For instance, last month researchers with the I-SPY 2 TRIAL published a study finding that triple-negative breast cancer patients who would not be candidates for treatment with the kinase inhibitor (Nerlynx) neratinib based on their protein expression levels, showed a strong response to treatment when they had elevated phosphorylation levels of two key cancer signaling proteins.
The PI3K iMALDI effort stems from work Borchers and his colleagues have been doing for drugmaker AstraZeneca, which is developing an AKT inhibitor as well as inhibitors targeting other members of the PI3K pathway.
AKT, Borchers said, is a good example of a protein target where genomic data is not particularly informative.
"AKT has only one main mutation, and it only occurs in very small percentage of [cancer] patients," he said. "So stratifying patients solely based on genomics is not very attractive in that case. So for that drug what we would like to know is very simple — what is the amount of these proteins in the pathway, and their functionality [according to their phosphorylation levels], and hopefully when we know this we will be able to identify the best responders to this therapy."
Antibody-based methods like immunohistochemistry are commonly used to measure protein expression in cancer samples, but IHC approaches are not as quantitative as mass spec and more limited in terms of the number of analytes they can measure.
On the other hand, mass spec, particular for proteomic measurements, has had difficulty making inroads to clinical applications, due in part to challenges around throughput and the expertise required to run such assays.
This has led Borchers and other researchers to explore MALDI for clinical applications. While not as sensitive as high-performance LC-MS methods, MALDI can achieve very high throughput due to the fact that it doesn't require upfront LC separations, and MALDI workflows are relatively simple, meaning they can potentially be run by clinicians with little mass spec expertise.
Additionally, Borchers and his colleagues have thus far done the PI3K assay development work on Bruker's Microflex instrument, which is used in the company's MALDI Biotyper clinical microbiology platform. Though not the highest performance MALDI system, the platform has won broad uptake from hospital and other clinical labs, which positions it well for additional clinical proteomics applications.
The iMALDI approach boosts the mass spec's sensitivity by first enriching target peptides via antibodies. The MRM researchers also use a method for measuring protein phosphorylation levels that he said can provide greater sensitivity.
Typically, researchers measure protein phosphorylation directly. However, Borchers noted that phosphorylated peptides are negatively charged, which makes them difficult to measure when the mass spec is operating in positive ion mode.
To improve the phosphorylation measurements of their PI3K assays, the researchers took an indirect approach, first measuring the amount of their target peptides that are not phosphorylated, then treating the sample with a phosphatase to remove any phosphorylations, and then measuring the sample again. The increase in non-phosphorylated peptides detected between the two measurements is the amount of phosphorylation that was initially present on the target peptides.
Borchers said he and his colleagues have not published on this approach yet but that they have been using it to measure phosphorylation levels in AKT1 and AKT2 and found that it improves the sensitivity of the assay compared to a direct approach.
Borchers said he expected the development of the additional PI3K pathway assays to take around two years, after which the researchers will use them to profile the 100 clinical samples provided by Exactis.
"Then, hopefully, we will see trends where the [assay] can predict the best responders, and then from there we will go into a validation study," he said.