University of North Carolina researchers have developed a chemical proteomics approach to measure global kinase activity and used it to investigate drug resistance in triple-negative breast cancer.
In the study, detailed in a paper in the current edition of Cell, the UNC team combined kinase capture via multiplexed inhibitor beads with MALDI TOF/TOF mass spec analysis to characterize kinome reprogramming in breast cancer in response to treatment with a targeted MEK inhibitor.
According to Gary Johnson, chair of pharmacology at the UNC School of Medicine and leader of the study, the technique allowed the researchers to identify key mechanisms in the development of MEK inhibitor resistance and predict drug combinations to overcome this problem.
Ultimately, Johnson told ProteoMonitor, the method could potentially be used to guide and monitor cancer therapy via personalized evaluations of patient kinomes.
Kinase inhibitors are a key area of drug development, with more than 130 such agents currently in clinical trials. Many of these drugs, however, become less effective over time, particularly when used singly, and so there is significant interest in developing combination therapies that make use of multiple inhibitors to cut off the alternate routes of kinase signaling that can lead to resistance.
Developing an improved understanding of cancer kinomes' responses to inhibition is key to this effort, Johnson said.
"The idea is that we can look at how a particular cancer type responds to a drug and how it reprograms its kinome," he said, noting that one of the primary surprises coming out of the study was just how extensive this reprogramming can be.
"The biggest surprise to me was how resilient the kinome was and how easily it bypassed single inhibition," Johnson said. "And this is true for virtually every single kinase inhibitor that's been developed today."
In the Cell study the researchers profiled kinome response to AstraZeneca's investigational MEK inhibitor AZD6244 in a triple-negative breast cancer cell line, mouse model, and human tumors. Based on these profiles, they predicted that a combination of AZD6244 with sorafenib – marketed by Bayer and Onyx Pharmaceuticals as Nexavar – would prove effective at overcoming resistance to MEK inhibition. They tested the combination in a C3Tag mouse model of triple-negative breast cancer and found that it led to tumor regression in 77 percent of the mice treated, compared to 30 percent for AZD6244 alone and zero percent for sorafenib alone.
"We rationally predicted a second drug to use in combination with the MEK inhibitor simply based on the kinome reprogramming profile that we saw," Johnson said. He added that his lab is now trying to develop similar profiles for other cancers and kinase inhibitors.
"We're in the process of trying to get [similar profiles] published for other very selective, targeted kinase inhibitors," he said. "We know that the kinase reprograms quite differently to different inhibitors, and so you can't just say that the kinase responds this way [in all circumstances.] It's going to respond uniquely for that tumor type and that specific inhibitor."
Johnson and his colleagues are now beginning a "window trial" analyzing kinome reprogramming in triple-negative breast cancer in response to an undisclosed drug, he said.
In the trial, patients will undergo a core biopsy of their tumor, which the researchers will analyze. The patients will then be put on a MEK inhibitor for a week to 10 days and then will have the primary tumor excised, which the researchers will also analyze and compare to the core biopsy taken pre-treatment to identify any kinome reprogramming that has taken place in response to the drug.
A second, similar trial is currently in the planning stages, Johnson noted. This one will also be in breast cancer, but, he said "there's interest in [applying the technique] to many other cancers."
George Mason University researcher Emanuel Petricoin – whose research focuses in large part on using phosphoproteomics to analyze cell signaling and drug response – called the paper "very elegant," noting that the UNC scientists "were able to come up with some really interesting findings that they were then able to validate."
However, he told ProteoMonitor, the amount of material typically required by such mass spec-based approaches could prove problematic clinically.
"You could argue that this is a great approach for preclinical cell line [work] where you're doing your drug screening to try to figure out which combinations might be effective," he said, but the difficulty of obtaining large quantities of tumor material in a clinical setting might limit its usefulness for such applications.
The study's workflow uses a range of targeted and pan-kinase inhibitors immobilized on beads, which capture the activated kinases present in the cell or tumor being investigated, allowing for their quantification via mass spec.
The technique, Johnson said, is similar to chemo-proteomic methods developed by companies including Cellzome and Kinaxo, but, he said, he and his colleagues had expanded its range, enabling capture of 60 percent to 70 percent of the kinome.
They did this, Johnson said, in part by expanding the range of inhibitors used as well as by optimizing the layering of these inhibitors within the columns to maximize coverage.
"We figured out how to layer the beads so that we put certain inhibitors [for abundant kinases] at the top and then used the pan-kinase inhibitors at the bottom of the column," he said. "What this does is absorb some of the higher abundance kinases and that allows you to capture more of the lower expressed kinases so that you expand your coverage."
Johnson added that his lab was currently working to expand the technique's kinome coverage, noting that his goal was to get it into the 90 percent range.
For the mass spec portion of their workflow the UNC scientists opted for MALDI TOF/TOF on an AB SCIEX 5800 machine – also a contrast with other chemo-proteomic approaches, which have typically relied on electrospray mass spec.
MALDI, which they combined with iTRAQ isobaric labeling, was appealing particularly for its multiplexing capability, Johnson said. "We needed to be able to do time courses and drug responses and measure certain aspects of dose responses and drug combinations … and you're somewhat limited in how you can do that with ESI."
Although MALDI for years had a reputation as being poorly suited to quantitative and clinical work due to high variability, the technology has made a comeback recently as a potential platform for such applications as researchers have come more and more to appreciate its ease and throughput (PM 1/20/2012).
"In terms of protein ID and high-throughput quantitation with the different isobaric tags that are available to do this, I think MALDI TOF/TOF is going to be very attractive to people," Johnson said.
According to Johnson, the researchers have patented the technique and are considering commercializing it, but, he said, he would be more interested in licensing it to an existing firm than launching a new company around it.
He said that the technology would likely be most effective as part of a service company, "and I'm not interested right now in starting a service company." He added that he has had conversations about the method with pharma companies including Novartis, Johnson & Johnson, and GlaxoSmithKline.
"GSK actually approached me a couple of years ago and wanted me to screen their whole library of kinase inhibitors [using the technique]," Johnson said. However, he said, no agreement was ever reached for that project.
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.