This story originally ran on Aug. 5.
CytoScale Diagnostics, a biotech startup formed in April by recent graduates of the University of California, Los Angeles, MBA program, has licensed a chip-based microfluidic image-cytometry platform and accompanying brain tumor biomarker assay from UCLA researchers that it plans to commercialize by the end of 2013.
The UCLA team described the use of the MIC-based platform for the simultaneous, single-cell quantitation of four signaling proteins associated with brain tumors in a study published in this month's issue of Cancer Research.
The platform can detect proteins in samples as small as 1,000 cells, potentially offering an approach for protein biomarker discovery and diagnostic work using highly limited tissue sources, Thomas Graeber, assistant professor of molecular and medical pharmacology at UCLA and one of the study's authors, told ProteoMonitor.
Diagnosing solid tumors is traditionally done via tissue immunohistochemistry, which offers only limited quantitative data. Flow cytometry, on the other hand, is able to provide quantitative data on multiple proteins in individual cells, but it requires a large amount of sample, making it impractical to analyze certain cancer types.
To solve this problem, the UCLA team turned to MIC, which combines the quantitative, multiparametric capabilities of flow cytometry with the low sample demands of a microfluidic format.
The device they developed, which Graeber said is automated to ensure reproducibility, consists of a 24-cell array chip fabricated by the attachment of a polydimethylsiloxane-based microfluidic component to a glass slide. Solid tumors are prepared as single-cell suspensions and introduced to the microfluidic component, where they are fixed on the slide and stained using fluorescently labeled antibodies for proteins of interest. After staining, the cells are examined and the proteins quantitated via a fluorescent microscope.
In the Cancer Research study, the researchers used the platform to determine the levels of the signaling proteins EGFR, PTEN, phosphor-Akt, and phosphor-S6 in individual cells from 19 human brain tumor biopsies.
The ability to quantify multiple proteins on a single-cell level allows researchers to examine subpopulations of tumor cells that might go unnoticed using conventional IHC techniques, Graeber said. This could help clinicians better predict disease progression and drug response, he added.
"This is targeted towards the idea of personalized medicine," he said. "When you stain for multiple markers on a single-cell level you might be able to identify subpopulations [within a tumor] that have a particular characteristic — high in one marker or low in another marker — where if you had averaged all the cells together you might not have noticed certain subpopulations as they got blended in with the other cells."
He added that a certain tumor type today "may be understood as five subtypes in the future where each of those subtypes is understood by what kind of molecular aberrations are driving the cancer in each case."
This sort of single-cell, multiparametric analysis generates a considerable amount of data, Graeber noted, requiring complementary bioinformatic approaches to render it clinically meaningful.
"How do you take measurements from 3,000 cells times four different markers and turn that into something you can interpret? We needed to develop bioinformatic processing for that data to reduce the dimensionality and make it more interpretable," he said.
To that end, the researchers applied a series of different bioinformatic techniques including self-organizing maps and hierarchical clustering to group the tumors based on their cell populations. Based on this work they were able to divide the 19 tumors into three clusters that correlated with patient survival and tumor progression.
"We were able to see what I would call preliminary differences," Graeber said. "It sets an indication that it's worth pursuing this further. It's a preliminary result that shows the relationship with a clinical outcome, and now we want to pursue it further."
Graeber said the researchers hoped to add more proteins to the panel and add samples to the study, testing the four-protein panel on additional, independent sets of patients to "find out how much information we can gather using this approach, and then coordinate that with clinicians to guide [clinical] development to really start to aid the diagnosis of tumors."
CytoScale Diagnostics signed a letter of agreement this May with the Regents of the University of California for exclusive licenses to the MIC platform and brain tumor biomarker assay. The company plans to sign a final licensing agreement for the technology by the end of this month, chief technology officer Kumar Duraiswamy told ProteoMonitor.
The licensing agreement also includes rights to a platform for collecting circulating tumor cells developed by another of the Cancer Research study's authors, UCLA associate professor of molecular and medical pharmacology Hsian-Rong Tseng. Graeber suggested his team's MIC device, with its small sample size requirements, is an obvious complement to this platform.
Researchers, he said, could couple the circulating tumor cell collection device with the MIC technology, allowing them to "all on a microfluidics platform collect those rare cells and then use [MIC] as a module for the characterization of the cells [they] collect."
According to Duraiswamy, CytoScale has no immediate plans to combine the two technologies into a single device, but will instead focus on developing them separately for commercialization — a decision he said is based on regulatory concerns.
The platforms "go hand-in-hand," he said, "but regulatory approval for a particular finite application is more doable than trying to get an entire platform."
Currently the company is working to develop the MIC-based brain tumor biomarker diagnostic for clinical use. CytoScale CEO David Franklin told ProteoMonitor that it hopes to have the test on the market by the end of 2013.
In the meantime, the company is also targeting the MIC and tumor cell collection platforms toward research laboratories that might be interested in them for non-clinical uses.
"It's a dual track," Franklin said. "Our end goal is clinical products, but we anticipate research collaborations, both academic and corporate, to begin within the next six months."