NEW YORK (360Dx) – A team led by researchers at the University of Massachusetts and University College London have developed a blood test for detecting liver fibrosis.
The test, which was detailed in a paper published last week in Advanced Materials, uses an array of fluorescent polymers to identify differences in serum proteins that can distinguish both between healthy patients and those with liver fibrosis and between patients with different degrees of fibrosis.
The assay has advantages over existing tests for fibrosis in that it can be easily miniaturized, can work with small sample volumes, and is stable at ambient conditions, making it well suited to low-cost point-of-care testing, especially in resource-constrained areas, said Vincent Rotello, professor of chemistry at UMass and senior author on the study.
The technology is, in theory, applicable to a range of other conditions, as well, he said, adding that he and his colleagues are now exploring its usefulness for detecting cardiovascular conditions.
The sensor array is built on a poly(oxanorborneneimide) (PONI) random copolymer scaffold, with the polymers containing benzoate monomers that allow for protein binding. The monomers also contain sites for fluorescent dyes to be attached. According to the authors, when proteins bind to the polymers, this causes changes to the ionic strength, pH, and supramolecular interactions of the dyes, which leads to changes in the intensity of the fluorescence emitted by the dyes. Different levels of serum protein binding therefore lead to different patterns of fluorescence, and these different patterns can be used to distinguish between cases and controls.
The approach is different from most protein biomarker research in that it is hypothesis-free, Rotello noted. While conventional proteomic experiments often start off with large unbiased searches comparing the levels of thousands of proteins, ultimately these efforts aim to identify a relatively small number of analytes that can be measured in a targeted way to detect disease or answer some other clinical question.
The polymer-based sensor developed by Rotello and his colleagues, on the other hand, doesn't rely on identification of the underlying analytes that lead to the changes in fluorescence patterns but simply uses those patterns to address the clinical question at hand — liver fibrosis, in the case of the Advanced Materials study. In this respect it is similar to tests like Biodesix's Veristrat, which relies on differences in MALDI spectra (as opposed to quantification of the underlying proteins) to guide cancer therapy, or Cleveland Diagnostics' IsoPSA test, which aims to improve prostate cancer diagnoses by looking not at levels of individual proteins or protein isoforms, but rather at bulk levels of protein isoforms associated with the disease.
Rotello said that despite not necessarily needing to know the underlying proteins that contribute to the test's performance, he and his collaborators are interested in identifying them.
"We tried to look at some of the obvious things that the sensor might be responding to, and we weren't able to figure it out," he said. "It seems like there is some real difference [between fibrosis patients and controls] that just isn't described well by the individual biomarkers. But we are interested in that, and that's something I'm certain the folks at UCL are going to be continuing to pursue."
Using a training set consisting of 50 patient samples (16 healthy controls, 17 with mild fibrosis, and 17 with severe fibrosis) and a test set of 15 samples (five healthy controls, five with mild fibrosis, and five with severe fibrosis), the researchers were able to distinguish between healthy controls and patients with fibrosis with 80 percent accuracy (sensitivity of 74 percent and specificity of 94 percent) and between mild and severe fibrosis cases with 60 percent accuracy.
Assuming the test's performance holds up in larger cohorts, Rotello suggested it could be useful as a simple and inexpensive screening test upfront of existing immunoassays like the enhanced liver fibrosis (ELF) test, which measures the serum analytes hyaluronic acid, PIIINP, and TIMP-1, and is the current gold standard among non-invasive tests for liver fibrosis.
Rotello said that in addition to setting up additional clinical studies of the test, his team is focused on the "widgetization" of the technology.
"Basically, we want to turn it into something that can be readily used both in the clinic and also at point of care," he said. "We're going for something that basically you can do with the same kind of thumb stick that you would do for measuring blood glucose levels."
He said that in the Advanced Materials study the researchers were able to run the test on blood samples equivalent in volume to a finger stick.
Rotello said the polymer sensor emerged from a roughly 15-year project in his lab focused on combining nanotechnology and materials science to "get selective recognition of biological systems and then take that selectivity and use it for [disease] sensing."
The key development, he said, was moving to a multichannel system featuring multiple polymers and dyes capable of emitting fluorescence in a ratiometric manner.
"We have done biosensing with nanoparticle systems [like gold nanoparticles] quite a bit, and also some sensing with polymers, but the polymers were all single-channel, and none of them had the same kind of sensitivity that this current family of polymers has," he said.
The sensor in the Advanced Materials study features four channels. "That means you get these four signals and it's the ratio of the four signals that gives you the output," Rotello said, adding that the fact that the channels are related to one another further improves the device's performance.
"If, for instance, you had four separate polymers in four separate wells, then you would get four pieces of unrelated data," he said. "In our system, all four are related, and that gives you better sensitivity to subtle changes because it cuts down on the noise of the system."
Rotello said he and his colleagues don't have immediate plans to commercialize the technology but will likely consider launching a company around it once they have developed a more clinic-ready version of the platform.