The reductionist approach to cancer detection—scrutinizing a particular set of biomarkers for each cancer—is fraught with obstacles. It may not be possible, in many instances, to rely on just a few biomarkers, given the complexity of the body’s response to disease. Worse, many biomarkers are elusive, so rare as to be all but undetectable, particularly after dilution in the bloodstream. Finally, in a typical reductionist scheme, the biomarkers that diagnosticians would have to follow would change from cancer to cancer, likely necessitating multiple detection platforms—hardly a convenience.
Is there a better way? Yes, say researchers at Arizona State University’s Biodesign Institute. These researchers, led by Phillip Stafford, Ph.D., advocate an approach they call immunosignaturing. It could, they say, enable early detection—always crucial in cancer treatment—and deliver universality. That is, a single platform could be used for any cancer.
Immunosignaturing relies on a multiplexed system that profiles, in snapshot fashion, the entire population of antibodies circulating in blood. This approach can leverage the response of antibodies to disease-related changes, as well as the inherent signal amplification associated with antigen-simulated B-cell proliferation.
In an article entitled, “Immunosignature system for diagnosis of cancer,” the researchers describe how their system can meet at least one of their self-imposed requirements—universality. This article, which appeared July 14 in the Proceedings of the National Academy of Sciences, promises that the other self-imposed requirement—early detection—will be addressed “separately.”
“To perform an immunosignature assay, the antibodies in diluted blood are incubated with a microarray of thousands of random sequence peptides,” write the authors. “The pattern of binding to these peptides is the immunosignature. Because the peptide sequences are completely random, the assay is effectively disease-agnostic, potentially providing a comprehensive diagnostic on multiple diseases simultaneously.”
When the authors refer to random sequences of peptides, they refer only to the amino acid sequences of individual peptides, not the placement of a peptide of a given sequence at a particular location in the array. That is, from one chip to the next, the peptide-to-peptide sequence is not random, and so binding at any particular location on a chip may be taken as likely being due to a particular antibody—even if the actual identity of the antibody is unknown.
“To explore the ability of an immunosignature to detect and identify multiple diseases simultaneously, 20 samples from each of five cancer cohorts collected from multiple sites and 20 noncancer samples (120 total) were used as a training set to develop a reference immunosignature,” the authors continue. “A blinded evaluation of 120 blinded samples covering the same diseases gave 95% classification accuracy.”
To further assess the diagnostic power of immunosignaturing, the researchers examined over 1,500 historical samples that comprised 14 different diseases, including 12 cancers. This part of the study consisted of a training phase, which used 75% of the samples, and blind-test phase, which used the remaining 25%. According to the researchers, an average diagnostic accuracy of over 98% was achieved.
Curious to learn more about immunosignaturing, GEN approached Dr. Stafford with a few questions.
The microarray chip used for the current study contains 10,000 imprinted peptides. To enhance the sensitivity and accuracy of immunosignaturing, Dr. Stafford’s group is currently developing a peptide imprinted with more than 100,000 peptides.