A new frontier in the fight against colorectal cancer might soon be on the clinical horizon, as Ohio State University scientists develop a machine learning tool capable of distinguishing the molecular profiles of those with the disease from healthy individuals, according to a recent study published in iMetaOmics through Ohio State News. Co-senior author Jiangjiang Zhu, an associate professor of human sciences, suggests the tool could be used for noninvasive diagnosis and disease monitoring, a significant leap forward in patient care.
By tapping into over a thousand biological samples collected from people afflicted with colorectal cancer as well as healthy controls the research team has identified metabolic shifts connected to disease severity and genetic mutations key to colorectal cancer risk, with their "biomarker discovery pipeline" pooling strengths from partial least squares-discriminant analysis and an artificial neural network culminating in what's dubbed as the PANDA platform, per the study details. Zhu told Ohio State News, "We believe this is a good tool for disease diagnostics and monitoring, especially because metabolic-based biomarker analysis could also be utilized to monitor treatment effectiveness."
Yet, Zhu clarifies, this algorithmic advancement isn't poised to kick colonoscopies to the curb; rather, it could augment the current standards, providing faster, potentially life-saving insights into how patients respond to treatment options. A myriad of metabolic and transcript data forms the backbone of this analytical approach, with the samples sourced from extensive research initiatives such as The Ohio Colorectal Cancer Prevention Initiative and a clinical laboratory biobank from the Ohio State Wexner Medical Center, ensuring a diverse and meaningful dataset for their work.
The possible biomarkers spotlighted in the research are linked to purines, compounds vital for DNA processes and notably more active in the bloodstream of those with cancer versus non-cancer patients, with such activity dropping off as cancer advances, these are not definitive indicators yet they shine a light on the intricate dance between metabolism and cancer mechanisms. The research, funded by the National Institute of General Medical Sciences and other sources, continues as Zhu's team refines PANDA and examines metabolic shifts associated with a spectrum of biological signals, aiming to elevate the capacity for early detection and personalized medicine strategies in the near future.
