Blood Fat Analysis Eyed for Cancer Screening
A new pilot study is exploring whether analyzing the full set of fats in blood — known as lipidomic profiling — could be a viable screening method for pancreatic cancer. This approach leverages big data and bioinformatics to find subtle patterns that might indicate early-stage disease, representing a shift toward data-intensive diagnostics.
Pancreatic cancer is notoriously difficult to detect early, as it often shows no specific symptoms until it's advanced. Currently, there is no general screening test; imaging like MRIs or endoscopic ultrasounds are only recommended for high-risk individuals, such as those with a strong family history or specific genetic mutations. The existing blood test for pancreatic cancer tracks a protein called CA 19-9, but it isn't reliable enough for early-stage screening due to low sensitivity and specificity. This limitation drives the search for new methods like lipidomics that can find clear, early signals of disease in the blood. The pilot study on lipidomics was led by a team of analytical chemists under Professor Michal Holčapek at the University of Pardubice in the Czech Republic. Their results showed the blood fat analysis could distinguish patients with pancreatic ductal adenocarcinoma (PDAC) from healthy individuals with over 95% accuracy, even detecting early-stage cases. A spin-off company, Lipidica, is now conducting a larger multicenter clinical trial. This research is a prime example of collaboration between lab scientists and computational biologists. A "wet lab" researcher would perform the mass spectrometry to generate the raw lipid data from blood samples. Then, a bioinformatician or computational biologist takes over, using programming languages like Python or R to clean the massive dataset, run statistical analyses, and build predictive models to find the patterns that signal cancer. A career in bioinformatics for this type of project involves less patient interaction and more "behind-the-computer" work, focusing on algorithms and data analysis to uncover biological insights. In contrast, a clinical researcher on the same project would be patient-facing, managing the trial, enrolling high-risk participants, collecting samples, and ensuring patient safety and data integrity throughout the study. The lipidomic test identified specific dysregulations in very long-chain sphingomyelins and ceramides in patients with pancreatic cancer. The study's reported sensitivity is approximately 30% higher than the CA 19-9 marker, a significant potential improvement for catching the disease sooner. This data-intensive approach is not just for cancer. Lipidomic profiling has also shown it can predict the risk of developing type 2 diabetes and cardiovascular disease years before the onset of symptoms, showcasing the broad potential of using metabolic data in preventative medicine.
