Virginia Tech researchers John Janiga, Dzenis Mahmutovic (seated), Katherine Brown, and lead investigator Carla Finkielstein work collectively within the Molecular Diagnostics Laboratory on the Fralin Biomedical Analysis Institute at VTC to develop a take a look at that would assist medical doctors establish most cancers sufferers in danger for extreme chemotherapy unintended effects. Credit score: Clayton Metz/Virginia Tech
Early outcomes from a brand new examine recommend {that a} easy saliva take a look at and highly effective synthetic intelligence algorithms might assist medical doctors establish most cancers sufferers at excessive danger for extreme unintended effects from a broadly used chemotherapy drug.
The drug, 5-fluorouracil (5-FU), has been in use because the Fifties—making it one of many oldest chemotherapy brokers nonetheless used in the present day. It stays a cornerstone of most cancers remedy, utilized in practically a 3rd of chemotherapy therapies for colorectal, breast, head and neck, pancreatic, and abdomen cancers.
5-FU disrupts how cells make and use DNA. Most cancers cells divide quickly and want to repeat their DNA constantly. The drug mimics DNA constructing blocks and jams the method, halting most cancers cell development.
However some sufferers carry genetic variations that forestall their our bodies from breaking down the drug correctly. In these instances, 5-FU can construct as much as poisonous ranges, inflicting severe and typically life-threatening issues.
About one in three folks has a diminished skill to course of 5-FU. Though the federal Meals and Drug Administration recommends screening, it’s not broadly applied.
“This isn’t just about improving outcomes, it’s about saving lives,” stated Carla Finkielstein, a professor on the Fralin Biomedical Analysis Institute at VTC and senior creator of the examine.
“For decades, 5-FU has been used with the assumption that it works the same for everyone. But we know it is not the case. Some individuals have a genetic susceptibility to this drug. Our approach adds a layer of precision to a treatment that’s been around for decades but still carries serious risks for certain patients.”
The analysis underscores the essential function of collaboration with oncologists and pathologists, whose medical insights contributed to the analysis route, Finkielstein stated.
The preliminary findings had been introduced by the main researchers of the mission, John Janiga and Dzenis Mahmutovic, on the 2025 ASCO Gastrointestinal Cancers Symposium. They had been additionally printed as an summary within the Journal of Medical Oncology, signaling early curiosity within the examine’s medical potential.
Scientists centered on mutations in a gene known as DPYD, which produces the enzyme accountable for breaking down 5-FU.
They initially analyzed DNA from wholesome volunteers and most cancers sufferers to find out whether or not saliva samples might reliably detect identified DPYD mutations. To increase past identified variants, the workforce later employed superior synthetic intelligence instruments alongside 3D protein modeling to judge 1000’s of samples.
“The AI-driven approach helped the team assess the structural and functional impact of previously unrecognized mutations in DPYD, uncovering potentially harmful variants that conventional methods might have missed,” stated Katherine Brown, the workforce’s lead bioinformatician.
A number of newly recognized DPYD mutations in colon most cancers sufferers had been predicted to impair 5-FU metabolism. Two had been labeled “pathogenic” by a number of predictive instruments and confirmed by a medical genetics database.
Some high-risk mutations had been additionally present in wholesome people, highlighting the necessity and feasibility of implementing saliva-based genetic screening for broader population-level danger evaluation, based on Finkielstein, who leads the Virginia Tech Molecular Diagnostics Laboratory on the Fralin Biomedical Analysis Institute.
“If we can flag high-risk patients early, we can tailor their treatment plans, reduce hospitalizations, and potentially avoid fatal complications,” stated Finkielstein, who can also be a professor of organic sciences within the Faculty of Science. “This is a real step forward in making cancer care safer, smarter, and more personalized.”
The early findings might increase present genetic testing and information safer, extra customized chemotherapy.
The examine was led by Finkielstein, a professor on the Fralin Biomedical Analysis Institute. Co-authors Janiga, Mahmutovic, Brown, Dulguun Myagmarsuren, and Clinton Roby are affiliated with the Molecular Diagnostics Laboratory. Douglas Grider is with the Virginia Tech Carilion Faculty of Drugs. Mark Kochenderfer is a medical oncologist with Blue Ridge Most cancers Care.
Extra info:
John M Janiga et al, Enhancing a low-cost, minimally invasive screening take a look at for dihydropyrimidine dehydrogenase mutations linked to 5-fluorouracil sensitivity by integrating computational mutation predictions, Journal of Medical Oncology (2025). DOI: 10.1200/JCO.2025.43.4_suppl.226
Supplied by
Virginia Tech
Quotation:
Saliva take a look at plus AI might flag chemotherapy danger, early examine outcomes recommend (2025, April 21)
retrieved 21 April 2025
from https://medicalxpress.com/information/2025-04-saliva-ai-flag-chemotherapy-early.html
This doc is topic to copyright. Other than any honest dealing for the aim of personal examine or analysis, no
half could also be reproduced with out the written permission. The content material is offered for info functions solely.
