Predictive models based on comparison of serum metabolomic patterns for patients at time of lung cancer diagnosis and matched controls
FRIDAY, Dec. 17, 2021 (HealthDay News) — Blood metabolomics has potential as a screening tool for early-stage lung cancer, according to a study published online Dec. 13 in the Proceedings of the National Academy of Sciences.
Tjada A. Schult, from Harvard Medical School in Boston, and colleagues used a training-validation-testing cohort design to assess metabolomics predictive models based on high-resolution magic angle spinning magnetic resonance spectroscopy of serum samples collected from patients prior to lung cancer diagnosis. Samples were collected from 79 patients within five years of and at lung cancer diagnosis and 79 healthy matched controls.
By comparing serum metabolomic patterns between the patients with lung cancer and matched healthy controls in the training cohort, disease predictive models were established. The researchers then applied these predictive models to assess serum samples, collected from patients before lung cancer diagnosis, from validation and testing cohorts. The predictive model yielded values for prior-to-detection serum samples that were intermediate between values for patients at time of diagnosis and healthy controls and that differed significantly from both these groups. The F1 score for cancer prediction was 0.628. For patients with localized disease, values from a metabolomics predictive model measured from prior-to-diagnosis sera could significantly predict five-year survival.
“Our study demonstrates the potential for developing a sensitive screening tool for the early detection of lung cancer,” a coauthor said in a statement.
Abstract/Full Text (subscription or payment may be required)
Copyright © 2021 HealthDay. All rights reserved.