VISE affiliates developing a novel integrative approach for early detection of lung cancer
Vanderbilt researchers have received a National Cancer Institute grant to develop a novel, integrative approach to detect early signs of lung cancer.
The four-year project builds on a related, recent study that established the value of using three separate measures—structural imaging, a protein marker and information available from electronic health records—to predict lung cancer in patients not yet exhibiting any symptoms of the disease.
The new study, which expands and integrates use of the different information sources, involves an innovative partnership that brings together Vanderbilt experts in pulmonary oncology, radiology, machine learning and data science.
Primary investigators are Bennett Landman, Professor of Electrical Engineering and Chancellor Faculty Fellow, and Cornelius Vanderbilt Chair in Medicine Pierre Massion.
“This collaboration advances technologies developed at Vanderbilt in both image processing and blood biomarkers. I’m excited our path to bring these technologies out of the research lab to positively impact patients,” said Landman.
Most lung cancers are first detected as small roundish, growths called indeterminate pulmonary nodules. While the majority of these nodules are benign, noninvasive strategies to pinpoint cancerous ones could reduce diagnostic time and mortality rates. Earlier, more accurate identification also would reduce the rate of false positives, researchers say.
The goal is to establish an integrative approach to early detection that leverages repeated measures of lung nodule imaging over time; changes in hsCYFRA 21-1, a high sensitivity blood biomarker; and longitudinal clinical patters from EHRs.
“Built upon strong preliminary data and unique resources from VUMC that include access to large imaging and EHR data sources, this novel integrative study has the potential to generate highly impactful and translatable results to reduce false positive rates among IPNs, and morbidity and mortality from lung cancer,” researchers said.
“With this research project we have a unique opportunity to diagnose early patients with lung cancer. Early is key. We will leverage IMAGE VU, deep learning methods, a novel high sensitive assay for protein biomarkers, and clinical patterns from the electronic health record over time to accomplish this early, and noninvasive diagnosis to lung cancer. This will be possible because of the creativity and commitment of experts like Drs. Bennett Landman, Tom Lasko, Kim Sandler and Michael Kammer,” Massion said.
Early detection among asymptomatic patients is key to improving survival rates; worldwide, lung cancer is responsible for more deaths than any other cancer.