Biomarkers & lung nodule follow-up: Refiguring the math

The National Lung Screening Trial (NLST) delivered some very positive statistics, capped by the 20 percent relative mortality risk reduction among high-risk individuals undergoing three rounds of CT screening. However, other statistics, particularly those associated with follow-up nodules, may be a bit more problematic. Various research teams are focusing on the development and validation of biomarkers, including imaging biomarkers, to address some of these challenges.

This feature is the second in a web series focused on the details of lung cancer screening programs. Look for additional features on upcoming Mondays in Health Imaging.

NLST data suggested that CT screening will depict lung nodules in approximately 30 percent of high-risk individuals undergoing three annual exams. As many as 96 percent of these nodules are benign. However, researchers and clinicians have not yet established optimal practices for determining which nodules have the greatest likelihood of malignancy.

Determining how to follow some nodules is a problem, said Pierre P. Massion, MD, of Vanderbilt-Ingram Cancer Center in Nashville, Tenn. Imaging follow-up or intervention may be recommended depending on nodule characteristics.

Current practices result in benign results for approximately one-third of lung biopsies, so some researchers are focusing on strategies to reduce the number of thoracotomies for benign disease. Biomarkers offer one approach.

“CT can be helpful at distinguishing between benign and malignant nodules for nodules 20 mm or larger, but it is not very specific with nodules less than 10 mm,” explained Massion, who is principal investigator of the Nashville Lung Cancer Screening Trial. The five-year project will enroll 480 high-risk individuals to undergo screening exams including CT and bronchoscopy. Researchers also are collecting blood, airway and urine samples, and studying a series of biomarkers identified over the last 10 years at Vanderbilt to determine if any are linked with lung cancer.

Biomarkers can answer several different questions. “We are interested in biomarkers to rule in cancer and biomarkers to rule out cancer. They are different in characteristics and complexity,” said Massion.

Vanderbilt researchers as well as scores of other research teams around the globe are employing genomics, proteomics and metabolomics to determine the biological composition of lung tumors and evaluate subjects’ blood, urine and breathe for their presence. The goal is to provide physicians sufficient data to affect the decision to intervene on a nodule. To date, no biomarker has met that criterion.

Imaging biomarkers may be the first foray into this model. Radiologists at Vanderbilt and elsewhere have focused on quantitative analysis of lung nodules. “Nodules,” said Massion, “are very different in size, shape and density, and in how they relate to the alveoli, bronchial tubes and pulmonary vessels. They can have unique growth characteristics.”

A nodule’s growth rate over time may provide a biomarker to help predict malignancy. The approach requires specific protocols and 3D analysis. Scans should be acquired at 1-1.25 mm slices; the right baseline data sets the stage for good follow-up, said Massion.

Better images and quantitative follow-up will allow radiologists to measure growth in accurate terms and differentiate which patients require invasive follow-up. “This will become standard of care,” predicted Massion, adding that imaging is likely to be the first of multiple biomarkers leveraged in lung cancer screening and follow-up.

Massion also emphasized that biomarker research and other screening efforts should not distract stakeholders from emphasizing the need to prevent smoking and to assist smokers to quit.