Blood Test Detects Lung Cancer

News | October 22, 2013 | Lung Cancer

By Anna Azvolinsky, PhD

A blood test that detects a combination of proteins can distinguish between early lung cancer and noncancerous lung nodules, according to a study published in Science Translational Medicine. The two types of nodules are currently difficult to distinguish using imaging. The test showed a 90% negative predictive value in a study of 104 patient samples from three clinical sites. Further validation on 37 samples from a single site showed a 94% negative predictive value.

This molecular test may complement the current tools used by clinicians to diagnose early-stage lung cancer.

The early lung cancer detection test is being developed by Seattle-based molecular diagnostic company Integrated Diagnostics (Indi). The test detects the levels of 13 proteins in a patient’s blood sample and could prevent unnecessary biopsies of lung nodules detected by CT scans.

“These studies suggest that Indi’s technology is capable of detecting the molecular signature of lung cancer by measuring the presence of multiple proteins in a patient’s blood,” said senior author Paul Kearney, PhD, president and chief science officer of Integrated Diagnostics, in a released statement.

The researchers used a systems biology approach, screening 371 blood-based proteins on 143 patient samples with either benign or stage 1A lung cancer that were matched for nodule size, age, gender, and clinical site.

Multiple reaction monitoring (MRM) mass spectrometry is used to analyze the relative concentrations on biomarkers. The technology allows simultaneous analysis of many protein levels.

Further analysis to understand the role of these biomarker proteins showed that all 13 are likely regulated by four transcription factors that bind to the regulatory elements of the 13 genes that encode the proteins. All four transcription factors have been associated with lung inflammation and lung cancer, as well as oxidative stress pathways.

The validation using 104 patient samples showed a test sensitivity of 71% and specificity of 44%. The study researchers assumed that the rate of cancer prevalence was 15%. At the same cancer prevalence rate, the sensitivity was 82% and the specificity was 66% in the discovery cohort of 143 samples.

The protein levels were found to be independent of known risk factors for pulmonary nodules: the size of the nodule detected, history of smoking, and age.

According to Kearney and fellow study authors, one-fifth of patients with detectable lung nodules who undergo biopsy or surgery actually have a malignant nodule. Therefore, a reliable test that can discriminate between a benign lung mass and a cancerous lung mass is needed to prevent unnecessary invasive procedures and surgery.

Still, the study cohorts used here are retrospective, and a prospective validation trial is needed to understand the full clinical potential of the blood test. Additionally, the authors note that the test results are not integrated with clinical risk factors, although they also note that “pulmonologists vary broadly in the use of clinical risk factors, and so, it is actually preferable to have a molecular diagnostic test that produces a score independent of clinical risk factors.”

Even if validated, it is likely that such a blood test will be part of the other information used by clinicians to make the best-informed decision for each individual patient.

Stephen Malkoski, MD, PhD, who specializes in pulmonary medicine at the University of Colorado Cancer Center in Aurora and who was not involved in the research, says interpreting the test results would likely depend on the calculated risk clinicians currently use to assess whether a nodule could be cancerous. Factors that go into the risk calculation include size of the nodule, its location, smoking history, age, emphysema, and family lung cancer history. “In a low-risk patient with a 1% to 2% risk of lung cancer, the patient would still likely need a follow-up CT scan since the blood test could not definitely exclude malignancy.”

Malkoski sees the test as potentially being the most helpful in intermediate-risk patients with a calculated risk of malignancy between 10% and 20%. “In this group, a negative test might obviate the need for additional workup, such as a PET scan or biopsy,” he notes. “This may make a clinician more comfortable opting for radiographic follow-up rather than an invasive procedure.”