Published in Cancer Detection and Prevention 2001; 25(4):336-343.

Integration of Peripheral Blood Biomarkers with Computed Tomography to Differentiate Benign from Malignant Pulmonary Opacities

Thomas Alola, MD,a Gerold Bepler, MD, PhD,d David Harpole, MD,a Philip C. Goodman, MD,b H. Page McAdams, MD,b Jeremy J. Erasmus, MD,b James E. Herndon, PhD,c and Edward F Patz, Jr, MDb

Duke University Medical Center, Departments of aSurgery and bRadiology and cCommunity and Family Medicine, Durham, NC, and dRoswell Park Cancer Institute, Departments of Medicine and Cancer Genetics, Buffalo, New York

Address all correspondence and reprint requests to: Edward F. Patz Jr. MD. Duke University Medical Center, Department of Radiology, Box 3808, Durham, NC 27710.

ABSTRACT: Our purpose was to determine whether peripheral blood biomarkers MUC1 and CK19 could be used to complement imaging studies in differentiating benign from malignant indeterminate pulmonary nodules or masses detected on computed tomography CT. One hundred and eighteen patients had a thoracic CT and blood drawn for tumor marker reverse transcriptase-polymerase chain reaction analysis. Thirty-five of the 118 patients had an indeterminate pulmonary nodular opacity on CT, and the findings then were correlated with the reverse transcriptase-polymerase chain reaction results. The sensitivity and specificity for the markers in determining malignancy was calculated. Thirteen of the 35 opacities on CT proved to be benign, and 22 proved to be lung cancer. Among the patients with indeterminate pulmonary abnormalities, polymorphic epithelial mucin protein 1 had a sensitivity and specificity for lung cancer of 100% and 46%. respectively. Cytokeratin 19 had a sensitivity and specificity for lung cancer of 95% and 8%. respectively. These preliminary data showed that serum biomarkers polymorphic epithelial mucin protein I and cytokeratin 19 were not specific for lung cancer. although patients with an indeterminate pulmonary abnormality and negative markers were unlikely to have lung cancer. Integration of imaging studies with the appropriate biomarkers may prove useful in evaluating indeterminate pulmonary nodules or masses.

KEY WORDS: biiomarkers, lung cancer, pulmonary nodule-mass.