A blood test for early detection of lung cancer

Nov. 21, 2016

An immune response can reveal cancer. As a normal cell undergoes tumorigenesis, it expresses aberrant protein antigens, known as tumor-associated antigens (TAAs). They are recognized as non-self by the immune system, thereby triggering a B-cell response. The production of autoantibodies by B cells is not dependent on the size of a tumor and can occur throughout all stages of the disease. Most important, autoantibodies are produced early in tumorigenesis, when just a few antigens may stimulate a prolific immune response.1 Autoantibodies, therefore, can serve as an early cancer marker.

Biomarkers for lung cancer

A blood test that measures a panel of seven autoantibodies associated with lung cancer is available to clinicians to further assess a patient’s risk of lung cancer being present.2-5 The test provides additional insight into the patient’s lung cancer risk beyond self-reported risk factors, such as age, smoking history, exposure to carcinogens (e.g., radon, asbestos), etc. A tube of blood is drawn from the patient, and the isolated serum is sent to a CLIA-certified laboratory, where autoantibodies are measured utilizing a proprietary platform technology based on indirect enzyme-linked immunosorbent assay (ELISA) principles.

The test results

Results are reported for each autoantibody as well as an overall test result.2-5 Two clinically-derived cut-offs for each autoantibody are utilized to determine whether a patient’s autoantibody levels are low, moderate, or high.3 A low level of autoantibodies does not rule out lung cancer; it simply means that the patient’s lung cancer risk has not changed appreciably from what was estimated by his or her self-reported risk factors.3 Conversely, patients with moderate or high levels of any one or more autoantibodies are at increased risk of having lung cancer, which may warrant additional testing consistent with the patient’s history and overall risk profile.3

Clinical validation

Clinical validation of the autoantibody test utilized serum from three cohorts of newly diagnosed lung cancer patients (n=655) and serum from risk-matched control patients. (Controls were not diagnosed with cancer and were risk-matched to lung cancer cases by gender, age, and smoking history.) The studies demonstrated that the autoantibody test detects all types and all stages of lung cancer equally.5 The panel of autoantibodies was subsequently changed from six to seven autoantibodies, which improved accuracy to 92 percent, PPV (positive predictive value) to one in eight (assumes lung cancer prevalence 2.4 percent) and specificity to 91 percent at 41 percent sensitivity.2 A second cut-off for each autoantibody was implemented to stratify the positive autoantibody results, with the primary advantage being the enhancement in specificity and PPV with a high level result to 98 percent and one in four (assumes lung cancer prevalence 2.7 percent), respectively.3 The performance characteristics demonstrated by these studies support the test’s utility in helping physicians identify which patients are at highest risk of having a lung cancer. (Note: The test is a rule-in test and cannot be used to rule-out a patient having lung cancer.)

A clinical audit was performed on 1,613 U.S. patients whose physicians had ordered the autoantibody test due to their patient’s high risk for lung cancer. Patients provided HIPAA authorization to release their medical records for the clinical audit. Records regarding cancer diagnosis were reviewed six months following testing for all patients with a positive or negative test result. Sixty-one patients (four percent) were identified with lung cancer, 25 of whom had a positive autoantibody test (sensitivity 41 percent). A positive autoantibody test was associated with a five-fold increase in risk of lung cancer versus a negative test. In the lung cancer patients with a positive autoantibody test, where stage was known, 8 of 14 (57 percent) were Stage I or II. The clinical audit also confirmed a highly statistically significant improvement in specificity of the seven-autoantibody panel and demonstrated that the autoantibody test performs the same in routine clinical practice as was shown in the case-control validation studies.4

In another clinical audit study, the autoantibody test was evaluated in 296 high-risk patients with non-calcified pulmonary nodules (lung cancer prevalence 25 percent). The patient’s lung cancer risk was calculated using the Swensen/Mayo Clinic Nodule Calculator.6 A positive autoantibody test resulted in a greater than two-fold increased relative risk of lung cancer.7 Using a “both positive rule” of combining binary tests, adding the autoantibody test to the Swensen risk model improved the diagnostic performance with high specificity (> 92 percent) and positive predictive value (> 70 percent). Accordingly, a positive autoantibody test reflects a significant increased risk of lung cancer in non-calcified nodules 4-20mm in largest diameter.7

Prospective trials

Currently, there are two large prospective trials underway evaluating the autoantibody test. In Scotland, the National Health Service (NHS) is conducting a randomized prospective trial of 12,000 individuals at high-risk for lung cancer. Participants are randomized, with half getting the autoantibody test and half not. Those with a positive autoantibody test are followed-up with a chest x-ray and CT chest scan. (Low-dose CT chest scans are not approved for screening in Scotland.) The goal of the study is to determine if screening with the autoantibody test reduces the number of patients being diagnosed with Stage III or IV lung cancer. This trial completed accrual in June 2016 and follow-up is ongoing.8

The second prospective trial, at National Jewish Health in Denver, Colorado, combines screening with low-dose CT chest scan and the autoantibody test. The goal of the study is to determine if screening with both modalities results in a higher rate of detecting early stage lung cancer.9 To date, more than 1,300 high-risk participants have been enrolled. Encouraging preliminary results of both studies were presented at the World Conference on Lung Cancer in Denver in late 2015.8,9

Currently, only 25 percent to 30 percent of all patients with lung cancer in the United States meet the criteria for lung cancer screening with low-dose CT chest scans.10 Clearly there is a need for additional tests, such as the autoantibody test described here, to help identify the remaining 70 percent to 75 percent of lung cancers while they are asymptomatic and early stage. Additionally, as the number of patients with non-calcified pulmonary nodules is rapidly growing through CT screening, there is an urgent need for supplemental tests to assist physicians in determining which are malignant and which are benign.


  1. Zhong L, Coe SP, Stromberg AJ, et al. Profiling tumor-associated antibodies for early detection of non-small cell lung cancer. J Thor Oncol. 2006;1(6):513-519.
  2. Chapman CJ, Healey GF, Murray A, et al. EarlyCDT-Lung test: improved clinical
    utility through additional autoantibody assays. Tumor Biol. 2012;33(5):1319-26.
  3. Healey GF, Lam S, Boyle P, et al. Signal stratification of autoantibody levels in
    serum samples and its application to the early detection of lung cancer. J Thorac Dis  2013;5(5):618-625.
  4. Jett JR, Peek LJ, Fredericks L, et al. Audit of the autoantibody test, EarlyCDT-Lung, in 1600 patients: An evaluation of its performance in routine clinical practice. Lung Cancer 2014;83(1):51-55.
  5. Boyle P, Chapman CJ, Holdenrieder S, et al. Clinical validation of an autoantibody test for lung cancer. Ann Oncol. 2011;22(2):383-389.
  6. Swensen SJ, Silverstein MD, Ilstrup DM, Schleck CD, Edell ES. The probability of malignancy in solitary pulmonary nodules. Application to small radiologically indeterminate nodules. Arch Intern Med.1997;157(8):849-855.
  7. Massion PP, Healey GF, Peek L, et al. Autoantibody signature enhances the positive predictive power of computed tomography and nodule-based risk models for detection of lung cancer. J Thorac Oncol. 2016. Published online Sept 2016. doi.org/10.1016/j.jtho. 2016.08-143.
  8. Sullivan F, Schembri S. Progress with an RCT of the detection of autoantibodies to tumor antigens in lung cancer using the EarlyCDT-Lung test in Scotland. J Thorac Oncol. 2015;10:S306.
  9. Jett JR, Dyer K, Kern JA. Screening for lung cancer with EarlyCDT-Lung and
    computed tomography. J Thorac Oncol. 2015;10:S306.
  10. Pinsky PF, Berg CD. Applying the national lung screening trial eligibility criteria to the us population: What percent of the population and of incident lung cancers would be covered? J Med Screen 2012;19(3):154-156.

Laura J. Peek, PhD, serves as Principal Scientist and Assistant Lab Director for Oncimmune (USA) LLC. She is in her tenth years with the company. Her responsibilities include the oversight of all aspects of the clinical laboratory operations.

James R. Jett, MD, serves as Chief Medical Officer for Oncimmune. His past affiliations include faculty appointments at Mayo Clinic, Rochester, Minnesota, and National Jewish Health, Denver, Colorado. His work focuses on the early detection, diagnosis, and treatment of lung cancer as well as other malignancies of the chest.