Lung cancer diagnosis is notoriously challenging, largely because the current standard of care frequently requires invasive methods to obtain surgical tissue and, thereby, enable an accurate diagnostic call. This challenge will become more pronounced as lung cancer screening programs reach more at-risk patients across the United States. This article addresses the specific, current limitations of lung cancer diagnosis, the impact of these limitations on patients and the healthcare system, and the demonstrated ability of genomic testing to help address these limitations. It also explores the genomic discoveries and scientific advances that soon may enable even earlier detection and treatment of this disease.
Challenges of lung cancer diagnosis
Lung cancer is the leading cause of cancer-related deaths in the U.S., killing more than 150,000 Americans each year.1 While early detection and diagnosis can significantly improve survival,2-4 presently only 16 percent of lung cancer cases are diagnosed at an early stage.5 In 2011, results from the U.S. National Lung Screening Trial (NLST) demonstrated that annual screening with low-dose computed tomography scan (LDCT) is an effective means to identify lung nodules and lesions earlier and thereby reduce lung cancer deaths among people at risk for the disease.6 As a result, in 2013 the United States Preventative Services Task Force (USPSTF) recommended annual LDCT screening for current and former smokers aged 55 to 74.7
Among the pulmonology community, there is significant uncertainty about the optimal work-up of pulmonary nodules and lesions, whether they are found incidentally or via LDCT screening. Current guidelines are complex and non-specific, particularly in cases where patients’ nodules are not clearly benign or cancerous based on imaging and clinical workup.8 Bronchoscopy is the most commonly used tool to evaluate lung nodules and lesions, generally preferred over transthoracic needle biopsy (TTNB) and surgical lung biopsy (SLB) because it is less invasive. Unfortunately, the benefits of bronchoscopy come with some drawbacks. It is challenging to reach peripheral nodules with this tool, so when a nodule comes back with a negative result, it is not always clear whether the nodule is truly benign or was just missed. Perhaps the most problematic issue with bronchoscopy is that among the estimated 350,000 patients who undergo the procedure each year as part of a pulmonary nodule work-up, 40 to 60 percent come back non-diagnostic.8-10
Lung nodules that are not clearly benign or malignant present a significant challenge to physicians, who often pursue aggressive approaches in order to secure a definitive result and avoid missing a lung cancer diagnosis. Published literature suggests that up to 41 percent of patients with inconclusive bronchoscopy results are referred to risky and expensive invasive procedures, including TTNB and SLB. Ultimately, up to 40 percent of these invasive procedures show that the nodule is benign, meaning patients were unnecessarily exposed to procedural risks and discomfort.11 These procedures also create additional financial burden to the healthcare system: costs for SLB, for example, can exceed $20,000.
These challenges and implications are likely to become more problematic as expanded lung-cancer screening is increasingly embraced nationwide.
The role of genomic testing
Genomic testing has demonstrated the ability to reduce ambiguity in lung cancer diagnosis and thereby reduce reliance on invasive and costly procedures such as biopsy. In 2015, a U.S.-based company introduced a 23-gene bronchial genomic classifier that complements bronchoscopy to increase its accuracy among patients undergoing a work-up for pulmonary nodules and lesions.
This classifier is based on research conducted by Dr. Avrum Spira and his team at Boston University School of Medicine, which demonstrated that cells in the central bronchial airways of the lung exhibit measurable cancer-associated gene-expression changes due to smoking. These collective genomic alterations comprise a “field of injury” which serves as a biomarker distinguishing ever-smokers with lung cancer from those with benign lung disease, independent of clinical risk factors.12 The genomic test developers combined gene expression data and machine learning to create a classifier that effectively identifies these field-of-injury molecular changes without the need to sample a lung nodule or lesion directly.
The genomic test uses cells obtained from standard cytology brushings taken from the mainstem bronchus during a lung cancer diagnostic bronchoscopy. Local pathologists review the cytology sample, and a sample is then sent to the test-maker’s CLIA-certified laboratory, where genomic testing is performed if the bronchoscopy result is inconclusive.
The classifier’s performance—including its ability to significantly enhance the diagnostic yield of bronchoscopy—has been verified in multiple clinical studies, including clinical validation data published in The New England Journal of Medicine. Findings demonstrate that the classifier improved the overall sensitivity of bronchoscopy, from 75 percent for bronchoscopy alone to 97 percent when paired with the test (p<0.001). Additionally, the genomic test has a high degree of accuracy (negative predictive value of 91 percent) in identifying patients who are at low (<10 percent) risk of cancer.13 Patients who are classified as low risk by the test can be monitored with CT imaging rather than directed to surgery.
The real-world implications of this genomic classifier have been confirmed through a prospective clinical utility study. Hogarth et al found that when the test classified patients as low risk for lung cancer, there was greater than 50 percent relative reduction in physician recommendations for invasive diagnostic procedures such as biopsy, as compared to recommendations made without genomic testing.14 In a second study, Ferguson et al concluded that a low-risk classifier result prompted a three-fold reduction (from 57 percent to 18 percent) in invasive procedure recommendations among pulmonologists, compared to when no genomic test results were available.15
In the current healthcare environment, the cost-effectiveness of any new technology must also be considered. Data published in 2017 suggest that use of the bronchial genomic classifier reduces invasive lung-cancer diagnostic procedures by 28 percent at one month and 18.3 percent at two years, and is cost-effective compared to the use of bronchoscopy alone.16
The science and discoveries that enabled the existing bronchial genomic classifier have created opportunities for even earlier lung cancer detection and treatment. In one very exciting recent development, researchers from Boston University demonstrated that the same field-of-injury genomic changes found in the main airway of current and former smokers with lung cancer can be detected in the nasal passages. Their findings, published in the Journal of the National Cancer Institute, provide evidence that molecular biomarkers used to determine lung cancer risk in cells from the bronchial airway could provide similar information as cells obtained from a simple nasal swab.17 Research is already underway to explore how this discovery could be translated into genomic tests that could enable earlier lung-cancer detection and treatment and ultimately help to reduce the number of associated deaths.
- American Cancer Society. Cancer Facts & Figures 2017. Atlanta, GA: American Cancer Society; 2017.
- The National Lung Screening Trial Research Team, Aberle DR, Berg CD, et al. The National Lung Screening Trial: overview and study design. Radiology. 2011;258(1):243-253.
- American Cancer Society. Small Cell Lung Cancer Survival Rates, by Stage.
- Wood DE, Eapen GA, Ettinger DS, et al. Lung cancer screening. J Natl Compr Canc Netw. 2012;10(2):240-265.
- U.S. National Institutes of Health. National Cancer Institute. SEER Cancer Statistics Review, 1975-2013.
- The National Lung Screening Trial Research Team, Aberle DR, Berg CD, et al. Results of initial low-dose computed tomographic screening for lung cancer. N Engl J Med. 2013;368:1980-1991.
- U.S. Preventive Services Task Force. Lung cancer: screening. https://www.uspreventiveservicestaskforce.org/Page/Document/UpdateSummaryFinal/lung-cancer-screening. December 2013.
- Gould MK, Donington J, Lynch WR, et al. Evaluation of individuals with pulmonary nodules: when is it lung cancer? Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. CHEST. 2013;143(5 Suppl):e93S-e120S.
- Memoli JSW, Nietert PJ, Silvestri GA. Meta-analysis of guided bronchoscopy for the evaluation of the pulmonary nodule. CHEST. 2012;142(2):385-393.
- Ost DE, Ernst A, Lei X, et. al. Diagnostic yield and complications of bronchoscopy for peripheral lung lesions: results of the AQuIRE Registry. AJRCCM. 2016;193(1):68-77.
- Tanner NT, Aggarwal J, Gould MK, et al. Management of pulmonary nodules by community pulmonologists a multicenter observational study. CHEST. 2015;148(6):1405–1414.
- Beane J, Sebastiani P, Whitfield TH, et al. A prediction model for lung cancer diagnosis that integrates genomic and clinical features. Cancer Prev Res. 2008;1(1):56-64.
- Silvestri GA, Vachani A, Whitney D, et al. A bronchial genomic classifier for the diagnostic evaluation of lung cancer. N Engl J Med. 2015;373:243-251.
- Hogarth DK, Dotson T, Lee H, Whitten P, Lenbyrg M. The Percepta® Registry: a prospective registry to evaluate percepta bronchial genomic classifier patient data. CHEST Annual Meeting: Abstract PC082.1.1610. 2016;150(4_S):1026A.
- Ferguson S, Van Wert R, Choi Y, et al. Impact of a bronchial genomic classifier on clinical decision making in patients undergoing diagnostic evaluation for lung cancer. BMC Pulm Med. 2016;16:66.
- Feller-Kopman D, Liu S, Geisler BP, DeCamp, MM, Pietzsch JB. Cost-effectiveness of a bronchial genomic classifier for the diagnostic evaluation of lung cancer. J Thoracic Oncol. 2017;12(8):1223-1232.
- AEGIS study team. Shared gene expression alterations in nasal and bronchial epithelium for lung cancer detection. JNCI: Journal of the National Cancer Institute. 2017;109(7).