Background: Lung cancer is the leading cancer killer in both men and women in the United States.1 More than half of people with lung cancer die within 1 year of being diagnosed.2 Only 15% of lung cancers are diagnosed at an early stage. The 5-year survival rate is 54% for localized disease, compared with 4% at advanced stages.2
Objective: Conduct an 18-month pilot study to evaluate whether emergency room (ER) visits for medical emergencies resulting in a low-dose computerized tomography (LDCT) of the chest could be used to identify patients and incidentally discover lung cancer, leading to an earlier diagnostic stage.
Methods: Natural Language Processing (NLP) software was used to review all thoracic LDCT scans obtained in 3 community-based Sarah Cannon Cancer Center–affiliated ERs in Nashville, TN, and Dallas, TX, from October 2014 through March 2016. All radiology reports from LDCTs of the chest containing worrisome terminology such as “nodule,” “opacity,” or “mass” were identified by the NLP software and automatically sent to a navigator. Plan of care or next course of action for these patients was determined via multidisciplinary tumor board (MDTB) review, utilizing NCCN guidelines.3
Results: NLP software identified 1381 CT scans for MDTB review, which led to 88,576 LDCT scans performed at participating Sarah Cannon centers; 1545 incidental findings needing MDTB review; 79 biopsies (5% of NLP-identified cases); 48 cancer diagnoses (3% of NLP-identified cases); 34 (71%) lung cancers; and 14 (29%) other cancer types.
Among the diagnosed lung cancers, 9 (26%) were early stage (TNM I or II) and were treated definitively with surgery.
Conclusions: This work indicates that the use of technology can improve diagnosis of early-stage lung cancer in the ER setting within the general population. These patients received an LDCT of their chest, looking for other medical conditions or worrisome findings. This technology led to lung cancers being incidentally discovered, potentially improving a patient’s earlier diagnosis. This pilot study is expanding to include multiple institutions within the Sarah Cannon network.
- Centers for Disease Control and Prevention. National Center for Health Statistics. CDC WONDER On-line Database, compiled from Compressed Mortality File 1999-2012 Series 20 No. 2R, 2014.
- National Cancer Institute. State Cancer Profiles. https://statecancerprofiles.cancer.gov/.
- National Comprehensive Cancer Center. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines). Lung Cancer Screening. www.nccn.org/professionals/physician_gls/pdf/lung_screening.pdf.