SUNDAY Lung Cancer Screening: Management of Incidental Findings Caroline Chiles, MD Incidental findings at lung screeningg CT Caroline Chiles MD Wake Forest University Health Sciences Winston-Salem, North Carolina additional costs patient anxiety iatrogenic injury potential benefit to patient STR 2014 Incidental findings at lung screening CT of mortality Causes o o ta ty in tthee NLST S Other 2000 Should we look for incidental findings? Should h ld we report incidental d l findings? f d Should we recommend further investigation? 343 1800 349 1600 12 175 7 226 1400 1200 486 470 416 Respiratory Cardiovascular 442 Other cancers 503 Lung cancer 200 427 Complications CV Other cancer Lung cancer 600 400 Other Respiratory p y 1000 800 Complications 0 CT Significant causes of mortality in the lung cancer screening population l ti Cardiovascular disease COPD Other cancers 144 CXR Significant causes of mortality i the in h llung cancer screening i population Cardiovascular C di l disease COPD Other cancers 4 Annualized Event Rates for Coronary A Artery Calcium C l i (CAC) Ri Riskk C Categories i CAC Risk category All-cause mortality CVD endpoint CHD endpoint 0 0.08 0.7 0.3 Case-cohort design 1 - 100 0.2 1.5 0.4 150 cases (d (death, th cardiac di event), t) 808 random d sample l 101 1,000 0.6 6 1.7 1.0 , > 1,000 1.1 6.1 33.2 NELSON trial i l Excluded subjects with history of CVD 4 level risk stratification based on Agatston score: score 0, 1 - 100, 101 - 1,000, > 1,000 SUNDAY Coronary artery calcium can predict all-cause ll mortality t lit and d cardiovascular di l events on LDCT screeningg for lungg cancer CAC is an independent predictor of all-cause mortality, fatal and nonfatal cardiovascular events,, and fatal and nonfatal coronary events in a lung cancer screening population. Jacobs PC, Gondrie MJ, van der Graaf Y, et al. Am J Roentgenol. 2012;198:505511. Annualized Event Rates for Coronary Artery Calcium (CAC) Risk Categories Jacobs PC et al., AJR 2012. 198(3): p. 505-11. CAC on LDCT A visual assessment of CAC can be used for risk 7 6 5 4 ACM 3 CVD CHD 2 1 0 0 1-100 101-1,000 >1,000 prediction of CHD death and all-cause mortality using non-gated non gated LDCT for lung cancer screening, and is comparable to Agatston scoring. Simplest scoring system of none/mild/moderate/heavy / ild/ d /h calcification l ifi i is i adequate d for risk assessment. Patients who are aware of high CAC scores are more likely to be treated with statins as a lipid-lowering intervention, and to initiate aspirin therapy, dietary changes and increased exercise. changes, exercise Agatston score Significant g causes of mortalityy in the lung cancer screening population Cardiovascular disease di COPD Other cancers 10 COPD Underdiagnosed disease 4th leading cause of death Infrequently self-reported even in older, heavy smokers in screening trials An A earlier li di diagnosis i off COPD can iimprove patient i management and result in fewer exacerbations Independent risk factory for lung cancer 145 SUNDAY Imaging phenotypes in COPD Emphysema-predominant COPD Qualitative assessment Quantitative assessment emphysema index = % of all lung voxels <950 HU (affected by reconstruction algorithm, section thickness, inspiration level, scanner, gravity, and radiation dose) Airway-predominant COPD Qualitative assessment wall thickness, luminal narrowing Quantitative assessment software analysis of average wall thickness, total wall area, inner perimeter length, length wall area percentage COPD exacerbations in COPDGene Emphysema-predominant group total percentage of emphysema > 35% segmental bronchial wall thickness < 11.75 75 mm mm. Airway-predominant group <35% 35 emphysema p y segmental bronchial wall thickness >1.75 mm. COPD exacerbations related to both emphysema severity i and d airway i wall ll thickness hi k At lower levels of emphysema, airway wall thickness became the predominant p factor. Han MK, Kazerooni EA, Lynch DA, et al. Radiology. 2011;261:274282 Identification of COPD at lung cancer screening (NELSON) 1140 men PFTs and inspiratory/expiratory CT on same date PFT diagnosis COPD = FEV1/FVC of < 70% Air trapping = RV/TLC > 95th% predicted value CT diagnosis emphysema p y = % of voxels < 95 950 HU air trapping = expiratory to inspiratory ratio of mean lung density Identification of COPD at lung cancer screening (NELSON) 437 (38%) had PFT dx of COPD Logistic g regression g used to develop p a multivariable model with FEV1/FVC ratio of PPV 76%, NPV 79% Mets OM et al, JAMA 2011; 306: 1775 Mets OM et al, JAMA 2011; 306: 1775 Author Significant causes of mortality in the lung cancer screening population Cardiovascular di disease COPD Other cancers MacRedmond 2003 Swensen 2003 van de Wiel 2007 Kucharczyk 2011 Rampinelli ll 2011 NLST 2011 Priola 2013 146 < 70% as the outcome 5 factors independently associated with COPD CT emphysema, CT air trapping, BMI, pack-years, and smoking status Automated dx with low dose CT LDCT Study Cohort 449 1,520 Subjects with extrapulmonary malignancy 0 Details of extrapulmonary malignancies 17 Renal cell (4), Breast (3), y p ((2), ), Bronchial carcinoid Lymphoma (2), Gastric (2) (1) = Ovarian, Spinal metastases, Pancreatic, Phaeochromocytoma Liver metastases (1.1%) 1 929 1,929 1 (0 05%) (0.05%) 4,073 7 (0.8%) 5,201 27 ( (0.5%) ) 26,722 519 0 Rib Plasmocytoma (2), Thyroid (1), Breast (4) Renall Cell ll ((5), ) Renall Clear l Cell ll ((2), ) Lymphoma (5), Thyroid (3) , Thymoma (2), Pancreas (2), (1) = Breast, Urinary tract, Schwannoma, Adrenal, Heptocellular, GI Stromal, Prostate , Ovary 416 (1.6%) ----- 6 Renal cell (3), Thymoma (2) (1.2%) Adrenal metastasis Chiles C, Paul N. J Thorac Imaging. 2013 (1) Nov;28(6):347-54 Clinical Implications and Added Costs of Incidental Findings SUNDAY Extrapulmonary malignancies detected at lung screening CT Continuous Observation of Smoking Subjects [COSMOS] Clinically relevant = findings that required further 5,201 heavy h smokers k aged d 50+ d investigation dx i i i or med/surg d/ i intervention i screened for lung cancer for 5 consecutive years 27 asymptomatic t ti extrapulmonary t l malignancies li i frequency of one case per 200 individuals (0.5%) Found in 63/519 (12%) subjects at baseline Found in 10 subjects in 4 years of follow follow-up up 52/63 participants had IFs not already known (30 thorax; 23 abdomen). Most common thyroid gland (23), kidney (10), adrenal glands (7). Rampinelli C, Preda L, Maniglio M, et al. Radiology 2011; 261: 293-299 Priola AM, Priola SM, Giaj-Levra M, et al.Clin Lung Cancer 2013; 14:139-148 Number and cost of additional imaging investigations Clinically relevant incidental findings Thyroid lesions (23) 18 multinodular l d l goiter 3 nodular goiter 2 benign nodule Renal lesions (10) 7 Cyst > 20 HU 2 Renal cell carcinoma 1 Oncocytoma 23 thyroid th id ultrasounds lt d (5 ( with ith bx) b ) 17 CT (with contrast) scans of chest or abd 2 MR 3 Mammography 5 CT guided g needle bx 4,644 ($6,575) Based on national reimbursement rate in Italy Adrenal Ad l glands l d ((7)) 4 Adenoma < 10 HU 2 Adenoma > 10 HU 1 Metastasis (lung cancer) Priola AM, Priola SM, Giaj-Levra M, et al.Clin Lung Cancer 2013; 14:139-148 Neglectable benefit of searching for incidental in the d l ffindings d h NELSON triall usingg low-dose multidetector CT Baseline scans of 1,929 participants Liver lesions (76) Multiple cysts (55), single cyst (10), hemangioma (6), FNH (1), hepatic steatosis (1), metastasis (1) Renal R l llesions i ((53)) Solitary cyst (40), multiple cysts (7), hydronephrosis (2), angiomyolipoma g y p ((1) Thyroid gland (9) Multinodular goiter (5), cyst/nodular goiter/benign nodule (1 each) van de Wiel JC, Wang Y, Xu DM, et al. Eur Radiol. 2007;17:14741482. Priola AM, Priola SM, Giaj-Levra M, et al.Clin Lung Cancer 2013; 14:139-148 Neglectable benefit of searching for incidental in the d l ffindings d h NELSON triall usingg low-dose multidetector CT The The one malignant lesion we found had no clinical benefit for this participant, because it was a metastasis from pancreatic cancer van de Wiel JC, Wang Y, Xu DM, et al. Eur Radiol. 2007;17:14741482 . 147 SUNDAY Summary Cardiovascular disease, COPD, and other cancers are significant causes of mortality in the lung cancer screening population 1. Report p coronaryy arteryy calcification none,, mild,, moderate, heavy. 2. Report the presence, severity of emphysema 3. Limited Li i d opportunity i to d detect extrapulmonary l malignancy. Insufficient evidence to recommend p of low attenuation lesions in thyroid, y kidneys, y follow-up liver. 148