Abstract
Objectives. To determine the lung cancer screening yield and stages in a union-sponsored low-dose computerized tomography scan program for nuclear weapons workers with diverse ages, smoking histories, and occupations.
Methods. We implemented a low-dose computerized tomography program among 7189 nuclear weapons workers in 9 nonmetropolitan US communities during 2000 to 2013. Eligibility criteria included age, smoking, occupation, radiographic asbestos-related fibrosis, and a positive beryllium lymphocyte proliferation test.
Results. The proportion with screen-detected lung cancer among smokers aged 50 years or older was 0.83% at baseline and 0.51% on annual scan. Of 80 lung cancers, 59% (n = 47) were stage I, and 10% (n = 8) were stage II. Screening yields of study subpopulations who met the National Lung Screening Trial or the National Comprehensive Cancer Network Group 2 eligibility criteria were similar to those found in the National Lung Screening Trial.
Conclusions. Computerized tomography screening for lung cancer among high-risk workers leads to a favorable yield of early-stage lung cancers.
Public Health Implications. Health equity and efficiency dictate that screening high-risk workers for lung cancer should be an important public health priority.
An ongoing challenge is identifying optimal populations for low-dose chest computerized tomography (CT) scan for the early detection of lung cancer.1–4 The National Lung Screening Trial (NLST) demonstrated that screening based on age and smoking reduces lung cancer mortality but does not address other lung cancer risk factors such as family history, chronic lung diseases, and occupational exposures, let alone a broader range of ages and smoking histories.5 Published risk-prediction models,2,3,6 online lung cancer–risk calculators, and professional organizations7,8 address additional lung cancer risk factors, but empirical studies that support their use are limited.
Exposures to occupational lung carcinogens such as asbestos, diesel exhaust, and silica are important, because they are common, may act synergistically with tobacco to raise lung cancer risk, and occur among blue-collar populations with the highest prevalence of cigarette smoking.9–11 Uncertainties about the magnitude of occupation-related lung cancer risk, however, complicate decision-making about eligibility for screening.6,12
Another unsettled question is whether the benefits of low-dose CT screening, heretofore demonstrated principally at large tertiary care medical centers, could be replicated in community-based screening.5
We examined the feasibility and screening yield of low-dose CT scanning for the detection of early-stage lung cancers among a large population of former nuclear weapons production, testing, and research workers, principally in nonurban settings in 5 states in the United States. In doing so, we included screening participants with a broad range of age and smoking histories to compare lung cancer screening results among population subsets with different combinations of age, smoking, and occupational risk.
METHODS
In 2000, we initiated low-dose CT scan screening as part of an ongoing voluntary union-sponsored multisite occupational medicine screening program for former US Department of Energy (DOE) workers at 3 DOE sites in 3 states and later expanded it to 9 DOE sites in 5 states. Participating DOE facilities included gaseous diffusion plants in Paducah, Kentucky; Portsmouth, Ohio; and Oak Ridge, Tennessee; the Fernald Feed Materials Production facility in Hamilton, Ohio; the Mound Site in Miamisburg, Ohio; Y-12 and the Oak Ridge National Laboratory in Oak Ridge, Tennessee; the Nevada Test Site near Las Vegas, Nevada; and the Idaho National Laboratory in Idaho Falls, Idaho. These sites performed diverse roles in nuclear weapons production and research with heterogeneous occupational exposures to chemical, mineral, and radiologic agents. Only participants who completed their baseline CT screening by March 31, 2013, are reported herein (n = 7189).
Lung Cancer Risk Factor Data
Occupational, smoking, and medical histories were obtained by self-administered questionnaire as part of the ongoing occupational medicine screening program and updated by telephone interview. Current smoking was defined as a positive response to “Do you smoke cigarettes now?” Never smoking was defined as a negative response to the following question: “Have you smoked more than 100 cigarettes (5 packs) during your entire life?” Chest x-rays (postero–anterior and lateral) were obtained in the occupational medicine screening program for 89.2% (6415/7189) of participants and were interpreted by a National Institute for Occupational Safety and Health–certified B reader, using the International Labor Organization classification system to identify the presence of asbestos-related parenchymal and pleural fibrosis.13 The beryllium lymphocyte proliferation test was performed by an expert laboratory (National Jewish Health, Denver, Colorado; Oak Ridge Associated Universities, Oak Ridge, Tennessee).
Occupational information included job titles, departments, and years worked, and a checklist of workplace exposures. No radiation dosimetry results were available, and no characterization of a history of radiation exposure of participants was feasible. Quantitative data on chemical and dust exposure at the plants were sparse over the relevant decades (1940s–1980s).
Participants were assigned by an industrial hygienist or occupational medicine physician to 1 of 5 occupational categories according to reported job titles: maintenance (Standard Occupational Classification [SOC] group 49), operations or production supervision (SOC groups 51 and 53), services (SOC groups 29, 31, 33, 35, 37, and 39), engineering (SOC groups 17 and 19), and administration (SOC groups 11, 13, 15, 25, and 43).14 Individuals who had more than 1 job category during their DOE tenure were assigned to the category with the presumed higher exposure category if they worked in that category at least 5 years. A priori, maintenance workers were considered likely to have the highest exposures, followed by operations or production supervision workers. The remaining categories were considered equivalent in terms of exposures for the purposes of assigning a single job category to individuals who worked in multiple job categories over their careers.
Currently smoking participants were given a brochure on smoking cessation and the telephone numbers of local and national smoking quit lines.
Eligibility Criteria
Eligibility criteria included age, smoking history, occupational category, chest x-ray evidence of asbestos-related parenchymal or pleural fibrosis, and positive beryllium lymphocyte proliferation test; criteria were modified somewhat over the 2000-to-2013 program period. This report details results on eligible workers aged 50 years or older who had smoked for 1 year or longer. Individuals were deemed ineligible if their spirometry showed a forced expiratory volume in 1 second (FEV1) of less than 1 liter or FEV1 less than 40% predicted (making them unlikely candidates for surgery), or if they had been treated for cancer within the past 5 years, excluding early prostate and bladder cancer, melanoma, and other skin cancers.
Scan Characteristics
The CT scans were obtained through the use of 2 dedicated mobile scanners (93%) or, beginning in 2012, at 2 imaging facilities in Las Vegas and Idaho Falls (7%).
From 2000 to 2006, a coach-mounted, single-slice General Electric HiSpeed DX/I (GE Medical Systems, Milwaukee, WI) was used to obtain full-chest, helical CT scans with a low-dose technique (120 kilovolt peak [kVp], 40–50 milliamps, pitch of 1.5, 7-millimeter collimation). From 2006 to 2013, we performed CT scanning with Siemens Emotion 16, 16-slice scanners (Siemens, Malvern, PA) at most program sites, supplemented by a Siemens Sensation 64 scanner in Las Vegas and a General Electric Lightspeed 16-slice scanner in Idaho Falls in 2012. We selected kilovolt peak, tube current, pitch, and other scanner parameters to produce a CT dose index for a standard-sized patient, which was within the guidelines for low-dose chest CT scan (1.5–2.3 mGy). The average estimated effective dose across all scanners was 1 millisieverts (range = 0.8–1.2 mSv).
Of 16 229 scans, 90% (14 674) were interpreted by a single senior academic thoracic radiologist (J. M.), and the remaining scans were read by experienced radiologists in DOE communities.
Scan Interpretation and Follow-Up
The CT reading and follow-up accorded with the International Early Lung Cancer Action Program protocol, including modifications, 2000 to 2011.15 Participants with CT-evident nodules suspicious for lung cancer (or other urgent findings) were referred to personal or local pulmonary physicians for follow-up. We provided follow-up of CT-detected indeterminate nodules with low-dose CT scans at 3, 6, or additional months to monitor nodule changes.
Baseline CT scans and follow-up scans of indeterminate nodules were offered beginning in 2000. One-time annual CT scans were offered at 1 year (range = 10–18 months) following the baseline scan to program participants beginning in 2002.
A lung cancer first identified as an indeterminate nodule on the baseline or annual CT scan, but only diagnosed after subsequent CT-detected nodule growth was considered a baseline or annual CT scan–detected lung cancer.
Individuals with nodules suspicious for lung cancer received diagnostic evaluations and treatment from their personal physicians. Medical and pathology records were obtained on individuals who underwent biopsy or surgery. Stage of cancer was determined from surgical findings, when performed.
Statistical Analysis
We examined the occurrence of lung cancer by relevant risk factors, using χ2 analyses. We conducted Cochran–Armitage tests for linear trends for age, pack-years, and years since smoking cessation. We calculated odds ratios and 95% confidence intervals for lung cancer for age, pack-years, smoking status, and job category with univariate and multivariable logistic regression models. We tested continuous risk factors (age and pack-years) included in the logistic regression model graphically for linearity; we addressed significant nonlinear relationships by treating the independent variable as categorical in the final regression model. We conducted all statistical tests at an α of 0.05 unless otherwise specified. It is noted that statistical tests and parameters were performed on results determined for a study population that was nonrandomly drawn from the underlying population of DOE workers. We conducted all analyses with SAS version 9.4 (SAS Institute, Cary, NC).
RESULTS
Between 2000 and 2013, 7189 individuals, mostly White (93%), male (88%), and of mean age 65 years (SD = 9.4), received a baseline low-dose chest CT scan. Only 20.4% (1468) of screened participants currently smoked. Mean cigarette consumption among all current and former smokers was 29 (SD = 23.3) pack-years. Mean time since quitting for former smokers was 25 (SD = 14.1) years; 75% had quit at least 15 years previously. Maintenance was the most common occupational category (n = 3224; 44.8%) among participants. Among the 6415 participants who had plain chest films before the CT scans, 10.9% (n = 700) had chest x-ray evidence of asbestos-related fibrosis.
The 1-time annual scan was initiated in 2002; it was offered to all participants beginning in August 2006. Of the 3759 participants who were first screened during or after August 2006, 83% (3110 participants) returned for an annual scan within 10 to 18 months after the baseline scan.
Screening Yield by Lung Cancer Risk Factor
Table 1 shows the occurrence of screen-detected lung cancers, or screening yield, on the baseline and annual scans for the 7189 participants, aged 50 years or older, who had ever smoked. The lung cancer screening yield was 0.83% among 7189 baseline participants and 0.53% among 3760 participants on annual scan. Lung cancer occurrence on baseline scan increased with advancing age (P < .01), current smoking (P < .01), greater pack-years (P < .01), and fewer years since quitting (P < .01). A similar pattern was seen on the annual scans, though smaller numbers limited statistical power.
TABLE 1—
Lung Cancer Screening Yield Among Nuclear Weapons Workers, by Lung Cancer Risk Factors: United States, 2000–2013
| Baseline |
Annual |
|||||
| Scans, No. (%) | Lung Cancer, No. (% Yield) | P a | Scans, No. (%) | Lung Cancer, No. (% Yield) | P a | |
| Total | 7189 (100) | 60 (0.83) | 3760 (52) | 20 (0.53) | ||
| Gender | .61 | ≥ .99 | ||||
| Male | 6315 (87.8) | 54 (0.86) | 3235 (86.0) | 18 (0.56) | ||
| Female | 874 (12.2) | 6 (0.69) | 525 (14.0) | 2 (0.38) | ||
| Age, y | < .01 | .17 | ||||
| 50–59 | 2607 (36.3) | 7 (0.27) | 1277 (34.0) | 3 (0.23) | ||
| 50–54 | 1264 (17.6) | 4 (0.32) | 582 (15.5) | 1 (0.17) | ||
| 55–59 | 1343 (18.7) | 3 (0.22) | 695 (18.5) | 2 (0.29) | ||
| 60–69 | 2241 (31.2) | 17 (0.76) | 1301 (34.6) | 10 (0.77) | ||
| 70–79 | 1876 (26.1) | 29 (1.55) | 988 (26.3) | 5 (0.51) | ||
| ≥ 80 | 465 (6.5) | 7 (1.51) | 194 (5.2) | 2 (1.03) | ||
| Race | .47 | .72 | ||||
| White | 6647 (93.1) | 57 (0.86) | 3462 (92.3) | 20 (0.58) | ||
| Black | 422 (5.9) | 2 (0.47) | 238 (6.4) | 0 (0) | ||
| Other | 74 (1.0) | 1 (1.35) | 49 (1.3) | 0 (0) | ||
| Smoke | < .01 | < .01 | ||||
| Current | 1468 (20.4) | 24 (1.63) | 717 (19.1) | 11 (1.53) | ||
| Former | 5721 (79.6) | 36 (0.63) | 3043 (80.9) | 9 (0.30) | ||
| Pack-years | < .01 | < .01 | ||||
| 1–9 | 1448 (20.8) | 2 (0.14) | 856 (23.4) | 1 (0.12) | ||
| 10–19 | 1372 (19.7) | 1 (0.07) | 749 (20.5) | 2 (0.27) | ||
| 20–29 | 1137 (16.3) | 9 (0.79) | 599 (16.4) | 2 (0.33) | ||
| 30–39 | 1098 (15.8) | 5 (0.46) | 559 (15.3) | 3 (0.54) | ||
| ≥ 40 | 1918 (27.5) | 41 (2.14) | 900 (24.6) | 12 (1.33) | ||
| Time since quit, y | .01 | .20 | ||||
| 0–9 | 982 (17.3) | 12 (1.22) | 551 (18.2) | 2 (0.36) | ||
| 10–19 | 1124 (19.8) | 9 (0.80) | 554 (18.3) | 3 (0.54) | ||
| 10–14 | 531 (9.4) | 4 (0.75) | 267 (8.8) | 1 (0.37) | ||
| 15–19 | 593 (10.5) | 5 (0.84) | 287 (9.5) | 2 (0.70) | ||
| 20–29 | 1381 (24.3) | 6 (0.43) | 728 (24.0) | 3 (0.41) | ||
| 30–39 | 1293 (22.8) | 6 (0.46) | 675 (22.3) | 0 (0) | ||
| ≥ 40 | 894 (15.8) | 1 (0.11) | 524 (17.3) | 1 (0.19) | ||
| Occupational history | .72 | .30 | ||||
| Administration/services | 1143 (16.0) | 7 (0.61) | 643 (17.2) | 4 (0.62) | ||
| Engineering | 823 (11.5) | 8 (0.97) | 521 (14.0) | 4 (0.77) | ||
| Maintenance | 3224 (45.2) | 30 (0.93) | 1549 (41.5) | 10 (0.65) | ||
| Operations/production supervision | 1951 (27.3) | 15 (0.77) | 1019 (27.3) | 2 (0.20) | ||
| Asbestos-related fibrosis (CXR) | .11 | .10 | ||||
| Total screened | 6415 (100) | 57 (0.89) | 3103 (100) | 19 (0.61) | ||
| Parenchymal disease (with or without pleural) | 168 (2.6) | 3 (1.79) | 65 (2.1) | 1 (1.54) | ||
| Pleural disease only | 532 (8.3) | 7 (1.32) | 226 (7.3) | 3 (1.33) | ||
| No evidence of fibrosis on x-ray | 5715 (89.1) | 47 (0.82) | 2812 (90.6) | 15 (0.53) | ||
| Beryllium sensitization | .26 | .55 | ||||
| Total screened | 6217 (100) | 47 (0.76) | 3121 (100) | 19 (0.61) | ||
| Sensitized (2 abnormal or 1 abnormal and 1 borderline) | 125 (2.0) | 0 (0.0) | 61 (2.0) | 0 (0.0) | ||
| Single abnormal | 132 (2.1) | 0 (0.0) | 68 (2.2) | 1 (1.47) | ||
| Normal | 5960 (95.9) | 47 (0.79) | 2992 (95.9) | 18 (0.60) | ||
Note. CXR = chest x-ray.
Chi square; linear trend for variables with > 2 categories.
Age.
The screening yield was highest among participants aged 70 years or older: 1.5% had a lung cancer detected on baseline scan. Six of the 9 cancers among participants aged 80 years or older were stage I.
Smoking.
Current smokers had a screening yield 3 to 5 times higher than former smokers on baseline and annual scans, respectively (Table 1). Participants who smoked 20 to 29 pack-years experienced a similar baseline yield of lung cancer compared with people with a 30 to 39 pack-year history (Table 1). Lung cancer yield decreased with increasing number of years since smoking cessation.
Occupational exposure.
Logistic regression analysis showed no significant differences in lung cancer yield among different occupational categories after adjustment for age and smoking (Table A, available as a supplement to the online version of this article at http://www.ajph.org). X-ray evidence of asbestos-related fibrosis was associated with an increase in lung cancer screening yield, though without statistical significance (Table 1). Workers with a single or double abnormal beryllium lymphocyte proliferation test did not have an increased lung cancer screening yield (Table 1).
Stage and Cell Type of Screen-Detected Cancers
Primary lung cancer was detected in 80 participants, including 60 on baseline and 20 cancers on annual scan. Of cancers detected on baseline and annual scan, 57% and 65% were stage I (1 was carcinoma in situ), respectively (Table 2). Of cancers detected on baseline and annual scans, 12% and 5% were stage II, respectively. One half of the cancers were adenocarcinomas, and one quarter were squamous cell carcinomas.
TABLE 2—
Cell Type and Stage of Screen-Detected Lung Cancers Among Nuclear Weapons Workers: United States, 2000–2013
| Screen-Detected Lung Cancers |
||
| Baseline, No. (%) | Annual, No. (%) | |
| Cell type | ||
| Adenocarcinoma | 30 (50.0) | 9 (45.0) |
| Squamous cell carcinoma | 12 (20.0) | 6 (30.0) |
| Adenosquamous carcinoma | 2 (3.3) | 1 (5.0) |
| Large cell carcinoma | 2 (3.3) | 1 (5.0) |
| Other non–small cell—unspecified | 3 (5.0) | 0 (0.0) |
| Sarcomatoid carcinoma | 1 (1.7) | 0 (0.0) |
| Small cell | 6 (10.0) | 2 (10.0) |
| Large cell endocrine | 3 (5.0) | 0 (0.0) |
| Missing | 1 (1.7) | 1 (5.0) |
| Total | 60 (100) | 20 (100.0) |
| Stage | ||
| Carcinoma in situ | 1 (1.7) | 0 (0.0) |
| IA | 20 (33.3) | 12 (60.0) |
| IB | 13 (21.7) | 1 (5.0) |
| IIA | 2 (3.3) | 1 (5.0) |
| IIB | 5 (8.3) | 0 (0.0) |
| IIIA | 4 (6.7) | 2 (10.0) |
| IIIB | 3 (5.0) | 0 (0.0) |
| IV | 5 (8.3) | 1 (10.0) |
| Limited (small cell) | 2 (5.0) | 0 (0.0) |
| Advanced (small cell) | 4 (6.7) | 2 (5.0) |
| Not staged | 0 (0.0) | 1 (5.0) |
| Total | 60 (100) | 20 (100) |
Source. Travis et al.16
The median time from CT scan that demonstrated a suspicious lesion and subsequent date of lung cancer diagnosis was 61.5 days.
Comparison of Screening Eligibility Criteria
We examined the screening yield of program participant subsets who met the eligibility criteria of the NLST study (age 55–74 years; ≥ 30 pack-years; ≤ 15 years since quitting) or the Group 2 recommended criteria of the National Comprehensive Cancer Network (NCCN; age ≥ 50 years; ≥ 20 pack-years; no time limit since quitting; and worked in maintenance, operations or production supervision, or laboratory; Table 3 and Figure 1). Only 20.5% of the study population met the NLST criteria, and they were similar to the NLST study population in age, current smoking prevalence, and mean pack-years (Table 3). Their baseline and annual screening yields were statistically similar to those of the NLST study population (1.5% vs 1.03% at baseline scan and 1.2% vs 0.68% at annual scan in the current study vs NLST study, respectively; Figure 1).
TABLE 3—
Lung Cancer Screening Yield of Current Study Subpopulations of Nuclear Weapons Workers, According to Different Screening Eligibility Criteria: United States, 2000–2013
| Current Study Population |
|||||
| Population Characteristics | NLST Study Population | Full Population | Subpopulation Who Met NLST Criteriaa | Subpopulation Who Met NCCN Group 2 Criteriab | Subpopulation Who Did Not Meet NLST or NCCN Group 2 Criteria |
| Demographics | |||||
| Age, y | |||||
| Mean (SD) | 61 (5)c | 65 (9.4) | 63 (5.6) | 66 (10.7) | 64 (9.8) |
| Median | 60c | 64 | 62 | 67 | 63 |
| Male/female, % | 59/41 | 88/12 | 84/16 | 94/6 | 86/14 |
| Smoking | |||||
| Current smokers, % | 48 | 20 | 46 | 19 | 11 |
| Pack-years | |||||
| Mean (SD) | 56 (NA) | 29 (23.3) | 52 (19.5) | 38 (19.3) | 15 (15.6) |
| Median | 48 | 25 | 46 | 32 | 12 |
| Former smokers, time since quit, y | |||||
| Mean (SD) | NA | 24 (14.1) | 7 (4.6) | 22 (10.9) | 30 (13.2) |
| Median | NA | 25 | 7 | 22 | 31 |
| Screening | |||||
| Baseline | |||||
| Screened, no. | 26 309 | 7 189 | 1 471 | 1 979 | 3 739 |
| Lung cancers, no. (% yield) | 270 (1.03) | 60 (0.83) | 22 (1.50) | 27 (1.36) | 11 (0.29) |
| Annual | |||||
| Screened, no. | 24 715 | 3 670 | 752 | 908 | 2 100 |
| Lung cancers, no. (% yield) | 168 (0.68) | 19 (0.52) | 9 (1.20) | 4 (0.44) | 6 (0.29) |
Note. NA = not available; NCCN = National Comprehensive Cancer Network; NLST = National Lung Screening Trial.
Source: NLST Research Team et al.5,17
NLST criteria: age 55–74 y, ≥ 30 pack years, ≤ 15 y since quitting.
NCCN criteria: age ≥ 50 y, ≥ 20 pack years, no time limit since quitting.
Approximated from NLST Research Team et al.17
FIGURE 1—
Screening Yields Among Nuclear Weapons Workers According to the Eligibility Criteria of the National Lung Screening Trial (NLST) and the National Comprehensive Care Network (NCCN) at (a) Baseline Scan and (b) First Annual Scan: United States, 2000–2013
An additional 27.5% of the current study population met the NCCN Group 2 criteria. Their screening yields were also similar to those of the original NLST study population at baseline scan and at annual scan (Table 3, Figure 1). For the 3739 (52.0%) participants in the current study who met neither the NLST nor NCCN Group 2 criteria, their lung cancer screening yields were significantly lower than lung cancer screening yields on baseline and annual scans compared with the original NLST study population. This group had fewer mean smoking pack-years (15 pack-years) and longer mean time since smoking cessation (30 years) than the groups that met the eligibility criteria of the NLST or the NCCN Group 2 (Table 3).
DISCUSSION
At present, low-dose CT scan screening for lung cancer is recommended by the US Preventive Services Task Force18 and covered by medical insurance19 only for people who meet certain age and smoking criteria in the United States. Current study results lend empirical support for the use of expanded age and smoking criteria for lung cancer screening as recommended in the Group 2 guidelines of the NCCN in an occupational population with variable exposure to workplace lung carcinogens. When we applied those criteria, the lung cancer screening yields in the current study were similar, subject to statistical uncertainty, to the screening yields of NLST. In addition, one third of detected lung cancer (24/69; 30.3%) occurred among people who had quit smoking at least 15 years previously, which would have been undetected if NLST criteria alone had been used. Our results were similar to those obtained at the Lahey Clinic20 and support an enhanced risk-based approach.3,4
This study also demonstrated that CT-based lung cancer screening of blue-collar workers in nonmetropolitan areas is well-accepted and highly feasible. The NLST was completed at 33 sites in the United States, 31 of which were in major metropolitan areas or at tertiary care medical centers affiliated with medical schools. The NLST population also had a higher education level than the US population as a whole.17 The current study was addressed to a workforce that lived almost entirely in rural areas and had less education on average than the population of the NLST study. Thus, success in recruitment and compliance in our study population was untested in the NLST study. Four of 5 (83%) current study participants returned for their 1-time annual low-dose CT scan. Mobile CT scanners were initially used in 2000, because local radiology facilities were unfamiliar with low-dose CT technique. We centralized CT scan reading to a single experienced thoracic radiologist, which may limit the generalizability of our findings. However, the success of NLST permitted a transition to the use of local radiology facilities, where expertise in low-dose CT has rapidly grown. Diligent and prompt communication between the screening program and local providers in the 9 study communities facilitated timely follow-up.
The lung cancer screening yields were somewhat higher among participants with nonmalignant asbestos-related fibrosis of the pleura and parenchyma than the screening yields of the original NLST study populations, though without statistical significance. These markers of a past history of significant exposure to asbestos are easily identified for use in determining screening eligibility. We did not reliably identify a history of asbestos exposure in the study population other than the presence of radiographic fibrosis and cannot comment on screening yield according to a history of asbestos exposure alone. It is well known that a history of asbestos exposure in the absence of fibrosis increases lung cancer risk10 and should be used as a risk factor in determining eligibility for lung cancer screening.
Study results show that participants who do not meet the NLST criteria or the NCCN Group 2 criteria, despite a history of work in the nuclear weapons complex, have a lung cancer screening yield that is approximately one third of that of the NLST study. There is no evidence at present that links low-dose CT screening at that lowered level of risk with reduced lung cancer mortality.
The lung cancer stage distribution of the current study was quite similar to that of the NLST results despite differences in study populations, health care settings, and lung cancer risk factor profiles. On baseline CT scans in the NLST, 62% of lung cancers detected were stage I cancers versus 57% in the current study.
Most lung cancer screening guidelines do not recommend screening of individuals aged 80 years or older. Of smokers aged 80 years or older who were in good health, we detected lung cancer on the baseline scan in 1 of every 66 participants. Two thirds of these cancers were stage I. Mean life expectancy at age 80 years in the United States in 2011 was 9.1 years.21 Our findings are based on small numbers, and the issue needs to be the subject of further study.
Study strengths include a large, novel study population; excellent credibility with the study population through labor union cosponsorship; implementation in community settings; high study compliance; excellent follow-up; and use of a protocol with demonstrated quality (International Early Lung Cancer Action Program protocol). Limitations include the lack of specific knowledge about occupational exposures, lack of data on mortality follow-up, and limited statistical power because of the relative infrequency of lung cancer, even in a sizable study population. In addition, our use of a study population that was nonrandomly selected limits inferences about a larger population of DOE workers. Generalizability is also limited by the dominance of men in the study cohort.
In conclusion, our study results provide evidence that occupation can be used in combination with age and smoking to identify populations that provide a similar screening yield and lung cancer stage distribution as was demonstrated in the landmark NLST study. Furthermore, we have demonstrated the utility of low-dose CT scan screening for a blue-collar population in nonmetropolitan settings that are not routinely served by tertiary care medical centers. This study is ongoing and should yield additional useful information with additional person-time experience and lung cancer cases detected.
ACKNOWLEDGMENTS
The study was supported by contracts from the US Department of Energy (DE-FC01-06EH0601 and DE-FC01-06EH06008).
The authors acknowledge Jennifer Mihalics for data management; John Lain and Lori Lanier for their work as computerized tomography technicians; Mark Griffon for assistance with job categorization; Brittany Dickens for article preparation; and Claudia Henschke, MD, and David Yankelevitz, MD, for their leadership of the International Early Lung Cancer Action Program.
Note. The views expressed in this article do not necessarily reflect those of the sponsor.
HUMAN PARTICIPANT PROTECTION
The screening program was reviewed and approved by the Oak Ridge Associated Universities institutional review board (IRB; 2000–2009), the Central Department of Energy IRB (2010–2013), the Queens College IRB (2000–2012), and the City University of New York IRB (2012–2013). All participants signed written informed consent at the time of the baseline computerized tomography scan.
Footnotes
REFERENCES
- 1.Kovalchik SA, Tammemagi M, Berg CD et al. Targeting of low-dose CT screening according to the risk of lung-cancer death. N Engl J Med. 2013;369(3):245–254. doi: 10.1056/NEJMoa1301851. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Tammemägi MC, Church TR, Hocking WG et al. Evaluation of the lung cancer risks at which to screen ever- and never-smokers: screening rules applied to the PLCO and NLST cohorts [erratum: PLoS Med. 2015] PLoS Med. 2014;11(12):e1001764. doi: 10.1371/journal.pmed.1001764. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Katki HA, Kovalchik SA, Berg CD, Cheung LC, Chaturvedi AK. Development and validation of risk models to select ever-smokers for CT lung cancer screening. JAMA. 2016;315(21):2300–2311. doi: 10.1001/jama.2016.6255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gould MK. Who should be screened for lung cancer? And who gets to decide? JAMA. 2016;315(21):2279–2281. doi: 10.1001/jama.2016.5986. [DOI] [PubMed] [Google Scholar]
- 5.National Lung Screening Trial Research Team. Aberle DR, Adams AM et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365(5):395–409. doi: 10.1056/NEJMoa1102873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tammemägi MC. Application of risk prediction models to lung cancer screening: a review. J Thorac Imaging. 2015;30(2):88–100. doi: 10.1097/RTI.0000000000000142. [DOI] [PubMed] [Google Scholar]
- 7.Wood DE. National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines for Lung Cancer Screening. Thorac Surg Clin. 2015;25(2):185–197. doi: 10.1016/j.thorsurg.2014.12.003. [DOI] [PubMed] [Google Scholar]
- 8.Jacobson FL, Austin JH, Field JK et al. Development of The American Association for Thoracic Surgery guidelines for low-dose computed tomography scans to screen for lung cancer in North America: recommendations of The American Association for Thoracic Surgery Task Force for Lung Cancer Screening and Surveillance. J Thorac Cardiovasc Surg. 2012;144(1):25–32. doi: 10.1016/j.jtcvs.2012.05.059. [DOI] [PubMed] [Google Scholar]
- 9.International Agency for Research on Cancer. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans: Volume 100c: Arsenic, Metals, Fibres and Dusts. Lyon, France: WHO Press; 2012. [PMC free article] [PubMed] [Google Scholar]
- 10.Markowitz SB, Levin SM, Miller A, Morabia A. Asbestos, asbestosis, smoking and lung cancer: new findings from the North American Insulator Cohort. Am J Respir Crit Care Med. 2013;188(1):90–96. doi: 10.1164/rccm.201302-0257OC. [DOI] [PubMed] [Google Scholar]
- 11.Syamlal G. Combustible tobacco and smokeless tobacco use among working adults—United States, 2012 to 2014. J Occup Environ Med. 2016;58(12):1185–1189. doi: 10.1097/JOM.0000000000000898. [DOI] [PubMed] [Google Scholar]
- 12.Wolff H, Vehmas T, Oksa P, Rantanen J, Vainio H. Asbestos, asbestosis, and cancer, the Helsinki criteria for diagnosis and attribution 2014: recommendations. Scand J Work Environ Health. 2015;41(1):5–15. doi: 10.5271/sjweh.3462. [DOI] [PubMed] [Google Scholar]
- 13.Guidelines for the Use of the ILO International Classification of Radiographs of Pneumoconioses. Rev ed. Geneva, Switzerland: International Labor Organization; 2011. [Google Scholar]
- 14.Bureau of Labor Statistics. 2010 Standard Occupational Classification System. Available at: https://www.bls.gov/soc/2010/2010_major_groups.htm. Accessed November 1, 2017.
- 15.Henschke CI. International Early Lung Cancer Action Program: enrollment and screening protocol. New York, NY: International Early Lung Cancer Action Program; 2011. [Google Scholar]
- 16.Travis WD, Brambilla E, Burke AP, Marx A, Nicholson AG, editors. WHO Classification of Tumours of Lung, Pleura, Thymus and Heart. 4th ed. Lyon, France: International Agency for Research on Cancer; 2015. [DOI] [PubMed] [Google Scholar]
- 17.National Lung Screening Trial Research Team. Aberle DR, Adams AM, Berg CD et al. Baseline characteristics of participants in the randomized National Lung Screening Trial. J Natl Cancer Inst. 2010;102(23):1771–1779. doi: 10.1093/jnci/djq434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Final update summary: lung cancer: screening. US Preventive Services Task Force. July 2015. Available at: https://www.uspreventiveservicestaskforce.org/Page/Document/UpdateSummaryFinal/lung-cancer-screening. Accessed February 20, 2018.
- 19.Centers for Medicare & Medicaid Services. Decision memo for screening for lung cancer with low dose computed tomography (CAG-00439N). 2015. Available at: https://www.cms.gov/medicare-coverage-database/details/nca-decision-memo.aspx?NCAId=274. Accessed February 20, 2018.
- 20.McKee BJ, Hashim JA, French RJ et al. Experience with a CT screening program for individuals at high risk for developing lung cancer. J Am Coll Radiol. 2016;13(2, suppl):R8–R13. doi: 10.1016/j.jacr.2015.12.006. [DOI] [PubMed] [Google Scholar]
- 21. Arias E. United States Life Tables, 2011. Natl Vital Stat Rep. 20152;64(11):1–63. [PubMed] [Google Scholar]