Pleural Plaques and the Risk of Pleural Mesothelioma

JNCI Journal of the National Cancer Institute Advance Access published January 25, 2013 DOI:10.1093/jnci/djs513 Article The Author 2013. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com. Pleural Plaques and the Risk of Pleural Mesothelioma Jean-Claude Pairon, François Laurent, Mickaël Rinaldo, Bénédicte Clin, Pascal Andujar, Jacques Ameille, Patrick Brochard, Soizick Chammings, Gilbert Ferretti, Françoise Galateau-Sallé, Antoine Gislard, Marc Letourneux, Amandine Luc, Evelyne Schorlé, Christophe Paris Manuscript received May 2, 2012; revised November 12, 2012; accepted November 13, 2012. Correspondence to: Jean-Claude Pairon, MD, PhD, Service de pneumologie et pathologie professionnelle, Centre Hospitalier Intercommunal de Créteil, 40 avenue de Verdun, 94010 Créteil cedex, France (e-mail: jc.pairon@chicreteil.fr). Background Methods Results Conclusion The association between pleural plaques and pleural mesothelioma remains controversial. The present study was designed to examine the association between pleural plaques on computed tomography (CT) scan and the risk of pleural mesothelioma in a follow-up study of asbestos-exposed workers. Retired or unemployed workers previously occupationally exposed to asbestos were invited to participate in a screening program for asbestos-related diseases, including CT scan, organized between October 2003 and December 2005 in four regions in France. Randomized, independent, double reading of CT scans by a panel of seven chest radiologists focused on benign asbestos-related abnormalities. A 7-year follow-up study was conducted in the 5287 male subjects for whom chest CT scan was available. Annual determination of the number of subjects eligible for free medical care because of pleural mesothelioma was carried out. Diagnosis certification was obtained from the French mesothelioma panel of pathologists. Survival regression based on the Cox model was used to estimate the risk of pleural mesothelioma associated with pleural plaques, with age as the main time variable and time-varying exposure variables, namely duration of exposure, time since first exposure, and cumulative exposure index to asbestos. All statistical tests were two-sided. A total of 17 incident cases of pleural mesothelioma were diagnosed. A statistically significant association was observed between mesothelioma and pleural plaques (unadjusted hazard ratio (HR) = 8.9, 95% confidence interval [CI] = 3.0 to 26.5; adjusted HR = 6.8, 95% CI = 2.2 to 21.4 after adjustment for time since first exposure and cumulative exposure index to asbestos). The presence of pleural plaques may be an independent risk factor for pleural mesothelioma. J Natl Cancer Inst Because asbestos has been widely used in Western countries, a substantial proportion of retired workers have been occupationally exposed to it. It has been estimated that approximately 20% to 25% of male retirees in France have been previously exposed to asbestos (1). Occupational asbestos exposure is known to be associated with several benign diseases of the pleura and lungs, including pleural plaques, pleurisy, fibrosis of the visceral pleura, rounded atelectasis, and asbestosis, as well as several malignancies mainly lung cancer and mesothelioma (2 4). The epidemiology of asbestos-related diseases has been largely documented over recent decades (2,3,5). However, several issues concerning asbestos-related diseases remain controversial, as underlined in a recent consensus expert statement (eg, the level of asbestos exposure required to cause lung cancer, the existence of a link between the extent of asbestos exposure and the presence and extent of pleural abnormalities, and the indication for chest computed tomography [CT] scanning to screen populations at risk for asbestos-related diseases) (6). Pleural plaques are circumscribed areas of thickening of parietal or diaphragmatic pleura composed of avascular collagen connective tissue. They are the most common disease in asbestos-exposed subjects (4). An association between pleural plaques and pleural mesothelioma has been reported in several consensus statements (4,6). Subjects with pleural plaques have been reported to have increased risks of lung cancer and pleural mesothelioma compared with the general population (7 9). However, these associations were mainly established by a small number of studies based on chest x-ray, with a limited number of subjects. It has been demonstrated that chest x-ray has a fairly poor sensitivity and specificity for the detection of pleural plaques compared with CT scan. The use of chest x-ray is therefore associated with a high risk of misclassification of subjects according to the presence or absence of pleural plaques. Moreover, only a few studies have taken cumulative exposure to asbestos into account. In a study of former workers and residents of the Australian crocidolite mining and milling town of Wittenoom, Reid et al. reported an association between benign pleural disease and an excess of peritoneal mesothelioma, but not pleural mesothelioma, JNCI Article Page 1 of 9

after adjustment for cumulative exposure to asbestos and time since first exposure (10). The potential link between the most common benign asbestos-related disease (ie, pleural plaques) and pleural mesothelioma, therefore, cannot be considered to be formally demonstrated (11). Because pleural plaques are very frequent in asbestos-exposed subjects, it is crucial to establish whether or not they confer an increased risk of pleural mesothelioma in comparison with individuals free of pleural plaques but with similar levels of asbestos exposure. A large-scale pilot screening program for asbestos-related diseases was initiated in four regions of France in 2001 after a national consensus conference on the clinical surveillance strategy for workers previously exposed to asbestos, which proposed periodic chest CT scans and pulmonary function tests for workers with moderate to high exposure to asbestos (12). This study was designed to examine the association between pleural plaques, detected on CT scan, and the risk of pleural mesothelioma in a 7-year follow-up study of formerly asbestos-exposed workers. Methods Study Population The Asbestos Post Exposure Survey (APEXS) screening program for asbestos-related diseases was organized at the request of the French Ministry of Labour and National Health Insurance between October 2003 and December 2005 in four regions of France: Aquitaine, Rhône-Alpes, Haute-Normandie, and Basse- Normandie. Retired or unemployed workers previously occupationally exposed to asbestos and covered by the General National Health Insurance fund (which covers more than 80% of the French population) were eligible for the screening program. They were invited to participate in the program in various ways according to the region (letters targeting age groups below 65 or 67 years and previous type of job; trade union, radio, television, and newspaper advertisements). As previously described, volunteers who agreed to participate had a free medical check-up that included chest CT scan and pulmonary function tests (12 14). Only male subjects with an available copy of their CT scan sent to our team centers were included in this study. The study was approved by the hospital ethics committee. All participants received information about the study and gave their written informed consent. Asbestos Exposure and Tobacco Consumption As described elsewhere, all subjects completed a standardized questionnaire, which allowed industrial hygienists to evaluate asbestos exposure on the basis of the complete work history of each subject (13). For each job associated with asbestos exposure, the duration (number of years) and dates of exposure were determined. The following weighting factors were attributed for the intensity level of exposure: low (passive exposure): 0.01; low intermediate: 0.1; high intermediate: 1; high: 10. A cumulative exposure index (CEI) to asbestos was calculated for each subject over his working life as the sum of exposures calculated for each exposed job (duration weighting factor). Because of the absence of atmospheric measurements and the lack of detailed information on the frequency of exposure to asbestos (percentage of the working time), the CEI was expressed in exposure units years rather than in fibers/ml years. The latency was defined as the time elapsed between the beginning of the first job considered to have exposure to asbestos and the date of CT scan. The questionnaire also collected information on tobacco consumption. Subjects were classified into three categories: current smokers, ex-smokers (defined as those who had stopped smoking for at least 1 year), and never smokers. CT Scanning Subjects underwent CT scan according to a specific protocol. The modalities for performing CT scan were established by a group of chest radiologists designated by the Société Française de Radiologie (French Radiology Society), with technical modalities designed to detect all CT scan changes related to asbestos-induced pleural and parenchymal diseases with a limited radiation burden. Briefly, the main parameters were as follows: the entire chest was screened in the supine position using a spiral acquisition without injection of contrast media; a slice thickness of 1.5 to 5 mm was used; a pitch of 1.5 to 2.0, 120 kv, 60 to 150 ma maximum was used; parenchymal images were reconstructed with sharp filters and visualized with a window width approximately 1600 Hounsfield Units (HU) and a window level of 600 HU; and soft tissue images were reconstructed with smooth filters and visualized with window width approximately 400 HU and window level of 50 HU. A minimum of five, high-resolution, 1-mm thick CT sections performed in the prone position between the carina and the pleural recess were also acquired (14). All available CT scans were submitted for randomized, independent, double reading (and triple reading in the case of disagreement) focused on benign asbestos-related abnormalities by a panel of seven chest radiologists trained in the interpretation of asbestosrelated CT abnormalities. (Two members of the panel are study authors [F. Laurent and G. Ferretti]. The five others are Y. Badachi, C. Beigelman, A. Jankowski, V. Latrabe, and M. Montaudon.) These experts were blinded to the subject s cumulative exposure to asbestos and to the results of the initial reading by the radiologists who performed the CT scans. A standardized form was used to register interstitial or pleural abnormalities according to the Fleichner Society glossary of terms (15). Pleural plaques were considered to be present in the case of circumscribed quadrangular pleural elevations, with sharp borders and tissue density, sometimes calcified, with usual topography for at least some of the images (16). Pleural plaque thickness was classified by taking into account the thickest plaque (four categories: <2 mm, 2 <5 mm, 5 <10 mm, and 10 mm or more). The cutpoints were chosen and adapted from the International Labour Office classification of radiographs of pneumoconiosis (17). In this classification, width/thickness of pleural abnormalities is categorized as a (3 5 mm), b (5 10 mm) or c (>10 mm). The extent of pleural plaques was assessed using a semiquantitative method. The cumulative extent of pleural plaques detected on each section was calculated and expressed as the percentage of the lateral chest contour as measured on a single axial section at the level of the carina. This cumulative extent was graded according to four categories: less than 1 cm, between 1 cm and less than 25% of the chest contour, between 25% and less than 50% of the lateral chest contour, and 50% or more of the lateral chest contour. Page 2 of 9 Article JNCI

Because there is no available consensus for classification of the extent of pleural plaques on CT scan, the categories were adapted from the International Labour Office classification of radiographs of pneumoconiosis (17). A fourth category for plaques with a cumulative extent less than 1 cm, very likely identified on CT scan only, was added to the three International Labour Office categories. The choice of adapting International Labour Office classification to CT scan has been previously used by Jarad et al. (18) in their proposal for a scoring system. Parietal pleural plaques were defined in our study as typical when they were bilateral, thicker than 2 mm, and with an extent greater than 1 cm, regardless of whether they were calcified. Mesothelioma Diagnosis A follow-up study was organized in subjects who had participated in the APEXS program using free cancer medical care data. In France, all cancers must be reported to the French National Health Insurance to provide full coverage of medical expenses, including treatment. Annual determination of all new subjects applying for free medical care for pleural mesothelioma was therefore carried out for the 7 years between January 1, 2004, and March 31, 2011, in subjects who had participated. Patients with mesothelioma were considered to be incident when there was no suspicion of mesothelioma on the baseline CT scan. The French mesothelioma panel of pathologists (MESOPATH) was asked to certify the diagnosis of mesothelioma in all mesothelioma patients of this study (19). Each mesothelioma patient submitted to MESOPATH underwent a standardized diagnostic confirmation procedure, including transmission to MESOPATH of histological slides or blocks. Morphological analysis on hematoxylin and eosin stained slides and systematic immunohistochemical analysis including at least two positive and two negative markers were then performed to maximize the reliability of diagnosis. Slides were reviewed by at least three expert members of MESOPATH (and a consensus opinion with a quorum of 10 experts in the case of disagreement between the first three experts), blinded to evaluation of asbestos exposure. Final pathological certification was classified as certain mesothelioma, uncertain mesothelioma (not definitively certified and not definitively excluded), or unclassifiable tumor because of inadequate material. In some patients, mesothelioma could be ruled out in favor of another diagnosis. Complementary analysis on available histological material was performed by a single pathologist for determination of fibrohyaline pleural plaques associated with the tumor on histology slides. This pathologist evaluated the presence or absence of hyaline fibrous plaques composed of hyaline acellular collagen with basket weave reticulin. Statistical Analysis The variables used to characterize asbestos exposure were duration of exposure to asbestos, CEI, and time since first exposure (TSFE) to asbestos until CT scan. Statistical associations between pleural plaques and incidence of pleural mesothelioma were studied using survival regression analysis based on the Cox proportional hazards model (34 091 subject-years). Age was the main time variable, thus accounting for age in a nonprespecified way, whereas duration of exposure, CEI [expressed as Ln(CEI +1)], and TSFE to asbestos were independent variables. Only the latter (TSFE to asbestos) was a time-varying variable because the subjects were no longer exposed at the time of inclusion. For each subject, date of diagnosis of mesothelioma or date of last update namely, March 31, 2011 was used. Proportionality assumption of Cox model was verified graphically. Unadjusted hazard ratios (HRs) and adjusted hazard ratios for these time-varying variables were calculated for the risk of pleural mesothelioma associated with typical pleural plaques and with other less typical pleural plaques, with subjects free of pleural plaques as reference subjects according to CT scan date. All mesothelioma cases were initially used in the analysis, and then only those definitely confirmed by MESOPATH. The Fisher Freeman Halton test was used to compare characteristics (age, smoking status, asbestos exposure parameters) of mesothelioma patients with those of nonmesothelioma subjects. Statistical analysis was carried out using SAS software version 9.2 (SAS Institute, Inc, Cary, NC) and STATA for survival analyses, release 11 (StataCorp, College Station, TX). All statistical tests were two-sided, and statistical significance was defined as P less than.05. Results Overall, 13 859 eligible subjects agreed to participate in the screening program, 7275 underwent CT scan, and 5825 of these CT scans were sent to our team centers. After exclusion of subjects with missing data for asbestos exposure parameters, subjects considered to be unexposed to asbestos by industrial hygienists, subjects with unreadable CT scan, and one subject considered to have a probable pleural tumor at the time of CT scan (effusion and irregular pleural nodular thickening, which was finally diagnosed as a prevalent mesothelioma), the study population consisted of 5287 male subjects (Figure 1). General characteristics of the study population according to the presence or absence of pleural plaques are shown in Table 1. Overall, 1078 subjects (20.4%) had parietal pleural plaques with or without diaphragmatic pleural plaques, and 20 subjects (0.4%) had isolated diaphragmatic pleural plaques. Subjects with typical plaques or with less-typical pleural plaques had longer mean duration of exposure to asbestos than subjects without pleural plaques (32.7 ± 9.7 years, 32.6 ± 9.7 years, and 30.5 ± 10.5 years, respectively; P <.0001 for both comparisons) and also higher CEI to asbestos (111.2 ± 124.7 unit of exposure years, 104.5 ± 128.0 unit of exposure years, and 54.8 ± 91.5 unit of exposure years, respectively; P <.0001 for both comparisons). In contrast, there was no statistically significant difference in the duration of exposure to asbestos or CEI to asbestos between subjects with typical pleural plaques and those with less-typical pleural plaques (P =.91 and P =.43, respectively). The characteristics of parietal pleural plaques for the various subjects are described in Table 2. More than 95% of subjects with parietal pleural plaques had plaques thicker than 2 mm and an extent greater than 1 cm. More than 60% of subjects with pleural plaques had several calcified plaques. Of the 1098 subjects with pleural plaques, 782 (71.2%) had typical pleural plaques. Very few patients had interstitial abnormalities consistent with asbestosis (n = 36, 0.7%) or diffuse pleural fibrosis (n = 87, 1.6%). JNCI Article Page 3 of 9

13 859 eligible subjects 7275 subjects with CT scan 5825 subjects with CT scan referred to the team centers 261 women 1 subject with missing data on duration of exposure to asbestos 53 subjects considered to be unexposed to asbestos by industrial hygienists 222 subjects with unreadable CT scan 1 subject with high suspicion of pleural tumor on CT scan Study population N = 5287 Figure 1. Study flow chart. CT = computed tomography. A total of 17 incident cases of pleural mesothelioma were diagnosed in this cohort between the date of CT scan and March 31, 2011 (one case was diagnosed in 2004, two in 2005, two in 2006, three in 2007, one in 2008, five in 2009, and three in 2010). Only one case was diagnosed in the year after CT scan. MESOPATH concluded that 14 of these cases were certain (13 were classified as epithelioid mesothelioma and one was classified as sarcomatoid mesothelioma), and the other three cases were considered uncertain or unclassifiable (generally because of inadequate material). No case was ruled out by MESOPATH. The review of available histological material for the determination of fibrohyaline pleural plaques on histology slides revealed that 13 of the 17 mesothelioma cases had pleural plaques on histology slides: seven of the 10 subjects with typical pleural plaques on CT scan, one of the two subjects with less-typical plaques on CT scan, and all five subjects without pleural plaques on CT scan. It should be emphasized that the absence of pleural plaques on histological material for two subjects with typical plaques on CT scan may be because of the very small size of the biopsy obtained by CT-guided pleural biopsy with only tumor material available. The characteristics of the subjects diagnosed with mesothelioma are summarized in Table 3. Mean age at the time of diagnosis of mesothelioma case subjects was 66.6 ± 6.6 years (range = 56 81 years). All mesothelioma subjects had an identified occupational exposure to asbestos, with a mean duration of exposure to asbestos of 33.7 ± 11.2 years (range = 3 46 years) and a mean latency (time elapsed between the beginning of the first job considered to have exposure to asbestos and the date of CT scan) of 47.2 ± 6.9 years (range = 33 58 years). Mesothelioma case subjects were slightly older and tended to have higher CEI to asbestos compared with subjects without mesothelioma. No mesothelioma case subject had interstitial abnormalities on CT scan. The unadjusted incidence of pleural mesothelioma according to the presence of pleural plaques presented in Figure 2 indicates that the evolution with age of the proportion of subjects without mesothelioma, represented by the survival curves, differed Page 4 of 9 Article JNCI

Table 1. Study population characteristics according to the presence of pleural plaques (N = 5287) Characteristic No plaques (n = 4189) Typical parietal pleural plaques* + diaphragmatic plaques (n = 802) Other less-typical plaques (n = 296) Age, years, mean ± SD 62.6 ± 5.5 65.4 ± 5.9 64.9 ± 6.1 <60 1004 (23.9%) 112 (14.0%) 45 (15.2%) 60 74 3079 (73.5%) 629 (78.4%) 229 (77.4%) 75 106 (2.6%) 61 (7.6%) 22 (7.4%) Smoking status Never smokers 1130 (27%) 163 (20.3%) 54 (18.2%) Ex-smokers 2422 (57.8%) 534 (66.6%) 198 (66.9%) Current smokers 299 (7.1%) 60 (7.5%) 23 (7.8%) Missing data 338 (8.1%) 45 (5.6%) 21 (7.1%) Duration of exposure to 30.5 ± 10.5 32.7 ± 9.7 32.6 ± 9.7 asbestos, years, mean ± SD 1 9 243 (5.8%) 32 (4%) 10 (3.4%) 10 19 433 (10.3%) 59 (7.4%) 20 (6.8%) 20 29 870 (20.8%) 127 (15.8%) 59 (19.9%) 30 39 1796 (42.8%) 377 (47.0%) 130 (43.9%) 40 847 (20.2%) 207 (25.8%) 77 (26%) Cumulative exposure index to asbestos (unit of exposure years) >0 3.2 967 (23.1%) 81 (10.1%) 39 (13.2%) 3.3 13.5 824 (19.7%) 81 (10.1%) 40 (13.5%) 13.6 31.9 841 (20.1%) 123 (15.3%) 44 (14.9%) 32 63.9 790 (18.8%) 183 (22.8%) 64 (21.6%) 64 767 (18.3%) 334 (41.6%) 109 (36.8%) * Parietal pleural plaques were considered to be typical when they were bilateral, thicker than 2 mm, and with an extent greater than 1 cm, regardless of whether they were calcified. Table 2. Characteristics of parietal pleural plaques (n = 1078 subjects) Unilateral Bilateral* All Characteristic n % n % n % Thickness, mm <2 27 10.0 26 3.2 53 4.9 2 4.9 195 72.2 453 56.1 648 60.1 5 9.9 39 14.4 277 34.3 316 29.3 10 9 3.3 52 6.4 61 5.7 Extent <1 cm 20 7.4 2 0.2 22 2.0 1 cm 24% of lateral chest wall 221 81.8 387 47.9 608 56.4 25% 49% of lateral chest wall 22 8.1 246 30.4 268 24.9 50% of lateral chest wall 7 2.6 173 21.4 180 16.7 Calcification No 141 52.2 279 34.5 420 39.0 Yes 129 47.8 529 65.5 658 61.0 * Parietal pleural plaques were considered to be typical when they were bilateral, thicker than 2 mm, and with an extent greater than 1 cm, regardless of whether they were calcified. statistically significantly among the three groups of subjects: subjects with typical pleural plaques, subjects with less-typical pleural plaques, and subjects without pleural plaques (log-rank test P <.0001). The Cox proportional hazards model revealed a statistically significant association between typical pleural plaques and pleural mesothelioma incidence in an unadjusted analysis (HR = 8.9, 95% confidence interval [CI] = 3.0 to 26.5) and after adjustment for TSFE and CEI to asbestos (HR = 6.8, 95% CI = 2.2 to 21.4) (Table 4). The risk of pleural mesothelioma also increased with increasing CEI (P =.06; HR = 1.4, 95% CI = 1.0 to 1.9). Less typical parietal pleural plaques were also associated with non-statistically significant increased risks of pleural mesothelioma. Similar associations were observed when the analysis was restricted to the 14 cases of mesothelioma certified by MESOPATH (crude HR for typical parietal or diaphragmatic pleural plaques = 6.5, 95% CI = 2.0 to 20.8; adjusted HR = 4.9, 95% CI = 1.5 to 16.6) after adjustment for TSFE and CEI to asbestos. Finally, no statistically significant interactions were observed between exposure variables and the presence or absence of pleural plaques as risk factor for mesothelioma (data not shown). Discussion The main finding of this study was the strong association between pleural plaques and pleural mesothelioma in a large cohort of JNCI Article Page 5 of 9

Table 3. Characteristics of mesothelioma patients and nonmesothelioma subjects Characteristic Mesothelioma (n = 17) Others (n = 5270) P* Age, years, mean ± SD 66.6 ± 6.6 63.1 ± 5.7.01 < 60 3 (17.6%) 1158 (22.0%).007 60 74 11 (64.8%) 3926 (74.5%) 75 3 (17.6%) 186 (3.5%) Smoking status Never smokers 4 (23.5%) 1343 (25.5%).07 Ex-smokers 7 (41.2%) 3147 (59.7%) Current smokers 2 (11.8%) 380 (7.2%) Missing data 4 (23.5%) 400 (7.6%) Duration of exposure to asbestos, years, mean ± SD 33.7 ± 11.2 31.0 ± 10.4.28 1 9 1 (5.9%) 284 (5.3%).66 10 19 1 (5.9%) 511 (9.7%) 20 29 2 (11.7%) 1054 (20.0%) 30 39 7 (41.2%) 2296 (43.6%) 40 6 (35.3%) 1125 (21.3%) Cumulative exposure index to asbestos (unit of exposure years) > 0 3.2 1 (5.9%) 1086 (20.6%).01 3.3 13.5 0 945 (17.9%) 13.6 31.9 7 (41.2%) 1001 (19.0%) 32 63.9 2 (11.8%) 1035 (19.6%) 64 7 (41.2%) 1203 (22.8%) Time since first exposure, years < 30 0 156 (3.0%).21 30 39 2 (11.8%) 1191 (22.6%) 40 49 8 (47.0%) 2874 (54.5%) 50 7 (41.2%) 1049 (19.9%) * Two-sided Fisher Freeman Halton test. Figure 2. Proportion of subjects without pleural mesothelioma at any given age according to the presence of pleural plaques on computed tomography (CT) scan (Kaplan Meier survival curve, log-rank test P <.0001). N = 34 091 subject-years. At-risk subjects at different ages were as follows for the different groups: subjects with no plaques on CT scan: 1870 at 65 years, 1646 at 70 years, 353 at 75 years, 113 at 80 years, and 23 at 85 years; subjects with typical parietal or diaphragmatic pleural plaques on CT scan: 297 at 65 years, 353 at 70 years, 146 at 75 years, 63 at 80 years, and 22 at 85 years; subjects with other less typical plaques on CT scan: 108 at 65 years, 134 at 70 years, 43 at 75 years, 22 at 80 years, and 8 at 85 years. Parietal pleural plaques were considered to be typical when they were bilateral, thicker than 2 mm, and with an extent greater than 1 cm, regardless of whether they were calcified. Page 6 of 9 Article JNCI

Table 4. Association between pleural plaques and mesothelioma: Cox model* HR (95% CI) CT scan pleural findings and asbestos exposure parameters Number of cases Unadjusted Adjusted for CEI Adjusted for duration Adjusted for TSFE Adjusted for TSFE and CEI No plaques on CT scan 5 1 (referent) 1 (referent) 1 (referent) 1 (referent) 1 (referent) Typical parietal or diaphragmatic 10 8.9 (3.0 to 26.5) 6.5 (2.1 to 20.2) 8.7 (2.9 to 26.2) 9.4 (3.1 to 28.3) 6.8 (2.2 to 21.4) pleural plaques on CT scan Other less-typical plaques on 2 4.9 (0.9 to 25.5) 3.8 (0.7 to 20.1) 4.8 (0.9 to 25.1) 5.1 (1.0 to 26.8) 4.0 (0.7 to 21.2) CT scan CEI NA 1.4 (1.0 to 1.9) 1.4 (1.0 to 2.0) Duration of exposure NA 1.0 (0.9 to 1.1) TSFE NA 1.0 (0.9 to 1.0) 0.9 (0.9 to 1.0) * CEI = cumulative exposure index to asbestos (Ln CEI + 1); CI = confidence interval; CT = computed tomography; HR = hazard ratio; TSFE = time since first exposure to asbestos; NA = not applicable; = parameter not included in the Cox model. Number of cases of pleural mesothelioma. Parietal pleural plaques were considered to be typical when they were bilateral, thicker than 2 mm, and with an extent greater than 1 cm, regardless of whether they were calcified. asbestos-exposed male subjects. The strengths of this study are the large number of subjects (n = 5287), individual estimation of cumulative occupational exposure to asbestos, accurate determination of pleural plaques detected on CT scan by thoracic radiology experts, and a rigorous procedure for certifying mesothelioma incident cases, with certification of certain mesothelioma for 14 of the 17 cases. Two possible weaknesses of this study need to be discussed. The first point concerns the possible misdiagnosis of mesothelioma as a pleural plaque. It should be emphasized that our definition of typical pleural plaques strictly required a quadrangular appearance and bilateral pleural abnormalities. It, therefore, appears highly improbable that some cases of mesothelioma could have been misdiagnosed as pleural plaques at the time of CT scan. Moreover, the association of mesothelioma with pleural plaques was statistically significant for typical pleural plaques. It should be noted that the hazard ratios associated with less-typical pleural plaques were also slightly elevated but not statistically significantly elevated. Finally, standardized CT scans did not reveal any other pleural abnormalities suggestive of possible mesothelioma, and all but one of the mesotheliomas occurred at least 1 year after baseline CT scan. The second point concerns the possibility of a differential mesothelioma misdiagnosis between the two groups (subjects with pleural plaques and those without pleural plaques). Identification of subjects developing pleural mesothelioma was based on the French National Health Insurance data and was strictly independent of CT scan data. The consulting physician must report all cases of mesothelioma to the French National Health Insurance system to ensure free medical care. The absence of report of a mesothelioma case for free medical care in this database would therefore be highly improbable. Moreover, cross-referencing with four local cancer registries (representing 25% of our population, with comprehensive case recording) did not reveal any new cases of mesothelioma. We, therefore, assume that the observed association between typical pleural plaques and incidence of pleural mesothelioma is valid. Regarding the role of chrysotile or amphiboles in the occurrence of asbestos-related pleural diseases, it was impossible to accurately determine the type of asbestos to which the subjects were exposed in this cohort. Because no autopsy was performed, it was also impossible to determine the type of asbestos and the asbestos load in subjects with mesothelioma in this study. Overall, the frequency of pleural plaques was only 20.4% among the subjects in this cohort. This fairly moderate frequency probably reflects the fact that the subjects had worked in various industries with various levels of cumulative asbestos exposure. Nevertheless a dose response relationship for the occurrence of pleural plaques has already been reported, with some industrial sectors and jobs associated with higher frequencies of pleural plaques in this cohort, up to more than 35% (12). To the best of our knowledge, no previous study has documented the association of pleural plaques and pleural mesothelioma using CT scan for the detection of pleural plaques. Indeed, several published studies have focused on the association between pleural plaques and mesothelioma, mainly pathological studies and studies based on chest x-ray. Two Swedish studies in asbestosexposed shipyard or construction workers reported no association between pleural plaques and pleural mesothelioma (20,21). In contrast, other authors have reported an excess risk of mesothelioma linked to the observation of pleural plaques on chest x-ray (8 10). However, in two of these studies (8,9) subjects with pleural plaques were compared with the general population. It was, therefore, not surprising to observe an excess of mesothelioma in the group with pleural plaques because all subjects had been exposed to asbestos, whereas asbestos exposure was presumably much less frequent in the general population. All radiological assessments in these studies were based on chest x-ray, which is known to have poor sensitivity and specificity for the detection of pleural plaques (22). Moreover, only one study took into account the TSFE (latency), age, and cumulative exposure to asbestos (10). In the follow-up of 1988 former workers of a crocidolite mine in Wittenoom, Australia (10), the authors found an increased risk of peritoneal mesothelioma, but not of pleural mesothelioma, associated with pleural thickening on chest x-ray. Autopsy studies have reported a high frequency (up to more than 70%) of pleural plaques in subjects with pleural mesothelioma (23 26). However, such studies are difficult to interpret for the link between pleural plaques and pleural mesothelioma because they JNCI Article Page 7 of 9

concern subjects who sometimes had high levels of exposure to asbestos (26). Moreover, in a series of 288 autopsies in men with sudden and unexpected death in Helsinki, Finland, pleural plaques were detected in 58% of subjects, indicating that the presence of pleural plaques at autopsy may be a common finding in the general population in some countries (27). Nevertheless and despite their limitations, pathological studies are useful to understand the potential mechanisms involved after inhalation of asbestos fibers and their translocation to the pleura with subsequent pathological reactions (28). It has been demonstrated that asbestos fibers translocate to pleural spaces with local accumulation in black spots, with potential subsequent induction of pleural fibrosis and tumors (29). The hypothesis of a link between pleural accumulation of fibrous particles and subsequent development of pleural fibrosis and cancer has been recently enhanced by experimental data in laboratory animals with asbestos fibers and carbon nanotubes that suggest a link between accumulation of fibers in pleural spaces and the subsequent development of pleural fibrosis and cancer (30). Pleural plaques might reflect a local (pleural) response to accumulation of fibers, with other potential pleural responses (including induction of tumor) induced by these fibers, although a specific study failed to establish a relationship between the predominant locations of black spots and hyaline pleural plaques in humans (31). Importantly, a specific review of available histological material in our series was performed by a single pathologist (F. Galateau-Sallé) for determination of fibrohyaline pleural plaques on histological slides. This review revealed that 13 of 17 mesothelioma case subjects had pleural plaques on histological slides, which indicates a close histological relationship between these two diseases. The topic of screening workers previously exposed to asbestos remains controversial. Very few screening programs have been conducted in asbestos workers using low-dose CT scan (32 35). These studies mainly focused on early detection of lung cancer, and no conclusions can be drawn on the link between pleural plaques and pleural mesothelioma because of their small sample sizes. It was recently reported that low-dose CT screening is associated with reduced lung cancer mortality in a cohort of heavy smokers (36). It, therefore, seems important to establish whether subjects with pleural plaques should also be considered to be at high risk of lung cancer and might also benefit from low-dose CT screening, but this remains a controversial issue at the present time (6,11). In the field of pleural mesothelioma, apart from the knowledge of previous occupational exposure to asbestos and the level of cumulative exposure, it is useful to identify additional criteria, including certain laboratory markers, to define high-risk populations, particularly those eligible for screening. However, in view of the frequency and prognosis of pleural mesothelioma at the present time, we believe that large-scale specific screening for pleural mesothelioma cannot be recommended in subjects previously exposed to asbestos with pleural plaques. The results of this study could be used for medicosocial purposes, such as compensation of asbestos-related diseases according to the specific rules used in each country, particularly for estimation of the health risks linked to this benign disease. References 1. Goldberg M, Banaei A, Goldberg S, et al. Past occupational exposure to asbestos among men in France. Scand J Work Environ Health. 2000;26(1):52 61. 2. Becklake MR, Bagatin E, Neder JA. Asbestos-related diseases of the lungs and pleura: uses, trends and management over the last century. Int J Tuberc Lung Dis. 2007;11(4):356 369. 3. Jamrozik E, de Klerk N, Musk AW. Asbestos-related disease. Intern Med J. 2011;41(5):372 380. 4. Anonymous. Diagnosis and initial management of nonmalignant diseases related to asbestos. Am J Respir Crit Care Med. 2004;170(6):691 715. 5. Berman DW, Crump KS. Update of potency factors for asbestos-related lung cancer and mesothelioma. Crit Rev Toxicol. 2008;38(Suppl 1):1 47. 6. Banks DE, Shi R, McLarty J, et al. American College of Chest Physicians consensus statement on the respiratory health effects of asbestos. Results of a Delphi study. Chest. 2009;135(6):1619 1627. 7. Hillerdal G, Henderson DW. Asbestos, asbestosis, pleural plaques and lung cancer. Scand J Work Environ Health. 1997;23(2):93 103. 8. Hillerdal G. Pleural plaques and risk for bronchial carcinoma and mesothelioma. A prospective study. Chest. 1994;105(1):144 150. 9. Karjalainen A, Pukkala E, Kauppinen T, Partanen T. Incidence of cancer among Finnish patients with asbestos-related pulmonary or pleural fibrosis. Cancer Causes Control. 1999;10(1):51 57. 10. Reid A, de Klerk N, Ambrosini G, et al. The additional risk of malignant mesothelioma in former workers and residents of Wittenoom with benign pleural disease or asbestosis. Occup Environ Med. 2005;62(10):665 669. 11. Ameille J, Brochard P, Letourneux M, Paris C, Pairon J-C. Asbestosrelated cancer risk in patients with asbestosis or pleural plaques. Rev Mal Respir. 2011;28(6):e11 e17. 12. Paris C, Thierry S, Brochard P, et al. Pleural plaques and asbestosis: dose- and time-response relationships based on HRCT data. Eur Respir J. 2009;34(1):72 79. 13. Ameille J, Letourneux M, Paris C, et al. Does asbestos exposure cause airway obstruction, in the absence of confirmed asbestosis? Am J Respir Crit Care Med. 2010;182(4):526 530. 14. Clin B, Paris C, Ameille J, et al. Do asbestos-related pleural plaques on HRCT scans cause restrictive impairment in the absence of pulmonary fibrosis? Thorax. 2011;66(11):985 991. 15. Hansell DM, Bankier AA, MacMahon H, et al. Fleischner Society: glossary of terms for thoracic imaging. Radiology. 2008;246(3):697 722. 16. Beigelman-Aubry C, Ferretti G, Mompoint D, et al. [Computed tomographic atlas of benign asbestos related pathology]. J Radiol. 2007;88(6):845 862. 17. International Labour Office (ILO). Guidelines for the Use of the ILO International Classification of Radiographs of Pneumoconioses, Revised Edition 2000. Occupational Safety and Health Series, no. 22 (Rev 2000). Geneva, Switzerland: International Labour Office; 2002. 18. Jarad NA, Wilkinson P, Pearson MC, Rudd RM. A new high resolution computed tomography scoring system for pulmonary fibrosis, pleural disease, and emphysema in patients with asbestos related disease. Br J Ind Med. 1992;49(2):73 84. 19. Goldberg M, Imbernon E, Rolland P, et al. The French National Mesothelioma Surveillance Program. Occup Environ Med. 2006;63(6):390 395. 20. Sandén A, Järvholm B. A study of possible predictors of mesothelioma in shipyard workers exposed to asbestos. J Occup Med. 1991;33(7):770 773. 21. Koskinen K, Pukkala E, Martikainen R, Reijula K, Karjalainen A. Different measures of asbestos exposure in estimating risk of lung cancer and mesothelioma among construction workers. J Occup Environ Med. 2002;44(12):1190 1196. 22. Aberle DR, Balmes JR. Computed tomography of asbestos-related pulmonary parenchymal and pleural diseases. Clin Chest Med. 1991;12(1):115 131. 23. Bianchi C, Brollo A, Ramani L, Zuch C. Pleural plaques as risk indicators for malignant pleural mesothelioma: a necropsy-based study. Am J Ind Med. 1997;32(5):445 449. 24. Gemba K, Fujimoto N, Kato K, et al. National survey of malignant mesothelioma and asbestos exposure in Japan. Cancer Sci. 2012;103(3):483 490. 25. Klebe S, Brownlee NA, Mahar A, et al. Sarcomatoid mesothelioma: a clinical-pathologic correlation of 326 cases. Mod Pathol. 2010;23(3):470 479. 26. Roggli VL, Sharma A, Butnor KJ, Sporn T, Vollmer RT. Malignant mesothelioma and occupational exposure to asbestos: a clinicopathological correlation of 1445 cases. Ultrastruct Pathol. 2002;26(2):55 65. 27. Karjalainen A, Karhunen PJ, Lalu K, et al. Pleural plaques and exposure to mineral fibres in a male urban necropsy population. Occup Environ Med. 1994;51(7):456 460. Page 8 of 9 Article JNCI

28. Broaddus VC, Everitt JI, Black B, Kane AB. Non-neoplastic and neoplastic pleural endpoints following fiber exposure. J Toxicol Environ Health B Crit Rev. 2011;14(1 4):153 178. 29. Boutin C, Dumortier P, Rey F, Viallat JR, De Vuyst P. Black spots concentrate oncogenic asbestos fibers in the parietal pleura. Thoracoscopic and mineralogic study. Am J Respir Crit Care Med. 1996;153(1):444 449. 30. Donaldson K, Murphy FA, Duffin R, Poland CA. Asbestos, carbon nanotubes and the pleural mesothelium: a review of the hypothesis regarding the role of long fibre retention in the parietal pleura, inflammation and mesothelioma. Part Fibre Toxicol. 2010;7:5. 31. Mitchev K, Dumortier P, De Vuyst P. Black spots and hyaline pleural plaques on the parietal pleura of 150 urban necropsy cases. Am J Surg Pathol. 2002;26(9):1198 1206. 32. Das M, Mühlenbruch G, Mahnken AH, et al. Asbestos Surveillance Program Aachen (ASPA): initial results from baseline screening for lung cancer in asbestos-exposed high-risk individuals using low-dose multidetector-row CT. Eur Radiol. 2007;17(5):1193 1199. 33. Mastrangelo G, Ballarin MN, Bellini E, et al. Feasibility of a screening programme for lung cancer in former asbestos workers. Occup Med (Lond). 2008;58(3):175 180. 34. Tiitola M, Kivisaari L, Huuskonen MS, et al. Computed tomography screening for lung cancer in asbestos-exposed workers. Lung Cancer. 2002;35(1):17 22. 35. Roberts HC, Patsios DA, Paul NS, et al. Screening for malignant pleural mesothelioma and lung cancer in individuals with a history of asbestos exposure. J Thorac Oncol. 2009;4(5):620 628. 36. Aberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365(5):395 409. Funding This work was supported by the French National Health Insurance (Occupational Risk Prevention Department); French Ministry of Labour and Social Relations; and the French Agency for Food, Environmental and Occupational Health & Safety (ANSES grant 07-CRD-51 and EST 2006/1/43). Notes All authors declare that they have no conflicts of interest. The sponsors of the study had no role in study design (except in the choice of the regions of France where the study was conducted), data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. We would like to thank P. Wild for his helpful statistical advice, the members of National Reference Center MESOPATH (I. Abd-Al-Samad, H. Begueret, E. Brambilla, F. Capron, M.C. Copin, C. Danel, A.Y. De Lajartre, A. Foulet- Roge, L. Garbe, O. Groussard, V. Hofman, S. Lantuejoul, J.M. Picquenot, I. Rouquette, C. Sagan, F. Thivolet-Bejui, J.M. Vignaud) for their contribution to mesothelioma certification, the chest radiologists who participated in the reading of CT scans (Y. Badachi, C. Beigelman-Aubry, A. Jankowski, V. Latrabe, M. Montaudon), and the members of the Association of the French Cancer Registries FRANCIM (S. Bara, F. Colombani, M. Colonna, A.V. Guizard, G. Launoy) for the data provided. We also thank the other members of the asbestos postexposure program for their contribution to the study design or data collection in the APEXS screening program: E. Abboud, B. Aubert, J. Baron, J. Benichou, A. Bergeret, A. Caillet, P. Catilina, G. Christ de Blasi, F. Conso, E. Guichard, N. Le Stang, M.F. Marquignon, M. Maurel, B. Millet, L. Mouchot, M. Pinet, A. Porte, J.L. Rehel, P. Reungoat, R. Ribero, M. Savès, A. Sobaszek, A. Stoufflet, F.X. Thomas, L. Thorel, and National Health Insurance personnel (Aquitaine, Haute-Normandie, Basse-Normandie, and Rhône-Alpes). Affiliations of authors: INSERM, U955, Créteil, France (JCP, PA); Université Paris-Est Créteil, France (JCP, PA); Centre Hospitalier Intercommunal, Service de pneumologie et pathologie professionnelle, Créteil, France (JCP, PA); Centre cardiothoracique INSERM 1045, Bordeaux, France (FL); Centre INSERM 897, Bordeaux, France (MR, PB); Université Segalen Bordeaux, CHU de Bordeaux, France (FL, MR, PB); Cancers et Populations, INSERM U1086, France (BC, FGS); CHU Caen, Service de santé au travail et pathologie professionnelle, Caen, France (BC, ML); MESOPATH National Reference Center (FGS); Faculté de Médecine de Caen, France (BC, ML, FGS); AP-HP, Hôpital Raymond Poincaré, Unité de pathologie professionnelle, Garches, France (JA); Institut Interuniversitaire de Médecine du Travail de Paris-Ile de France, Paris, France (SC); INSERM U823, Grenoble, France (GF); Université J Fourrier, Grenoble, France (GF); CHU Grenoble, Clinique universitaire de radiologie et imagerie médicale, Grenoble, France (GF); CHU Rouen, Service des maladies professionnelles, Rouen, France (AG); INSERM U954, Nancy, France (AL, CP); Université de Lorraine, Faculté de Médecine de Nancy, Nancy, France (CP); CHU Nancy, Nancy, France (AL, CP); ERSM Rhône-Alpes, Lyon, France (ES). JNCI Article Page 9 of 9