Malignant pleural mesothelioma risk among nuclear ...

Home

Search

Collections

Journals

About

Contact us

My IOPscience

Malignant pleural mesothelioma risk among nuclear workers: a review

This content has been downloaded from IOPscience. Please scroll down to see the full text. 2011 J. Radiol. Prot. 31 9 (http://iopscience.iop.org/0952-4746/31/1/R01) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 212.234.146.173 This content was downloaded on 11/01/2016 at 14:57

Please note that terms and conditions apply.

IOP PUBLISHING

JOURNAL OF RADIOLOGICAL PROTECTION

J. Radiol. Prot. 31 (2011) 9–23

doi:10.1088/0952-4746/31/1/R01

REVIEW

Malignant pleural mesothelioma risk among nuclear workers: a review C Metz-Flamant1 , I Guseva Canu and D Laurier Laboratory of Epidemiology, Institute of Radiological Protection and Nuclear Safety (IRSN), Fontenay-aux-Roses, France E-mail: [email protected]

Received 7 June 2010, in final form 30 August 2010, accepted for publication 27 October 2010 Published 23 February 2011 Online at stacks.iop.org/JRP/31/9 Abstract Exposure to ionising radiation has been suggested as a causal risk factor for malignant pleural mesothelioma (MPM). Studies of patients treated by radiotherapy for primary cancers have suggested that radiation contributes to the development of secondary MPM. Here we examined the risk to nuclear workers of MPM related to exposure to low doses of occupational radiation at low dose rates. All results concerning MPM risk in published studies of nuclear workers were examined for their association with radiation exposure and potential confounders. We found 19 relevant studies. Elevated risks of pleural cancer were reported in most (15/17) of these studies. Eight reported risks higher for radiation monitored workers than for other workers. However, of 12 studies that looked at associations with ionising radiation, only one reported a significant dose–risk association. Asbestos was an important confounder in most studies. We conclude that studies of nuclear workers have not detected an association between ionising radiation exposure and MPM. Further investigations should improve the consideration of asbestos exposure at the same time as they address the risk of MPM related to occupational exposure of nuclear workers to low doses of ionising radiation at low dose rates. (Some figures in this article are in colour only in the electronic version)

1. Introduction Pleural mesothelioma is the most common type of malignant mesothelioma, accounting for about 70% of all cases [1]. Incidence rates rose steeply until the 1990s in most European countries and in the United States [2, 3], mainly due to the widespread use of asbestos at the 1 Address for correspondence: IRSN/DRPH/SRBE/LEPID, BP17, 92262 Fontenay-aux-Roses Cedex, France.

0952-4746/11/010009+15$33.00 © 2011 IOP Publishing Ltd

Printed in the UK

10

C Metz-Flamant et al

beginning of the last century [4]. Although asbestos exposure remains the major risk factor for pleural mesothelioma, it fails to explain the origin of about 20% of cases [2]. Several reviews have suggested that radiation exposure may be involved in the onset of malignant mesothelioma [1, 5–7]. Evidence from numerous studies demonstrates the development of malignant mesothelioma in organs close to areas of the body treated with therapeutic radiation. In particular, post-radiation malignant mesothelioma has been reported in several large-scale retrospective cohort studies of patients after radiation therapy for breast cancer and Hodgkin’s disease, both of which require relatively high doses of radiation delivered at a high dose rate to specific tissues [5]. Epidemiological data about the effects of occupational exposure at low doses and low dose rates are very sparse. Studies of nuclear plant workers are especially interesting because many of them are exposed in this way, and their exposure is carefully monitored over time with personal dosimeters. It was recently suggested that these workers’ radiation exposure plays a causal role in the development of malignant pleural mesothelioma (MPM) [5]. In view of the steadily growing demand for nuclear energy, which implies the employment of hundreds of thousands of workers around the world likely to be exposed to radiation, this is an important public health issue. This review therefore aims to assess the effects of low doses of ionising radiation on the risk of MPM in nuclear workers.

2. Methods 2.1. Definition of malignant pleural mesothelioma Until 1999 all malignant neoplasms of the pleura were coded with malignant neoplasms of the respiratory and intrathoracic organs according to their point of origin, without morphological distinction. Accordingly, most studies of nuclear workers report results for ‘pleural cancer’ (code 163 of the International Classification of Diseases, ninth revision; ICD-9). Mesothelioma became a separate morphological group of malignant neoplasms only in 1999, in the tenth revision (ICD-10, code C45) [8]. This includes MPM (C45.0) but excludes other malignant neoplasms of the pleura (C38.4). Only a few studies of nuclear workers have focused on MPM as a disease variable. In those cases, results are often listed together with those for many other cancers and health outcomes and would therefore be difficult to find in a key word search.

2.2. Search strategy We used the following keywords to search the epidemiological literature: cancer, malignancy, pleural cancer, mesothelioma, ionising radiation, occupational exposure and nuclear workers. Research was performed in the Medline and Scopus databases for the period 1980–2009. Each paper was then assessed for relevant information on MPM risk and its description of its nuclear workers. Several studies included results about pleural cancer or mesothelioma among a large number of results for different cancer sites. Because many studies of nuclear workers were not found by the keyword search, we conducted additional research in all published epidemiological studies of nuclear workers, checking the completeness of the bibliographic selection against the reference list from the last United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) report [9], which included the most important of these studies. When several studies focused on the same population, only the last one was used for reporting results. Nevertheless, previous studies were reviewed for additional information (dosimetry, asbestos exposure, . . .).

Review

11

2.3. Evaluation criteria for studies The capacity of the selected studies to detect associations between occupational exposure to ionising radiation and MPM was evaluated according to three criteria: (1) Ability to evaluate the risk of MPM. The number of cases of or deaths from MPM is important information that indicates the study’s power. Descriptive results are often reported as the standardised mortality ratio (SMR) or standardised incidence ratio (SIR), calculations that compare the mortality or incidence of the cohort to that of the national population reference, usually standardised for sex, age and calendar year. This estimator makes it possible to detect a statistical excess of MPM in the study population, should it exist. Several studies also assessed risk by radiation status, that is, whether workers were thought to be exposed to occupational radiation and required to wear dosimeters (radiation workers are defined as monitored workers; they are compared to ‘non-radiation’, i.e. non-monitored workers). This distinction was retained only as a descriptive result: radiation worker status is not a sufficient criterion for assessing risk according to ionising radiation exposure. (2) Precision of exposure measurements for examining association with ionising radiation. The standard method for assessing the effect of ionising radiation consists of a dose– response analysis that computes the estimated risk by level of radiation exposure. The most informative studies therefore quantify the risk for a given dose, for example by calculating the excess relative risk (ERR) per sievert (Sv). (3) Assessment of potential confounders, mainly asbestos. Asbestos is currently the only well-established specific occupational risk factor for MPM [1, 6]. Because it was widely used in the early years of the nuclear industry, especially for thermal insulation, many workers may have had jobs that led to nonnegligible asbestos exposure. We therefore focused specifically on the available information reported in published papers on potential asbestos exposure. 3. Results 3.1. Search results We found 19 studies that reported results for pleural cancer or MPM (table 1). Most of them included workers in facilities producing nuclear power or specialising in activities such as research, waste management or the production of nuclear fuel, isotopes and weapons. Only one considered workers from naval shipyards. Most subjects were men; the percentage of women varied between studies from 0 to 34%. Both radiation and non-radiation workers were generally included in the studies [10–22]. 3.2. Disease variable As table 1 shows, the specific disease variable examined varied between studies. Two US studies did not distinguish between pleural and peritoneal cancer [21, 23], another focused on mesothelioma [15], and the Australian studies of nuclear workers [13, 14] considered cancer of the pleura and other thoracic organs as a whole. Nevertheless, outcome was generally defined homogeneously, and most of the studies (14/19) focused on pleural cancer (ICD9 code 163). Of the 19 published studies, one focused on incidence data [14] and five presented results based on mortality as well as incidence [16–19, 24]. For these five studies, the incidence results were not detailed in this review because of their similarity to the mortality results. All the other studies reported only mortality data.

Cohort and countrya

Main activities

Study period

Number of workers (percentage of women)

Average length of follow-up in years

Percentage of radiation workers

Research

1972–1998

4 717 (27.9)

17.4

53

Research

1974–1996

4 523 (27.7)

16.2

53

Research

1968–1994

58 023 (23.3)

23

—c

CEA-Areva NC [34]

Research, fuel reprocessing and manufacturing

1968–1994

29 204 (21.3)

18

100

Areva NC Pierrelatte [12]

Uranium enrichment and conversion

1968–2005

2 709 (0)

Research, reprocessing and storing spent fuel and military applications Research, reprocessing and storing spent fuel and military applications Plutonium processing and reprocessing Uranium processing, purification and enrichment

1946–1988

United Kingdom Sellafield + AEA + AWE [11] Sellafield + AEA + AWE [26] Sellafield [19] Capenhurst [17]

Disease variable

Workers ever employed between 1972 and 1998 Workers ever employed between 1957 and 1998

Pleura and other thoracic organs, ICD9 codes:163, 164

Workers employed for more than 1 year between 1946 and 1994 Monitored workers ever employed between 1946 and 1994 Male workers employed for more than 6 months between 1960 and 2005

Pleural cancer ICD9 code 163 Pleural cancer ICD9 code 163

28.0

53

75 006 (25)

24

54

Workers employed at one of the three sites since 1946

Pleural cancer ICD9 code 163

1947–1988

40 761 (8)

—

100

Workers employed at one of the three sites since 1946


1947–1993

14 385 (19)

29

72

1946–1995

12 540 (11)

26.7

26

Workers employed by BNFL between 1947 and 1976 Workers ever employed before 1996




Australia LHSTC Mortality study [13] LHSTC Incidence study [14] France CEA [22]

Study population

12

Table 1. Description of the studies providing information about malignant pleural mesothelioma risk among nuclear workers. (Note: LHSTC, Lucas Heights Science and Technology Centre; CEA, Commissariat a` l’Energie Atomique; Areva NC, French nuclear cycle company; AEA, Atomic Energy Agency; AWE, Atomic Weapons Establishment; NRRW, National Registry for Radiation Workers; INEEL, Idaho National Engineering and Environmental Laboratory; BNFL, British Nuclear Fuels Ltd (now plc).)

Review

Table 1. (Continued.) Number of workers (percentage of women)

Average length of follow-up in years

Percentage of radiation workers

Cohort and countrya

Main activities

Study period

Chapelcross [18]

1955–1995

2 628 (14)

24.3

84

Springfields [16]

Reactors operation and tritium production Uranium processing

1946–1995

19 454 (12)

24.6

72

AEA [10]

Research

1946–1997

51 367 (29)

26.7

51

NRRW [24]

All activities including medical field

1955–2001

174 541 (10)

Research, nuclear reactor design and testing, chemical processing, research, and testing of navy ship reactors Uranium processing

1949–1999

63 129 (18)

1948–1999

5 801 (8)

28

100

Uranium processing

1950–2002

18 883 (19)

34

—

Naval shipyard

1957–1982

71 815 (0)

13

54

All activities except energy production

1943–1988

95 673 (15)

22

100

All activities

Differed between studies

12.7

100

United States INEEL [21]

Rocketdyne Atomics International [23] Savannah River Site [20] Naval shipyard [15] International 3-country [27]

15-country [28, 35]

21.1b

All results are based on mortality data, except the LHSCT incidence study. b Median. c Not reported.

100

57.3

Disease variable

Workers ever employed before 1996 Workers ever employed before 1996 Workers ever employed between 1946 and 1979 Workers included in the NRRW beginning radiation work in or after 1976

Pleural cancer ICD9 code 163 Pleural cancer ICD9 code 163 Pleural cancer ICD9 code 163 Pleural cancer ICD9 code 163

Workers ever employed by the Department of Energy at INEEL between 1949 and 1991 Workers employed for more than 6 months after 1948

Cancer of pleura and peritoneum, ICD9 codes: 158, 163 Cancer of pleura and peritoneum, ICD9 codes: 158.8, 158.9, 163 Pleural cancer ICD9 code 163 Mesothelioma, ICD codes not available

Workers employed for more than 3 months before 1987 Male workers employed for more than 1 year in one of the eight shipyards after 1957 Workers employed for more than 6 months by nuclear industry after 1943 Workers employed for more than 1 year with exclusively external exposure


13

a

407 391 (10)

22

Study population

14


Figure 1. Risk of malignant pleural mesothelioma among nuclear workers in epidemiological studies. a Incident cases and SIR instead of deaths and SMR.

3.3. Risk of malignant pleural mesothelioma among nuclear workers Figure 1 shows the number of observed deaths and the SMRs for MPM and their associated 95% confidence intervals (CIs) in the epidemiological studies of nuclear workers. The LHSTC incidence study [14] reported the number of cases and the SIR. When the publication did not report the confidence intervals, we calculated the 95%CI as described in [25]. The joint study of Sellafield, AEA and AWE workers [26] did not report a combined SMR for the entire cohort. Therefore, we include in figure 1 for this study the SMR for radiation workers who were not monitored for internal radiation exposure. None of the two international workers studies [27, 28] is included in the figure, as they did not calculate combined SMRs; the 3country study [27] included 20 pleural cancer deaths and the 15-country study 39 [28]. Although most of the studies included a population of more than a thousand workers, with a follow-up often longer than 20 years (table 1), there were relatively few observed cases or deaths. Only four studies reported 20 or more deaths from pleural cancer [22, 24, 27, 28]. Nonetheless, the risk of this cancer was elevated in most studies (15/17), including nine that reported a significant excess [13–15, 17, 19–22, 24] (figure 1). Eleven studies calculated SMR or SIR by radiation status (figure 2). As in figure 1, when the publication did not provide confidence intervals, we calculated the 95%CI as described in [25]. Studies were not included in this figure unless they reported SMRs or SIRs for both radiation and non-radiation workers [10–19, 23]. Among these eleven, eight studies reported a higher risk of pleural cancer in radiation workers than in their non-radiation counterparts [10, 11, 13, 15–19].

Review

15

Figure 2. Comparison of the risk of malignant pleural mesothelioma between radiation and nonradiation workers. a Incident cases and SIR instead of deaths and SMR. Radiation workers. Non-radiation workers.

3.4. Association between malignant pleural mesothelioma and ionising radiation exposure Table 2 describes workers’ exposure to external and internal ionising radiation and the risk of MPM associated with ionising radiation exposure in the studies. Dose–response analyses conducted in these studies were usually based on external doses only, usually expressed as a whole-body equivalent external dose in millisievert (mSv). Sometimes it also included the tritium (beta emitter) dose, since tritium is uniformly distributed throughout the body. The average cumulative career-long external dose per worker ranged from 8.3 to 130 mSv between studies. Of 19 studies reporting MPM results, 12 included a dose–response analysis or computed trend tests according to radiation doses (table 2), and only one [16] of them reported a significant association between radiation dose and MPM ( p = 0.01), driven by one case. As shown in table 2, all analyses are based on very few numbers of MPM deaths, except the latest analysis on UK National Registry for Radiation Workers (NRRW) including 112 deaths. Even the 3-country and 15-country international combined studies are based respectively on only 20 and 39 deaths. Only two studies quantified the risk for a given dose [24, 28] and both found an ERR greater than 1 but not statistically significant and associated with large confidence intervals. Workers in most of the studies were potentially exposed to internal radiation (table 2). Only three studies conducted analyses according to internal monitoring status [19, 23, 26], and only two estimated internal doses [19, 23]. In the Sellafield study, trend tests were computed with both external doses alone and with external and internal doses, and the results were similar [19]. One study partially assessed the effect of internal exposure on MPM risk by calculating SMR

16 Table 2. Estimated risk of malignant pleural mesothelioma associated with exposure to ionising radiation. (Note: LHSTC, Lucas Heights Science and Technology Centre; CEA, Commissariat a` l’Energie Atomique; Areva NC, French nuclear cycle company; AEA, Atomic Energy Agency; AWE, Atomic Weapons Establishment; NRRW, National Registry for Radiation Workers; INEEL, Idaho National Engineering and Environmental Laboratory.) External radiation exposure Cohort and country Australia LHSTC mortality study [13] LHSTC incidence study [14] France CEA [22]

Association of MPMa with cumulative doseb of ionising radiation

Dose estimationc

—d —

Mean cumulative dose (mSv)

15.04 Photon

18.7

—

Internal contamination

Main radionuclides involved

Dose estimation

Mainly uranium and thorium radionuclides, fission and activation products and transuranic elements

Not included in dose

Not assessed

9 deaths, p-trend = 0.43

Photon

8.3

Mainly uranium and plutonium

Identification of the potentially exposed workers Not included in dose

Areva NC Pierrelatte [12]

—

Photon

1.7

Natural and reprocessed uranium compounds

Identification of the potentially exposed workers Not included in dose

17 deaths, p-trende = 0.59

Photon, neutron (