IMAGING OF MESOTHELIOMA

Introduction

Malignant pleural mesothelioma (MPM) is an asbestos-related neoplasm that is refractory to current therapies and associated with poor prognosis. The disease originates in pleural mesothelial cells and progresses locally along the pleural reflections until it encases the lungs and mediastinum, ultimately causing death. MPM has been designated as a worldwide epidemic, which is predicted to peak in the next decade (2015–2019) in most Western countries Patients with mesothelioma have an average survival of 7–12 months; however, trimodality therapy with cytoreductive surgery followed by radiotherapy and chemotherapy can prolong survival  The three distinct histologic subtypes – epithelial, sarcomatoid (sarcomatous), and mixed (biphasic) – cannot be distinguished by imaging. Even though contrast-enhanced CT is the preferred technique for evaluating suspected malignant pleural disease, histological sampling and immunohistocytochemistry can only reliably

diagnose MPM. The complex morphology and growth pattern of MPM make it an imaging enigma. This chapter aims to highlight the practical aspects of imaging of MPM with an emphasis on guiding management.

Patterns of Presentation and Imaging Features

MPM has varied and nonspecific imaging appearances ranging from pleural effusion, focal pleural thickening, diffuse circumferential pleural thickening, pleural nodularity to pleural masses. Calcified and noncalcified bilateral pleural plaques coexist with pleural thickening. Pleural thickening can be focal or circumferential

and extends along the mediastinal, diaphragmatic surface of the pleura and along fissures. Nodal involvement and contiguous invasion of adjacent chest wall and direct intra-diaphragmatic extension can be seen in later stages. Contralateral disease can be in the form of pleural effusion or pulmonary nodules. Brain and osseous metastases can be seen in later stages, as well. The constellation of findings ranges from unilateral pleural effusion, circumferential nodular pleural thickening, pleural masses, and invasion of adjacent structures, to adenopathy, osseous, pulmonary and distant metastases in the later stages. Pleural thickening and/or effusions also represent early presentation and are nonspecific without histological confirmation. Rind-like circumferential pleural thickening is seen as the disease progresses, with the disease process often starting from the diaphragmatic surface of the pleura extending upward. Apical involvement is considered a bad prognostic factor and is seen in later stages. Volume loss and mediastinal shift can be seen secondary to encasement of the lung. Sixty percent of the time the disease is seen on the right and is only bilateral in 10% cases. Biphasic and sarcomatoid subtypes have more aggressive behavior and can present with distant and osseous metastases in early stages of the disease.

Preoperative Evaluation of MPM
MPM patients are considered surgical candidates if the disease is confined to the ipsilateral hemithorax and there is no evidence of spread to mediastinal lymph nodes (N = 0) or distant metastases (M = 0). Current methods for predicting resectability of patients undergoing extrapleural pneumonectomy for macroscopic complete resection of MPM are limited. Despite improvements in diagnostic imaging over several decades, the proportion of patients who are unable to complete resection after thoracotomy remains high at 25%  Using current methods of preoperative evaluation for patients with malignant pleural mesothelioma, evidence of local invasion of contiguous structures, transdiaphragmatic or transmediastinal invasion, and diffuse chest wall invasion are clear indicators of unresectability. Computed tomography (CT) is the mainstay in preoperative evaluation and is complemented by magnetic resonance imaging and 18F-FDG positron tomography Plain radiography plays a limited role due to varied and nonspecific appearances ranging from pleural effusion to lobulated pleural thickening and pleural masses. Pleural plaques, the hallmark of asbestos exposure, further limit evaluation on radiographs and can potentially obscure contralateral involvement and can obscure pulmonary nodules. CT continues to be the initial and primary modality for diagnosis, staging, and monitoringof therapeutic response in MPM [25]. Even though CT can easily depict the overall extent of

the pleural abnormality, early chest wall invasion, peritoneal involvement, and lymph node metastases can be challenging even on a contrast- enhanced CT scan. Subtle transdiaphragmatic extension can also be difficult to identify on CT.
CT image data can also be effectively reconstructed in three-dimensional planes to yield multi-planar reformats and volume rendered images to simulate the anatomical detail for surgical planning. Three-dimensional (3-D) volume rendered images are increasingly becoming popular to show association with adjacent structures and encasement or encroachment of vascular structures by the tumor [13, 15]. Maximum intensity projections depict the course of vessels encased by the pleural rind and are helpful during
surgery. The 3-D images are intuitive and provide the surgeons an overview of the tumor in vitro, thereby aiding the surgeons during resection. These images also provide patients an overview and extent of their disease during management discussions.
Furthermore, volumetric assessment of MPM can be easily acquired by serially segmenting the tumor using Hounsfield thresholding. Tumor and lung volumes can be generated and have been proven to be prognostically significant.
Ultrasound has a limited role in diagnosis and management of MPM; however, the fluid attenuation of the tumor provides a diagnostic window for the ultrasound, thus enabling ultrasound-guided biopsy and thoracentesis, thereby improving the diagnostic yield of pleural biopsy . MRI is superior to CT both in the differentiation of malignant from benign pleural disease due to its superior signal-to-noise ratio and is the modality of choice in the assessment of chest wall and diaphragmatic invasion by MPM. Dynamic contrast-enhanced (DCE) MRI is a promising technique and has the ability to correlate histology and pathology(Giesel 2008).

MRI not only confirms the CT findings such as diffuse pleural thickening and pleural effusion, but is superior in delineating contiguous invasion of adjacent structures. MPM has intermediate to slightly high signal intensity on T1-weighted images (T1-WI) and moderately high signal intensity on T2-weighted images (T2-WI) as compared to adjacent chest wall musculature and shows moderate enhancement after administration of gadolinium. MRI has a higher sensitivity and specificity to CT in detecting early chest wall and subdiaphragmatic involvement. Linear enhancing foci in the chest wall depicting sites of previous biopsy, thoracotomy, or chest tube tracts are also relatively more easily seen on MRI than on CT.


Postoperative Evaluation
Curative treatment for MPM is with extrapleural pneumonectomy. Localized disease or minimal disease is treated with local resection or radical pleurectomy or pleural decortication. Radiographs are used to follow patients postoperatively, reserving CT for evaluating complications.
After pneumonectomy, the pneumonectomy space fills up with fluid, generally at the rate of one intercostal space per 7 days, and can be
monitored by serial radiographs. Controlled filling of the pneumonectomy space helps control mediastinal shift [35]. Rapid filling of the pneumonectomy space is worrisome and is of concern for hemorrhage within the pneumonectomy space or a Chyle leak. Slow filling of the pneumonectomy space or decreasing fluid
level is worrisome for a bronchopleural fistula, or leakage of fluid into the abdomen along the diaphragmatic reconstruction, both these scenarios are secondary to infection
MDCT with the help of multi-planar reformats and 3-D imaging can help delineate the BPF. The data can also be interpolated to provide measurements for personalized stents . Ventilation scans can help delineate a tiny central BPF. Marsupialization of the pneumonectomy space and Clagette window creation are the treatments of choice for a central BPF. The pneumonectomy space is opened and cleaned and packed with antibiotic soaked packing in an attempt to heal the infection and then the cavity is closed and packed with a muscle flap, generally the latissimus dorsi or the omentum.
CT and PET 18F-Fluorodeoxyglucose scans are also used to identify and biopsy possible sites of recurrence. Another complication seen especially with a left-sided pneumonectomy is herniation of stomach along the medial aspect of the pneumonectomy space. Plain radiographs are the best at depicting the herniation of the gastric bubble above the gortex reconstruction, usually seen on the first postoperative radiograph.
Post-pneumonectomy syndrome, another rare complication, can also be assessed by CT. The left main stem bronchus gets stretched over the vertebral body due to severe mediastinal shift to the right postpneumonectomy . The mediastinal shift can be corrected by putting in a saline-filled implant into the pneumonectomy space, with an aim to displacing the mediastinal structure. MR is a very useful modality when Chyle leak is suspected and helps in identifying the site of leak and the thoracic duct prior to embolization. Recurrence and/or progressive metastatic disease are generally evaluated by contrast-enhanced CT scan. Multiple patterns of recurrence are seen mostly as enlarging soft tissue masses along the resection margins, ascites, and peritoneal thickening, which is a manifestation of intra-abdominal disease, new pulmonary nodules, and increasing size of mediastinal nodes. FDG/ PET is very useful in restaging and also monitoring response to therapy.
Post-radical pleurectomy, the granulation tissue along resection margins can be irregular and nodular, thus often raising concern for recurrence; however, serial FDG/PET can help distinguish between the two by semiquantitative evaluation of tracer uptake. Tumor will show progressive increase in uptake of tracer as opposed to granulation tumor, which slowly, over a period of time, will either regress or remain stable.


Future Directions
Dynamic contrast-enhanced MRI can be used to map the heterogeneity of microcirculation in MPM and can be used to predict therapeuticresponse and stratify survival. The development of such a quantitative technique will bring new measures essential to the diagnosis and management of patients with MPM, and will enable an objective assessment of new pharmacologic agents and serve as a possible tumor biomarker enabling prediction of outcomes. Diffusion MRI, combined with DCE MRI, can be a powerful tool. ADC maps derived by plotting intensity from multiple b values can be used to measure tumor cellularity. However, these techniques need to be validated and studied before they can be adapted into clinical practice.



Summary
Imaging plays a key role in diagnosis, management, and follow-up of patients with MPM. CT is the primary diagnostic modality in diagnosis, staging, and posttreatment management of MPM. MRI and PET provide additional and complementary information to CT. Optimization of current MR protocols will provide more efficient and valuable MR applications and potentially serve as an imaging biomarker. Larger population studies and correlation of imaging to pathology and genomic profiles can help improve survival.

MESOTHELIOMA AND EXPOSURE TO ASBESTOS 1962-2000IN SOUTH AFRICA

Three minerals, three epidemics
All three of the major commercial forms of asbestos – Cape crocidolite (blue asbestos),
amosite (brown asbestos) and chrysotile (white asbestos) occur in South Africa. Amosite,
together with Cape and Transvaal crocidolite, belongs to the amphibole family of asbestos
minerals, whereas chrysotile is a serpentine asbestos

In their monograph on asbestos mining and disease in South Africa, Felix et al.1
have traced the history of the mining of each of the these forms of asbestos, and demonstrated
their links with local epidemics of asbestos-related diseases, and through
exportation of massive amounts of these minerals, with disease in workers and communities
around the globe.
The South African asbestos trade began in the early nineteenth century following
the discovery and mining of crocidolite near Prieska in the Northern Cape in 1806. This
town is at the southern end of a region with numerous asbestos occurrences, extending
some 50 km in width and 400 km in length northwards to South Africa’s border with
Botswana, where the now defunct Pomfret, the largest of the crocidolite mines, is situated.

Initially, and well into the first half of the twentieth century, crocidolite mining
was outcrop and small digging operations run by tributers and farmers. Mining and
milling were highly labour-intensive. Fibre was cobbed from rock by hand-held hammers,
sieved by hand, sorted by a combination of a hand and mechanised method, and
transported in Hessian sacks to its destination. Women and child labour was extensively
employed in sorting and cobbing.
Trans-national companies became predominant in the mining of asbestos for
international markets, a process that began with the establishment of the United Kingdom
based Cape Asbestos Company that began mining and milling operations at
Prieska in 1893. By 1919 more than 3000 tons of Cape crocidolite fibre were being
produced per year. By the 1930s trans-national corporations dominated the asbestos
mining scene in South Africa.
Amosite – a name derived from the acronym for Asbestos Mines of South Africa –
occurs in the Lydenburg district, with the major deposit being at Penge where mining was
carried out from 1914 to 1992. Numerous deposits occur north-west of Penge in a 45 km
arc and during the mid-twentieth century were mined for amosite and Transvaal crocidolite
in numerous small operations using crude technologies that caused extensive environmental
pollution and exposed the labour force to high levels of asbestos dust.
Chrysotile mining in the Barberton district of Mpumalanga Province began at a
number of sites after 1915. Exploitation of the major deposit at Msauli began in 1937.
Msauli is currently South Africa’s only operational asbestos mine, now employing less
than 350 people.
A recent review of South African asbestos production, exports and destinations
from 1959–932 confirms that sales of crocidolite and amosite reached their peaks
in 1977 when South Africa exported a total of 380 000 tons of asbestos (Fig. 5.1),
making it the third biggest supplier in the world in that year.

Mesothelioma and asbestos – the beginnings
The association between asbestos and mesothelioma was first correctly made in people
from the Northern Cape Province crocidolite fields, an association that was subsequently
confirmed and accepted worldwide.4,5 Considering that crocidolite had already been exploited
for 150 years and that thousands of people had been exposed, many at an early
age and under crude conditions, it is perhaps surprising that it had not been considered
earlier. It appears that there were sufficient cases for Sleggs, Wagner and Marchand to
assemble their series of 33 cases of this usually rare tumour in only four years!
In reviewing developments subsequent to the discovery of this association, Felix et al.
recount that at the 1959 Pneumoconiosis Conference held in Johannesburg a resolution
was adopted to further investigate the relationship between asbestos exposure and mesothelioma,
and in 1961 the Pneumoconiosis Research Unit (PRU) of the governmentsponsored
Council for Scientific and Industrial Research (CSIR) embarked on a field survey
in the northern Cape and at Penge. The first year of the study cost 12 000 rand. Asbestos
mining companies contributed 8000 rand and the South African Cancer Association
contributed 4000. It must have soon been clear to the researchers that they were dealing
with a serious instance of environmental pollution. Flynn6 recounts that Webster was charged
with the unpleasant task of informing the asbestos industry of the environmental disaster
in the Northern Cape and that crocidolite was the carcinogenic factor. This information
apparently evoked a negative response which included a campaign to denigrate the scientists
involved, accusing them of “trying to destroy a valuable export industry for selfaggrandisement”.
6 Funding of the project by the industry and the Cancer Association was
not renewed for a second year of the study, and consequently the field-work was never
completed. However, the CSIR contributed 10 000 rand to finalise the research report on
the work completed, but subject to an undertaking by the Research Advisory Committee
of the PRU that “such a ‘report’ would not be published or made available outside the unit
[PRU], other than to sponsors and the various members of the working committees that
had been concerned with the conduct of the ‘survey

Mesothelioma: reporting of cases and trends,
1955–92
After the 1959 Pneumoconiosis Conference in Johannesburg, the National Research
Institute of Occupational Diseases (NIROD – subsequently the National Centre for
Occupational Health [NCOH]) in association with the Asbestos Tumour Reference
Panel – a panel of expert histopathologists established in 1965 – continued to record
and check each diagnosis of mesothelioma reported to it. In 1973 Webster published
the group’s experience of the first 232 cases.9 Later, Zwi et al. released results of a case
series of mesothelioma compiled from a variety of sources, provided an estimate of
mesothelioma incidence and described characteristics of 1347 cases identified in South
Africa between 1976 and 1984.10 Other case series have also been reported.11–13
In 1986, a South African National Cancer Registry (SANCR) was established within
the South African Institute for Medical Research (SAIMR). The SANCR essentially
incorporated the work of the Asbestos Tumour Reference Panel (ATRP) and centralised
reporting of cancer diagnoses made by pathologists at all major public and private institutions
in South Africa.14 Between 1986 and 1992, 1158 cases were reported to the SANCR

Asbestos exposure and mesothelioma
Four case series have been published that detail the source of exposure in 505 cases of
histologically proven mesothelioma in South Africa (Table 5.2).9,11–13 Most of the cases
where exposure was not known or where it was believed that there was no exposure
are from the first case series. If these 61 cases are excluded, of the remaining 444, in
118 or 26.6% the exposure was confirmed as being environmental only, whilst in the
remainder a source of occupational exposure was identified with a slight majority of
cases being reported from secondary industry.
Occupational mesothelioma
Mining-related exposures in asbestos mines or mills contribute the largest proportion of
cases (145 cases or 44.5%) of those mesotheliomas where exposure is known. Mesothelioma
has been described in relation to both crocidolite and amosite exposure, but given

mesothelioma
Prior to 1993, compensation for miners developing occupational diseases, including mesothelioma,
was racially discriminatory. The Occupational Diseases in Mines and Works
Amendment Act of 1993 abolished racial discrimination in the amounts paid, but awards
are modest – usually about 70 000 rand given as a lump sum. The funding of future
awards is of some concern as payments are made from the miners’ Compensation Fund,
maintained by current mining employers pro rata for the number of miners that they
employ and their level of risk. If a mine has ceased operation the Act can require the
state to make up the cost to the Fund of that mine’s compensation. Since the asbestos
mining industry is essentially defunct, all compensation costs of former miners in South
Africa have come either from general revenues or from current mining employers.

Environmental mesothelioma
Evidence from mesothelioma case studies
In the vast majority of the cases of mesothelioma in South Africa where there was
known to have only been environmental exposure to asbestos, this exposure has occurred
in the Northern Cape in proximity to mines, mills and dumps.
A high proportion of cases of environmental origin (26.6%) is unique to South
Africa. The only comparable example is Australia, the only other country to have mined
crocidolite in significant amounts. Ferguson et al.27 found that in 726 cases of mesothelioma
registered in Western Australia during 1980–85, 43 cases or 6% had environmental
exposure only, and only in six cases (<1%) was environmental asbestos exposure
due to residence in an asbestos mining region (Wittenoom)

Compensation for patients with environmental
mesothelioma
There is currently no form of financial compensation in South Africa for people who develop
mesothelioma from environmental exposure. It has at various times in the past been
mooted that the Department of Health should establish such a fund, but nothing has been
forthcoming. It also appears that there has been no successful civil legal action for personal
injury damages in this context. As we discuss below, this situation may be changing.
Preventing occupational mesothelioma
Since 1987 the use of asbestos in South African secondary industry has followed Asbestos
Regulations framed in terms of the Occupational Health and Safety Act.27 Among the
provisions are a Permissible Exposure Limit (PEL) of 1 f/mL and an Action Exposure
Limit (AEL) of 0.5 f/mL for all types of asbestos. More importantly and in line with
trends elsewhere, there has been major substitution of most of the former uses of
asbestos, particularly amphibole asbestos.
Most of the occupations listed in Table 5.2 are of historical interest only and are
of little importance as current prevention strategies. For example, steam locomotives
are being replaced. The asbestos cement industry used significant amounts of crocidolite
in its products until the mid-1980s when substitution with chrysotile began.
In the future the main occupational asbestos hazard in industry will be the removal
of asbestos, particularly in situations where it has been used as insulation on
boilers, steam pipes and in buildings. It is not clear that this risk is adequately controlled
by the current Asbestos Regulations and it is important that contractors are
monitored to prevent unacceptable practices and methods of removal such as using
untrained, daily paid workers to do this kind of work.
Although amphibole asbestos is no longer mined in South Africa, it is sobering to read
Felix et al.’s account1 of regulation of the health effects of asbestos mining by the Department
of Minerals. Standards applied in South Africa have lagged behind those applied in Europe
and North America. Sluis-Cremer17 estimated that in surface workings in 1945, dust counts
at crocidolite mines were 30–160 f/mL, and were 8–30 f/mL in 1970. In the 1970s, at a time
when European countries were beginning to ban the importation of crocidolite, the PEL for
asbestos surface workings was 10 f/mL. In 1984 the current uniform surface and underground
mining standard of 2 f/mL was introduced.1 In 2001 the Department is proposing
that this limit be reduced to 1 f/mL for all types of asbestos.

Preventing environmental mesothelioma
The focus for the prevention of environmental mesothelioma has to be twofold: the rehabilitation
of the areas polluted by asbestos during mining and milling of the mineral and
the education of the affected communities about the asbestos hazard. The extent of the
problem of environmental pollution by asbestos in the affected regions is enormous. Living
for the most part in complete ignorance of the health threat posed by asbestos, many
communities have been seriously polluted with asbestos. Although initial remedial efforts
by local and Provincial governments were slow and sporadic, a more concerted effort has
been made by the Department of Minerals in recent years. The government has spent 44
million rand on this process, and estimates that a further 52 million is needed to complete
this task. Companies have contributed less than 5% of the rehabilitation costs

Treating mesothelioma
People who develop this cancer have a poor prognosis and, internationally, experience
with treating mesothelioma has been discouraging. It is not clear that any treatment
available to us confers benefit above that provided by simple palliation. This grim
outlook seems to have discouraged clinical interest in this condition in South Africa.
Specialised oncology therapies are only available in major centres and none of these
are particularly close to the asbestos fields. Individual clinicians have over time developed
considerable experience in the palliation of this cancer, but little of this experience
has been published.36
The nihilism that surrounds the clinical management of mesothelioma has other
consequences. In high-prevalence areas a typical case of mesothelioma (a painful pleural
effusion that does not respond to anti-tuberculous therapy) is often not subjected to
pleural biopsy unless drainage and pleurodesis becomes nescessary. Pleural biopsy is
viewed as conferring no benefit other than confirming the diagnosis and carries a not
insubstantial risk of tumour recurrence at the biopsy site – a troublesome complication
that requires radiotherapy and distant referral.
Future prospects for mesothelioma
The future prospects for mesothelioma in South Africa are not encouraging. It could
be expected that the occurrence of mesothelioma will peak some 20 to 30 years after

the peak of crocidolite exploitation. This peak production was in 1977 and we may
well be approaching this point in the epidemic. It can be expected that the epidemic
will continue at least for the lifetime of those large numbers of people exposed to
crocidolite in mining and industry up until the late 1980s. In view of the lack of timely,
effective environmental rehabilitation in most affected communities it can be expected
that mesothelioma will still be a spectre in the lives of children being born into these
communities even today.
The asbestos tragedy has been described as “one of the most colossal blunders of
the twentieth century” by Bill Sells,2 a former executive of Johns-Manville, once a
major US manufacturer of asbestos products. Sells wrote in 1994, “In my opinion the
blunder that cost thousands of lives and destroyed an industry was a management
blunder, and the blunder was denial,” instead of the “responsibility . . . and product
stewardship” demanded of so serious a situation.
Although asbestos in other countries could be considered a colossal blunder of
management denial it is perhaps easier to understand the South African citizenry as
the naive victims of a superbly conducted confidence trick carried out in the names of
jobs, development and prosperity. Even now the con victim is scratching his head,
looking at the dud cheque and saying to himself, “How did they do it?”

MESOTHELIOMA IN EUROPE

In 1991, Wagner introduced the Mesothelioma Conference in Paris with an historical
review, from the 1870s to the 1930s, when European and American pathologists were
discussing the exact origin and nature of the so-called primary neoplasms of the pleura.
Mesothelioma as a real pathologic entity
Klemperer and Rabin used the word ‘mesothelioma’ for the first time in 1931.In this
early period, the biphasic pattern of this tumour, epithelial and mesenchymatous, had
been demonstrated by different pathologists.Thereafter, we have to wait until the
1950s to find case reports of ‘primary diffuse pleural mesotheliomas’ described by
different authors in Europe and in North America.
Finally, in the 1960s, the pioneer work of European and North American pathologists
reached a consensus for considering that diffuse malignant mesothelioma, located
mainly in the pleura and less frequently in the peritoneum, was a primary neoplasm
arising from the pluripotential mesothelial cells. The pathological diagnosis of mesothelioma
appeared difficult, so that, in most serious epidemiological studies, the diagnosis
was ascertained by a panel of national or supranational expert pathologists.At present,
most European countries have a mesothelioma panel of trained pathologists.
Discovery of the relationship with asbestos exposure
Another major step in the history of mesothelioma was the studies which demonstrated
that asbestos exposure was a strong causal factor for the development of this
malignant tumour. In fact, it took about 30 years for this to be demonstrated! While a
case of mesothelioma associated with asbestos exposure was published in 1943 by
Wedler in Germany, this tumour was very rare in Europe and North America, so its
relationship with asbestos exposure was not identified in industry, possibly in relation with the growth of asbestos uses. By contrast the
tumour was rare in females. Actually, the year 1960 is of historical note as, 5 years after
the publication by Richard Doll10 demonstrating the link between lung cancer and occupational
exposure to asbestos, Wagner et al. reported 33 cases of MM in Northwest Cape
(South Africa) among crocidolite miners and their family contacts. Thus, the causal
relationship between mesothelioma and exposure to asbestos was demonstrated. Five
years later in France, Turiaf et al. reported the first case of pleural mesothelioma in a 54-
year-old man with a 30-year history of occupational exposure to asbestos.
Incidence of malignant mesothelioma (MM)
As mesothelioma, whatever its two main locations (pleura and peritoneum), represents
a severe disease, highly related to asbestos exposure, all industrialised countries
are presently concerned with the evaluation of the true incidence of this cancer and its
relationships with the different types of asbestos exposure. This section will cover the
incidence of mesothelioma in European countries and the role played by accurate
registration of cases for assessing this fundamental parameter.
Incidence in industrialised countries
In industrialised countries, data on mortality from mesothelioma can be obtained from
death certificates and incidence rates from cancer registries. Actually, the problem is
to know the accuracy of mesothelioma cases registration.
As recently as the 1970s, in order to establish the incidence of this malignancy
and its causal relationships, several industrialised countries decided to collect data on
pleural and peritoneal cases for the whole country (United Kingdom, Scandinavian
countries) or only for some specific areas (USA, Canada). From these studies, the
‘background’ level of mesothelioma incidence could be estimated at around 1 to 2 per
million per year in industrialised countries
Incidence rates vary in different countries, apparently in relation with tonnages
and types of asbestos production and/or consumption: 28.3 cases per million among
males in Australia14 and 33 per million in South Africa, both countries being producers
of crocidolite and amosite. In European countries, various incidence rates have
been observed: for instance, the mortality rate in males was reported to be 17.5 per
million in Great Britain16 and 20.9 per million in the Netherlands.
During the past 30 to 40 years, most industrialised countries (e.g., North America,
Europe, Australia) have observed an increase in the annual incidence of MM (at about
7% to 10% per annum), with an obvious predominance among males. Such trends suggested
the causal role of specific occupational asbestos exposures (particularly to amphiboles),
no other causes having been identified in this early period, but possibly also
in relation to a real improvement in diagnosis and registration of this rare type of cancer.
In the 1960s, definite mesothelioma cases showed a marked clustering in areas
where there was substantial industrial use of asbestos. Thus, in the USA, Connelly et

al. observed that the highest rate was in shipyard areas (Seattle, San Francisco, Hawaii).
 However, from the mesothelioma register in Great Britain, Peto et al. found
that mesothelioma deaths were still increasing and will continue for at least 15 to 25
years. As the deceased’s job (probably the last one) was mentioned on United Kingdom
death certificates, the authors could calculate the trend of mesothelioma in different
jobs; workers in construction and maintenance of buildings containing asbestos
accounted for the largest proportion of these deaths. This important finding stresses
the usefulness of a nationwide mesothelioma register as the most relevant tool for the
surveillance of possible risks of malignancy related to low-dose asbestos exposures, for
instance in asbestos-containing buildings.
As a definite diagnosis of MM is difficult to establish with certitude, requiring
the opinion of trained clinicians and pathologists, it is probable that estimates
from death certificates do not match those obtained from cancer registries. It is probable
that presently this tumour is underreported, although the reverse is also possible!
Thus, for the period 1967–68, 413 cases from England, Wales and Scotland
were notified as mesothelioma to the national register. After revision of the slides by
the British panel of pathologists, 246 cases were accepted as definite and 76 cases as
definitely not mesothelioma. Such discrepancies have been reported from other
industrialised countries where both registry-based incidence data and mortality data
were available. In their review, Iwatsubo et al. have shown that, for different studies,
variations in the percentage of pathologically confirmed diagnosis of mesothelioma
ranged from 26% to 96%

Incidence of mesothelioma in Europe from case registration
The background level of mesothelioma was assumed to be as low as 1 to 2 per million
inhabitants, but since the 1950s, this incidence has been increasing in the general
population of most industrialised countries.
Since mesothelioma is usually a rapidly fatal malignancy, mortality rates based on
the underlying cause of death, as recorded on death certificates, have often been used as a
close approximation of incidence rates. But large discrepancies have been observed in some
cases,19 so that we should distinguish the studies according to the origins of the data:
either mesothelioma registries, general cancer registries or death certificates.
In non-asbestos-producing European countries, it appears that an accurate evaluation
of the true incidence of mesothelioma cases and their causal relationships was
the best epidemiological tool for assessing the types (amphiboles versus chrysotile) of
imported asbestos and determining the cumulated tonnes of past exposures. Nevertheless,
this aim had some limitations, as only a limited number of European countries
(Scandinavian countries, United Kingdom) decided early in the 1970s to register
all pleural and peritoneal mesotheliomas cases.
In the United Kingdom, a mesothelioma register was set up in 1967.Mesothelioma
cases were identified from the death certificates mentioning ‘pleural or peritoneal
mesothelioma’, information provided by registrations from the Cancer Bureaux,
the Pneumoconiosis Panel and also from chest physicians, surgeons, pathologists and

to the UICC Panel of Pathologists. For the year 1967–68, 412 cases were notified to
the register, of which 245 were considered as definite mesothelioma, leading to a rate
of 2.29/million/year for England, Wales and Scotland.
Gardner et al. examined the time trend of mortality by pleural cancer, i.e., death
coded into the category 163 of the ICD, 9th revision, in England and Wales for the
period 1968–78 from the death records (Office of Population Censuses and Surveys).27
For the entire period, the mortality rate by pleural cancer was 5 per million in men
and 2 per million in women.
A later publication of the mesothelioma register concerning the period 1968–
8316 showed an increase of about 10% in men. The mesothelioma mortality in 1983
was 17.5 per million in men and 3.2 per million in women.
In the Netherlands, Meijers et al. examined the mortality trend of pleural malignancies
between 1970 and 1987.17 The coded underlying cause of death was provided from the
Dutch Central Bureau of Statistics. In men, the average pleural cancer mortality increased
from 10.7 per million for the period 1970–78 to 20.9 per million for the period 1979–87. In
women, these rates were, respectively, 2.5 per million and 3.6 per million.
In Scandinavian countries (Finland, Sweden, Denmark, Norway), the existence
of a national cancer register of all deaths from cancer facilitated the task of
examining the time trend of cancer incidence.
In Finland, a nationwide Finnish Cancer Registry was established in the 1950s.
This Cancer Registry allowed Karjalainen et al. to study the trend of mesothelioma incidence
in Finland between 1960 and 1995.29 In that country, anthophyllite asbestos was
produced and widely used from 1918 to 1975. The age-adjusted incidence of mesothelioma
was under 1 per million in both sexes around 1960, and then rose steeply in 1975–
90. In 1990–94, the age adjusted incidence of mesothelioma was 10 per million in men
and 2.9 per million in women. The overall pattern of mesothelioma seems to be stable in
the very recent period. This plateauing could be related to a significant decrease in the
use of amphiboles. Nevertheless, it seems that the mesothelioma risk related to
anthophyllite asbestos is low,30 but crocidolite was also used from the late 1960s.
In Sweden, Järvholm et al. studied the incidence of pleural mesothelioma between
1958 and 1995 with respect to preventive measures taken to reduce occupational
exposure to asbestos. There were about 10 cases of pleural mesothelioma in
men and no case in women in 1958. In 1995, 92 cases in men and 15 in women were
observed. An increasing incidence was found in recent birth cohorts in men.
In Denmark, Andersson and Olsen described the time trend and the distribution
of MM since 1942. The registration to the National Cancer Registry was based
on reports from hospital departments, pathology institutes, notifications from practising
physicians and death certificates. For the entire period, the authors observed a
regular increase in both sexes. The incidence rates for the latest period, 1978–80,
were 14.7 per million in men and 7 per million in women.
In Norway, Mowé et al. examined the time trends of mesothelioma incidence
between 1960 and 1988.34 The investigation was based on data from the Cancer Registry

of Norway to which all new cases of cancer are reported from hospitals and pathology
departments. The age-adjusted incidence of mesothelioma increased during the observed
period in men: 4 per million for 1960–69 to 13 per million for 1980–87. In women, mesothelioma
incidence remained at the same level, 1 per million, during the whole period.
In several other European countries (France, Germany, Italy), up to now, case
registration was limited to some specific areas.
In France, all data obtained from death certificates are collected by the Service
Commun N°8 (SC8) of Institut National de la Santé et de la Recherche Médicale
(INSERM). Since 1968, information on the underlying medical causes of death mentioned
by the practitioners on death certificates has been coded by INSERM SC8 according
to the International Classification of Diseases (ICD 8 or 9). The numbers of deaths due to
pleural malignancies were those classified in category 163 of the ICD 8 (i.e., malignant
neoplasm of pleura, stated or presumed to be primary) and category 158 of the ICD 9
(i.e., malignant neoplasm of peritoneum, stated or presumed to be primary).
A ‘registration’ of mesothelioma cases was set up in 1975. Actually, as death
certificates were totally confidential, MM cases could only be collected through pathologists
and clinicians. Thus, this ‘register’ was mainly made up of a panel of French pathologists
(still working) trained to confirm the diagnosis of mesothelioma, eventually
with the contribution of the EEC mesothelioma panel. From the death certificates, for
the period 1968–92, the mortality with mention of pleural malignancies increased from
8.2 per million to 22.5 per million in men and 4.7 per million to 9.2 per million in women.
The average increase during this period was 4.3% in men and 2.8% in women.
In such conditions, the French ‘register’ was not able to provide valid information
on the true incidence of MM. In January 1987, the mesothelioma panel interrupted
its random case collection on a national basis, to conduct a case-control study
in 5 regions of France, with a double objective: evaluation of the dose–response for
different occupational exposures and, eventually, identification of other asbestos-related
jobs (cf. below).
Recently, Ménegoz et al. (Réseau France CIM) examined the trend of mesothelioma
incidence from data obtained by 7 departmental registries for the period 1979–
93.37 These registries cover about 9.5% of the French population. For the entire period,
the increase was 25% over 3 years. In men, the incidence increased from 7 per million/
year for the period 1979–81 to 16 per million/year for 1991–93.
The progression of MM in France was estimated at around 7 to 8% per year. It
is worth comparing these data to the trend observed in the UK, the national register
of this country indicating a mesothelioma rate about three times higher than F rance.
This difference correlates with a much lower consumption of amphiboles in France
than in the UK

Relationship of mesothelioma to asbestos
exposure
As underlined by Howell et al., there are several routes of exposure to asbestos fibres:
occupational, paraoccupational (domestic contamination from asbestos workers), residential
contamination near dusty industrial sources, incidental, related contact with
asbestos products for domestic uses or hobbies, and general environmental exposure
from asbestos sediments at the surface of soil.
Occupational exposures
In Europe, most mesothelioma cases are related to asbestos exposures in occupational
activities in various industries, some of them not apparent. However, staff
in two industries were particularly exposed to asbestos dusts: shipyard workers
and construction workers.
The excess number of mesothelioma cases in coastal areas of European countries
is consistent with shipyard-related exposures (workers involved in shipbuilding
and repair). The situation applied in the Scandinavian countries (Sweden, Norway,
Finland, Denmark), The Netherlands, the United Kingdom, the Loire–Atlantique
and Normandy areas of France,and the Province of Trieste in Italy.
On the other hand, recently, Peto et al. identified from the UK mesothelioma
register an excess of mesothelioma in workers involved in construction and building
maintenance, particularly plumbers, gas fitters, carpenters and electricians, who appeared
as the largest high-risk group.

Dose–response relationships
Several authors, when evaluating cohort studies and case-control studies focusing
on mesothelioma, have reported a dose–response relationship.
Jones et al. examined the occurrence of respiratory malignancies among workers
in a gas-mask factory in the UK. The production of gas masks lasted 4.5 years from
September 1940. Crocidolite as well as chrysotile was used in production. A total of
951 women were known to have worked in this job. Up to the end of 1978, 17 cases of
mesothelioma were observed,  in those exposed to crocidolite only and one in those
exposed to both crocidolite and chrysotile. Among the 16 cases exposed to crocidolite
only, a dose–response relationship was observed with duration of exposure.
Raffn et al. studied the incidence of cancer and the mortality among employees
in the asbestos cement industry in Denmark for the period 1943–84.53 Subjects included
were those exposed between 1928 and 1984. The estimated exposure levels
varied greatly during the study period: in 1948, between 50 and 800 fibres per millilitre
(f/mL); in 1957, between 10 and 100 f/mL; in 1973, 41% of the measurements were
above 2 f/mL. Among the 7996 men and 517 women studied, 10 pleural mesotheliomas
were observed. The incidence of pleural mesothelioma increased with the duration
of exposure among subjects with 15 years or more of latency (SIR=3.77 for less
than 5 years of exposure versus 13.56 for more than 5 years of exposure).
Peto et al. observed that the risk of mesothelioma in an occupationally exposed
cohort (north American insulators) was best described by a mathematical model in which
the risk increases with the third or fourth power of time since first exposure;20 their data
were compatible with a linear dose–response relationship. Peto et al. found also this type of
relationship when considering the mortality of subjects (men and women) employed at the
Rochdale asbestos textile factory. A total of 18 mesothelioma cases in men (7 in the first
group and 11 in the third group ) were observed during the study period. Despite the small
number of deaths observed (10 occurred in the main cohort), the observed and predicted
numbers were in reasonable agreement for different times since first exposure.
Newhouse et al. studied the mortality of asbestos factory workers (asbestos textiles
and others asbestos products such as asbestos cement) in east London. The population
concerned 3000 male factory workers, 1400 laggers and 700 women factory workers. The
men were first exposed between 1933 and 1964 and the women between 1936 and 1942.
Crocidolite asbestos as well as amosite and chrysotile was used in the factory. Exposure
to asbestos was classified into 4 categories according to degree (light or moderate versus
severe) and duration (<2 years versus 􀀵 2 years). During the study period, 38 pleural
and 35 peritoneal mesotheliomas were observed in men and 14 pleural and 11 peritoneal
mesotheliomas in women. Among factory workers, the mesothelioma death rates
increased according to both duration and severity of asbestos exposure.

Environmental and neighbourhood exposure to asbestos
It seems likely that mesothelioma may arise from domestic and environmental exposures,
either natural or para-occupational. This is of concern particularly among family
members of asbestos workers but also in the general population who inhale fibres
in the vicinity of asbestos plants.
In the mesothelioma series of the London area published by Newhouse and
Thompson, there were 25 cases without evidence of any exposure to asbestos. Among
them, 11 cases used to live within 800 metres of an asbestos factory, a number significantly
greater than among a control group of patients without mesothelioma. However,
at that early time, the authors seemed to be reluctant to accept this relationship,
waiting for more evidence in the future.
Subsequently, the same kinds of observation have been made in other European
countries: in Finland, in relation with dwelling in the vicinity of an anthophyllite mine;
and in north-western Italy, where an increased incidence of histologically confirmed
mesothelioma was reported in the vicinity of a large asbestos cement factory at Casale
Monferrato, in operation from 1907 to 1985.
Endemic cases of mesothelioma have been also observed in rural areas in several
European countries, Metsovo in Greece61 and Corsica in France, and also, far from
Europe, New Caledonia, with a high incidence of mesothelioma, particularly in rural
areas. The asbestos fibres concerned are mainly of the tremolite type, except in Finland,
where anthophyllite was invoved. Those cases are related to outdoor and indoor
contamination from environmental geologic sources of asbestos dusts.
Mesothelioma cases related to indoor exposures are well known, particularly among
family members of asbestos workers coming home with contaminated clothes. This issue
might also be relevant to indoor contamination by fibres in buildings with asbestos
containing materials, fibres being eventually released at very low concentrations (0.001–
0.0001 f/mL of air). Presently, there is no sufficiently large epidemiological survey to
confirm such a relationship. We must refer with caution to the mathematical projections
published in the two documents: HEI-AR, 1991 and INSERM, 1997

Role of fibre types
Numerous epidemiological studies, discussed at length in HEI-AR, have compared
the incidence of lung cancer and mesothelioma in various situations. Moreover, the
ratio of mesothelioma/lung cancer numbers has been compared according to fibre
types. However, the observed discrepancies between cohorts may be due to the fact
that it was retrospectively difficult to know exactly what types of asbestos have been
used in these plants

What about the future?
Prediction of mortality from mesothelioma
A major concern in public health is to know the trend of the incidence rate of mesothelioma:
is it decreasing, stable or increasing? What about the evolution in the next 20 years?
Recently, three publications have provided estimates of the mortality from mesothelioma
in both sexes which can be predicted in the next few years.
In 1995, Peto et al., analysing the data from the British Mesothelioma Register,
forecast a peak of deaths in 2020.21 They estimated the total size of the 1996–2020
epidemic of deaths due to mesothelioma in males as 62 000 (which extrapolates to 166
males per million male inhabitants of Britain).
In 1997 Price, analysing current trends of mesothelioma incidence in the USA,81
concluded that the peak of mesothelioma was just going to appear (1997) and that the
projected number of future mesothelioma cases suggests a peak in the annual number
of cases at 2300 for males, before the year 2000. Then, the number of male cases will
drop during the next 50–60 years toward 500.
In 1998, from the available data in France, Gilg Soit Ilg et al. predicted that the
nationwide peak will occur between 2020 and 2060, and, for men, the total size of the
epidemic will be 20 000 over the period 1998–2020, which extrapolates to 43 per million.22
Such studies are important, because it is the death toll which predominantly
builds the perception of the risk in the general population. They are necessarily based
on the past and present mortality data.

MESOTHELIOMA AND EXPOSURE ASBESTOS IN SOUTH AFRICA

Asbestos
South Africa has deposits of three types of asbestos of value and in certain regions these
ore bodies can be exploited commercially. It is important to note that at certain times
the amount of the three types of asbestos produced has been comparable, and so have
the workforces. By far the most mined asbestos fibre in the world (accounting for 95%)
is the white, chrysotile, found in serpentine rock, the ore of which can be mined in the
Eastern Transvaal and Swaziland. Amosite (accounting for 1% of the world market)
occurs in the Northern Transvaal, is greyish-brown in colour, is an amphibole and is
found in banded ironstone. Crocidolite, or blue asbestos (accounting for 2% of the worlds
market), at one stage was colour graded, the lavender-blue being the most valuable. It is
also an amphibole occurring in banded ironstone. It is found in the north-west Cape
(Griqualand West). It is also found in the Northern Transvaal, where in some situations,
amosite and crocidolite occur in the same seam. Exploitation of the Transvaal deposits
of crocidolite peaked during the Second World War and then declined.
The mining of crocidolite in Griqualand West (north-west Cape) was begun south of
Prieska, the southern most town in these asbestos fields, in 1893. In that year 100 tons of
crude ore was produced. The early method of mining was open-cast quarrying. In the
Cape, the black workers (unlike elsewhere in South Africa) carried out the blasting. They
also did the quarrying, whilst their women cobbed the fibre, and their children recovered
any asbestos fibres that had been discarded. Cobbing, or knapping as it was originally
called, consists in separating the rock from the fibre by striking the fragments with a
square faced hammer. The women then sorted the fibre into different lengths. The work
was supervised by the whites. The fibre was bought by companies on the tribute system.
Gradually, as the overseas market for crocidolite increased, underground mining
started at the richer strikes, so that by 1918 there were mines in the south near Prieska
and 150 miles to the north at Kuruman. In 1930 Hall listed 104 sites in the asbestos mountains,
but there were only 12 actual mines. Most of these mines were worked by black
workers, but white farmers were working outcrops, especially during times of drought

There was a greatly increased demand for crocidolite immediately before and
during the First World War, with mills being built in the towns of Prieska and Kuruman.
Following a lull, the industry revived in the late 1930s, increasing further during the
Second World War. Then came another major increase between 1950 and 1960, production
rising from 40 000 tons to 100 000 tons. By this time, the mills were modernised
and mining occurred on a large scale with underground shafts. Only in Prieska did the
tribute system continue and there was an active mill in the town. In other areas the
mill was attached to the mine.
By 1962 as the mills became larger the amount of dust increased, as did the size
of the tailings dumps, which released more dust during dry and windy conditions.
Furthermore, the demand from the manufacturing industry for fine crocidolite fibre
increased the amount of dust. The large tailings dumps contained about 20% fibre in
the vicinity of active and previous sites of mills.

Mesotheliomas
During the Second World War, a Royal Air Force hospital was built in Kimberly, which
is in the northern part of the Cape Province. After the war, this hospital was presented
to the South African government, which decided it should be used for the treatment of
tuberculosis. Dr C. A. Sleggs was appointed as Superintendent. The area covered by
the hospital stretched from the Orange Free State in the east, to what is now Namibia
in the west. Before 1950, tuberculosis was endemic in the whole of this area. In the
next few years, with the advent of specific anti-tuberculous treatment, Sleggs noticed
that there was a difference in the response to treatment of those with tuberculous
pleurisy. Patients from the eastern part of the area showed a good response, whilst
there was a poor response from some of the cases from the west. Twelve of these latter
were seen by thoracic surgeons in Capetown, Pretoria or Johannesburg. In all cases a
diagnosis of metastatic carcinoma from an unknown primary site was made. By February
1956, there were a further 6 cases in this (west-end) hospital.
In February 1956 in Johannesburg, a black male patient came to necropsy examination.
He was thought to have had tuberculous pleurisy and had been treated
in the main mine hospital in Johannesburg but did not respond to treatment. Aspiration
of his pleural cavity had failed, as the fluid was thick and sticky. A diagnosis of
empyema was made and subsequently given as the cause of death. However, at postmortem
examination the findings were different with the presence of a huge gelatinous
tumour filling the right thoracic cavity, completely surrounding and compressing
the right lung, infiltrating the pericardium and displacing the mediastinum.
Mesothelioma was a rare tumour. At this time most pathologists agreed with R. A.
Willis5 that these tumours did not exist and that any tumour found in the pleural
cavity was a metastasis from a primary site elsewhere.6 Wagner carried out the postmortem,
and he consulted the professor of pathology at Witwatersrand University
(B. J. P. Becker). Becker suggested that a detailed examination should be made to
eliminate any other primary tumour. None was found and the diagnosis of asbestos
bodies were observed in the lung tissue, but no evidence of asbestosis.
This case was presented to the local thoracic society group by Olaf Martiny7 and
Chris Wagner about 2 months later. In the meantime, Dr Sleggs was still perturbed
over his cases, and consulted L. Fatti, senior thoracic surgeon, and his partner, P.
Marchand, both of whom visited Kimberly. The marked similarity of clinical and radiological
findings of Dr Sleggs’ cases and the case that Wagner and Martiny had
presented was noted by both these surgeons. Dr Sleggs sent 2 needle biopsies from his
other cases, but the material was insufficient for a definite diagnosis. Marchand carried
out open biopsies on the cases in the West-End hospital. By the end of 1956 pathology
had been seen from 10 large pleural biopsies and 2 post-mortem examinations.
Obviously, a definite diagnosis could only be given after death, followed by a full
post-mortem examination. All material was shown to Professor P. Steiner from Chicago,
who had seen a number of mesotheliomas previously, and he agreed with the
diagnosis that had been made. He asked why there were so many of these rare tumours.
An association with asbestos was considered at an early stage, partly because
asbestos bodies had been found in the first case and the range of the asbestos mountains
was 90 miles west of Kimberly. Other possible aetiologies were suggested. Was there a
virus implicated, the situation being similar to that in East Africa, with the Epstein-Barr
virus? Could radioactivity play a part as monazite, the ore of thorium was known to
occur in the region? Was there a local genetic disposition, associated with tuberculosis?

Of the first 16 cases collected only 4 had worked in the asbestos industry. The
biopsies from these cases were mainly from the parietal pleura, but from 3 there were
fragments of lung tissue that contained a few asbestos bodies. The majority of these
16 people had lived in the Kuruman district.
Evidence against the implication that asbestos was responsible for the development
of these tumours was that only a quarter of the cases had admitted working with
asbestos. Secondly, since asbestos mining had been in progress for more than 50 years
in this region, it would be expected that the tumours could have been recognised at an
earlier date. Later it was discovered from the records that a pleural endothelioma was
first reported in 1917 and that several other tumours had been notified but always
considered to be from a primary site elsewhere. Thirdly, none of these tumours had
been observed where other types of asbestos were produced. Fourthly, Doll had shown
a dose response associated with other asbestos diseases, occurring in heavily exposed
(for more that 20 years) individuals. The majority of the mesothelioma cases denied
having worked with asbestos at all. Their occupations were diverse, including housewives,
domestic servants, cattle herders, farmers, a water-bailiff, an accountant and
later an international goal-keeper.
It was not until the middle of 1958 when Paul Marchand interviewed two brothers
that it was realised that a different question must be asked. One brother had a
mesothelioma, while the other had a suspicious x-ray. Marchand asked the latter
whether either of them had worked in the asbestos industry, to which he replied that
his father had owned a small asbestos mine and that they used to play on the dumps as
children. All the cases were then re-interviewed (or the relatives), mainly by Dr Sleggs.
From this information, it was clear that the majority had had exposure to blue asbestos.
They had lived in the vicinity of either the mills or the tailings dumps. Exposure
could be as little as 6 months even 40 years before. Some were exposed as infants and
others at school where the neighbouring dumps made excellent play slides. The first
patient diagnosed with mesothelioma whose occupation was given as a bath attendant
was found to have herded sheep in the Kuruman area when young.
Could these tumours have occurred in those countries where the Cape blue had
been exported? Preliminary enquiries in Europe during 1957–58 by Wagner elicited
evidence of one tumour in the Midlands, United Kingdom, and also one in Turin, Italy.
E. McCaughey reported 15 cases, which he had seen in Belfast, Northern Ireland, but
no association with asbestos was considered. (Webster in 1959 stated that he had seen
two further cases at the London Hospital, United Kingdom; personal communication).
In 1958 Dr Harold S. Stewart, Head of Cancer Research at the National Institutes
of Health, Washington, DC, and an authority on geographical distribution of cancer, was
keenly interested in these findings and he persuaded Dr A. J. Orenstein, Director of the
South African Pneumoconiosis Unit, that Wagner and Sleggs should present papers at
the International Pneumoconiosis Conference, to be held in Johannesburg in 1959. These
papers were ‘The pathological aspects of asbestosis in South Africa’ and ‘Clinical aspects
of asbestosis in the Northern Cape’. The main topics of this conference were the
diseases associated with the major industries of gold and coal mining. On the whole, no
one appeared strongly interested in the mesothelioma problem, except for Dr J. C. Gilson,
Director of the British Pneumoconiosis Unit.

MALIGNAN MESOTHELIOMA IN NORTH AMERICAN

Incidence
In the United States
In 1890 Biggs reported a case of ‘endothelioma’ of the pleura. This report may have
been the first recognised case of malignant mesothelioma in North America. Since then
the incidence of mesothelioma in North America and the world has steadily climbed.
Studies of the incidence of mesothelioma in North America have been hampered
by a paucity of data. Before 1988, the United States did not even have a specific code for
mesothelioma, so many cases were misclassified on death certificates as lung cancers or
abdominal cancers. In the United States, the best estimates of mesothelioma incidence
are derived from the Surveillance, Epidemiology and End Results (SEER) Program of
the National Cancer Institute. The SEER database2 includes about 9.5 per cent of the
United States population. It covers 10 regional areas, in five states (Connecticut, Iowa,
New Mexico, Utah and Hawaii), and five major urban areas – San Francisco–Oakland,
New Orleans, Seattle, Atlanta, and Detroit. Although the SEER regions are reasonably
representative of the United States population in terms of demographic and epidemiological
factors, the programme may not accurately reflect the country as a whole. It
includes some shipbuilding areas, but large urban areas where asbestos was used in
manufacturing and construction are underrepresented. The database is organised by
case; each case is identified by age, sex, race, date of diagnosis and cancer type. The data
for mesothelioma are published only intermittently. However, this database provides
the most comprehensive national incidence data available for this disease.
In 1997 Price analysed the SEER data for mesothelioma. He divided the data into
five-year age groups in each diagnosis year. He found a consistently higher rate of mesothelioma
in men than in women. The rate for women remained relatively constant over
the years. On the other hand, the rate for men increased until 1992, when it peaked at 1.9per 100 000 people. Since then the incidence rate in men has been slowly decreasing
 This trend is presumably due to occupational asbestos exposure, which was
highest during the years 1930–60. The highest lifetime risk was for the 1925–30 birth
cohort – a group that would have been at work in shipyards, manufacturing and construction
during the years 1930–60. In recent years in the United States, the incidence has been
approximately 2000–3000 cases per year, or 11.4 cases per million men and 2.8 cases per
million women. The mortality rate in people with prolonged heavy exposure to asbestos
varies from 2 to 10 per cent in different studies, and the latency period between initial
exposure and manifestation of disease is usually 20–50 years. From 1987 to 1996 an average
of 520 people died per year in the United States of malignant mesothelioma

Data from the United States Department of Health and Human Services5 show that
states with the highest incidence of mesothelioma are all coastal or Great Lakes States.
Florida has displaced New York as the State with the highest number of deaths per year
from mesothelioma. In 1996, 78 people died of mesothelioma in Florida. The states with
the highest age-adjusted mortality rates were Washington and Oregon, probably due to
the presence of shipyards. The most frequently recorded occupation on death certificates
of people with mesothelioma in the United States was homemaker (10.6 per cent of all
deaths), followed by managers and administrators, plumbers, pipefitters and steamfitters,
production supervisors, labourers, electricians, farmers, carpenters and machinists. The
most common industry was construction, followed by ship building and railroads. Other
areas with significant mesothelioma incidence were schools and government.

In Canada
In Canada several large epidemiological studies have provided data for estimates
of the incidence of mesothelioma. Morrison and colleagues in 19846 looked at all cases
of pleural mesothelioma diagnosed in British Columbia from 1973 through 1980 and

reported to the National Cancer Incidence Reporting System (NCIRS) of Statistics
Canada. They identified 64 cases (54 men and 10 women). Almost all cases were clustered
in Cowichan Valley, Capital and Greater Vancouver counties. They noted an
overall increasing incidence with age. Among men, but not women, they found an
increasing incidence over time, and correlated this with a relatively high level of shipbuilding
and ship repair activity in Vancouver and Victoria thirty to forty years before
the study. Similar increased rates of mesothelioma have been observed in shipbuilding
centers in the United States and Great Britain. The authors also compared the incidence
of mesothelioma obtained from NCIRS data to that obtained by looking at death
records. They found that only 60 percent of the cases were identified by looking at
mortality data – so death records are therefore a poor substitute for incidence records,
since mesothelioma is often not recorded as the cause of death.
In 1985 Churg and colleagues7 surveyed all pathologists in British Columbia, in an
attempt to identify all cases of mesothelioma diagnosed in 1982. Occupational histories
were obtained when possible, the pathology slides were reviewed, and when tissue was
available the lung was analysed for asbestos content. They identified 19 cases (17 men
and 2 women) of confirmed mesothelioma; obtained occupational histories for 16, and
analysed lung tissue fibre content in 7. The calculated incidence per year based on their
data was 17 cases per million men and 1.9 cases per million women over age 15. Compared
with data from 1966–1975, this was a marked increase in the incidence rate for
men, but no obvious increase for women – similar to the observations by Price in the
United States. Fourteen of fifteen men had a history of occupational asbestos exposure,
mostly in shipyards, or in construction or insulation work. In the six men whose lung
tissue was analysed, the pulmonary content of chrysotile asbestos was within the range
of the general population, but the values for amosite and crocidolite were elevated on
average 300-fold compared to a reference population. No commercial amphibole was
found in the lungs of the one woman analysed. They concluded that the cases in women
may not have been associated with asbestos, and may represent the background nonasbestos
associated mesothelioma rate in the general population.

Causes of malignant mesothelioma
Asbestos
The adverse effects of asbestos were first observed in the early 1900s and the relationship
to mesothelioma was suggested in the 1940s. One of the earliest reports linking
mesothelioma to occupational asbestos exposure came out of the medical clinic at an
asbestos mine in Canada. At a scientific meeting in 1952 Cartier,10 then in charge of the
industrial medical clinic at Thetford Mines, Quebec, reported eight cases of respiratory
cancer, two of which he described as pleural tumors. He declared that two such rare
cancers in a small series of only eight cases suggested an occupational origin. By 1960
the scientific community generally recognised asbestos as a cause of mesothelioma.
However, the issue was far from settled. In a national survey of mesothelioma in
Canada from 1960 to 1968, McDonald and colleagues11 found a history of asbestos
contact in a relatively small proportion of cases – mostly in insulation and allied trades
rather than in the asbestos-producing industry. They surveyed pathologists across the
country to find all cases of mesothelioma after 1959. They found 165 cases (111 pleural,
47 peritoneal, 3 both and 4 pericardial). Updates in subsequent years showed a
slightly increased association with asbestos exposure, but still lower than expected.
When pathologists reviewed the cases12 the diagnosis was confirmed in only about 50
per cent of the cases, among whom the incidence of asbestos exposure was also higher.
A follow-up analysis added a chrysotile mining industry cohort and two small groups of
employees in gas mask factories to the survey data. They found 254 fatal cases of
mesothelioma (181 men, 73 women) in Quebec from 1960 to 1978. They were able to
obtain occupational and residential histories for the majority, and found that only
about 40 per cent of the male cases and 5.4 per cent of the female cases were attributable
to occupational asbestos exposure (asbestos manufacture, production, insulation,
heating trades, shipyards, and construction). Six people probably had household exposure.
The intervals between first employment and death from mesothelioma were
longer for miners and millers than for manufacturing workers. All the miners and
millers had pleural mesothelioma, while the factory workers included 8 with peritoneal
mesothelioma. The incidence did not clearly increase over this time period.

In 1994 Spirtas and colleagues15 published a study aimed at defining the attributable
risk of asbestos exposure in the United States. They used the Los Angeles County
Cancer Surveillance Program, the New York State Cancer Registry (excluding New York
City), and 39 large Veterans Administration hospitals to identify people diagnosed with
mesothelioma from 1975 to 1980. They identified 208 cases of pathologically confirmed
mesothelioma. Controls were 533 people who had died of other causes. They interviewed
immediate family members to obtain asbestos exposure history. Among men the attributable
risk for asbestos exposure was 88 per cent, among women it was 23 per cent
(although the confidence interval was very wide at 3–72%). The increasing incidence of
mesothelioma in the United States is primarily due to the increased incidence among
men, thus probably reflecting more occupational exposure.
Most patients with mesothelioma do have a history of exposure to asbestos, although
it may have been brief and remote in time. The variability in percentages of
cases with identifiable asbestos exposure may be related to several factors: incomplete
history taking, unknown or hidden occupational exposures, or environmental exposure.

Other possible causes
Although it is clear that asbestos causes mesothelioma, other factors also play a role.
It is not clear why only a relatively small proportion of people exposed to asbestos
develop mesothelioma, or why anywhere from 20 to 60 per cent of people with mesothelioma
in different studies lack a known history of asbestos exposure. There
may be as-yet unidentified cofactors that act as co-carcinogens with asbestos in the
induction of mesothelioma.
Evidence that not all cases of mesothelioma are caused by asbestos is threefold.
First, we have evidence that a background rate of mesothelioma does exist.18 Historically,
primary malignant tumours of the pleura were recognised by pathologists at autopsy
in the late nineteenth century, before the industrial use of asbestos could have
been responsible, and without any link to occupation. Epidemiological evidence from
mortality statistics for mesothelioma over the past 50 years in the United States and
Canada also support a background rate of mesothelioma. Many studies have shown that

disease incidence has increased much more rapidly in men than in women, reflecting
occupational asbestos exposure among the men. The rate among women has remained
relatively stable, indicating that it may be the background rate of mesothelioma. Backward
extrapolation suggests that before the diverging pattern began, mortality was about
1–2 per million population in both sexes. This possibility is supported by data from regions
with low mesothelioma mortality where both male and female rates are at about
this level, and by data for California, after exclusion of occupationally related cases.
However, it is also possible that waterborne or airborne fibres originating from naturally
occurring asbestos deposits are the cause of this background rate of mesothelioma.
Second, since the latency period after asbestos exposure is normally at least 14
years, the occurrence of malignant mesothelioma in childhood may be evidence of a
genetic predisposition, or an environmental exposure which may cause mesothelioma
in some cases. In a Canadian survey, from 1960 to 1980, four fatal cases in children
were found through systematic inquiry of all pathologists – a rate of about 0.7 per 10
million per annum. A similar figure can be derived from the 13 cases identified by
Grundy and Miller from death certificates in the United States from 1965 to 1968.
And data from the SEER programme in the United States (1973–84) reveal an incidence
among children of about 0.5 per 10 million per year.
Third, lung burden analyses provide some evidence that not all cases are caused
by asbestos. In several studies in both the United States and Canada, a certain number
of cases could not be attributed to either amphibole or chrysotile fibres. Various
other agents have been suggested as additional causes of mesothelioma. A report from
India suggested that organic fibres may cause mesothelioma in sugar cane workers.
Reports of cases of mesothelioma in sugar cane workers in southern Louisiana provided
some support for this hypothesis. After the harvest, the cane is burned off and
long, thin, respirable silica fibres form, a process that might also occur after forest
fires and other circumstances. Theoretically, these fibres could cause mesothelioma in
much the same way that asbestos fibres do.

United States and Canadian asbestos use and
production
Historically, Canada has been one of the largest asbestos miners, and the United States
has been one of the largest asbestos consumers. In 1877 extensive chrysotile asbestos
deposits were discovered in eastern Quebec. These areas were developed in the 1880s,
and soon half of all the world’s asbestos was mined in Canada. A small amount was
also mined in the United States. The asbestos industry grew rapidly in the 1930s and
1940s and the use of asbestos in North America was at its peak from the 1950s through
1970. In 1970 the United States used almost a quarter of the asbestos produced throughout
the world

In contrast to other areas of the world, the United States used predominantly
the serpentine chrysotile type asbestos. Approximately 95 per cent of United States
consumption has been chrysotile. Currently it accounts for more than 99 per cent of
United States asbestos consumption.The United States imported 165 000 tons of
chrysotile in 1935 and 650 000 tons in 1962. Crocidolite was used much less, and much
later. Less than 500 tons of crocidolite were imported in 1935, and only 20 000 tons by
1962. In 1997, it was reported that 238 tons were imported, but it has been estimated
that only 5 tons were used. Small amounts of amosite have also been used in the
United States, averaging about 16 000 tons per year in the 1960s.
Starting in 1970, United States consumption of asbestos declined precipitously,
largely as a result of mounting evidence of its toxicity. During the early 1970s the
number of asbestos lawsuits increased dramatically. World production of asbestos has
declined in the 1990s to approximately 2.7 million tons per year – about 50 per cent of
the peak production in 1973. Canada is now the second largest miner of asbestos, after
Russia. In the United States, consumption of asbestos in 1997 was only 21 000 metric
tons

Occupational exposures and the fibre-type debate
Over the years investigators have looked at the incidence of mesothelioma in different
cohorts of workers in North America, attempting to answer several questions about
asbestos and mesothelioma. Initially studies were aimed merely at documenting the

relationship between asbestos and mesothelioma. Subsequently, debate became heated
over whether different types of asbestos fibres were more or less carcinogenic than
others – particularly whether the serpentine type (chrysotile) was as carcinogenic as
the amphiboles (crocidolite, amosite and tremolite). North America was the perfect
forum for this debate, since the mines and mills in Canada seemed to provide a means
to study workers exposed only to chrysotile, and in the United States chrysotile has
been the predominant exposure. Mossman, Gee, McDonald and Wagner, who favored
crocidolite as the sole cause of mesothelioma, squared off against Selikoff, Smith, Cullen
and others, who believed that all forms of asbestos were a threat


Insulators
Asbestos has been used as an insulation material since 1866. In the United States and
Canada, until the early 1940s, the only asbestos used was chrysotile, almost exclusively
from the Canadian mines. During the 1930s the United States began to use small quantities
of amosite asbestos imported from South Africa. Amosite was mixed with chrysotile
in making insulation blocks. The insulators likely had heavy though intermittent
exposure. Exposures were especially high in shipyards where they ripped out and replaced
old asbestos insulation in the confines of boiler rooms and submarine hulls.
The insulators have been a well-organised group. The first insulation workers’
union was formed in New York City in 1883 – the “Salamander Association of Boiler and
Pipe Felters” (the name came from a story about Marco Polo, who was shown some
inflammable cloth while traveling through Siberia, and was told that it was made from
salamander’s wool). In 1910 the Salamander Association merged with other independent
locals in the United States and Canada and became the International Association
of Heat and Frost Insulators and Asbestos Workers. The union rolls thus provided the
ideal cohort to study the relationship between asbestos and mesothelioma.

Miners and millers
The miners and millers of chrysotile asbestos in Canada seemed to be the perfect
group in which to pursue these questions, as they were thought to represent a group
exposed to only one type of fibre. However, the situation was not so simple. Canada
and Russia have produced most of the world’s asbestos. Chrysotile asbestos was first
mined in 1878 in the Eastern Townships region of Quebec, and is now the overwhelming
type mined in Canada. The two main mining areas are now the towns of Asbestos
and Thetford. Most of the mines were open pit mines, where 13-metre (40-foot) holes
were drilled in asbestos-containing ore deposits, and were then blasted with explosives.
Workers loaded the loose ore onto trucks and then transported it to a crushing
machine. Moisture was removed by heating the crushed ore in huge dryers. During
milling the ore went through stages of screening and vacuum separation in which
loose asbestos fibre was lifted from the rock. Conveyor belts, chutes, and vacuumexhaust
pipes transported the ore and fibre through the mills. Most employees worked
in the mills, cleaning up spilled asbestos fibre, feeding asbestos into storage bins, and
maintaining equipment. Before 1970, only total dust counts were made – asbestos
fibre concentrations were not measured separately. Virtually all jobs in this industry
were very dusty. In the 1970s asbestos fibre concentrations were measured, and it was
found that average asbestos-fibre concentrations were significant, especially in the
drying and crushing operation, and greatly exceeded the asbestos standard of 5 fibres
per cubic centimetre (5 f/cm3) of air then in effect in the United States

The shipyards
Currently the highest age adjusted incidence of mesothelioma in the United States is in
Washington and Oregon, probably due to the shipbuilding industry. Asbestos exposure
in shipyards frequently involves massive exposures of short duration in enclosed spaces,
as opposed to the chronic exposure in mining and manufacturing processes. Thus, shipyard
workers represent a group where the effects of short-term intense exposure might
be seen. Asbestos has been used to insulate the steam pipes and boilers of ships since the
turn of the century. The asbestos may be woven in a mattress of cloth and wrapped
around piping, applied as a cement-like material, or sprayed as fireproofing. Not only
the insulators were exposed – bystander exposure in this situation was probably significant,
since the work was done in the enclosed spaces of ship boiler rooms.
Shipyard worker populations fluctuate greatly. In 1918 there were 318 500 shipyard
workers in the United States, compared to 75 000 during the 1920s and 1930s. With
the approach of World War II, the shipyards underwent a tremendous expansion and
shipbuilding became the largest manufacturing industry in the United States for a time.
By the end of 1943, 1 722 500 people were employed in shipyards. In some yards between

10 and 20 per cent of these workers were women. Turnover was very high, so it has been
estimated that about 4.5 million people were employed in shipyards during World War
II. After the war the number of shipyard workers rapidly decreased.
Since no dust counts were taken, no one really knows what levels of asbestos
fibre existed. They were likely very high at times. In 1975 the use of asbestos in shipbuilding
and ship repair was largely abandoned. However, ship repair remained a problem
because asbestos previously used as insulation on steam pipes and boilers had to
be ripped out and replaced during renovation and repair work.

Factory workers
Authors on both sides of the fibre-type debate have investigated factory workers in
various industries. In some workplaces, measurements of dust exposure were made,
and both the exact fibre type and degree of exposure could be evaluated. Thus these
factories proved to be optimal forums to continue the debates over fibre type versus
fibre size versus amount of exposure.
In 1967, Enterline and Kendrik conducted a study of the effect of asbestos exposure
on factory workers. They recorded the causes of deaths among 21 755 white men aged 15 to
65 who worked in several different asbestos-products plants in the United States at some
time during the period 1948 to 1951, identified from social security tax returns filed with the
United States Bureau of Internal Revenue for the first quarters of 1948, 1949, 1950 and
1951. Deaths were identified from death claims filed with the social security administration
through 1963, and death certificates were reviewed. A cohort of 6281 cotton textile workers
was used as a control group for the asbestos textile workers. Among 1853 death certificates
examined, only one listed mesothelioma. Although respiratory cancer rates were increased
overall, mortality was actually lower than for the general population. At most the follow-up
was only 17 years, so the full effect of asbestos exposure had probably not yet been seen.

Construction industry
In the past, the construction industry accounted for an estimated 70–80 per cent of
total United States consumption of asbestos fibre, and an enormous number of workers
in various construction trades have been exposed to varying amounts of asbestos.
From 1958 to 1973, until the Environmental Protection Agency abolished the practice,
asbestos fireproofing material was sprayed in more than half the multistoried buildings
constructed in the United States. About 25 per cent of the sprayed material would
fail to adhere, and was released into the air. All workers on the site during and after
the spraying were exposed. Thirty per cent of the water distribution pipe sold in the
United States in 1974 was asbestos cement.
A mortality study of the members of the union of plumbers and pipefitters in the
United States found significant excesses in proportional mortality ratios for malignancies,
including 7 deaths due to mesothelioma.72 Two studies of sheet-metal workers in New York
City found significantly increased mortality, and mesothelioma was recorded on death
certificates in 9 out of 716 total deaths (1.3 per cent).73,74 Studies in construction workers
are difficult since specific exposures often are not known, and workers change jobs frequently.
Epidemiological studies, however, consistently show that construction workers
are one of the groups most at risk for asbestos-related disease and mesothelioma.
Miscellaneous exposed groups
Several other groups in North America have been studied to determine the effects of asbestos
exposure. Railroad machinists were exposed to significant amounts of mostly chrysotile
asbestos, especially during the application and removal of asbestos insulation on the boilers
of steam locomotives. The last steam engine repairs took place in the late 1950s. In 1988
Mancuso75 published a study investigating the risk of mesothelioma among railroad machinists.
His cohort was railroad machinists employed by a company before 1935, and still
alive in 1945. He analysed the data for 181 people hired from1920 to 1929. By 1986, 156 were
identified as dead. There were 41 cancer deaths, 14 with mesothelioma. The relative risk was
1 mesothelioma in every 13 machinists hired. Similarly, in 1986 Schenker et al.76 completed
a case-control analysis of mesothelioma among United States railroad employees. The United
States Railroad Retirement Board notified investigators of all deaths among male railroad
workers occurring in 1981–82. They reported 15 059 deaths during the study year. Twenty
cases of mesothelioma were identified. More cases occurred among the workers in asbestosexposed
job categories. The latency period was 30 years or more.
In the jewellery industry, asbestos has been used to make

Non-occupational exposure
Several studies have documented that non-occupational exposure to asbestos is also
associated with mesothelioma. Given that chrysotile is thought to be less carcinogenic
than the amphiboles, the question was then raised as to whether the low levels of
exposure seen with non-occupational situations can cause mesothelioma or other asbestos-
related diseases.
Researchers at Mt Sinai Hospital in New York City studied the health experience
of 679 household contacts of the 1664 workers employed at the Paterson, New
Jersey, factory between 1941 and 1954. They discovered 5 cases of mesothelioma in
family members of factory workers. Asbestos dust was also found in the homes of former
Paterson factory workers 20 years after the factory was shut down. There was some
indication that both the occurrence of mesothelioma and the length of the latent period
were dose related.
The impact of neighbourhood asbestos exposure has also been investigated in
relation to the Paterson factory. In 1979, Hammond and Selikoff traced 2447 men
who were living within 800 metres of the factory in 1942, and compared them to residents
of a neighbourhood a few kilometres away. They found that dust samples collected
from houses located near the asbestos factory contained appreciable amounts
of asbestos fibre even many years after the factory closed. So far, however, they have
not reported any increased mortality.
In 1996, Berry published a study of the incidence of mesothelioma in Somerville
County, New Jersey, the location of the Manville asbestos plant. This plant produced
multiple asbestos products, primarily (95 per cent) using chrysotile asbestos. A previous
study in 1987 indicated that the rate of mesothelioma was significantly higher in Somerset
County than in the rest of New Jersey. Berry expanded on this study. He used the New
Jersey State Cancer Registry to identify cases of mesothelioma from 1979 through 1990,
and compared the incidence rates in Manville, Somerset County excluding Manville,
and the rest of New Jersey. He then excluded all who worked at the Manville plant.