The Western Australian Mesothelioma Registry: Analysis of 60 years of cases

Australia introduced a partial ban on asbestos consumption in 1984. There is continuing concern about exposure to asbestos in the built environment and non‐occupational exposures. The aim of this study was to describe epidemiological trends of mesothelioma in Western Australia (WA) over the 60 years since the first case was recorded.


INTRODUCTION
Mesothelioma is an incurable, universally fatal disease predominantly of the pleural and peritoneal mesothelium, primarily caused by exposure to asbestos fibres. 1 Global asbestos consumption peaked in the last century and despite a ban on asbestos import and utilization in more than 60 countries, it is estimated that more than a million tonnes of asbestos was exported to predominantly developing countries in 2019. 2,3ross developed countries, the 'first wave' of the mesothelioma was caused by exposure to raw asbestos fibres from the mining, milling and handling of asbestos. 1ustralia both imported and mined asbestos including chrysotile in New South Wales, and crocidolite in Western Australia (WA). 4 Between 1937 and 1966, approximately 150,000 tonnes of crocidolite was mined from the Wittenoom region WA. 4 The nearby town of Wittenoom (established for miners and their families) was only completely closed in 2022.
The 'second wave' of mesothelioma was caused by workers who used asbestos-containing products in industry, typically construction, boiler workers, carpenters, railways and dockyard workers.With a partial ban on asbestos containing products in Australia in 1984, there has been much interest and concern about the 'third wave' of mesothelioma; people diagnosed with mesothelioma after short term or low level exposure to residual asbestos in the built environment at home or at work. 5 A previous report using data from the Western Australian Mesothelioma Registry (WAMR) demonstrated a rising number of mesothelioma cases attributed to home renovation and do it yourself (DIY) projects. 6he recent report from the Australian Mesothelioma Registry (AMR) demonstrated for the first time a decline in age-standardized incidence and death rates in Australia, peaking in the early 2000's. 7The purpose of this current study was to describe epidemiological trends over the 60 years of mesothelioma in Western Australia, with particular analysis of cases with home renovation/DIY exposure.

METHODS
A detailed report on the establishment and methods of the WAMR is described online and elsewhere 6 and is described briefly below.

Case ascertainment
The WAMR is linked to the Western Australian Cancer Registry.Clinical data is sourced from pathology laboratories, hospitals and clinics and all possible cases are reviewed by the committee that includes an expert pathologist, a respiratory physician, an epidemiologist and an occupational physician.Data on all confirmed cases in the registry were extracted for this analysis.

Pathological diagnosis
Every pathology report is reviewed to confirm the diagnosis with all available cytology, histology, necropsy, immunohistochemistry and/or electron-microscopy reports assessed.The subtype of mesothelioma was recorded as epithelioid, sarcomatoid, biphasic or not otherwise specified (NOS). 8Cases diagnosed by effusion cytology are accepted as true cases and recorded as epithelioid. 9,10Where there was uncertainty, original slides are retrieved and reviewed before a final determination was made.

Classification of exposure
The WAMR assigns a primary exposure code dependent the main source of exposure and year of first exposure.Where there is more than one source of exposure the committee considers the most significant exposure accounting for duration and likely intensity of exposure.
The WAMR has 55 exposure codes (including 48 occupational and 5 non-occupational).These were condensed the into seven categories: two occupational: asbestos miners and millers from Wittenoom (Wittenoom workers); all other asbestos workers (other asbestos workers); three non-occupational: residents (and visitors) from the town of Wittenoom (Wittenoom residents); home maintenance (home renovators); other non-occupational exposure (other non-occupational); and people whose exposure could not be identified (unknown) or who were known to have no exposure after interview (no known).Appendix S1 in the Supporting Information includes Table S1 of the exposure code matrix.

Data analysis
Data were extracted on all confirmed mesothelioma cases available on the WAMR.Overall survival was calculated from the date of diagnosis (from pathological report) to date of recorded death on the WA Cancer Registry; the date of censoring was 01 August 2020.
All statistical tests reported are two-tailed.Statistical calculations and data manipulation were performed using the SPSS (Version 26.0.Armonk, NY: IBM Corp).Survival between groups was compared using the Log Rank test and logistic regression was performed to examine for any association with survival more than 5 years including the variables sex, age, time since first exposure (TSFE), exposure source, mesothelioma morphology and mesothelioma site.
Incidence rates for cases with DIY exposure, for the whole group and for men and women separately in 5-year age groups (from 40 years onwards to 85 years and older) and 5-year calendar periods (from 1980) excepting the last period 2015-2018 were created.
Data were stratified by decades for various analyses and due to case low numbers, the 1960-1969 and 1970-1979 where combined.Due to a reporting lag to the WAMR, some calculations are censored at 2018 (considered the last complete reporting year) and others utilized all available data, through to 2020.
The study was approved by the Department of Health WA Human Research Ethics Committee (REF DOH HREC 2010/40).

SUMMARY AT A GLANCE
This study describes the exposure and outcomes of 2796 mesothelioma cases.The median survival was 298 days and latency 46 years, both doubling over the study period.Most females have a nonoccupational exposure to asbestos.The incidence of do it yourself (DIY)/home renovation exposure has fallen, suggesting the asbestos ban has been effective.

Population
In total, 2796 cases of mesothelioma were identified in Western Australia.The median (IQR) age of the population at diagnosis was 70 (62-78) years, and a median TSFE of 47 (38-55) years.As expected, males comprised the majority of the population (n = 2368, 84.7%).Overall survival across the population was a median of 299 (128-585) days.The population was primarily non-indigenous (n = 2712, 97%), with only 20 (0.3%) being of Aboriginal descent, and the remainder unknown or with unspecified backgrounds.
Pleural mesothelioma was recorded in 2620 (93.7%) cases, with 169 (6%) identified as peritoneal mesothelioma and 7 (0.1%) originating from the testes or the mediastinum.The epithelioid subtype was the most prevalent (n = 1730, 61.9%).There was a statistically significant association between epithelioid and sarcomatoid containing mesothelioma subtype and sex observed with a higher proportion of females with epithelioid mesothelioma (χ 2 = 21.83,df = 3, p < 0.001).The majority of cases were confirmed with biopsy (69.9%).Table 1 provides a summary of the data.The number of cases diagnosed each year has increased to >100 per year since 2008, peaking at 132 cases in 2017.

Source of asbestos exposure
Overall, workers exposed to manufactured asbestos products comprised the majority of cases (n = 1591, 56.9%), see Table 2 and Figure 1.There were 375 (13.4%)Wittenoom workers recorded, and 106 (3.8%) ex-residents of Wittenoom.The proportion of cases with non-occupational exposure has increased over the duration of the study and was proportionally higher in females (n = 225, 52.6%), compared with males, (n = 224, 9.5%), see Table 2. DIY/home renovation was recorded as the primary exposure in 227 (8.1%) cases.The recorded pathological subtype changed over the study period, with mesothelioma NOS initially accounting for 26.3% of cases, falling to 2.6% in the final decade.Similarly, epithelioid initially accounted for 30.3% and has increased to 67% of all cases, see Table 3.

DIY/home renovation exposure
There were 227 cases of mesothelioma whose source of exposure was DIY, with the first case recorded in 1981, and four (1.5% of DIY cases) with first exposure after 1984.DIY was proportionally the most common source of asbestos exposure for women but accounted for only 5.6% (n = 133) of cases in men (Table 2), with the most recent year of first exposure being 1991.
Adjusted incidence rates for DIY exposure are presented in Table 4 and Figure 2. The incident rate ratio (comparing subsequent calendar periods with the baseline 1980-1984) was highest in 2005-2009, followed by some attenuation in rates in the subsequent two calendar periods.

DISCUSSION
To the authors knowledge this is the longest running mesothelioma cohort study in the literature and uniquely describes changing patterns of asbestos exposure with the first exposure more than 100 years ago and 60 years of cases.This report provides evidence for the effectiveness of the ban on use of asbestos containing products in Australia with a fall in incidence of DIY/renovation exposure.Whilst mesothelioma remains a disease predominantly of males, there is a clear distinction between sexes in the source of exposure, with the majority of females with non-occupational exposure(s).
As expected for a predominantly occupational disease, most (84.7%)mesothelioma cases are in males, however, this study demonstrates some important difference in exposure and outcome data between sexes.For instance, the proportion of females with a non-occupational source of exposure is much higher (more than half, with one in five females having DIY exposure, compared to one in twenty males).This has previously been recognized and is likely a reflection of historical social and work-related factors, and how other household members can be affected by DIY exposure.The recent AMR report described just 7% of females having an occupational exposure, 7 in this WA cohort 12.8% of females had a definite occupational exposure.It should also be noted that over a quarter (n = 111) of mesothelioma in females did not have a clear source of exposure apparent after reviewing the clinical notes.This is likely a reflection of limited exposure history taking and perhaps a lack of appreciation of para-occupational exposure as an important source, particularly for family members, of exposed workers. 1 The high proportion of 'unknowns' means that actual numbers of exposure through other non-occupational sources may be much higher.
This longitudinal cohort provides an insight into changing source of asbestos exposure over time with the earliest exposure recorded as 1920.The absolute numbers of Wittenoom workers developing mesothelioma has not changed much since the 1980s, but there has been a steady fall as a proportion of all cases since this time.As expected, the 'second wave' of individuals working with processed asbestos containing products comprises the largest proportion of all cases (56.7%).There has been significant concern about the 'third wave' of exposure from DIY home renovation 5,6 and the data in the present study demonstrate that both the proportion of DIY cases has not risen in the last decade and, the incident rate ratio for DIY cases has fallen (Figure 2).Just 4 DIY cases and 16 other mesothelioma cases had a recorded first year of exposure as 1984 or beyond, perhaps as a reflection of the legacy of asbestos containing material in the built environment. 5This provides some of the first evidence that the asbestos ban in Australian enacted in 1984 may have been successful in reducing exposure to the wider community.This complements the recent Australia-wide data demonstrating a decline in age-standardized mesothelioma incidence rates in Australia, 7 although it may be an artefact due to longer latency for low-level or short-term exposures.A large Netherlands-based population study recently reported a fall in mesothelioma incidence for males younger than 80 years old, thought to be a reflection of the nation-wide ban on asbestos from 1993. 11Increasing evidence of the effectiveness of bans on asbestos utilization should be noted by policy makers in countries that continue to import and utilize asbestos.Time since first exposure (i.e., latency) has increased in each decade to a median of 52 years.At the same time, the median year of first exposure has only changed from 1950 to 1962, a reflection of the historic peak utilization of asbestos in WA (see Figure S2 in the Supporting Information).3][14] The increasing age at diagnosis over time is also likely a reflection the relatively fixed time point as to when asbestos utilization was most widespread.
While overall survival is around 10 months for the whole cohort, this article demonstrates a doubling in survival time over the 60-year period (median 156-360 days).This apparent improvement will be predominately be due to a lead time bias from earlier diagnosis, with chemotherapy regimes only being adopted into clinical practice from around 2005 onwards (with only a modest survival benefit of 2.4 months in the original trial population 15 ) and approximately 60% of the WA mesothelioma population receiving systemic therapy. 16The recent advances in immune-based therapy will not be reflected in the current data. 17mprovements and advances in diagnostic techniques including immunohistochemical and chromosomal loss analyses 8,18-20 also likely account for the apparent change in tumour subtypes observed over the decades, with a falling proportion of mesothelioma NOS and increasing numbers with an epithelioid subtype.The World Health Organization recently revised the classification of tumours of the pleura with retention of the subtypes but recommendation to no longer use the prefix 'malignant' for localized or diffuse mesothelioma. 82][23] To the author's knowledge, this is the first time that the proportion of females with an epithelioid subtype is statistically significantly different to males.The reasons for this association is not clear, it is unlikely due to exposure to different asbestos fibre type 24 and is perhaps a true signal only noted in this large dataset.There is no known biological reason for the better survival observed in females and, while the analysis of the long survivors (beyond 5 years) demonstrated a higher proportion of females and epithelioid subtype, neither of these variables was statistically associated with long survival after multivariable analysis.
The strengths of these data are the size of the database and its longitudinal nature, together with the completeness of data with all cases of mesothelioma being notified to the state cancer registry.The weaknesses of the data are the lack of comorbidity or treatment data, although the WAMR was not designed to be a clinical outcomes registry.Staging data is similarly not available and is rarely used in clinical settings in Australia.The reporting lag to the cancer registry and/or subsequent confirmation through the WAMR is noted, although all mesothelioma cases are ultimately reviewed and confirmed.
In conclusion, there is a clear difference of exposure source(s) and outcomes between sexes with mesothelioma in this large longitudinal cohort.There is further evidence of effectiveness of the Australia-wide ban in use of asbestos in 1984 with a falling incidence rate of home renovation exposure leading to mesothelioma and very few cases with first exposure after this time.Together, these data should inform policy makers to protect further generations from asbestos containing products in the built environment.

F I G U R E 1
Cumulative mesothelioma cases in Western Australia 1960-2020 stratified by exposure category.

F
I G U R E 2 Relative change in mesothelioma incidence from do it yourself (DIY) asbestos exposure for males and females by calendar period, and gender, 1980-2018.investigation (equal); project administration (equal); validation (equal); writingoriginal draft (equal); writingreview and editing (equal).
The Western Australian Mesothelioma Registry population characteristics stratified by sex.
T A B L E 1Abbreviations: IQR, interquartile range; NOS, not otherwise stated.T A B L E 2 Sources of primary asbestos exposures for WAMR mesothelioma cases.Abbreviation: DIY, do it yourself.a Most common categories.
T A B L E 3 Trends over decades of survival, latency (TSFE), year of first exposure (all median IQR) and pathological subtype for all mesothelioma cases (n = 2796).