If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author. Address for correspondence: Pietro Gino Barbieri, MD, Occupational Health Unit, Local Health Authority and Mesothelioma Registry of Brescia, Via Foro Boario 8/b, 25124 Brescia, Italy.
Malignant pleural mesothelioma is an asbestos-related disease with poor survival. The prognostic role of histologic subtype is well established. Some studies (without a biological hypothesis) suggested that higher asbestos lung burden is associated with reduced survival.
We selected subjects from two series of necropsies: residents in Brescia province (North-West Italy) and workers (or persons living with them) employed in the Monfalcone shipyards (North-East Italy). Asbestos fibers and asbestos bodies in lung samples were counted using a scanning electron and an optical microscope, respectively. Separately in the two series, we analyzed median survival time and fitted multivariable Cox regression models (adjusted for sex, period and age at diagnosis, and histopatholocical diagnosis) to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for three levels of asbestos fiber counts (reference: <1 million fibers/g of dry lung tissue).
We analyzed 185 necropsies, 83 in Brescia and 102 in Monfalcone. Despite a much higher lung burden in Monfalcone patients, median survival was slightly shorter in Brescia (8.3 mo) than in Monfalcone (10.2 mo). In Brescia, medium (1.0–9.9) and high (10+) fiber burden HRs were 0.91 (95% CI: 0.54–1.53) and 1.23 (95% CI: 0.41–3.70), respectively. In Monfalcone, the corresponding HRs were 1.18 (95% CI: 0.59–2.35) and 1.63 (95% CI: 0.77–3.45), respectively.
No relationship between asbestos lung burden and survival was found. Histologic subtype was the strongest prognostic determinant.
Malignant pleural mesothelioma (MPM), like other malignant mesotheliomas (MMs), is a cancer causally associated with asbestos exposure. Despite current available therapy, the prognosis is poor, with estimated median survival time of approximately 10 months in population-based settings.
To the best of our knowledge, six studies evaluated the association of survival in patients with MM or MPM with lung asbestos burden, measured as concentration of asbestos fibers (AFs) or asbestos bodies (ABs).
We are aware of no plausible biological hypothesis to support a role of lung asbestos burden in affecting survival after mesothelioma occurrence. To shed light on this controversy, we exploited existing information from two series of necropsies performed on patients with MPM in the context of criminal trials in Northern Italy. The specific aim of our study was to evaluate the relationship between asbestos lung AF burden and survival separately within these two groups, which together represent the second largest necropsy series with information on survival and lung asbestos burden. AB counts are presented for descriptive purposes, but they were not used in survival because they were unavailable for a substantial proportion of subjects.
Materials and Methods
Subjects were selected from two series of necropsies performed from 1994 to 2017. The first group of 83 autopsies was performed in the context of the activity of the Malignant Mesothelioma Registry of Brescia Province (Lombardy region, North-West Italy), as described in detail elsewhere.
All individuals in Brescia had a histologically confirmed MPM diagnosis.
The second group of necropsies was performed in the context of criminal trials regarding a large shipyard in Monfalcone (Friùli Venezia-Giulia region, North-East Italy) on patients affected by MPM (definite or suspected) and previously exposed to asbestos (shipyard workers or their family members).
In particular, this work was originally carried out in the framework of an expert testimony to the public prosecution office by one of us (PGB). From this series, we selected necropsy reports for 102 subjects with definite (histologically confirmed) MPM.
For both series, with histologically confirmed diagnosis in life, information on demographics, age at MPM diagnosis and death, histologic subtype, and pleural plaques was collected. AF counts and percentage of amphibole fibers were available for all cases, whereas AB counts were available for only 22 (26.5%) individuals in Brescia and 90 (88.2%) in Monfalcone.
Information on presence of asbestosis at histologic examination was also available in the shipyard group. According to the criteria of Asbestosis Committee of the College of American Pathologists and Pulmonary Pathology Society, asbestosis diagnosis was done in the presence of two or more ABs per cm2 of a 5-μm-thick fresh lung section, in combination with interstitial fibrosis of the appropriate pattern.
In both series, necropsy had been ordered by the public prosecutor in the context of criminal trials. For this reason, informed consent was not required; institutional review board approval was obtained.
Analysis of ABs and AFs
Following the guidelines proposed by the European Respiratory Society, lung tissue samples of 1 cm3 were collected from different parts of the lung and stored in a formaldehyde solution.
Samples were then prepared by lyophilization, plasma asher digestion, and filtration. The analyses were carried out using a scanning electron microscope equipped with x-ray fluorescence microanalyzer at 12,000 magnifications (ZEISS EVO 40; Zeiss and Oxford XMAX microanalyzer; Oxford) as described in detail elsewhere.
Fiber concentration was expressed as total number of fibers (amphibole: amosite and crocidolite, and chrysotile; noncommercial amphibole: tremolite and actinolite) with length of more than 1 μm/g of dry lung tissue with 95% confidence interval (CI). This method does not allow differentiation of amphibole fibers if they have a diameter of less than 0.08 μm; for this reason, crocidolite and amosite were counted together and reported as amphiboles.
In the same laboratory, 50 mg freeze-dried lung sample was treated according the method ISTISAN 17/12
to count AB with optical microscope at 500 magnifications. The concentration was expressed as number of AB/g of dry lung tissue with 95% CI. The laboratory is accredited according to the International Organization for Standardization 17025 by ACCREDIA on both methods (laboratory n. 1324 sede E). AB and AF concentrations were expressed in thousands and millions per gram of dry lung tissue, respectively. Both AF and AB analyses were performed between 2000 and 2017 blindly of subjects’ disease and occupational history.
We compared subjects from Brescia MPM Registry and Monfalcone shipyard using chi-square (for categorical variables) or Wilcoxon-Mann-Whitney (for quantitative variables) tests. AF lung burden in these two series was compared with that previously measured in 13 subjects from the Brescia province never exposed to asbestos (11 men, two women, aged 54–67 y) who died from cardiovascular disease.
We informally compared median survival times between the series of Brescia (lower exposure) and Monfalcone (higher exposure). Nevertheless, because the two groups largely differed for several potential confounding variables (including the period of diagnosis), formal survival analyses were performed separately within the two series of necropsies. Survival time was computed starting from the date of histologically confirmed MPM diagnosis. We fitted univariate and multivariable (adjusted for gender, period of diagnosis, age at diagnosis, and MPM morphology) Cox proportional-hazards regression models to calculate hazard ratios (HRs) and 95% CIs for three levels of AF counts. We tested the proportional-hazards assumption on the basis of Schoenfeld residuals. Given that AB counts were missing for a number of subjects, we did not use them in Cox analyses. Statistical analyses were performed with Stata 17 (StataCorp, 2021).
Subjects from Brescia MPM Registry included more women, were younger, and had less pleural plaques than subjects from Monfalcone shipyard (Table 1). Distribution of histologic MPM types was similar in the two groups. Approximately one-fourth of subjects from Monfalcone had histologically confirmed asbestosis. Subjects from Brescia MPM Registry had much lower AB and AF counts. In both series, the large majority of AF were amphiboles, but in lung tissue samples from Brescia, patients were found to have less amphibole fibers, either in absolute or relative terms. Nevertheless, a negligible difference (1.9 mo) in median survival time was observed between the two groups of subjects (8.3 mo in Brescia, 10.2 mo in Monfalcone). Both series had higher lung concentrations of AFs compared with unexposed controls (Fig. 1). AF and AB counts were highly correlated (Pearson’s correlation coefficient on log-transformed counts: r = 0.83, p < 0.001).
Table 1Characteristics of 185 Subjects With MPM Included in a Necropsy Study, Italy, 1994 to 2017
Among subjects from Brescia MPM Registry, the adjusted HRs were high for those aged 75 or more years at diagnosis and 3 to 4 times elevated HRs for people with biphasic and sarcomatoid MPM, but there were no important differences according to AF counts (HR = 1.23 only in the highest AF category with a wide CI, p = 0.71) (Table 2).
Table 2HRs and 95% CIs According to Selected Variables in 185 Subjects With MPM Included in a Necropsy Study, Italy, 1994 to 2017
In Monfalcone shipyard workers, the adjusted HRs were much higher for those diagnosed in recent periods, elevated in subjects with biphasic MPM (HR = 1.89) and strongly elevated in those with sarcomatoid histological subtype (HR = 4.34). Again, no important differences according to AF counts were found (HR = 1.63 only for those in the highest AF category, with a wide CI, p = 0.20). In neither group, we found evidence of violation of proportional hazards. Multivariable Cox models in which we analyzed categories of amphiboles yielded similar HRs (results not shown). In Monfalcone subjects, we found no association between survival and presence of histologically confirmed asbestosis (adjusted HR = 1.10, 95% CI: 0.64–1.88, p = 0.73).
In this study, in two necropsy series, we found no association between asbestos lung burden and survival, either in univariate or multivariable Cox analysis. Furthermore, survival was similar between the two groups of patients, despite much higher lung concentrations of AFs, amphibole fibers, and ABs in the Monfalcone shipyard group compared with the Brescia MPM Registry group. Finally, within the Monfalcone shipyard series, no association was found between survival and histologic asbestosis (a marker of high asbestos exposure).
Previous studies that evaluated the role of asbestos lung burden or asbestos exposure as a prognostic factor in patients with MPM found controversial results. Of the seven studies (including the present one) based on lung AF and AB concentrations, four studies found a positive association (Table 3, upper part).
Table 3Studies That Evaluated the Association Between MM and/or MPM Survival and Lung Asbestos Burden or Asbestos Exposure History
Study (First Author, Country, y)
No. of Cases
AB or AF Lung Tissue Counts
Lung asbestos burden
Tammilehto, Finland, 1992
27 MM (26 MPM)
AF g/dry t. SEM-EDS/TEM
Kayser, Germany, 1999
AF g/wet t. LM
Neumann, Germany, 2001
404 MM (387 MPM)
National MM registry
AB/cm3 LM or g/wet t. LM
Christensen, USA, 2008
AB g/wet t. LM
Visonà, Italy, 2021
AF > 5 μg/dry t. SEM-EDS
Laaksonen, Finland, 2022
590 MPM (540 analyzed)
National MM registry
AF g/dry t. TEM
This study, Italy, 2022
AF >1 μg/dry t. SEM-EDS
Asbestos exposure history
Musk, Australia, 1982
Law, UK, 1983
Spirtas, USA, 1988
Ruffie, Canada, 1989
Edwards, UK, 2000
Gorini, Italy, 2005
Regional MM registry
Flores, USA, 2007
Ak, Turkey, 2009
Montanaro, Italy, 2009
National MM registry
Nojiri, Japan, 2010
Berardi, Italy, 2016
AB, asbestos bodies; AF, asbestos fibers; LM, light microscopy; MM, malignant mesothelioma; MPM, malignant pleural mesothelioma; NA, not applicable; SEER, Surveillance, Epidemiology and End Results Program of the National Cancer Institute; SEM-EDS, scanning electron microscope–energy-dispersive x-ray spectroscopy; t., lung tissue; TEM, transmission electron microscope; UK, United Kingdom; USA, United States of America.
The first was a Finnish study on 41 patients with mesothelioma, 40 with MPM and one with peritoneal mesothelioma. The main finding was a better survival (p = 0.031, Mantel-Cox test) in eight patients with low lung asbestos burden (<1 million fibers/g), compared with 19 with high burden (≥1 million fibers/g). Moreover, survival was worst (p = 0.021) in 15 patients with greater than or equal to 50% crocidolite/amosite fibers, best in four patients with greater than or equal to 50 antophyllite fibers and intermediate in eight patients with greater than or equal to 50% other fibers.
In a second study performed in Germany on 73 MPM, survival was shorter in patients with high lung fiber concentration (≥70 AF/g) compared with those with less than 70 AF/g (35 versus 50 wk, respectively, p = 0.04).
Nevertheless, we note that in that study a very unusual choice of the reference category (the moderate instead of the low one) was made when performing a multivariable Cox analysis. As a matter of fact, no trend in reduced survival with increasing AF lung burden was found: survival was worse in the low exposure category than in the moderate one. Considering the width of CIs, the adjusted HR in the low burden (<100 AB/g) category (HR = 3.0, CI: 0.95–5.0, 37 patients) was not much different from the HR in the high (>1099 AB/g lung) category (HR = 4.8, CI: 1.5–15.0, 22 patients).
The fourth recent study in Finland is the largest conducted so far, being based on 590 subjects (527 men and 43 women) identified within the national mesothelioma registry. After excluding 50 subjects with MPM diagnosed at autopsy, an association between reduced survival and lung AF burden was found in multivariable Cox models. More specifically, the authors found a reduced HR (i.e., higher survival in those with high lung burden) using the initial lung asbestos burden (referred to as “initial HR”) and a high HR (i.e., lower survival in those with high lung burden) using lung asbestos burden during the follow-up period (referred to as “HR change”). It is unclear to us how the authors performed a time-dependent analysis given that asbestos lung burden was measured at a single time point (death).
Conversely, two studies found no association between lung asbestos burden and survival. A large study (404 patients, 387 [95.8%] with MPM, 16 with peritoneal, and one with pericardial mesothelioma) with available information on asbestos lung burden was performed within the German mesothelioma register (1987–1999).
Survival in 345 MM cases with high lung burden (>22 AB/cm3) was 13.2 months, almost identical to that in 59 cases with less than 22 AB (13.3 mo) (p = 0.59, log-rank test). Recently, a study in Italy evaluated survival in 59 MM and found no correlation between survival and lung AF burden (Spearman’s rho = 0.09, p = 0.45).
The authors had information on AB counts for a subset of 88 patients, but these data were not used in survival analyses. A large study in the United States found ecological associations between shorter survival and presence of shipbuilding as a major industry.
The authors concluded that there was “weak support for the hypothesis that asbestos-exposed cases of mesothelioma have worse survival experience than other cases.” Finally, a study in Canada on 188 MPM with complete data found that asbestos exposure was an important predictor of survival at multivariate analysis.
In general, these studies are much less informative, because history of asbestos exposure is a crude proxy for asbestos lung burden.
Our study had several strengths. First, this is one of the largest necropsy studies with documented AF lung burden. The lung content of AFs is a widely accepted index of lifetime cumulative asbestos exposure.
Second, it included two series with very different asbestos lung burdens. Third, accurate lung burden assessment was performed in a reference regional laboratory. Our study also had some limitations. First, AB concentrations were missing in many patients. Nevertheless, AF counts were available for all patients with MPM. Second, we did not have information on patient’s treatment or BAP1 germline mutations, which could differ between the two series. We performed, however, separated analyses, and we found no association between lung asbestos burden and survival in neither of the two necropsy series.
We were not able to find in the literature convincing hypotheses on the role of lung asbestos burden on survival. One could hypothesize that a higher burden is a marker of a higher degree of lung fibrosis, which might affect lung function. Although we find a negligible relationship between histologic asbestosis and survival in shipyard patients, one should consider that histologically documented fibrosis rarely implies relevant lung function loss; in fact, only a few subjects had radiologically confirmed asbestosis. Because germline testing was not performed in any of the studies, the discrepancies among different studies might in theory be explained by different proportions of subjects with BAP1 mutations in the different series. Nevertheless, a relationship between BAP1 and lung asbestos burden would be necessary to confound the association between lung asbestos burden and survival.
In conclusion, the results of our study, based on a large series of autopsies and high-quality laboratory data, do not confirm the inverse relationship between asbestos lung burden or exposure and survival observed in previous studies. In contrary, our results are in agreement with other studies revealing lack of association between lung asbestos burden or exposure and survival. We confirmed that the strongest predictor is histologic type. Results of this study may be relevant for clinicians and in medico-legal settings.
CRediT Authorship Contribution Statement
Pietro Gino Barbieri: Conceptualization, Investigation, Data curation.
Dario Consonni: Methodology, Software, Formal analysis, Writing—review and editing.
Anna Somigliana: Lung sample analysis.
The study was approved by the Ethical Committee of the Friuli Venezia-Giulia region (n. 2019/13818, May 21, 2019).
Survival of pleural malignant mesothelioma in Italy: a population-based study.
Disclosure: Dr. Barbieri served as a consultant for the prosecutor or for the victims’ families in litigations concerning asbestos-related diseases. Dr. Consonni served as a consultant for the court in litigations concerning asbestos-related diseases. Dr. Somigliana declares no conflict of interest.
Two recent papers in the Journal of Thoracic Oncology propose two opposite hypotheses: that the amount of asbestos in the lungs may directly correlate with shorter survival in patients with mesothelioma1 and, on the opposite, that the amount of asbestos in the lungs does not influence survival.2