Quantitative structure-mesothelioma potency model optimization for complex mixtures of elongated particles in rat pleura: A retrospective study

Our plain-English summary of Quantitative structure-mesothelioma potency model optimization for complex mixtures of elongated particles in rat pleura: A retrospective study, by P.M. Cook, J. Swintek, T.D. Dawson, et al (Journal of Toxicology and Environmental Health. Part B, Critical Reviews, 2016).

The study

• Authors: P.M. Cook, J. Swintek, T.D. Dawson, D. Chapman, M.A. Etterson, D. Hoff
• Journal: Journal of Toxicology and Environmental Health. Part B, Critical Reviews
• Published: 2016;19(5-6):266-288
• Cited by: 8 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Cancer potencies of mineral and synthetic elongated particle mixtures, including asbestos fibers, are influenced by changes in fiber dose composition, bioavailability, and biodurability in combination with relevant cytotoxic dose-response relationships. An extensive rat intrapleural dose characterization data set with a wide variety of elongated particles physicochemical properties facilitated statistical analyses of pleural mesothelioma response data combined from several studies for evaluation of alternative dose-response models. Utilizing logistic regression of individual elongated particle dimensional variations within each test sample, four major findings emerged: (1) Mild acid leaching provides superior prediction of tumor incidence compared to samples that were not leached; (2) sum of the elongated particle surface areas from mildly acid-leached samples provides the optimum holistic dose-response model; (3) progressive removal of dose associated with very short and/or thin elongated particles significantly degrades the resultant particle count and surface area dose-based predictive model fits; and (4) alternative biologically plausible model adjustments provide evidence for reduced potency of elongated particles with aspect ratios less than 8 and lengths greater than 80 µm. Regardless of these adjustments, the optimum predictive models strongly incorporate potency attributable to abundant short elongated particles in proportion to their surface area. Transmission electron microscopy analyses of low-temperature-ashed pleural membrane and lung tissues 5.5 mo post intrapleural exposures do not support hypotheses that short elongated particles that reach the pleural space are rapidly eliminated. Low-aspect-ratio elongated particles were still abundant in pleural membrane tissues but may have reduced potencies due to aggregation tendencies and therefore lower potential for intracellular presence.

In plain English

This was a retrospective statistical reanalysis. The authors pooled rat mesothelioma results from five published studies and one unpublished study, in which particles were injected into the pleural space around the lungs. They matched these with archived electron-microscope measurements of individual particles from a U.S. EPA laboratory in Duluth, Minnesota (1978–1986). The final data set had 50 test results covering 10 fiber types. These included amosite, crocidolite, anthophyllite and tremolite asbestos, plus non-asbestos and synthetic particles; chrysotile was not studied. The authors' preferred model used the total surface area of particles after mild acid leaching, meant to imitate changes in the body. It counted particles with aspect ratio at least 8 and length 80 micrometres or less. Some alternative models fit better but excluded more of the dose. Leaving out short particles worsened surface-area models but improved particle-count models. Tissue examined 5.5 months after injection still contained short particles. The authors concluded that short particles' potency should not be ignored. They stated that their models show correlation, not causation, and do not predict mesothelioma rates in humans.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Retrospective statistical reanalysis (logistic regression with model comparison by AIC) of pooled rat intrapleural injection data from earlier studies, combined with archived electron-microscope particle measurements. It also includes electron-microscope analysis of rat tissues taken after injection.
• Who was studied: Rats given intrapleural injections (Osborne-Mendel, Wistar and Fischer 344 strains). There were 50 test results from 26 unique samples, drawn from Stanton and Wrench (1972), Wagner et al. (1973), Stanton et al. (1981), Coffin et al. (1982, 1992) and one unpublished Coffin study. Particle measurement data were collected at the U.S. EPA Mid-Continent Ecology Division laboratory in Duluth, Minnesota, during 1978–1986. Subsets analyzed separately were Stanton-only (n = 22) and amphibole-only (n = 38).
• Exposure: Intrapleural doses of 0.5 to 40 mg. The ten fiber types were:
• Four amphibole asbestos types: amosite, crocidolite, anthophyllite and tremolite.
• Five non-amphiboles: alumina, dawsonite, nonfibrous grunerite, ferroactinolite and silicon carbide.
• One synthetic: potassium octatitanate (fybex).
  The data set was heavily weighted toward crocidolite. Samples were analyzed both unleached and after mild acid leaching intended to simulate changes in the body.
• Main results:
  • In the Stanton-only subset and the final data set, models using acid-leached samples predicted mesothelioma incidence better than models using unleached samples.
  • In the amphibole-only subset, there was no observed difference between leached and unleached models.
  • The authors' proposed model predicted mesothelioma rates in rats from the total surface area of particles with an aspect ratio of 8 or more and a length of 80 micrometres or less.
  • Five models showed better predictive capability than the proposed model. The authors did not propose them:
• Four of them removed more of the particle dose: on average 52.1% to 62.2%, compared with 39.4% for the proposed model.
• The fifth was a theoretical model that added dose.
  • The four better-fitting models that removed extra dose discarded more of the particles than the proposed model did.
  • Excluding short particles had opposite effects depending on the dose measure. It improved the predictability of models based on particle counts, but worsened the fit of models based on surface area.
  • Among leached surface-area models in the final data set, those that excluded short particles fit worse (higher AIC differences, where lower is better):
• Counting only particles longer than 5 micrometres gave an AIC difference of 126.
• Counting only phase-contrast-visible (PCME) fibers gave 147.
  • Within the Stanton-only subset (all 40-mg doses), a count of Stanton 'index fibers' (longer than 8 micrometres, 0.25 micrometres wide or thinner) performed as well as the proposed surface-area model. In the final data set it performed worse, with AIC differences of 88 and 26 (see text problems on which is leached).
  • For the Stanton tremolite sample, counting only Stanton index fibers would ignore 99.5% of its elongated particles and 98.5% of its total surface area.
  • The authors reported that once leaching was taken into account, asbestos had the same potency on a surface-area basis as the other mineral and synthetic particles.
  • 5.5 months after injection, about 7% of the UICC crocidolite particle-number dose remained in the pleural membranes and associated outer lung tissue. The figure for leached UICC amosite was 6.5%.
  • Short particles (under 5 micrometres, and 5 to under 10 micrometres) were retained in tissues for UICC amosite, UICC crocidolite and ferroactinolite.
  • Particles with low aspect ratios were retained relatively less.
  • The number of rats strongly affected how precise a study could be. At a dose of 10.5 log square micrometres, the predicted range of observed mesothelioma was 10% to 80% with 10–20 rats, versus about 30% to 50% with 40–50 rats.
• The authors' conclusion: The authors concluded that the potency of short elongated particles should not be ignored. They proposed a dose metric for predicting mesothelioma rates, based on AIC, that includes the highest proportion of observed particles. This metric is the total surface area of leached particles, with aspect ratio and length adjustments.
• Limitations the authors state:
  • The model was developed from rat studies, and the authors do not provide a model to predict mesothelioma rates in humans.
  • The authors state that their models show correlation, and they make no claim about whether any subset of particles causes or does not cause mesothelioma.
  • Chrysotile was not included.
  • Acid leaching only roughly simulated changes in the body and did not account for movement (translocation) of fibers.
  • Particle thickness values were the least certain measurements.
  • Pleural tissue could not be cleanly separated from the outer lung. Doses measured in pleural tissue may therefore include an unknown amount that had moved into lung tissue.
  • The data could not distinguish a sharp cutoff from a gradual change in potency with aspect ratio.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Mesothelioma.

Medical terms in this study, explained in our Lexicon: Mesothelioma, Pleura, Dose-response relationship, Pleural Mesothelioma, Aspect Ratio (Length-to-diameter ratio), Transmission Electron Microscopy (TEM).

Related studies

• Mixtures research at NIEHS: an evolving program — C.V. Rider, D.J. Carlin, et al. (2013)
• International symposium on the health hazard evaluation of fibrous particles associated with taconite and the adjacent duluth complex. Introduction — R.P. Nolan (2008)
• Final results from a 90-day quantitative inhalation toxicology study evaluating the dose-response and fate in the lung and pleura of chrysotile-containing brake dust compared to TiO2, chrysotile, crocidolite or amosite asbestos: Histopathological examination, confocal microscopy and collagen quantification of the lung and pleural cavity — D.M. Bernstein, B. Toth, et al. (2021)
• Evaluation of the dose-response and fate in the lung and pleura of chrysotile-containing brake dust compared to chrysotile or crocidolite asbestos in a 28-day quantitative inhalation toxicology study — D.M. Bernstein, B. Toth, et al. (2018)
• Evaluation of the exposure, dose-response and fate in the lung and pleura of chrysotile-containing brake dust compared to TiO2, chrysotile, crocidolite or amosite asbestos in a 90-day quantitative inhalation toxicology study - Interim results Part 1: Experimental design, aerosol exposure, lung burdens and BAL — D.M. Bernstein, B. Toth, et al. (2020)
• Evaluation of the dose-response and fate in the lung and pleura of chrysotile-containing brake dust compared to TiO2, chrysotile, crocidolite or amosite asbestos in a 90-day quantitative inhalation toxicology study - Interim results Part 2: Histopathological examination, Confocal microscopy and collagen quantification of the lung and pleural cavity — D.M. Bernstein, B. Toth, et al. (2020)

Browse more research in Journal of Toxicology and Environmental Health. Part B, Critical Reviews, or return to medical research.