Serum Proteomics and Plasma Fibulin-3 in Differentiation of Mesothelioma From Asbestos-Exposed Controls and Patients With Other Pleural Diseases

Our plain-English summary of Serum Proteomics and Plasma Fibulin-3 in Differentiation of Mesothelioma From Asbestos-Exposed Controls and Patients With Other Pleural Diseases, by S. Tsim, L. Alexander, C. Kelly, et al (Journal of Thoracic Oncology, 2021).

The study

• Authors: S. Tsim, L. Alexander, C. Kelly, A. Shaw, S. Hinsley, S. Clark, M. Evison, J. Holme, E.J. Cameron, D. Sharma, A. Wright, S. Grundy, D. Grieve, A. Ionescu, D.P. Breen, E. Paramasivam, I. Psallidas, D. Mukherjee, M. Chetty, G. Cox, A. Hart-Thomas, R. Naseer, J. Edwards, C. Daneshvar, R. Panchal, M. Munavvar, R. Ostroff, L. Alexander, H. Hall, M. Neilson, C. Miller, C. McCormick, F. Thomson, A.J. Chalmers, N.A. Maskell, K.G. Blyth
• Journal: Journal of Thoracic Oncology
• Published: 2021;16(10):1705-1717
• Cited by: 20 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Introduction: Malignant pleural mesothelioma (MPM) is difficult to diagnose. An accurate blood biomarker could prompt specialist referral or be deployed in future screening. In earlier retrospective studies, SOMAscan proteomics (Somalogic, Boulder, CO) and fibulin-3 seemed highly accurate, but SOMAscan has not been validated prospectively and subsequent fibulin-3 data have been contradictory.

Methods: A multicenter prospective observational study was performed in 22 centers, generating a large intention-to-diagnose cohort. Blood sampling, processing, and diagnostic assessment were standardized, including a 1-year follow-up. Plasma fibulin-3 was measured using two enzyme-linked immunosorbent assays (CloudClone [used in previous studies] and BosterBio, Pleasanton, CA). Serum proteomics was measured using the SOMAscan assay. Diagnostic performance (sensitivity at 95% specificity, area under the curve [AUC]) was benchmarked against serum mesothelin (Mesomark, Fujirebio Diagnostics, Malvern, PA). Biomarkers were correlated against primary tumor volume, inflammatory markers, and asbestos exposure.

Results: A total of 638 patients with suspected pleural malignancy (SPM) and 110 asbestos-exposed controls (AECs) were recruited. SOMAscan reliably differentiated MPM from AECs (75% sensitivity, 88.2% specificity, validation cohort AUC 0.855) but was not useful in patients with differentiating non-MPM SPM. Fibulin-3 (by BosterBio after failed CloudClone validation) revealed 7.4% and 11.9% sensitivity at 95% specificity in MPM versus non-MPM SPM and AECs, respectively (associated AUCs 0.611 [0.557-0.664], p = 0.0015) and 0.516 [0.443-0.589], p = 0.671), both inferior to mesothelin. SOMAscan proteins correlated with inflammatory markers but not with asbestos exposure. Neither biomarker correlated with tumor volume.

Conclusions: SOMAscan may prove useful as a future screening test for MPM in asbestos-exposed persons. Neither fibulin-3 nor SOMAscan should be used for diagnosis or pathway stratification.

Copyright © 2021 International Association for the Study of Lung Cancer. All rights reserved.

In plain English

This prospective multicenter study (DIAPHRAGM) tested whether two blood tests, the SOMAscan protein panel and fibulin-3, could distinguish malignant pleural mesothelioma (MPM) from other conditions. Serum mesothelin served as a comparison. From December 2013 to December 2016, 22 centers in the United Kingdom and Ireland recruited 638 patients with suspected pleural malignancy. Asbestos-exposed controls with pleural plaques, asbestosis or diffuse pleural thickening were recruited in Glasgow. MPM was the final diagnosis in 152 of 622 analyzed patients. In an internal validation set, the 13-protein SOMAscan signature separated MPM from asbestos-exposed controls (AUC 0.855, 95% confidence interval 0.741 to 0.970; 75% sensitivity; 88.2% specificity). It could not separate MPM from other pleural diseases. At 95% specificity, fibulin-3 sensitivity was 7.4% against other pleural diseases and 11.9% against controls, lower than mesothelin (20.4% and 37.2%). Among 26 MPM patients with MRI, no marker showed a statistically significant correlation with tumor volume. The authors linked SOMAscan protein differences to inflammation, not asbestos exposure. They concluded SOMAscan may prove useful for future screening of asbestos-exposed persons, but neither test proved useful for diagnosis.

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: Prospective, multicenter, observational diagnostic accuracy study (DIAPHRAGM) of blood biomarkers, benchmarked against serum mesothelin
• Who was studied: 638 patients with suspected pleural malignancy (a new unilateral pleural effusion or mass), recruited at 22 centers in the United Kingdom and Ireland between December 31, 2013 and December 31, 2016. Asbestos-exposed controls were recruited in Glasgow: 112 according to the Results text, and 110 in the fibulin-3 analysis. After 16 'evolvers' were excluded, final diagnoses among 622 patients were MPM in 152 (24.4%), secondary pleural malignancy in 218 (35.1%) and benign pleural disease in 252 (40.5%).
• Exposure: The study evaluated blood biomarkers (SOMAscan serum proteomics, plasma fibulin-3 and serum mesothelin), not asbestos doses. Asbestos exposure defined the control group: controls needed a history of asbestos exposure with pleural plaques, asbestosis or diffuse pleural thickening. Fiber types were not reported.
• Main results:
  • SOMAscan, MPM versus asbestos-exposed controls: the 13-protein signature had a training set AUC of 0.955 (0.926 to 0.984), 87.4% sensitivity and 92.4% specificity. In the internal validation set (an 80-to-20 split of the study's own data), it had an AUC of 0.855 (0.741 to 0.970), 75% sensitivity and 88.2% specificity.
  • SOMAscan could not differentiate MPM from patients with other pleural diseases (non-MPM suspected pleural malignancy, including other cancers and benign disease).
  • The authors interpreted SOMAscan's performance in MPM versus asbestos-exposed controls as supporting a future role in screening.
  • Fibulin-3 (BosterBio assay): at the prespecified 95% specificity, sensitivity was 7.4% for MPM versus non-MPM suspected pleural malignancy and 11.9% for MPM versus asbestos-exposed controls.
  • Fibulin-3 AUC was 0.611 (0.557 to 0.664), p = 0.0015, for MPM versus non-MPM suspected pleural malignancy, and 0.516 (0.443 to 0.589), p = 0.671, for MPM versus asbestos-exposed controls.
  • Fibulin-3 levels were significantly higher in MPM than in non-MPM suspected pleural malignancy (p = 0.0001). They were not significantly different between MPM and asbestos-exposed controls (p > 0.99).
  • Mesothelin (comparison test): at 95% specificity, sensitivity was 20.4% for MPM versus non-MPM suspected pleural malignancy and 37.2% for MPM versus asbestos-exposed controls. The AUCs were 0.707 (0.649 to 0.765) and 0.766 (0.702 to 0.830), respectively.
  • In the MRI substudy (26 of 152 MPM patients), no statistically significant correlations were observed between primary tumor volume and any of the 13 SOMAscan proteins, fibulin-3 (r = 0.0137, p = 0.502) or mesothelin (r = 0.031, p = 0.893).
  • C-reactive protein, white cell count and a history of asbestos exposure correlated with 10, 9 and 11 of the 13 SOMAscan proteins, respectively. In this analysis, the asbestos exposure group included any suspected pleural malignancy patient reporting exposure, plus all controls.
  • The authors interpreted the SOMAscan protein differences between MPM and asbestos-exposed controls as not driven by asbestos exposure itself. They identified systemic inflammation as an important confounder.
  • Mesothelin values were not associated with sex, asbestos exposure or the timing of blood sampling. Fibulin-3 showed a weak positive association with known asbestos exposure (p < 0.0001).
• The authors' conclusion: The authors concluded that SOMAscan may prove useful as a future screening test for MPM in asbestos-exposed persons, but that neither fibulin-3 nor SOMAscan proved useful in the diagnosis of MPM.
• Limitations the authors state:
  • The SOMAscan assay was revised to version 3 during recruitment, so the planned validation of the original version was not possible.
  • The fibulin-3 assay used in earlier studies (CloudClone) had unacceptable reproducibility. A different assay was therefore used, and the planned validation of the CloudClone assay and its cut-point could not be done.
  • SOMAscan's moderate specificity (88.2%) may limit future screening use, because positive predictive value would be lower in low-prevalence settings.
  • The MPM incidence in the study (24.4%) was higher than in many countries.
  • Budget limited the number of samples analyzed by SOMAscan.
  • The MRI method used to measure tumor volume is indirect.

Topics

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

Medical terms in this study, explained in our Lexicon: Fibulin-3, Mesothelioma, Pleural Mesothelioma, Biomarker, Mesothelin, Asbestos Exposure.

Related studies

• Diagnostic and Prognostic Biomarkers in the Rational Assessment of Mesothelioma (DIAPHRAGM) study: protocol of a prospective, multicentre, observational study — S. Tsim, C. Kelly, et al. (2016)
• Serum soluble mesothelin-related protein (SMRP) and fibulin-3 levels correlate with baseline malignant pleural mesothelioma (MPM) tumor volumes but are not useful as biomarkers of response in an immunotherapy trial — S.I. Katz, L. Roshkovan, et al. (2021)
• Plasma Fibulin-3 as a Potential Biomarker for Patients with Asbestos-Related Diseases in the Han Population — Z. Jiang, S. Ying, et al. (2017)
• Comparison of mesothelin and fibulin-3 in pleural fluid and serum as markers in malignant mesothelioma — J. Creaney, I.M. Dick, et al. (2015)
• Comparison of fibulin-3 and mesothelin as markers in malignant mesothelioma — J. Creaney, I.M. Dick, et al. (2014)
• Fibulin-3 immunoexpression in malignant mesothelioma due to fluoro-edenite: a preliminary report — R. Caltabiano, C. Loreto, et al. (2018)

Browse more research in Journal of Thoracic Oncology, or return to medical research.