Glycan Fingerprint of Malignant Pleural Mesothelioma

Our plain-English summary of Glycan Fingerprint of Malignant Pleural Mesothelioma, by L. Kavur, T.S. Klarić, N. Mraz, et al (International Journal of Molecular Sciences, 2026).

The study

• Authors: L. Kavur, T.S. Klarić, N. Mraz, N. Šimunić-Briški, D. Lalić, G. Lauc, M. Martinić, L. Batelja Vuletić, M. Martinić Kavur, S. Seiwerth, O. Gamulin
• Journal: International Journal of Molecular Sciences
• Published: 2026;27(14)
• Read it: PubMed · Publisher · Free full text

Abstract

Malignant pleural mesothelioma (MPM) is an aggressive pleural tumor associated with asbestos exposure. Poor clinical outcome of MPM is often driven by late-stage diagnosis due to non-specific clinical presentation, similarity to pleural lesions (e.g., inflammatory changes, metastatic adenocarcinoma), and limitations of current diagnostic methods. We employed Fourier transform infrared (FTIR) spectroscopy combined with convolutional neural networks (CNNs) to analyze formalin-fixed paraffin-embedded (FFPE) pleural tissue samples from patients with MPM, metastatic adenocarcinoma, pleural inflammation, and normal (healthy) pleura. Glycan analysis of FFPE normal pleura and MPM was performed using ultra-high-performance liquid chromatography (UPLC) and mass spectrometry (MS). Our FTIR-spectral analysis uncovered a strong spectral fingerprint of MPM that was especially apparent in the region typical for C-O and C-C stretches as well as local symmetry region typical for deformation vibrations of CH2 and C-OH groups, all appearing in carbohydrates. Our orthogonal validation of these findings through a targeted glycomics approach using UPLC confirmed that the MPM N-glycome exhibits a distinct fingerprint that distinguishes it from normal pleural tissue. Through utilization of MS for identifying the exact structures of differentially expressed N-glycan peaks, we also identified two high-mannose N-glycan structures that show a specific biomarker potential for MPM and need to be examined in future studies.

In plain English

This laboratory study used archived pleural tissue from the University of Zagreb pathology archives to look for a chemical fingerprint of malignant pleural mesothelioma (MPM). The authors re-analyzed infrared (FTIR) spectra from an earlier study of 32 patients (10 each with MPM, pleural inflammation or metastatic adenocarcinoma, and 2 with normal pleura) using a computer model. The model sorted test samples into the four tissue types with 93.42% accuracy. In two instances pleural inflammation was wrongly classed as MPM, and no MPM was missed. The most informative spectral range was one linked to carbohydrates. The authors then measured sugar chains (N-glycans) in normal pleura and MPM samples, of which 25 and 24 passed quality checks. Sixteen of 49 glycan peaks differed significantly. Two high-mannose glycans, H5N2 and H7N2, were lower in MPM than in normal pleura. Comparisons with adenocarcinoma and inflammation at this step came from other published studies. The authors cited the small sample size and risk of overfitting, and concluded these glycans should be examined in larger studies. The article did not report the participants' asbestos-exposure histories.

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: Laboratory study of archived formalin-fixed paraffin-embedded pleural tissue. It had two parts: a re-analysis, with a convolutional neural network, of infrared (FTIR) spectra from the authors' earlier study; and a separate N-glycan analysis (UPLC and mass spectrometry) comparing MPM with normal pleura.
• Who was studied: FTIR part: 32 patients (MPM, pleural inflammation and metastatic adenocarcinoma, 10 per class, plus 2 with normal pleura), with 690 tissue sections, from the Department of Pathology archives, University of Zagreb School of Medicine. Glycan part: 30 normal pleura samples from 30 individuals (93% male, median age 65.5 years, range 50–77) and 30 MPM samples from 22 patients (90% male, median age 63 years, range 36–82). This set included 10 MPM patients and 2 normal-pleura individuals from the FTIR study. After quality control, 25 normal pleura and 24 MPM samples were analyzed.
• Exposure: Not reported for the participants. The study compared tissue types (MPM, pleural inflammation, metastatic adenocarcinoma, normal pleura). The introduction describes MPM as primarily caused by asbestos exposure, but no asbestos-exposure histories or measurements were given for the patients or controls.
• Main results:
  • For classifying all four tissue types, the model reached a test accuracy of 93.42%, precision of 93.61%, sensitivity of 93.42% and specificity of 93.87%.
  • Pleural inflammation was falsely classified as MPM in two events. The false negative rate for MPM was 0.
  • The spectral region from 1150 to 1200 cm-1, which Table 1 associates with carbohydrates and glycans, achieved 97.37% accuracy on its own.
  • Of 49 N-glycan peaks, 16 (33%) differed significantly between MPM and normal pleura.
  • Nine peaks were significantly lower in the MPM group (peaks 9, 14, 20, 23, 30, 33, 37, 40 and 45). Seven peaks were significantly lower in the control (normal pleura) group (peaks 11, 17, 27, 31, 34, 35 and 36).
  • Peak 23, whose dominant structure was the high-mannose glycan H7N2, made up 6.7% of the total N-glycome in MPM compared with 8.9% in normal pleura.
  • The high-mannose glycan H5N2 was significantly lower in MPM than in control tissue.
  • A multivariate analysis (PLS-DA) of all glycan peaks clearly separated control and MPM samples.
  • MPM samples showed more variation from person to person than control samples.
• The authors' conclusion: The authors concluded that FTIR combined with UPLC and mass spectrometry identified a distinct glycan fingerprint of MPM. In their view, the high-mannose glycans H5N2 and H7N2, together with isomer and phosphorylation changes in H6N2 and H8N2, have biomarker potential for distinguishing MPM and should be examined in larger studies.
• Limitations the authors state:
  • The sample size was relatively small, and the findings would need to be validated in larger cohorts.
  • The small sample number carried a high risk of overfitting.
  • The computer model did not use early stopping based on a held-out validation set or probability calibration. The authors advise adding these in larger studies before any clinical use.
  • Ageing may lower high-mannose glycans, so cases and controls should be carefully age-matched in future studies.
  • The glycan structures within all detected peaks were not identified.
  • The glycan validation compared only MPM with normal pleura. Comparisons with adenocarcinoma and inflammation rely on previously published studies.

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: Pleural Mesothelioma, Asbestos Exposure, Adenocarcinoma, Inflammation, Pleura, Biomarker.

Related studies

• Processing and analysis of serum antibody binding signals from Printed Glycan Arrays for diagnostic and prognostic applications — M.I. Vuskovic, H. Xu, et al. (2011)
• Differentiating Between Malignant Mesothelioma and Other Pleural Lesions Using Fourier Transform Infrared Spectroscopy — F. Sadiku-Zehri, O. Gamulin, et al. (2020)
• Molecular resistance fingerprint of pemetrexed and platinum in a long-term survivor of mesothelioma — O.D. Røe, A. Szulkin, et al. (2012)
• Molecular Fingerprints of Malignant Pleural Mesothelioma: Not Just a Matter of Genetic Alterations — E. Lorenzini, A. Ciarrocchi, et al. (2021)
• DNA fingerprint analysis reveals differences in mutational patterns in experimentally induced rat peritoneal tumors, depending on the type of environmental mutagen — N. Kociok, K. Unfried, et al. (1999)
• The transcriptomic fingerprint of cancer response to Tumor Treating Fields (TTFields) — K. Wainer-Katsir, A. Haber, et al. (2025)

Browse more research in International Journal of Molecular Sciences, or return to medical research.