A novel mesothelioma molecular classification based on malignant cell differentiation

Our plain-English summary of A novel mesothelioma molecular classification based on malignant cell differentiation, by J. Liu, Y. Liu, Y. Lu, et al (Cancer Cell International, 2025).

The study

• Authors: J. Liu, Y. Liu, Y. Lu, W. Zhang, J. Yan, B. Lu, Y. Yao, S. Xian, D. Lyu, J. Shi, Y. Li, X. Wu, C. Bai, J. Zhang, Y. Zhang
• Journal: Cancer Cell International
• Published: 2025;25(1):235
• Cited by: 1 article, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Background: The high heterogeneity and multi-directional poor differentiation of tumor cells in mesothelioma (MESO) contributes to tumor growth and malignant biological behaviors. However, a molecular classification based on differentiated states of tumor cells remains void.

Methods: We performed dimensionality reduction analysis on the single-cell RNA sequencing profiles available from the GEO database, to visualize the cell types in MESO. Multi-omics analysis was done to supplement the plausibility of classification. We also constructed regulatory networks to detect the function of important tumor cell differential genes (TCDGs) in the MESO.

Results: Following twice dimensionality reduction analysis and clustering, eight malignant cell subtypes in the MESO were visualized. According to the expression of TCDGs, MESO was classified into three subtypes (Malignant differentiation-related MESO, Benign differentiation-related MESO, and Neutral differentiation-related MESO) with prognostic differences. The prediction model was built by 12 key TCDGs (ALDH2, HP, CASP1, RTP4, PDZK1IP1, TOP2A, LOXL2, CKS2, SPARC, TLCD3A, C6orf99, and SERPINH1) and validated with high accuracy. In the regulatory networks of MESO subtypes, RTP4, CASP1, MYO1B, SLC7A5, LOXL2, and GHR were labeled as key genes. A total of 14 potential inhibitors were predicted. Clinical specimens validated the reliability of the clinical subtyping of MESO patients.

Conclusion: The novel molecular classification system and the prognostic prediction model might benefit the management of MESO patients.

© 2025. The Author(s).

In plain English

This was a computer-based (bioinformatics) analysis of gene activity in mesothelioma. It did not study asbestos exposure. The authors analysed single-cell RNA sequencing data from three samples in a public database (two treated with a compound called ADU-S100 and one control). They combined these with gene data from 84 mesothelioma patients in the TCGA-MESO cohort. They identified 32 genes linked both to tumor cell differentiation and to survival, and used them to sort patients into three groups. Patients in the group the authors named benign differentiation-related had the best survival, and those in the malignant differentiation-related group the poorest (p less than 0.001). A 12-gene risk model had AUC values of 0.936 in the whole cohort and 0.946 in the training and test sets. The analysis also predicted 14 possible drugs. In tumor tissue from 11 patients at a Shanghai hospital, differences between the groups were not statistically significant, which the authors attributed mainly to the small sample. The authors concluded the model was expected to improve patient management, but said the findings need experimental validation and larger cohorts.

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: Bioinformatics analysis of public single-cell and bulk gene-expression datasets, with immunohistochemical staining and a retrospective chi-square comparison in a small clinical cohort
• Who was studied: Single-cell data from three samples (ADU-S100_10um, ADU-S100_50um and Control) from database GSE201925. Bulk data from 84 MESO patients in the TCGA-MESO cohort. Further public validation datasets. Tumor tissue from 11 patients diagnosed with malignant MESO between 1 January 2020 and 31 December 2022 at Shanghai Pulmonary Hospital affiliated to Tongji University.
• Exposure: Not studied. Asbestos is mentioned only in the introduction, citing other work. The study examined tumor gene expression, not exposure.
• Main results:
  • 32 genes, which the authors call tumor cell differential genes, were linked both to tumor cell differentiation and to overall survival in TCGA-MESO patients. Of these, 7 were favorable factors and 25 were risk factors.
  • Sorting TCGA-MESO patients into three clusters, cluster 2 (named benign differentiation-related) had the best outcome and cluster 1 (named malignant differentiation-related) the poorest.
  • A 12-gene prognostic model had an AUC of 0.936 in the whole TCGA-MESO cohort and 0.946 in both the train and test cohorts.
  • In multivariate Cox regression, the risk score was a significant risk factor (HR = 1896.628, 95% CI 246.758 to 14,577.850). The univariate HR was 732.833 (95% CI 135.823 to 3954.010).
  • The analysis predicted 14 inhibitors. The malignant differentiation-related cluster was predicted to have the highest sensitivity to them.
  • In the 11-patient clinical cohort, differences between clusters in differentiation, immune infiltration and metastasis (each p = 0.07) and AJCC stage (p = 0.32) were not statistically significant.
• The authors' conclusion: The authors concluded that the new gene classification and prognostic model were expected to improve the treatment and management of MESO patients. They said the clinical specimens gave only preliminary validation.
• Limitations the authors state:
  • The findings come from bioinformatics analysis, which has inherent biases, and need experimental validation.
  • The clinical validation sample was small, and the classification needs testing in a larger cohort.
  • The predicted inhibitors have not been tested and need animal and clinical studies.

Topics

Medical terms in this study, explained in our Lexicon: Mesothelioma.

Related studies

• A gene signature linked to fibroblast differentiation for prognostic prediction of mesothelioma — J. Liu, Y. Lu, et al. (2024)
• Chinese expert consensus on the diagnosis and treatment of malignant pleural mesothelioma — Q. Wang, C. Xu, et al. (2023)
• Pleural Mesothelioma With Unusual Squamous Differentiation: A Clinicopathological Case Report With Molecular Insights From Whole-Exome Sequencing — J. Bodemeyer, M.E. Mierzwa, et al. (2026)
• Identification of DAB2 and Intelectin-1 as Novel Positive Immunohistochemical Markers of Epithelioid Mesothelioma by Transcriptome Microarray Analysis for Its Differentiation From Pulmonary Adenocarcinoma — M. Kuraoka, V.J. Amatya, et al. (2017)
• Large-scale analysis of BAP1 expression reveals novel associations with clinical and molecular features of malignant pleural mesothelioma — A. De Rienzo, L.R. Chirieac, et al. (2021)
• Molecular Mechanisms of Tumor Progression and Novel Therapeutic and Diagnostic Strategies in Mesothelioma — T. Kato, I. Tanaka, et al. (2025)

Browse more research in Cancer Cell International, or return to medical research.