Integrated Analysis of Clinical Outcome of Mesenchymal Stem Cell-related Genes in Pan-cancer

Our plain-English summary of Integrated Analysis of Clinical Outcome of Mesenchymal Stem Cell-related Genes in Pan-cancer, by M. Jiang, D. Zhu, D. Zhao, et al (Current Genomics, 2024).

The study

• Authors: M. Jiang, D. Zhu, D. Zhao, Y. Liu, J. Li, Z. Zheng
• Journal: Current Genomics
• Published: 2024;25(4):298-315
• Cited by: 18 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Background: Although the application of mesenchymal stem cells (MSCs) in engineered medicine, such as tissue regeneration, is well known, new evidence is emerging that shows that MSCs can also promote cancer progression, metastasis, and drug resistance. However, no large-scale cohort analysis of MSCs has been conducted to reveal their impact on the prognosis of cancer patients.

Objectives: We propose the MSC score as a novel surrogate for poor prognosis in pan-cancer.

Methods: We used single sample gene set enrichment analysis to quantify MSC-related genes into a signature score and identify the signature score as a potential independent prognostic marker for cancer using multivariate Cox regression analysis. TIDE algorithm and neural network were utilized to assess the predictive accuracy of MSC-related genes for immunotherapy.

Results: MSC-related gene expression significantly differed between normal and tumor samples across the 33 cancer types. Cox regression analysis suggested the MSC score as an independent prognostic marker for kidney renal papillary cell carcinoma, mesothelioma, glioma, and stomach adenocarcinoma. The abundance of fibroblasts was also more representative of the MSC score than the stromal score. Our findings supported the combined use of the TIDE algorithm and neural network to predict the accuracy of MSC-related genes for immunotherapy.

Conclusion: We comprehensively characterized the transcriptome, genome, and epigenetics of MSCs in pan-cancer and revealed the crosstalk of MSCs in the tumor microenvironment, especially with cancer-related fibroblasts. It is suggested that this may be one of the key sources of resistance to cancer immunotherapy.

Bentham Science Publishers.

In plain English

This was a computer-based analysis of public gene databases. It was not a study of asbestos exposure, and asbestos is not mentioned. The authors used tumor data from The Cancer Genome Atlas covering 33 cancer types, one of which was mesothelioma. The text does not say whether the tumors were pleural or peritoneal and does not give patient numbers. The authors combined the activity of 59 genes linked to mesenchymal stem cells into an 'MSC score' and split patients at the median score. For mesothelioma, after adjusting for clinical factors, the score was identified as a potential independent predictor of overall survival. The hazard ratio was 9.41 (95% confidence interval 3.11 to 28.50) in the Cancer Genome Atlas data and 4.70 (1.13 to 19.20) in a second mesothelioma dataset, GSE29354. Similar findings were reported for kidney papillary cancer, glioma and stomach cancer. In computer predictions, and in seven immunotherapy cohorts of other cancers, patients with low scores responded more often. The authors concluded that these genes show potential as markers of prognosis and immunotherapy response. They noted that the immunotherapy cohorts were retrospective and that some confounding factors in them remain unclear.

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: Computational (bioinformatics) analysis of publicly available tumor gene-expression, mutation and clinical databases, with validation in additional public datasets
• Who was studied: Tumor samples from 33 cancer types in The Cancer Genome Atlas, including mesothelioma (MESO), with normal tissue from GTEx; the mesothelioma validation dataset was GSE29354. The text gives no patient counts and does not specify the mesothelioma site.
• Exposure: No asbestos or fiber exposure was studied. The variable examined was an MSC score built from the expression of 59 mesenchymal stem cell-related genes, with patients grouped as high or low at the median score.
• Main results:
  • Overall, Cox regression suggested the MSC score as an independent prognostic marker for kidney renal papillary cell carcinoma, mesothelioma, glioma and stomach adenocarcinoma
  • In unadjusted (univariate) analysis, the MSC score had prognostic significance for mesothelioma in overall survival and disease-specific survival
  • In multivariate analysis of mesothelioma in The Cancer Genome Atlas data, overall survival HR = 9.41, 95% CI = 3.11 to 28.50, P less than 0.0001
  • In the mesothelioma validation dataset GSE29354, overall survival HR = 4.70, 95% CI = 1.13 to 19.20, P = 0.033
  • Using the TIDE prediction algorithm, patients with a low MSC score had higher predicted response rates to immune-checkpoint drugs than patients with a high score in all 33 cancers
  • In seven real cohorts of other cancers (not mesothelioma) treated with PD-1 or PD-L1 antagonists, patients with low-score tumors were more responsive than those with high scores
  • The neural network model had an area under the ROC curve of 1.00 for mesothelioma, with no misidentified samples in the mesothelioma test set
  • The MSC score was an indicator of fibroblast abundance in all 33 cancer types
• The authors' conclusion: The authors concluded that they had characterized MSCs across multiple cancer types and highlighted their potential as predictive biomarkers for immunotherapy response and prognosis. They suggested that crosstalk between MSCs and cancer-related fibroblasts may be one source of immunotherapy resistance.
• Limitations the authors state:
  • Large-scale clinical immune-checkpoint therapy trials could not be performed, so only retrospective studies of existing immunotherapy cohorts could be used
  • Some confounding factors that may affect immunotherapy in these cohorts remain unclear

Topics

Medical terms in this study, explained in our Lexicon: STEM, Metastasis, Prognosis, Immunotherapy, Carcinoma, Mesothelioma, Adenocarcinoma, Fibroblast.

Related studies

• Mesenchymal stem cell-mediated delivery boosts the efficacy of suicide gene therapy based on retroviral replicating vectors in peritoneally disseminated cancer — S. Kubo, Y. Takeuchi, et al. (2026)
• Systemic but not topical TRAIL-expressing mesenchymal stem cells reduce tumour growth in malignant mesothelioma — E.K. Sage, K.K. Kolluri, et al. (2014)
• Human lung-derived mesenchymal stem cell-conditioned medium exerts in vitro antitumor effects in malignant pleural mesothelioma cell lines — L. Cortes-Dericks, L. Froment, et al. (2016)
• Mesenchymal stem cells modulate lung injury — A.R. Brody, K.D. Salazar, et al. (2010)
• Boosting therapeutic efficacy of mesenchymal stem cells in pulmonary fibrosis: The role of genetic modification and preconditioning strategies — M. Mehrabani, S. Mohammadyar, et al. (2023)
• Circulating stem-like exhausted CD8 T cells point to better outcomes in lung cancer: a brief report — L. Ye, I. Dick, et al. (2025)

Browse return to medical research.