A molecular phenotypic map of malignant pleural mesothelioma

Our plain-English summary of A molecular phenotypic map of malignant pleural mesothelioma, by A. Di Genova, L. Mangiante, A. Sexton-Oates, et al (GigaScience, 2022).

The study

• Authors: A. Di Genova, L. Mangiante, A. Sexton-Oates, C. Voegele, L. Fernandez-Cuesta, N. Alcala, M. Foll
• Journal: GigaScience
• Published: 2022;12
• Cited by: 6 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Background: Malignant pleural mesothelioma (MPM) is a rare understudied cancer associated with exposure to asbestos. So far, MPM patients have benefited marginally from the genomics medicine revolution due to the limited size or breadth of existing molecular studies. In the context of the MESOMICS project, we have performed the most comprehensive molecular characterization of MPM to date, with the underlying dataset made of the largest whole-genome sequencing series yet reported, together with transcriptome sequencing and methylation arrays for 120 MPM patients.

Results: We first provide comprehensive quality controls for all samples, of both raw and processed data. Due to the difficulty in collecting specimens from such rare tumors, a part of the cohort does not include matched normal material. We provide a detailed analysis of data processing of these tumor-only samples, showing that all somatic alteration calls match very stringent criteria of precision and recall. Finally, integrating our data with previously published multiomic MPM datasets (n = 374 in total), we provide an extensive molecular phenotype map of MPM based on the multitask theory. The generated map can be interactively explored and interrogated on the UCSC TumorMap portal (https://tumormap.ucsc.edu/?p=RCG_MESOMICS/MPM_Archetypes ).

Conclusions: This new high-quality MPM multiomics dataset, together with the state-of-art bioinformatics and interactive visualization tools we provide, will support the development of precision medicine in MPM that is particularly challenging to implement in rare cancers due to limited molecular studies.

© World Health Organization, [2023]. All rights reserved. The World Health Organization has granted the Publisher permission for the reproduction of this article.

In plain English

This data note described a research dataset and how its quality was checked. It did not test whether anything causes disease. The MESOMICS project sequenced 120 malignant pleural mesothelioma tumors from the French MESOBANK; 105 had all three data types: whole-genome, RNA and DNA methylation. Patients were diagnosed between 1998 and 2017. The dataset records asbestos and smoking exposure; asbestos measures were checked against the methylation data in quality control, but no asbestos-related risk was estimated. On average, 98.93% of whole-genome reads aligned to the human genome. Part of the cohort lacked matched normal tissue, so tumor-only methods were checked against matched samples: 57 pairs for copy number calls and 46 tumors for the point-mutation model. Copy number results agreed closely (purity correlation R = 0.988), and point-mutation models reached accuracy of 0.926 for single-letter changes and 0.924 for small insertions and deletions. Cross-checks found and corrected mislabeled samples. Combining their data with two earlier studies (374 samples in total), the authors mapped three specialized tumor profiles: cell division, tumor–immune interaction and acinar. The authors concluded they provided a high-quality dataset.

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: Data note: description and quality control of a multiomic (whole-genome, RNA sequencing, DNA methylation) dataset, benchmarking of tumor-only analysis methods, and integration with two earlier published datasets
• Who was studied: 120 surgically resected malignant pleural mesothelioma (MPM) tumors from the French MESOBANK (105 with all 3 data types), from patients diagnosed between 1998 and 2017; later combined with the Bueno et al. (n = 181) and Hmeljak et al. (n = 73) cohorts for 374 samples in total
• Exposure: Not an exposure study. The dataset includes asbestos exposure variables (exposure, probability, frequency, intensity) and smoking data. The asbestos variables were among the covariables tested for association with methylation principal components during quality control; the article gives no asbestos-related risk estimates.
• Main results:
  • Whole-genome sequencing was done on 130 fresh-frozen MPMs plus 54 matched-normal tissue or blood samples. The mean percentage of aligned reads was 98.93% ± 0.81%.
  • RNA sequencing of 126 samples: on average, 96.8% ± 1.2% of reads mapped to the reference genome.
  • DNA methylation was analyzed for 135 MPM samples from 122 patients. One poor-quality sample was removed. Normalization reduced technical batch effects while keeping biological effects such as histologic type.
  • For copy number calling, 57 matched pairs were used as the evaluation set. Tumor-only copy number calls agreed closely with tumor/normal calls: purity R = 0.988, ploidy R = 1, number of segments R = 0.981, and gene-level copy number R > 0.94. Artifactual peaks near telomeric and centromeric regions were removed.
  • The random forest model for tumor-only point mutations was trained using 46 tumors with matched normal whole-genome sequences.
  • Random forest models for tumor-only point mutations reached an accuracy of 0.926 for single-nucleotide variants and 0.924 for indels. Of 1,454,942 variants classified, 217,436 were classified as somatic.
  • Structural variant models reached precision 0.905 ± 0.009, recall 0.87 ± 0.016 and accuracy 0.889 ± 0.010. The authors observed no significant difference between structural variants called in tumor-only and matched samples. The total was 12,914 structural variants.
  • Sample-matching checks found two whole-genome samples from the same patient that had been labeled differently, and an RNA-seq misannotation; both were corrected. Sex annotation was confirmed for all samples.
  • Task specialization (Pareto) analysis found a triangle with 3 archetypes. These were interpreted as cell division, tumor–immune interaction and acinar phenotype.
  • Reprocessing two earlier published studies with the same methods produced a harmonized dataset of 374 samples in total, shared on the UCSC TumorMap portal.
• The authors' conclusion: The authors concluded that they provide a high-quality multiomic dataset of malignant pleural mesothelioma, harmonized with two earlier series. They also said future studies will be needed at second- and third-line treatment time points to develop precision medicine strategies.
• Limitations the authors state:
  • Part of the cohort had no matched normal tissue, so tumor-only analysis methods were needed.
  • Samples were retrospective. Patients were diagnosed in 1998–2017, before recent clinical trials and before nivolumab and ipilimumab were authorized, so few had precision treatments.
  • Comorbidity data were not available.
  • Tumor-only copy number calls produced artifactual peaks near telomeric and centromeric regions; these segments were removed.
  • Four samples could not have genomic purity estimated and were excluded from that analysis.

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.

Related studies

• Multiomic analysis of malignant pleural mesothelioma identifies molecular axes and specialized tumor profiles driving intertumor heterogeneity — L. Mangiante, N. Alcala, et al. (2023)
• Autotaxin signaling governs phenotypic heterogeneity in visceral and parietal mesothelia — E.L. Shelton, C.L. Galindo, et al. (2013)
• Phenotypic characterisation of peripheral blood lymphoid cells in people exposed to fibrous zeolite — M. Ozesmi, A. Karlsson-Parra, et al. (1986)
• Phenotypic stability and metastatic behaviour of serially xenografted rat mesotheliomas — R.E. Edwards, R.J. Hill, et al. (1990)
• Highlights of the 14th international mesothelioma interest group meeting: Pathologic separation of benign from malignant mesothelial proliferations and histologic/molecular analysis of malignant mesothelioma subtypes — A. Churg, K. Nabeshima, et al. (2018)
• Effect of insulin and epidermal growth factors on the synthesis of glycosaminoglycans/proteoglycans in cultured human malignant mesothelioma cells of different phenotypic morphology — G.N. Tzanakakis, N.K. Karamanos, et al. (1996)

Browse return to medical research.