Distant Metastatic Pattern and Its Prognostic Significance in Malignant Pleural Mesothelioma: A Population-Based Study Based on a Machine Learning Model

Our plain-English summary of Distant Metastatic Pattern and Its Prognostic Significance in Malignant Pleural Mesothelioma: A Population-Based Study Based on a Machine Learning Model, by J. Yu, C. Peng, Q. Ye, et al (The Clinical Respiratory Journal, 2025).

The study

• Authors: J. Yu, C. Peng, Q. Ye, C. Huo, B. Gao, Q. Wei, Y. Li, K. Yang
• Journal: The Clinical Respiratory Journal
• Published: 2025;19(11):e70133
• Cited by: 1 article, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Background: Malignant pleural mesothelioma (MPM) is an insidious and aggressive tumor, often hindering timely clinical interventions. Despite its clinical relevance, epidemiological research focusing on MPM metastases remains limited.

Methods: We conducted a retrospective review of MPM cases with site-specific metastasis records from the Surveillance, Epidemiology, and End Results (SEER) between 2010 and 2019. Propensity Score Matching was employed to minimize bias between distant metastasis and non-distant metastasis groups. A prognostic model for predicting overall survival was established using clinical variables derived from Lasso regression. Variable importance for survival outcomes was estimated using the Random Survival Forests algorithm. The performance of the nomogram was evaluated using the receiver operating characteristic (ROC) curves and calibration plots.

Results: The presence of distant metastasis significantly reduced median overall survival from 10.5 to 7 months, with further detriment observed in cases with sarcomatoid histology and without chemotherapy intervention. Multivariable analysis identified sarcomatoid subtype, T4 stage, N1+ nodal involvement, and bilateral disease as significant predictors of increased metastatic potential. Histology, surgery, and metastasis status emerged as the top three clinical variables influencing survival. The nomogram demonstrated strong discrimination and calibration for predicting the 1-year and 3-year overall survival in both training and validation cohorts. The contralateral lung was the most frequent site of distant metastasis, with lymph node metastasis presenting a significantly better prognosis than that observed in patients with metastases to other organs.

Conclusions: The large population-based analysis provides a comprehensive characterization of site-specific metastases in MPM. The identified risk factors can help stratify patients at higher risk for metastatic progression and support early, targeted clinical decision-making.

© 2025 The Author(s). The Clinical Respiratory Journal published by John Wiley & Sons Ltd.

In plain English

This retrospective study used the US SEER cancer registry to look at people with malignant pleural mesothelioma (MPM) diagnosed from 2010 to 2019. It asked how cancer spread to distant parts of the body (distant metastasis) related to survival. The authors reported 2059 patients with complete records, 369 of them with distant metastases, although their table gives different group sizes. They then matched 336 patients with distant metastasis to 336 without on other characteristics. Median overall survival was 7 months in the metastasis group and 10.5 months in the group without. In a multivariable survival analysis, distant metastasis was linked to a higher risk of death (hazard ratio 1.61, 95% CI 1.38–1.89). Sarcomatoid histology, T4 tumor stage, lymph node involvement (N1+) and bilateral disease were associated with distant metastasis. The contralateral lung was the most frequent metastatic site. Patients with multiple distant lesions had a median survival of 6 months, compared with 9 months for those with a single lesion. The registry did not record asbestos exposure. The authors concluded that distant metastasis is an independent prognostic factor in MPM.

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: Retrospective population-based registry study (SEER database), using propensity score matching, logistic and Cox regression, and a machine-learning survival model (Random Survival Forest)
• Who was studied: Patients with microscopically confirmed malignant pleural mesothelioma in the US SEER 17 registries, diagnosed 2010–2019. There were 2059 patients with complete records, 369 of them with distant metastases. After matching, 336 patients were in each group.
• Exposure: The factor studied was distant metastasis status and its sites, not asbestos. Asbestos exposure was not available in the SEER data.
• Main results:
  • In the matched groups (336 per group), median overall survival was 7 months with distant metastasis and 10.5 months without (log-rank p less than 0.001).
  • In the multivariable Cox regression, distant metastasis (yes vs no) had a hazard ratio for death of 1.61 (95% CI 1.38–1.89).
  • In multivariable logistic regression, the following were independent risk factors for distant metastasis: sarcomatoid subtype (OR 1.92, 95% CI 1.24–2.98, vs biphasic), T4 stage (OR 2.05, 1.52–2.77, vs T1), N1+ (OR 2.12, 1.66–2.71, vs N0) and bilateral disease.
  • In the Random Survival Forest analysis, histology was the most important factor for survival (importance score 0.127), followed by surgery (0.066) and metastasis status (0.057).
  • Reported frequency of metastatic sites: contralateral lung 95 (18.96%), distant lymph nodes 89 (17.76%), bone 72 (14.40%), liver 34 (6.78%), brain 8 (1.60%). The denominator for these percentages is not stated.
  • Median survival was 6 months for patients with multiple distant lesions and 9 months for those with a single metastatic lesion (log-rank p less than 0.001). Patients with a solitary distant lymph node metastasis had the most favorable prognosis.
  • The survival prediction model had 1-year and 3-year AUC values of 0.739 and 0.727 in the training set, and 0.718 and 0.713 in the validation set.
• The authors' conclusion: The authors concluded that distant metastasis is an independent prognostic factor in MPM with substantial influence on overall survival. They also concluded that histologic subtype, T-stage, N-stage and laterality are key risk factors for distant metastasis.
• Limitations the authors state:
  • SEER lacks detailed information on metastases to organs other than bone, brain, liver and lung.
  • Missing data and the exclusion of patients without pathological confirmation may introduce selection bias.
  • SEER does not record asbestos exposure, primary tumor site or smoking, which may affect the assessment of other variables.
  • The timing of metastasis (before or after surgery) is not captured.

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, Metastasis, Epidemiology, SEER, Overall Survival, Median Overall Survival, Sarcomatoid, Histology.

Related studies

• The use of a next-generation sequencing-derived machine-learning risk-prediction model (OncoCast-MPM) for malignant pleural mesothelioma: a retrospective study — M.G. Zauderer, A. Martin, et al. (2021)
• Risk factors, prognostic factors, and nomograms for distant metastasis in patients with malignant pleural mesothelioma: A population-based retrospective study — Q. Lu, C. Liu, et al. (2025)
• Cancer-Associated Fibroblasts Influence Survival in Pleural Mesothelioma: Digital Gene Expression Analysis and Supervised Machine Learning Model — S. Borchert, A. Mathilakathu, et al. (2023)
• Prediction of Radiation Pneumonitis With Machine Learning in Stage III Lung Cancer: A Pilot Study — M. Yakar, D. Etiz, et al. (2021)
• Cancer Risk Prediction Using Machine Learning for Supporting Early Cancer Diagnosis in Symptomatic Patients: A Systematic Review of Model Types — F. Pennisi, S. Borlini, et al. (2025)
• Near-infrared spectroscopy combined with machine learning for plasma-based discriminant diagnosis of malignant mesothelioma: a retrospective study — Y. Gu, R. Wang, et al. (2025)

Browse more research in The Clinical Respiratory Journal, or return to medical research.