Pre-treatment Tumor Volume Growth Rate and Baseline Volume Measured by an Artificial Intelligence Program Predict Overall Survival in Patients With Pleural Mesothelioma Undergoing Deferred Systemic Therapy

Our plain-English summary of Pre-treatment Tumor Volume Growth Rate and Baseline Volume Measured by an Artificial Intelligence Program Predict Overall Survival in Patients With Pleural Mesothelioma Undergoing Deferred Systemic Therapy, by I. Smesseim, V. Pugliese, J. Burgers, et al (JTO Clinical and Research Reports, 2026).

The study

• Authors: I. Smesseim, V. Pugliese, J. Burgers, K.G. Lipman
• Journal: JTO Clinical and Research Reports
• Published: 2026;7(9):101040
• Read it: PubMed · Publisher · Free full text

Abstract

Introduction: The optimal timing of systemic therapy in pleural mesothelioma remains debated, particularly for asymptomatic patients with a limited disease burden. Prior studies suggest that deferred treatment may not compromise overall survival (OS) in selected patients. Reliable prognostic markers to guide treatment timing are lacking. We investigated whether baseline tumor volume and pretreatment tumor growth rate, quantified using an artificial intelligence (AI)-based model, predict OS in patients deferring systemic therapy.

Methods: We conducted a single-center retrospective study including patients with pleural mesothelioma diagnosed between 2010 and 2025 who deferred systemic therapy for more than or equal to 3 months and had at least two pretreatment computed tomography scans. Total tumor volume and tumor growth rate (mL/mo) were automatically measured using the AI-based ARTIMES model. Optimal cutoffs were determined using receiver operating characteristic analysis. OS was analyzed using Kaplan-Meier estimates and Cox proportional hazards models.

Results: A total of 71 patients were included (median age 66.8 y; 80.3% male; Eastern Cooperative Oncology Group 0-1). The median deferral time was 8.0 months. Optimal cutoffs were 56.5 mL for baseline tumor volume and 67.8 mL/mo for tumor growth rate. Patients with a low baseline tumor volume had a significantly longer median OS compared with those with a high volume (30.95 versus 16.89 mo; p = 0.0066). Similarly, a low tumor growth rate was associated with improved OS (29.40 versus 14.59 mo; p = 0.00057). In multivariable analysis, sarcomatoid histology remained independently associated with worse OS (hazard ratio = 3.28, 95% confidence interval: 1.38-7.75, p = 0.007).

Conclusions: AI-derived baseline tumor volume and tumor growth rate are strong prognostic factors for OS in patients with pleural mesothelioma deferring systemic therapy. Patients with low baseline tumor volume and those with low tumor growth rate were significantly associated with a longer OS.

© 2026 by the International Association for the Study of Lung Cancer.

In plain English

This single-center retrospective study at the Netherlands Cancer Institute, Amsterdam, examined 71 patients with pleural mesothelioma, referred from 2010 to 2025, who postponed systemic (drug) treatment for at least 3 months and had at least two pretreatment CT scans. Median postponement was 8.0 months. An artificial intelligence program, ARTIMES, measured total tumor volume in milliliters and its growth rate in milliliters per month. The study did not examine asbestos exposure. Patients with a baseline volume at or below the 56.5 mL cutoff had a median overall survival of 30.95 months, versus 16.89 months for those above it (p = 0.0066). Patients with a growth rate at or below the 67.8 mL per month cutoff had a median of 29.40 months, versus 14.59 months for those above it (p = 0.00057). In a multivariable model, sarcomatoid histology was independently associated with worse survival (hazard ratio 3.28, 95% CI 1.38 to 7.75). The authors concluded that lower volume and slower growth were associated with improved survival. They noted the study was small and retrospective and could not show whether postponing treatment itself changes outcomes.

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: Single-center retrospective study
• Who was studied: Patients with pleural mesothelioma referred to the Netherlands Cancer Institute, Amsterdam, between January 2010 and January 2025. Of 113 patients with at least two pretreatment scans, 71 were included. All had ECOG performance status 0 to 1, deferred systemic therapy for at least 3 months, and later received at least one line of therapy. The median age was 66.8 years, 57 of 71 (80.3%) were male, and 52 (73.2%) had epithelioid histology.
• Exposure: No asbestos exposure was studied. The factors examined were baseline total tumor volume (mL) and pretreatment tumor growth rate (mL/mo), both measured automatically on CT scans by the AI model ARTIMES. Other candidate prognostic factors were also tested: age, sex, ECOG, histology, pleurodesis, tumor side and deferral time.
• Main results:
  • The cutoff for baseline tumor volume was chosen for predicting 12-month mortality. It was 56.5 mL, with sensitivity 84.6% and specificity 58.6%. The low-volume group was defined as 56.5 mL or less.
  • The cutoff for tumor growth rate was 67.8, written as mL in this passage and as mL per month elsewhere in the article. Sensitivity was 76.9% and specificity 77.6%. The slow-growth group was defined as 67.8 mL per month or less.
  • Median overall survival was 30.95 months in patients with baseline tumor volume at or below 56.5 mL. In patients with higher baseline volume it was 16.89 months.
  • Median overall survival in the low baseline volume group was 30.95 months (IQR 19.19 to 57.92). In the high baseline volume group it was 16.89 months (IQR 9.40 to 31.67). The difference had p = 0.0066.
  • Median overall survival was 29.40 months in patients with tumor growth rates at or below 67.8 mL per month. In patients with faster growth it was 14.59 months.
  • Median overall survival in the low growth rate group was 29.40 months (IQR 17.35 to 57.46). In the high growth rate group it was 14.59 months (IQR 9.00 to 30.03). The difference had p = 0.00057.
  • In univariate Cox analysis, a low growth rate was associated with a lower hazard of death (HR 0.40, 95% CI 0.23 to 0.68, p = 0.00083).
  • A low baseline volume was associated with a lower hazard of death (HR 0.52, 95% CI 0.32 to 0.84, p = 0.0074).
  • In univariate analysis, sarcomatoid and biphasic histology had higher hazards than epithelioid histology (HR 2.24, 95% CI 1.07 to 4.69, p = 0.033).
  • In univariate analysis, longer deferral time was associated with a lower hazard of death (HR 0.98, 95% CI 0.97 to 1.00, p = 0.013).
  • In univariate analysis, pleurodesis was associated with a lower hazard of death (HR 0.46, 95% CI 0.24 to 0.88, p = 0.018).
  • The multivariable Cox model included histology, deferral time, pleurodesis, growth rate and baseline volume. In it, sarcomatoid histology was independently associated with worse overall survival (HR 3.28, 95% CI 1.38 to 7.75, p = 0.007). Multivariable results for volume and growth rate are not given in the main text.
  • Histologic subtype distribution did not differ significantly between the low (36/71) and high (35/71) baseline volume groups (p = 1.00).
  • Histologic subtype distribution did not differ significantly between the low (48/71) and high (23/71) growth rate groups (p = 0.107).
• The authors' conclusion: The authors concluded that lower baseline tumor volume and slower tumor growth are associated with improved survival in patients with pleural mesothelioma who defer systemic therapy.
• Limitations the authors state:
  • The retrospective, single-center design introduces potential selection and information bias.
  • The sample size was relatively small (71 patients).
  • Treatment decisions were influenced by clinician judgment, patient preference and referral timing.
  • Patients who deteriorated during deferral and became too unwell for therapy are not included. The study may therefore underestimate how often deferral fails.
  • The study cannot determine whether deferral affects outcomes compared with immediate treatment or no treatment.
  • The results do not apply to patients needing immediate therapy or with more aggressive, symptomatic disease.

Topics

Medical terms in this study, explained in our Lexicon: Overall Survival, Pleural Mesothelioma, Systemic Therapy, Sarcomatoid, Histology, Hazard Ratio, confidence interval.

Related studies

• Progression free survival rate at 9 and 18 weeks predict overall survival in patients with malignant pleural mesothelioma: an individual patient pooled analysis of 10 European Organisation for Research and Treatment of Cancer Lung Cancer Group studies and an independent study validation — B. Hasan, L. Greillier, et al. (2014)
• The Mesothelioma Systemic Inflammation Score Is Independently Associated with Overall Survival and Predicts Benefit of Multimodality Treatment in Pleural Mesothelioma — B. Mosleh, K. Sinn, et al. (2025)
• Artificial Intelligence-based Quantification of Pleural Plaque Volume and Association With Lung Function in Asbestos-exposed Patients — K.B.W. Groot Lipman, T.N. Boellaard, et al. (2024)
• A Real-World Analysis of the Use of Systemic Therapy in Malignant Pleural Mesothelioma and the Differential Impacts on Overall Survival by Practice Pattern — K. Kerrigan, Y. Jo, et al. (2022)
• Artificial Intelligence Techniques to Predict the Airway Disorders Illness: A Systematic Review — A. Koul, R.K. Bawa, et al. (2023)
• Pre-treatment soluble PD-L1 as a predictor of overall survival for immune checkpoint inhibitor therapy: a systematic review and meta-analysis — Á. Széles, T. Fazekas, et al. (2023)

Browse more research in JTO Clinical and Research Reports, or return to medical research.