A novel prognostic nomogram for predicting survival in diffuse pleural mesothelioma

Area of Science:

• Oncology
• Pathology
• Medical Informatics

Background:

• Advances in diffuse pleural mesothelioma (DPM) management necessitate improved prognostication.
• Combination immunotherapy benefits may be greatest in patients with poorer prognoses.
• Existing grading schemes like the Mesothelioma Weighted Grading Scheme (MWGS) exist, but deep learning offers superior predictive modeling.

Purpose of the Study:

• To develop and validate a prognostic nomogram for diffuse pleural mesothelioma (DPM).
• To create a predictive tool incorporating established prognostic markers for DPM.
• To aid in risk stratification and inform clinical management decisions for DPM patients.

Main Methods:

• Utilized data from 369 DPM patients in independent training and validation cohorts.
• Developed a prognostic tool incorporating age, sex, histological type, nuclear atypia, mitotic count, necrosis, and BAP1 immunohistochemistry.
• Assessed model discrimination and calibration using risk stratification into four groups, calculating AUC, C-index, and D-index.

Main Results:

• The developed nomogram achieved an AUC of 0.75 based on 5-year ROC analysis.
• The model demonstrated a C-index of 0.67 (95% CI 0.53-0.79) and a D-index of 2.40 (95% CI 1.69-3.43).
• The nomogram incorporates routinely assessed pathological factors, indicating excellent predictive capability.

Conclusions:

• This novel prognostic nomogram for DPM is the first of its kind.
• The model effectively integrates established prognostic markers, offering strong predictive performance.
• It is anticipated to assist in prognostication and guide critical management decisions, including patient selection for novel therapies.