A novel prognostic model for malignant mesothelioma incorporating quantitative FDG-PET imaging with clinical parameters

摘要

Purpose: Existing prognostic systems for malignant pleural mesothelioma do not incorporate imaging information. We aimed to identify the contribution of quantitative fluorodeoxyglucose positron emission tomography (FDG-PET) analysis to other prognostic variables in this disease. Experimental Design: Patients with malignant pleural mesothelioma underwent helical thoracoabdomin0al computed tomography and FDG-PET scans at baseline. Patients were treated as clinically indicated and followed for survival. FDG-PET variables derived included total glycolytic volume, a composite of tumor volume and glycolytic activity. Results: Ninety-three patients were accrued from 2003 to 2006. Of 89 eligible assessable patients, 28 had undergone pleurodesis before enrolment. Seventeen patients remained alive at analysis; median survival is 15.4 months. On univariate analysis, significant prognostic factors were: total glycolytic volume on FDG-PET (P = 0.003), sarcomatoid histology (P < 0.0005), weight loss (P = 0.031), computed tomography stage (P = 0.015), and European Organization for Research and Treatment of Cancer good prognostic score (P = 0.049). In patients with epithelioid or biphasic histology, baseline total glycolytic volume remained predictive of survival in patients with (P = 0.01) or without (P = 0.018) previous pleurodesis. In multivariate analysis, no variable other than histology contributed to the model in patients with sarcomatoid histology; total glycolytic volume and weight loss contributed to the models in patients with nonsarcomatoid histology. computed tomography-assessed tumor-node-metastasis stage did not contribute to the model. A nomogram, which incorporates quantitative PET parameters and pleurodesis into prognostic information, is presented. Conclusions: Sarcomatoid histology remains the strongest prognostic factor. In patients with non sarcomatoid disease, volumetric FDG-PET parameters are more predictive of survival than tumor-node-metastasis staging, suggesting that tumor volume and glycolytic activity may be more important determinants of prognosis in malignant pleural mesothelioma than anatomic extent of disease.