Objective: To compare modified RECIST (mRECIST), EORTC criteria, and PERCIST for response evaluation and prognosis prediction in advanced malignant pleural mesothelioma (MPM) patients treated with chemotherapy.
Methods: Patients with MPM and not curative surgery candidates (n = 75) underwent standard chemotherapy with cisplatin and pemetrexed. CT and [F]fluorodeoxyglucose PET/CT scans were performed at baseline and after three chemotherapy cycles. Chemotherapeutic response was evaluated according to mRECIST, EORTC, and PERCIST, then concordance among those was assessed using Cohen's κ coefficient. PFS and OS were examined using log-rank and Cox methods.
Results: With EORTC, 27 patients had PMD, 23 SMD, 17 PMR, and eight CMR, while with PERCIST those were 28, 22, 11, and 14, respectively. Using mRECIST, 28 had PD, 29 SD, 18 PR, and 0 CR. Although there was high concordance between EORTC and PERCIST (82.7% of patients; κ = 0.83), that between mRECIST and EORTC (38.7%; κ = 0.27) and mRECIST and PERCIST (36.0%; κ = 0.26) was low. According to both EORTC and PERCIST, patients with no progression (CMR/PMR/SMD) showed significantly longer PFS and OS than PMD patients (EORTC: P = 0.0024 and P = 0.039, respectively, PERCIST: P = 0.0012 and P = 0.024, respectively), while according to mRECIST, those who achieved no progression (PR/SD) showed significantly longer PFS than PD patients (P = 0.011), but not significantly longer OS (P = 0.11).
Conclusion: EORTC and PERCIST are more accurate than mRECIST for evaluation of tumor response to chemotherapy and predicting prognosis in unresectable MPM patients.