Yuki Kataoka1, Yosuke Yamamoto2, Taiichiro Otsuki3, Mariko Shinomiya4, Takayuki Terada5, Shingo Fukuma6, Shin Yamazaki2, Masataka Hirabayashi4, Takashi Nakano5, Shunichi Fukuhara7. 1. Department of Healthcare Epidemiology, Graduate School of Medicine and Public Health, Kyoto University, Kyoto Department of Respiratory Medicine, Hyogo Prefectural Amagasaki Hospital, Hyogo. 2. Department of Healthcare Epidemiology, Graduate School of Medicine and Public Health, Kyoto University, Kyoto. 3. Cancer Center, Hyogo College of Medicine, Hyogo. 4. Department of Respiratory Medicine, Hyogo Prefectural Amagasaki Hospital, Hyogo. 5. Division of Respiratory Medicine, Hyogo College of Medicine, Hyogo. 6. Department of Healthcare Epidemiology, Graduate School of Medicine and Public Health, Kyoto University, Kyoto Center for Innovative Research in Clinical Evaluative Science (CiRCLE), Fukushima Medical University, Hikarigaoka, Fukushima, Japan. 7. Department of Healthcare Epidemiology, Graduate School of Medicine and Public Health, Kyoto University, Kyoto Center for Innovative Research in Clinical Evaluative Science (CiRCLE), Fukushima Medical University, Hikarigaoka, Fukushima, Japan fukuhara.shunichi.6m@kyoto-u.ac.jp.
Abstract
OBJECTIVE: Existing prognostic indices for malignant pleural mesothelioma do not incorporate the recent advances in oncology care. The purpose of this study was to provide a prognostic index for overall survival in malignant pleural mesothelioma patients treated with chemotherapy with pemetrexed or best supportive care in the recent clinical setting. METHODS: A retrospective cohort study was performed in two hospitals in Japan (2007-13). The primary outcome was overall survival. The Cox proportional hazards model was used for multivariable analyses to identify prognostic factors. A final model was chosen based on both clinical and statistical significance. RESULTS: A total of 283 patients (chemotherapy: n = 228, best supportive care: n = 55) were enrolled in the study. On multivariate analysis, regimen including platinum plus pemetrexed, a performance status >0, non-epithelial histological type and Stage IV disease predicted poor overall survival in chemotherapy patients. As hazard ratios of individual risk factors were approximately similar, a prognostic index for overall survival was constructed by counting the risk factors. Median overall survival in chemotherapy patients decreased by each one-point increase in this count: 1030 days for zero; 658 days for one; 373 days for two; 327 days for three; 125 days for four. Internal validation using the bootstrapping technique showed robustness of the model (c-index, 0.677; 95% confidence interval, 0.624-0.729). Further, the discrimination was consistent in best supportive care patients (c-index, 0.799; 95% confidence interval, 0.725-0.874). CONCLUSIONS: This novel index can provide clinicians and malignant pleural mesothelioma patients with a better framework for discussing prognosis at the time of diagnosis.
OBJECTIVE: Existing prognostic indices for malignant pleural mesothelioma do not incorporate the recent advances in oncology care. The purpose of this study was to provide a prognostic index for overall survival in malignant pleural mesotheliomapatients treated with chemotherapy with pemetrexed or best supportive care in the recent clinical setting. METHODS: A retrospective cohort study was performed in two hospitals in Japan (2007-13). The primary outcome was overall survival. The Cox proportional hazards model was used for multivariable analyses to identify prognostic factors. A final model was chosen based on both clinical and statistical significance. RESULTS: A total of 283 patients (chemotherapy: n = 228, best supportive care: n = 55) were enrolled in the study. On multivariate analysis, regimen including platinum plus pemetrexed, a performance status >0, non-epithelial histological type and Stage IV disease predicted poor overall survival in chemotherapy patients. As hazard ratios of individual risk factors were approximately similar, a prognostic index for overall survival was constructed by counting the risk factors. Median overall survival in chemotherapy patients decreased by each one-point increase in this count: 1030 days for zero; 658 days for one; 373 days for two; 327 days for three; 125 days for four. Internal validation using the bootstrapping technique showed robustness of the model (c-index, 0.677; 95% confidence interval, 0.624-0.729). Further, the discrimination was consistent in best supportive care patients (c-index, 0.799; 95% confidence interval, 0.725-0.874). CONCLUSIONS: This novel index can provide clinicians and malignant pleural mesotheliomapatients with a better framework for discussing prognosis at the time of diagnosis.