Fang Jiang1, Xi-Run Wan2, Tao Xu3, Feng-Zhi Feng1, Tong Ren1, Jun-Jun Yang1, Jun Zhao1, Tao Yang4, Yang Xiang5. 1. Department of Obstetrics & Gynecology, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing 100730, China. 2. Department of Obstetrics & Gynecology, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing 100730, China. Electronic address: wanxr@sina.com. 3. Department of Epidemiology and Statistics Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences and School of Basic Medicine, Peking Union Medical College, Beijing 100005, China. 4. Department of Medical Record, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing 100730, China. 5. Department of Obstetrics & Gynecology, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing 100730, China. Electronic address: XiangY@pumch.cn.
Abstract
BACKGROUND: To re-evaluate the efficacy of the prognostic factors currently employed in the treatment of malignant gestational trophoblastic neoplasia. METHODS: Clinical data from the Gestational Trophoblastic Disease (GTD) Center at Peking Union Medical Hospital (PUMCH) collected between January 2002 and December 2013 were retrospectively analyzed. Univariate and multivariate analyses of prognostic factors were performed using the Cox proportional hazards model. A new hazard ratio (HR)-based prognostic scoring scale was established and compared with the original scoring system. RESULTS: In total, 1420 cases were included in the study (median follow-up=40months, overall complete remission (CR) rate=95.8%, relapse rate=7.1%, mortality rate=5.5%, median disease-free survival (DFS)=36months). Low-risk (0-6 points) and high-risk (≥6 points) patients exhibited CR rates of 99.8% (915/917) and 88.5% (445/503) and mortality rates of 0.3% and 15.1% (P<0.001), respectively. Univariate and multivariate analyses showed that age, pretreatment serum levels of human chorionic gonadotropin beta-subunit (β-hCG) and maximum tumor diameter were not independent prognostic risk factors. Antecedent pregnancy, the interval from the index pregnancy, the number of metastases and a history of failed chemotherapy treatments were independent prognostic risk factors. By modifying the scoring system based on the variables identified in a Cox analysis, we significantly increased the area under the receiver operating characteristics (ROC) curve. CONCLUSION: Though effective, the accuracy of the International Federation of Gynecology and Obstetrics (FIGO) 2000 Trophoblastic Neoplasia Staging System requires improvement. Irrelevant prognostic factors should be removed, and the weights of other factors should be adjusted appropriately.
BACKGROUND: To re-evaluate the efficacy of the prognostic factors currently employed in the treatment of malignant gestational trophoblastic neoplasia. METHODS: Clinical data from the Gestational Trophoblastic Disease (GTD) Center at Peking Union Medical Hospital (PUMCH) collected between January 2002 and December 2013 were retrospectively analyzed. Univariate and multivariate analyses of prognostic factors were performed using the Cox proportional hazards model. A new hazard ratio (HR)-based prognostic scoring scale was established and compared with the original scoring system. RESULTS: In total, 1420 cases were included in the study (median follow-up=40months, overall complete remission (CR) rate=95.8%, relapse rate=7.1%, mortality rate=5.5%, median disease-free survival (DFS)=36months). Low-risk (0-6 points) and high-risk (≥6 points) patients exhibited CR rates of 99.8% (915/917) and 88.5% (445/503) and mortality rates of 0.3% and 15.1% (P<0.001), respectively. Univariate and multivariate analyses showed that age, pretreatment serum levels of humanchorionic gonadotropin beta-subunit (β-hCG) and maximum tumor diameter were not independent prognostic risk factors. Antecedent pregnancy, the interval from the index pregnancy, the number of metastases and a history of failed chemotherapy treatments were independent prognostic risk factors. By modifying the scoring system based on the variables identified in a Cox analysis, we significantly increased the area under the receiver operating characteristics (ROC) curve. CONCLUSION: Though effective, the accuracy of the International Federation of Gynecology and Obstetrics (FIGO) 2000 Trophoblastic Neoplasia Staging System requires improvement. Irrelevant prognostic factors should be removed, and the weights of other factors should be adjusted appropriately.