Human chorionic gonadotrophin regression rate as a predictive factor of postmolar gestational trophoblastic neoplasm in high-risk hydatidiform mole: A case-control study

Objective: The aim of this study was early prediction of postmolar gestational trophoblastic neoplasm (GTN) after evacuation of high-risk mole, by comparison of human chorionic gonadotrophin (hCG) regression rates. Study design: Fifty patients with a high-risk mole initially and spontaneously regressing after molar evacuation were selected from January 1, 1996 to May 31, 2010 (spontaneous regression group). Fifty patients with a high-risk mole initially and progressing to postmolar GTN after molar evacuation were selected (postmolar GTN group). hCG regression rates represented as hCG/initial hCG were compared between the two groups. The sensitivity and specificity of these rates for prediction of postmolar GTN were assessed using receiver operating characteristic curves. Multivariate analyses of associations between risk factors and postmolar GTN progression were performed. Results: The mean regression rate of hCG between the two groups was compared. hCG regression rates represented as hCG/initial hCG (%) were 0.36% in the spontaneous regression group and 1.45% in the postmolar GTN group in the second week (p = 0.003). Prediction of postmolar GTN by hCG regression rate revealed a sensitivity of 48.0% and specificity of 89.5% with a cut-off value of 0.716% and area under the curve (AUC) of 0.759 in the 2nd week (p < 0.001). In patients with an hCG regression rate over 0.716% in the 2nd week, the hazard ratio for progression to postmolar GTN was 3.00 by multivariate analysis (p < 0.001). Conclusion: Differences in hCG regression rates between spontaneous regression and postmolar GTN groups became evident from the second week following molar evacuation. The occurrence of postmolar GTN could be predicted as early as the second week by comparing regression rates. hCG regression rate is easily obtainable and a predictive factor for postmolar GTN.