Guidelines for the stratification of patients recruited to trials of therapy for low-risk gestational trophoblastic tumor

Objective. The aim of the study was to initiate a search for factors which might independently predict the need for salvage therapy in patients with low-risk gestational trophoblastic tumor. Methods. The independent effect of six factors on the need for salvage chemotherapy was assessed in patients with low-risk gestational trophoblastic tumor who were treated with low-dose methotrexate and folinic acid. The accuracies of World Health Organization and Charing Cross Hospital scores were also compared. Results. Age, pretreatment βhCG, antecedent pregnancy-treatment interval, and the presence of chest metastases detected on chest X ray were not significantly predictive. The size of tumor (P = 0.001) and the presence of chest metastases on chest computerized tomography (P = 0.00028) had independent, statistically significant predictive power, and a simple prognostic index was derived from these variables. The World Health Organization score was found to be significantly better than the Charing Cross Hospital score. The accuracy of the simple prognostic index was slightly greater than that of the World Health Organization score, although this was not statistically significant. Conclusions. These results confirm that patients entered into studies of different therapies for low-risk gestational trophoblastic tumor should be stratified and that a simple score, derived from the results of tumor size and chest computerized tomography, is potentially as good as the World Health Organization score for predicting the need for salvage therapy. (C) 2000 Academic Press.