Prognostic significance of early serum tumor marker half-life in metastatic testicular teratoma.

Abstract

PURPOSE To determine whether the initial regression rates of serum tumor marker concentration (alpha-fetoprotein [AFP] and beta-human chorionic gonadotrophin [HCG]) were important prognostic factors after chemotherapy for germ cell tumors. PATIENTS AND METHODS The analysis was confined to patients with at least two precise marker assay results between days 7 and 22 from start of platinum-based combination chemotherapy, with at least 7 days between markers. One hundred eighty-three patients were eligible and marker half-life (MHL) was evaluated for AFP in 142 and for HCG in 111 cases. MHL was calculated from the following formula: MHL = Ln1/2/G, where G was the gradient of the marker slope on a plot of Ln marker concentration versus time. MHL was regarded as prolonged if more than 3 days for HCG or more than 7 days for AFP. RESULTS The median AFP MHL was 6 days (range, 2.7 to 237) and the median HCG MHL was 2.6 days (range, 1.7 to 37.5). Forty-nine of 142 patients (35%) had a prolonged AFP MHL; 39 of 111 patients (35%) had a prolonged HCG-MHL. A prolonged MHL did not identify relapse after front-line chemotherapy. The positive predictive value of MHL tests in identifying patients who progressed after front-line therapy was 18% for HCG, 20% for AFP, and 18% for either marker. A prolonged MHL did indicate a higher risk of mortality (hazards ratio [HR], 2.4; P = .016), but again the positive predictive value of this test was only 23%. CONCLUSION Early evaluation of MHL by this method does not predict patients at higher risk of progression after front-line chemotherapy, and also is a poor guide to long-term prognosis.