PET scan’s clinical utility in detecting residual disease in postchemotherapy germ cell cancer patients

Abstract

4574 Background: The goal for metastatic germ cell tumors is cure. Cisplatin-based chemotherapy is effective in achieving this goal. Salvage therapy, including high dose chemotherapy, can cure some refractory patients. After completion of chemotherapy, residual masses may be seen on CT Scans. PET scans may help determine if these masses reflect persistent disease, which helps decide which patients need further intervention, especially as it relates to metastatic seminoma. Methods: Retrospective analysis of PET Scans in 30 germ cell cancer patients was done at Indiana University. 17 had seminoma; 13 had nonseminomatous tumors. PET Scans were done a minimum of 6 weeks post-chemotherapy. 40 PET Scans were evaluated - 14 after primary chemotherapy and 26 after salvage chemotherapy. Results: The sensitivity and specificity of PET Scans in detecting residual disease were 95% and 90%, respectively. For PET scans conducted after initial chemotherapy, sensitivity and specificity were 100% and 88%, compared to 92% and 92% after salvage chemotherapy. With seminomas, sensitivity was 100% and specificity was 87%. However, of the seven seminoma patients with positive PET scans, three had only necrosis found at surgery. Of these three patients, one relapsed in the same PET positive area and thus the active seminoma was not detected during prior surgery. The positive predictive value of our seminoma PET scans was 71% (5/7). With nonseminonatous disease, sensitivity was 92% and specificity was 100%. One nonseminomatous patient with a negative PET scan relapsed 7 months later. Conclusion: Postchemotherapy PET scans have a high sensitivity and specificity in decision analysis of residual radiographic masses. A positive PET scan supports persistent tumor requiring further therapy, however in our small series of PET-positive seminoma, false-positive results were observed. No significant financial relationships to disclose.