e17014 Background: Studies of plasma from nasopharyngeal carcinoma (NPC) patients from endemic regions demonstrate that circulating DNA derived from EBV may be used as a tumor marker, but data from non-endemic regions or in whole blood are limited. Methods: ViraCor Laboratories (Lee's Summit, MO) performed PCR for EBNA, EBER, and LMP on both plasma and whole blood of 11 NPC patients undergoing therapy at the Hospital of the University of Pennsylvania. Results were correlated with disease status via review of chart records and imaging reports. Results: 7 of 11 patients tested were positive for circulating EBV derived DNA. 2 patients were treated with induction chemotherapy with suppression to 0 copies of circulating DNA for all plasma measures, but not whole blood measures (LMP 0, EBER 100 and EBNA 200 copies for one patient and LMP 100, EBNA 100 and EBER 300 copies for the other). The first of these patients remains free of disease but the other retains a large mass believed to be active disease, but which has not grown over 8 months following therapy. 5 patients, including the 2 patients treated with induction therapy, were treated with chemoradiotherapy. PCR following therapy was negative on all plasma measures for the 4 patients who remain free of disease (at intervals of 7, 12, 13 and 16 months). 2 of these patients retained low-level EBV on at least 1 whole-blood measure–1 for EBNA alone (100 copies) and the other for both EBER and EBNA (300 and 100 copies). 2 patients were treated for metastatic disease. 3 episodes of progression were all accompanied by increases in plasma EBV (EBNA from 13,700 to 21,100 with EBER 1600 to 9400 and EBNA 1800 to 2200 to 3000 with EBER 100 to 100 to 500). Conclusions: This pilot data on a population treated in the United States is consistent with larger data sets obtained in endemic regions demonstrating that EBV is an effective tumor marker for prognosis and to follow therapy. It is the first to demonstrate greater specificity of plasma measures over whole blood measures, and it raises the hypothesis that whole blood measures may be more sensitive. Further prospective study is warranted in non-endemic populations both to validate the measure and to facilitate translational research into biologic factors driving the behavior of these tumors. [Table: see text]