Plasma EBV-DNA and peripheral blood mononuclear cell EBV-DNA have disparate clinical relevance in patients with extranodal NK/T-cell lymphoma.

• PBMC EBV-DNA poorly correlated with imaging-based disease assessment and plasma EBV-DNA in patients with ENKTCL, while Plasma EBV-DNA strongly correlated with disease status. • Plasma EBV-DNA was more sensitive than PET/CT and CT/MRI in response evaluation and relapse monitoring. Regardless of imaging-based assessment, a substantial increase of plasma EBV-DNA indicated disease progression during treatment, and relapse was almost inevitable in patients with positive plasma EBV-DNA at the end of treatment. • When imaging-based disease assessment was discordant with plasma EBV-based assessment, the kinetic patterns of EBV-DNA were helpful to identify at-risk patients. Relapse occurred only in patient with intermittently or persistently positive plasma EBV-DNA, and occasionally positive EBV-DNA did not correlate with relapse. Extranodal NK/T-cell lymphoma (ENKTCL) is an Epstein-Barr virus (EBV)-related hematological malignancy. The presence of EBV-DNA in peripheral blood is a widely used ENKTCL tumor marker. However, there is no consensus on the preferred blood specimen type for EBV testing. Furthermore, discordance between EBV-based and imaging-based disease assessments is common, and how to interpret this discordance is important. We retrospectively analyzed the data of ENKTCL patients in the Affiliated Cancer Hospital of Zhengzhou university and Sun Yat-sen University Cancer Center. All EBV-DNA and imaging-based disease assessment data were collected at diagnosis, during treatment, at the end of treatment, and during follow-up. We compared matched plasma EBV-DNA and peripheral blood mononuclear cell (PBMC) EBV-DNA and matched EBV-based and imaging-based assessments to uncover their clinical relevance. A total of 450 patients with adequate data were included, of whom 278 had plasma EBV-DNA data, 250 had PBMC EBV-DNA data, and 78 had matched plasma and PBMC EBV-DNA data. No significant correlations were found between PBMC and plasma EBV-DNA and between PBMC EBV-DNA and imaging-based assessment, but patients with positive PBMC EBV-DNA at diagnosis or intermittently/persistently positive PBMC EBV-DNA during follow-up had poorer survival. In contrast, plasma EBV-DNA strongly correlated with lymphoma status. Detectable pre- and post-treatment plasma EBV-DNA was associated with significantly worse survival. Patients with early-stage disease who had detectable plasma EBV-DNA at the end of treatment shared similar survival to those with advanced-stage disease, even if their imaging-based assessments were negative. For disease relapse monitoring, 78 (55.7%) episodes of relapse were detected by both imaging and plasma EBV-DNA; 58 (41.4%) detected by plasma EBV-DNA earlier than imaging, with a median time of 9.3 (0.3 - 37.8) months; and only 4 (2.9%) detected by plasma EBV-DNA later than imaging. The sensitivities of plasma EBV-DNA, PET/CT, and CT/MRI were 97.1%, 76.8%, and 45.1%, respectively, and their specificities were 91.7%, 84.2%, and 96.7%, respectively. Analysis of EBV kinetic patterns in EBV+/imaging- episodes revealed that relapse occurred only in patients with intermittently/persistently positive plasma EBV-DNA. Persistent plasma EBV+ was also seen in patients after autologous hematopoietic stem cell transplantation. Occasional EBV+ was not associated with relapse. Plasma and PBMC EBV-DNA have different clinical relevance in ENKTCL patients. PBMC EBV-DNA does not correlate with imaging-based disease assessment. PBMC or even whole blood should not be used for response evaluation and relapse monitoring. However, PBMC EBV-DNA still has prognostic value. Plasma EBV-DNA is strongly related to tumor status and is not only a prognosticator at diagnosis and end of treatment, but also a sensitive marker in relapse monitoring compared to PET/CT and CT/MRI. The specificity of plasma EBV-DNA is relatively low, but when EBV-DNA kinetic patterns are considered, it can identify at-risk patients. [ABSTRACT FROM AUTHOR]