Epstein‐Barr virus (EBV), characterized as an omnipresent virus, has been found able to infect NK cells and leads to NK‐cell type EBV‐positive lymphoproliferative diseases (EBV‐NK‐LPDs). We retrospective analyzed 202 EBV‐NK‐LPDs (including 64 CAEBV‐NK, 27 aggressive natural killer‐cell leukemia (ANKL), and 111 extranodal NK/T‐cell lymphoma (ENKTL)) patients' relationships between EBV DNA copies laboratory test results and clinical features. In CAEBV‐NK cohort, EBV DNA loads in either plasma or PBMCs had significant differences between the active state and the inactive state. Receiver operating characteristic curves were used to measure the diagnosis accuracy of EBV DNA copies. After comparing the area under the curve, EBV DNA loads in plasma had significantly higher accuracy in distinguishing disease activation than in PBMCs. Therefore, we propose redefining CAEBV‐NK diagnosis criteria as increased EBV DNA copies in plasma (over 7.1 × 102 copies/ml) instead of in peripheral blood. In ANKL and ENKTL cohorts, patients who received effective therapy had significantly lower EBV DNA copies in plasma & PBMCs than in those with ineffective therapy. The significant and consistent decline indicated EBV DNA loads in plasma being a more sensitive biomarker in monitoring EBV‐NK‐LPDs therapy responses. Hemophagocytic lymphohistiocytosis (HLH) can occur secondary to EBV‐NK‐LPDs, mostly associated with a poor prognosis, so we try to estimate the combination of HLH by monitoring EBV DNA copies. When comparing the Receiver operating characteristic curves of EBV DNA copies, EBV DNA loads in plasma had higher diagnosis accuracy. When the copies level over 4.16 × 103 copies/ml, it might indicate combining with HLH.

No abstract

IntroductionChronic active Epstein-Barr virus (CAEBV) disease is a high-mortality disease, which is characterized by persistent infectious mononucleosis-like symptoms. There is no standard treatment for CAEBV and allogeneic hematopoietic stem cell transplantation (HSCT) was considered the only potentially therapeutic approach. PD-1 inhibitors have achieved high response in many Epstein-Barr virus-related diseases. In this single-center retrospective analysis, we report the outcomes of PD-1 inhibitors treatment of CAEBV.MethodsAll CAEBV patients without hemophagocytic lymphohistiocytosis (HLH), who were treated with PD-1 inhibitors in our center between 6/1/2017 and 12/31/2021, were retrospectively analyzed. The efficacy and safety of the PD-1 inhibitors were evaluated.ResultsAmong the sixteen patients with a median age at onset of 33 years (range, 11-67 years), twelve patients responded to PD-1 inhibitors and the median progression-free survival (PFS) was 11.1 months (range, 4.9-54.8 months). Three achieved clinical complete response (clinical CR), as well as molecular CR. Five patients achieved and remained partial response (PR), and four converted from PR to no response (NR). For three CR patients, the median time and cycles from the first application of PD-1 inhibitor to clinical CR were 6 weeks (range, 4-10 weeks) and 3 cycles (range, 2-4 cycles), and molecular CR was achieved after a median of 16.7 weeks (range, 6.1-18.4 weeks) and 5 cycles (range, 3-6 cycles) of PD-1 inhibitor infusion. No immune-related adverse events have been observed except for one patient who suffered immune-related pancreatitis. There was no correlation of treatment outcome with blood count, liver function, LDH, cytokine or ferritin levels. NK cell function, PD-L1 expression in tumor tissue and gene mutation possibly correlated with treatment response.DiscussionIn patients with CAEBV, PD-1 inhibitors have tolerable toxicity and comparable outcomes while improving quality of life and financial toxicity. Larger prospective studies and longer follow-up time is needed to be conducted.