Abstract
Background: Hemophagocytic lymphohistiocytosis (HLH) is a severe systemic inflammatory disease which is life-threatening if not detected earlier and treated appropriately. The diagnosis of HLH is sometimes confused due to other similar febrile diseases with cytopenia such as severe infection, flare-up of autoimmune disease, and malignant conditions. Decreased natural-killer cell (NK) cytotoxicity is an important diagnostic parameter for HLH, but we do not know the level and role in similar febrile diseases.
Aim: We tried to identify the different levels in variable febrile diseases and find out the role for diagnosis and prognostication in adult patients with HLH compared to other similar diseases.
Methods: We prospectively enrolled 41 patients from 2015 to 2017. Adult patients older than 18 years with fever>38℃ presenting cytopenia in at least two lineages (neutrophil<1,000/㎕, platelet<100,000/㎕, Hemoglobin<9.0/dL) were included. Patients with previous hematological diseases were excluded. Diagnosis of HLH was based on HLH2004 criteria. Infection was managed according to the protocol and HLH-suspected patients were initially treated with 10mg/BSA of dexamethasone, and etoposide was considered if clinical improvement was not observed within 7 days after dexamethasone. Patients other than HLH were treated with disease-specified therapy. NK cytotoxicity was calculated at diagnosis by antibody-dependent Raji-cell cytotoxicity (ADCC) assay and K562-cell direct lysis using flow cytometry. Concomitantly, CD107a expression and NK-induced interferon gamma were also calculated.
Results: HLH was diagnosed in 29 patients caused by viral infection (n=8), malignancies (n=6), autoimmune diseases (n=6), bacterial infection (n=2), malaria (n=1), and unidentified origin (n=6). Febrile diseases other than HLH (n=12) were diagnosed with hematological disease (n=7), infectious mononucleosis (n=2), SLE (n=2), Kikuchi disease (n=1). The results of both K562 lysis and ADCC assay were highly correlated but the diagnostic power was greater in ADCC assay. In the entire group, ADCC lower than 40% was related with poor 1-year OS (70.2% vs. 100%, P=0.062). Median ADCC level was lower in HLH (20.5% vs. 37.7%, P=0.045) and in HLH with poor dexamethasone response (12.0% vs. 30.7%, P=0.009). ROC curve revealed that ADCC lower than 22.8% was useful for diagnosis of HLH and lower than 12.8% was related with poor treatment response. Unfortunately, CD107a expression and NK-induced interferon gamma was not useful for diagnosis and prognostication.
Conclusion: Lower NK cytotoxicity was diagnostic for HLH compared to other febrile disease with cytopenia, and we identified ADCC lower than 40% was related with poor OS and lower than 12.8% was predictable for severe HLH presenting poor treatment outcome.
No relevant conflicts of interest to declare.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal