Background Rheumatoid arthritis (RA) is associated with an increased risk of lymphoma linked to activity of the disease. Immunosuppressive drugs have been suspected to induce an additional risk. Since, NK cells have been recently shown to participate to anti–lymphoma immunosurveillance, we aimed to assess if anti-TNF might impact their anti-lymphoma activity.
Methods NK cells have been assessed ex vivo in patients with RA treated with methotrexate (MTX) with (n=19) or without (n=20) anti-TNF. Phenotype has been studied by flow cytometry and function (degranulation and IFNg production) has been assessed after NKp30-cross linking. Then, we have studied the consequences of in vitro exposure of NK to anti-TNF, to TNF-R inhibitors or to controls during 6 days: phenotype has been studied and then cytotoxicity against 2 B non-Hodgkin lymphoma cell lines [Farage (EBV+) and SU-DHL4 (EBV-)] was assessed. Results are shown as median [interquartile range (IGR)]. The Mann Whitney test was used to compare independent samples.
Results Firstly, in patients, we found that the proportion of activated NK cells, assessed by the expression of CD69, was significantly decreased in patients treated with anti-TNF + MTX compared to patients treated with MTX alone (8.9% [2.9–40.7] vs 48.4% [27.4–58.3], p=0.005). Moreover, we found that NK cells exhibited an impaired function in patients treated with anti-TNF compared to patients treated with MTX alone as assessed by the percentage of degranulation (20.9% [18.5–32.9] vs 31.3% [21.5–49.1], p=0.04) and the loss of capacity of IFN-γ secretion ((17.4% [8.9–25.9] vs to 29.7% [22.5–43.1], p=0.007). Secondly, we have confirmed that in vitro exposure to anti-TNF negatively impact NK cells activation and function leading to an impaired anti-lymphoma activity (figure 1A). In all these experiments, no difference was observed between etanercept and monoclonal anti-TNF. Last, we have demonstrated that negative impact of anti-TNF on NK cells may be the consequence of inhibition of TNF-R1 signaling (figure 1B).
Conclusions Even if meta-analysis of randomized controlled trials and of registries have not demonstrated to date an increased risk of lymphoma with anti-TNF, cautious must be pursued concerning this possible side effect in patients with long-term exposure. Moreover, negative impact of anti-TNF on NK cells may participate to infectious adverse events.
Disclosure of Interest None declared