Tocilizumab, a humanized antihuman interleukin-6 receptor antibody, has proven therapeutically effective for rheumatoid arthritis (RA) [1], systemic-onset juvenile idiopathic arthritis [2], Crohn’s disease [3] and Castleman’s disease [4]. However, infection is a major adverse effect of tocilizumab treatment. Inhibition by tocilizumab of interleukin-6 signaling in response to infection may also prevent fever induction and the elevation of acutephase reactants such as CRP. We previously reported that quantification of CD64 (FccR I) expressed on peripheral neutrophils is a useful marker for infection in patients with RA [5]. Neutrophil CD64 expression can be upregulated by complications due to vasculitis or interstitial pneumonia, but it is not affected by the disease activity of RA. Upregulation of CD64 on neutrophils can be observed in infections caused by any kind of pathogen, such as bacteria, viruses, fungi, and mycobacteria [5]. Based on this background information, we monitored neutrophil CD64 in an RA patient undergoing treatment with tocilizumab for 4 years who had repeatedly suffered from different infections but had little fever or elevated CRP (Fig. 1). A 53-year-old female patient with RA received tocilizumab (8 mg/kg, every 4 weeks) starting July 2003 in a clinical trial. She responded with a decrease in her clinical disease activity index (CDAI; from 148 in July to 93 in September to 35 in December of 2003), and her CRP remained less than 0.02 mg/dl. In January 2004, she had gastroenteritis with fever (38.4 C) and increased CRP (4.16 mg/dl). In February 2004, she complained of mild sore throat and slight fever (37.2 C). We suspected the common cold, but influenza A antigen in a pharyngeal swab tested positive by immunological detection. Although CRP was not elevated (0.05 mg/dl), neutrophil CD64 was increased to 5500 molecules/cell (normal range \2000 [5]). Thereafter, she suffered bronchitis and upper respiratory tract infection without fever or increased CRP, but again with upregulated CD64 (approximately 2800 molecules/cell). Afterwards, despite repeatedly suffering different infections (upper respiratory tract infection, bronchitis, gastroenteritis), she remained afebrile without CRP elevation except for the periods when the tocilizumab treatment was interrupted between clinical trials (CDAI flared up to 133) and for sinusitis surgery. Both of these occasions were accompanied by transient RA flares. CD64 levels remained at 400–1000 molecules/cell during periods without infection, but each infectious episode led to [50% elevation from the baseline, although it did not necessarily exceed the normal upper limit. In this RA patient, tocilizumab masked fever and elevation of CRP in all infectious episodes except the first gastroenteritis. When treating patients with this agent, we must keep in mind its possible masking of the signs of infection, which could lead to a delay in diagnosis. Monitoring neutrophil CD64 may facilitate the early detection of infection by using the definition of ‘‘CD64 T. Matsui (&) A. Komiya K. Shimada H. Nakayama S. Tohma Department of Rheumatology, Sagamihara National Hospital, National Hospital Organization, 18-1 Sakuradai, Sagamihara, Kanagawa 228-8522, Japan e-mail: t-matsui@sagamihara-hosp.gr.jp