It is traditionally held that human IgG4 MoAbs should not deplete target cells in vivo, as this isotype is inactive in a number of in vitro assays that measure effector function. We have previously challenged this dogma, and the current study was designed to investigate the in vivo biological effects in humans of a MoAb of human IgG4 isotype. Nine patients with refractory rheumatoid arthritis (RA) fulfilling ARA criteria, and one with ankylosing spondylitis (AS) received a human IgG4 Campath‐1 MoAb (with specificity against the pan‐lymphocyte antigen CD52) as part of a two‐stage therapeutic protocol. In stage 1, patients received a single dose of this MoAb. Stage 2, starting 48 h later, comprised a 5‐day course of a human IgG1 Campath‐1 MoAb with identical V‐region (CAMPATH‐1H), as previously used in the management of RA patients. The intervening 48 h provided a window of opportunity to monitor the biological effects of the IgG4 MoAb for comparison with the IgG1. The two MoAbs were also compared for in vitro biological activity. IgG4 depleted peripheral blood lymphocytes (PBL), albeit less efficiently than IgG1. It produced a first‐dose reaction of similar intensity, although associated circulating tumour necrosis factor‐alpha (TNF‐α) levels were lower. TNF‐α release from whole blood in vitro was also greater with the IgG1 MoAb. The study design did not permit conclusions to be drawn regarding therapeutic efficacy of the IgG4 MoAb. In summary, a human IgG4 Campath‐1 MoAb depletes target cells in vivo. Importantly, this study demonstrates for the first time in humans that in vitro assays may not predict the in vivo effector function of therapeutic MoAbs.