There is overwhelming evidence for interleukin-1β (IL-1β ) having a role in the pathogenesis of rheumatoid arthritis (Dinarello, 2002), and the use of antibodies to neutralise mediators of disease, such as cytokines, has shown considerable utility in the clinic. .Supply of material for chronic administration in diseases such as RA is an important consideration, and production of antibody fragments through microbial expression and their pegylation offers potential advantages in th is respect. In order to compare the efficacy of a whole antibody versus a pegylated Fab ’ fragment (Fab’ PEG), reagents in both formats with comparable affinity for murine IL-1β (approximately 400pM) were generated and tested .

The aim was to examine the effects of a whole antibody and a Fab’ PEG specific for mouse IL-1β in murine collagen induced arthritis (CIA), a model widely used in the pre-clinical assessment of anti-arthritic drugs.

Male DBA/1 mice (approximately 22g), n=11-15/group , were sensitised to type II collagen in CFA (Day 0) and boosted with antigen on Day 14. Dosing commenced Day -1 and was weekly thereafter. The whole antibodies (anti IL-1β and control) and the Fab’ PEGs (anti IL-1β and control) were dosed at 10mg/kg s.c. and the experiments were terminated on Day 40. Limbs were scored daily from first signs of arthritis according to the following criteria: 0 normal, 1 wrist/ankle affected, 2 additional pad involvement, 3 additional digit involvement. Thus an animal could achieve a maximum score of 12. At the end of the studies, limbs were removed and fixed in 10% buffered formol saline. Tissues were processed for paraffin wax embedding. Sections were cut and stained with haematoxylin and eosin and assessed in a blinded manner according to the following criteria: 0 normal, 1 synovial thickening, 2 cartilage and bone erosion, 3 loss of joint architecture. An AUC disease score was calculated for each animal and these data were expressed as means ± s.e.m. for each group. Statistical analysis was by Student’s T. Analysis of the histopathological scores was by Mann Whitney U test. P<0.05 was considered statistically significant.

In the whole antibody study, 93% of animals in the control group had disease symptoms by Day 40 compared with 13% in the anti IL-1β group. When the clinical data were expressed as area under curve (AUC) , there was a 99.6% reduction (p<0.001) in AUC when the anti IL-1β group was compared with the control group. Histological examination of the joints indicated that the mean histopathological score for anti IL-1β treated animals was 66% lower (p<0.001) than for the control group. In the Fab’ PEG experiment, disease incidence was 93% in the control group by Day 40 and 13% in the anti IL-1β group. There was a 92% reduction (p<0.001) in AUC when the anti IL-1β group was compared to the control group. The histopathological score for the anti IL-1β group was 44% lower (p<0.01) than for the control group.

The data show a significant anti-arthritic effect of IL-1β blockade in murine CIA and shows the effectiveness of using a Fab’ PEG to target cytokines in this model.