Once the clinical symptoms developed the treatment is just reducing the severity of diesease. Therefore you might not see what you are looking for. See the data from  the companies FDA approval.

"Animal models of autoimmunity and inflammation including arthritis were used to demonstrate the activity of TNF:Fc. The collagen induced-afthritis model in mice was used in 2 mouse strains with three sources of collagen (bovine, porcine, and chicken) to demonstrate biological activity. In the bovine collagen induced model daily ip administration (50 ug/mouse) of the fusion protein reduced the incidence of disease (joint inflammation and tissue degradation) to 28% in comparison to 86% in saline treated controls. When treatment was delayed until the onset of disease, a less severe arthritis in mice given albumin was observed. The effects of TNF:Fc were dose dependent. TNFR:Fc was also effective in a rat model of antigen induced arthritis. In this model decreased joint swelling and less joint damage was observed in animals given an intra-articular injection of the TNF binding protein before or after antigen challenge (methylated BSA)."

Another possibility is that Enbrel is designed for human use and may not bind as well to mouse TNF-alpha, thus failing to fully prevent its inflammatory activity.

Enbrel does cross react with mouse and rat TNFa.  However, while in rat CIA we achieved 85% inhibition at 10mg/kg i.p., no inhibition was observed in mouse CIA at 3mg given i.p. twice per week.  So your data are in line with these mouse findings. 

As long as you have an Ab control for your therapeutic Ab's, you can use dexamethasone as your efficacy control for the experiment itself.  A dose of 0.3mg/kg p.o. will be very effective. Get the very soluble 21-phosphate disodium salt (D1159), you could also give it i.p.

Enbrel does work in mouse antigen-induced arthritis, but the doses needed  for similar efficacy are at least 5 to 10-fold higher than in the rat.  Hence you could try a dose of perhaps 50-100mg/kg Enbrel in the mouse CIA model.

Thank you for all your answers. We did use i.p dexamethasone in one group with excellent results, but the Enbrel group got just as sick as the control group (PBS i.p). Your answers clearly gives me a better understanding of why Enbrel didn't work. As an Ab-treatment control Enbrel will not be the best choice anyway, as some of you already mentioned, since it is not really a therapeutic Ab. Alternatively we could try increasing the Enbrel dose 10-fold, in addition to isotype control and dexametasone groups.