P362 Agalactosyl IgG predicts therapeutic effect of anti-TNF agents in patients with Crohn's Disease
M. Araki*, S. Shinzaki, H. Iijima, T. Yamaguchi, S. Kawai, S. Hiyama, T. Inoue, Y. Hayashi, K. Watabe, M. Tsujii, E. Miyoshi1, T. Takehara
Osaka University Graduate School of Medicine, Department of Gastroenterology and Hepatology, 2-2 Yamadaoka, Suita, Japan, 1Department of Molecular Biochemistry and Clinical Investigation, Osaka University Graduate School of Medicine
Recent studies reported that clinical response to anti-TNF agents is associated with trough drug level and C-reactive protein (CRP) levels after treatment, however, biomarkers to predict therapeutic effect of anti-TNF agents before treatment are scarce. We previously reported that serum agalactosyl IgG level is an effective biomarker of disease activity in Crohn's disease (CD) (Shinzaki S, et al. Am J Gastroenterol 2008). In the present study, we investigated if agalactosyl IgG can predict long-term effect of anti-TNF agents.
Forty-six CD patients who met following criteria were enrolled; (1) scheduled maintenance treatment with anti-TNF agents (infliximab or adalimumab) was received, and (2) sera were collected within 1 month prior to administering initial treatment, and (3) followed-up more than 1 year after the initial treatment. Treatment change within 54 weeks was defined as follows; (1) increase in agents, or (2) shortening of treatment interval, or (3) switch to another anti-TNF agent, or (4) receiving surgery. Sustained clinical remission was defined as follows; (1) Crohn's disease activity index (CDAI) of < 150 at week 14, 38 and 54, and (2) no treatment change within 54weeks, and (3) CRP level of < 0.5 mg/dL at week 54. Serum IgG oligosaccharide structures were analyzed by high performance liquid chromatography and the ratio of agalactosyl peak and fully galactosyl peak in the fucosyl oligosaccharides (G0F/G2F) was calculated. G0F/G2F ratio of 1.4 was used as cut-off value to discriminate the G0F/G2F high and low group as previously reported. Endoscopic activity was assessed using modified Rutgeerts score.
Before starting anti-TNF agents, 32 patients were in high G0F/G2F levels. Patients with high G0F/G2F tended to show higher CDAI and endoscopic activity than those with low G0F/G2F before treatment (215.3 vs 156.3; p = 0.07, and 2.4 vs 2.1; p = 0.08, respectively). Patients with high G0F/G2F showed significantly higher CDAI and CRP levels at week 14 than those with low G0F/G2F (126.7 vs 83.1; p = 0.015, and 1.03 mg/dL vs 0.44 mg/dL; p = 0.029, respectively). CRP levels and endoscopic scores at week 54 were significantly higher in high G0F/G2F group than in low G0F/G2F group (1.60 mg/dL vs 0.57 mg/dL; p = 0.028, and 2.5 vs 1.1; p = 0.035). Treatment change rate was comparable in two groups (21.4% vs 25.0%; p = 0.65). Sustained clinical remission rate in high G0F/G2F group was significantly lower than that in low G0F/G2F group (18.8 % vs 50.0 %; p = 0.030).
Patients with high G0F/G2F showed higher disease activity and poor response to anti-TNF agents, suggesting that agalactosyl IgG (G0F/G2F) is a potential predictive biomarker of poor long-term response by anti-TNF agents.