P446 Characterization of ulcerative colitis patients in the Golimumab PURSUIT-Maintenance study: post-hoc analyses of patients who maintained and did not maintain clinical response through week 54

Abstract

Background: Assess clinical characteristics, serum GLM concentrations (conc), & immunogenicity to GLM of moderate-severe UC pts who maintained clinical response (MCR) & did not maintain clinical response (nMCR) to GLM through wk54 of PURSUIT. Methods: In the PURSUIT program, 456 GLM induction responders were randomized at wk0 of maintenance (maint) to PBO (n=154), GLM 50mg (n=151), & GLM 100mg (n=151) q4wks maint treatment through wk52. Partial Mayo scores (pMS) were assessed q4wks; pts with an increase from baseline (BL) (Wk0 of maint) in pMS ≥2 with an absolute pMS ≥4 or an absolute pMS ≥7 were considered in clinical flare & required to undergo endoscopy to confirm if pt lost response (i.e. no longer demonstrated a decrease from wk0 of induction of >30% & >3 points in the Mayo score with a decrease in the rectal bleeding score >1 or a rectal bleeding score of 0 or 1). In this post-hoc analysis, the clinical characteristics (at BL & wk6 of induction), serum GLM conc, & anti-GLM antibody status (during maint) of nMCR & MCR pts were compared. Results: There were no significant differences in pt demographics or the majority of UC disease characteristics (e.g. Mayo score or extent of disease) between nMCR & MCR pts. However, among nMCR pts, median conc of fecal inflammatory markers fCal & fLac were significantly higher at induction BL & wk6 following induction GLM therapy (Table 1). No significant difference was seen for median conc of CRP. At induction BL, the proportion of pts with elevated fCal (>250 mg/kg) & CRP (>8 mg/L) was significantly higher for nMCR pts vs MCR pts, 82.4% vs 71.4% (p=0.03) & 36.7% vs 24.8% (p=0.029), respectively. Among pts with elevated fCal & CRP at BL, the proportion of pts who normalized at wk6 of induction was not significantly different between nMCR & MCR pts. Median serum GLM levels during maint were significantly lower for nMCR pts vs MCR pts (p<0.05);no significant differences were observed during induction. The incidence of pts positive for anti-GLM antibodies was low overall; among nMCR pts, 7 (4.7%) were positive for anti-GLM antibodies vs 2 (1.4%) of MCR pts. [Table Presented] Conclusions: Pts who did not maintain clinical response to GLM therapy had higher median fCal & fLac levels prior to & following GLM induction treatment. Median CRP conc did not distinguish these pts; however, there was a higher proportion of pts with elevated CRP conc at BL. Other UC disease characteristics did not distinguish between nMCR & MCR pts. During maint treatment, serum GLM conc were lower among pts who did not maintain clinical response compared to those who did.