Serum Amyloid a as a Surrogate Marker for Mucosal and Histologic Inflammation in Patients With Inflammatory Bowel Disease

Abstract

Introduction: While the level of the C-reactive protein (CRP) has been used as a biomarker in the assessment of patients with inflammatory bowel diseases (IBD), its correlation with mucosal inflammation is poor in some patients. Serum amyloid A (SAA) is also an acute phase protein, but its role as a biomarker of disease activity in IBD is unclear. The aim of the study was to assess the correlation between SAA and mucosal/histological inflammation in patients with IBD and to investigate if this marker might be useful in patients that do not present with elevated CRP despite having active disease. We also sought to assess if other inflammatory cytokines can be used as surrogate markers of IBD activity. Methods: Cross-sectional study including Crohn's disease (CD) and ulcerative colitis (UC) patients that underwent colonoscopies for assessment of disease activity. Predictive variables included demographics, phenotype of disease, IBD medications and laboratories (SAA, CRP, intercellular adhesion molecule-1, vascular cell adhesion molecule-1, interleukins 8, 1β and 6). The primary outcome was the presence of active endoscopic or microscopic inflammation. Quartile analysis and receiving operating curves (ROC) were performed to evaluate the association between CRP, SAA and other cytokines with mucosal inflammation. Results: 94 patients were included (43% were women, 92% had CD). 59 (62.8%) had active endoscopic disease and 71 (75.5%) had active microscopic inflammation. All patients were receiving a biologic, 23% were on prednisone and 39% were on combination therapy (biologic plus immunomodulator). Differences between the groups with and without active disease are shown in Table 1. SAA was the test that best predicted mucosal inflammation (Table 2). When stratifying by location of disease, SAA predicted inflammation in both isolated small bowel disease (ROC: 0.8) and colonic disease (ROC=0.78). High SAA levels (≥2.4 mg/mL) identified 83% (19/23) and 81% (13/16) of those patients with microscopic and macroscopic inflammation but normal CRP ( < 4mg/mL).Table 1: Differences between those patients with and without inflammationTable 2: Performance of the studied biomarkers for predicting macroscopic or microscopic disease activityConclusion: High circulating SAA levels can accurately predict endoscopic and microscopic inflammation and may be used as a surrogate marker for disease activity, especially on those patients in which CRP levels do not correlate with their disease activity. Further studies looking into how SAA can predict response to treatment and its role in long-term disease monitoring are warranted.Figure 1