P221 Evolution of serological markers and their predictive value before and after right hemicolectomy with ileocolonic anastomosis in patients with Crohn's disease
M. Noben*1, A. de Buck van Overstraeten2, S. Lockton3, B. Verstockt1, G. De Hertogh4, F. Princen3, A. Wolthuis2, G. Van Assche1, S. Vermeire1, S. Singh3, A. D'Hoore2, M. Ferrante1
1University Hospitals Leuven, Department of Gastroenterology, Leuven, Belgium, 2University Hospitals Leuven, Department of Abdominal Surgery, Leuven, Belgium, 3Prometheus Laboratories, Department of Research and Development, San Diego, United States, 4University Hospitals Leuven, Department of Pathology, Leuven, Belgium
Preventing postoperative endoscopic (ER) and clinical recurrence (CR) remains challenging in patients with Crohn's disease (CD) undergoing an intestinal resection. We previously identified a pre-operative risk panel (including smoking behaviour, Fla2, and pANCA) which may guide postoperative prophylactic therapy. We aimed to evaluate the post-operative evolution of serological markers and its association with both ER and CR.
The study population consisted of 71 consecutive patients (33 males, 20 active smokers, median age 42.7 years) undergoing an ileal resection with ileocolonic anastomosis for refractory CD, in whom a serum sample was available both within 1 week prior to surgery and 6 months thereafter. ER was defined as an endoscopic recurrence score of i3 or i4 at month 6. Sera were analysed blindly at Prometheus laboratories Inc. for the expression of anti-Saccharomyces cerevisiae IgA and IgG antibodies, three different anti-flagellin antibodies (CBir1, Fla2 and FlaX), antibodies to the outer-membrane porin C of Escherichia coli (OmpC), and atypical perinuclear antineutrophilic cytoplasmic antibodies (pANCA). The Q3 value of each individual marker was defined as the cut-off point.
Post-operative evolution of serological markers
|Pre-operative, median (IQR)||Post-operative, median (IQR)||Wilcoxon signed rank p-value|
|ASCA IgA||34.05 (9.56–89.34)||21.39 (11.59–73.18)||p<0.001|
|ASCA IgG||29.07 (11.10–63.86)||25.22 (12.00–35.91)||p<0.001|
|CBir1||30.31 (16.78–89.28)||24.91 (16.37–49.60)||p<0.001|
|Fla2||31.01 (16.04–66.79)||42.54 (19.61–55.95)||p=0.363|
|FlaX||52.49 (25.63–95.80)||51.47 (23.49–73.87)||p=0.001|
|OmpC||6.58 (2.63–13.17)||12.66 (9.29–22.38)||p<0.001|
At month 6, ER and CR were observed in 20 (28%) and 12 (17%) patients, respectively. During a median (IQR) follow-up of 26.8 (18.1-39.2) months, 24 (34%) of patients developed CR. As shown in Table 1, a significant decrease of ASCA IgA and IgG, CBir1 and FlaX was demonstrated post-operatively, while a significant increase was noted for OmpC. The absolute and relative post-operative changes of these markers were not associated with ER or CR. However, active smoking, ASCA IgA > 72 EU, OmpC > 23 EU and positive pANCA (n=15) at month 6, were associated with ER. In multivariate analysis, OmpC > 23 EU was associated with ER [Odds ratio 4.398 (1.379-14.028), p=0.012]. In 59 patients without CR at month 6, Cox regression multivariate analysis, revealed that both ER [7.926 (2.256-27.848), p=0.001] and pANCA [4.741 (1.378-16.313), p=0.014] were independent predictors of long-term CR.
In contrast to most previous findings, we observed a clear evolution of serological markers in the postoperative phase. This observation is probably reflecting a changing microbial environment. At 6 months, OmpC antibodies were independently associated with postoperative ER. Interestingly, not only ER but also presence of pANCA at month 6 was an independent predictor of long-term CR. Validation of these Results in an independent cohort is warranted.