Search in the Abstract Database

Abstracts Search 2017

* = Presenting author

P437 Autologous hematopoietic stem cell transplantation for refractory Crohn's disease: predictive factors of relapse

Lόpez García A.*1, Rovira M.2, Jauregui Amezaga A.1, Marin P.3, Barastegui R.1, Salas A.1, Feu F.1, Elizalde J.1, Fernández-Avilés F.2, Martínez C.4, Gutierrez G.2, Rosiñol L.2, Carreras E.2, Urbano A.2, Lozano M.3, Cid J.3, Suárez-Lledό M.2, Masamunt M.C.1, Comas D.1, Giner A.1, Gallego M.1, Alfaro I.1, Ordás I.1, Panés J.1, Ricart E.1

1Hospital Clinic Barcelona, Gastroenterology, Barcelona, Spain 2Hospital Clinic Barcelona, Hematology, Barcelona, Spain 3Hospital Clinic Barcelona, Biomedic Diagnostic Center, Barcelona, Spain 4Hospital Clinic Barcelona, Gynecology, Barcelona, Spain

Background

Autologous hematopoietic stem cell transplantation (HSCT) is considered as an option for patients with severe refractory Crohn's disease (CD). Although >80% of patients achieve remission early after HSCT, a proportion of relapses occur over time. The aim of the study was to establish predictive factors of relapse after HSCT.

Methods

Patients with refractory CD were prospectively enrolled between 2007–2016. Patients underwent monthly evaluations of clinical status and biomarkers. Endoscopic and/or magnetic resonance (MR) were performed at years 0.5, 1, 2 and 4, or when relapse was clinically suspected. Relapse was defined by presence of ulcers at endoscopy and/or MR. Variables analyzed included demographic factors, disease characteristics, and transplant related events. Univariate analysis was used to explore statistically relevant variables. The best predictive models were obtained by using a model for all possible equations.

Results

Thirty patients were included (70% F, mean age 28.6) with a median follow up (FU) of 80 weeks (27/30 patients with at least one year of FU). Relapse occurred in 19/30 patients (63%), 12 of them during the first year after HSCT. Variables with statistical significance at univariate analysis for relapse during the first year were: perianal disease (p=0.01), Erythrocyte Sedimentation Rate (ESR) (p=0.04) and C-Reactive Protein (p=0.05). The univariate analysis considering 5 years of FU included: albumin (p=0.01), ESR (p=0.01), number of days of neutropenia after conditioning (p=0.01) and perianal disease (p=0.05). The table shows the best predictive models for relapse

Table 1. Best predictive models for relapse.

Conclusion

Patients with perianal disease, ileal and/or upper disease, and short time of neutropenia after conditioning are at higher risk of relapse after HSCT. Early reintroduction of conventional therapy after HSCT should be considered in this subset of patients and explored in future trials.