This study is part of the ongoing prospective transregional study Advanced Study of Inflammatory Bowel disease (ASIB- study). All patients included in this follow-up were previously treated to remission with anti-TNF, subsequent discontinuation, and retreatment in case of relapse. In the first report of clinical outcomes in UC patients following this algorithm, 116 patients were included. Of these 116 patients, 96 patients obtained remission and observed until August 2015 . The patients included and follow-up are described in Fig. 1. In addition, 24 healthy controls (HC) and 9 patients with UC that relapses within 12 months after discontinuation of IFX were also recruited (early relapse group). The HC and the early relapse, were included to measure mucosal gene transcripts to compare with patients from the follow up with > 3 years in remission to find potential predictors of remission.
Categorization of long-term outcome after stop of anti-TNF
Patients were divided into predefined groups termed “Relapse” or “Remission”. The “Relapse” group was further divided into subgroups determined by the highest treatment level needed to obtain remission in the last 3 years; Biological therapy, colectomy or non-biological therapy including; corticosteroids, immunomodulators, 5-ASA. Corticosteroid use was short term courses when needed. No patients received integrin blockers or JAK inhibitors. Patients in the Remission group, had no relapse during the study period following discontinuation of the last anti TNF treatment. The remission group was divided into two subgroups, long-term remission without any treatment other than 5-ASA (LTR) and with immunomodulating drugs (LTR + imids). The definition of LTR was minimum 3 years in clinical remission after stopping/discontinuing anti-TNF. Clinical remission was defined by a combination of global assessment using ulcerative colitis Clinical Score (UCCS) less than 3 and faecal calprotectin < 100 mg/kg. Patients in LTR were invited to a follow-up endoscopy and clinical evaluation with UCDAI score.
Criteria for discontinuation of anti-TNF treatment
The criteria of discontinuation of anti-TNF treatment were endoscopic remission with a Mayo endoscopic sub-score of 0–1, until these 4 years extended observation time whereas an additional criterium of clinical remission > 6 months and normalized mucosal TNF gene expression were included in 2014 [24, 25].
Relapse was defined as a clinical, biochemical, endoscopic signs of disease activity leading to a therapeutic intervention as escalation of medical therapy or surgery.
Healthy control group
We included 24 healthy controls, 8 and 16 females and males respectively, with median (25–75 percentile) of 57 (42–67) years. The patients were recruited in the time period March 2014 to March 2018. The healthy control included patients referred for cancer screening where colonoscopy was normal. Exclusion criteria were serious medical conditions including immunological disorders, irritable bowel symptoms, polyps or cancer and abnormal histology in colonic biopsies.
We performed tissue sampling by using standard forceps and retrieving two mucosal biopsies in witch we performed all cytokine measurements. Colonic mucosal biopsies were sampled from the region with most severe inflammation in patients with active inflammation. In patients in remission, biopsies were sampled from the previously most inflamed region. Biopsies were sampled from the sigmoid in healthy controls. The biopsies were histologically assessed by an experienced pathologist (SWS) using Robarts histopathology index (RHI) score . The samples for RNA extraction were immediately immersed in RNA later (Qiagen), and stored at room temperature overnight, then at – 20 °C until RNA isolation.
Real-time PCR procedures have previously been described in detail [25, 27, 28]. Total RNA was isolated from mucosal biopsies using the Allprep DNA/RNA Mini Kit (Qiagen, Hilden, Germany, Cat No: 80204) and the automated QIAcube instrument (Qiagen, Hilden, Germany) according to the manufacturer’s recommendations. Quantity and purity of the extracted RNA were determined using the Qubit 3 Fluorometer (Cat No: Q33216; Invitrogen by Thermo Fisher Scientific, Waltham, MA, USA). Reverse transcription of the total RNA was performed using the QuantiTect Reverse Transcription Kit (Cat. No: 205314; Qiagen, Hilden, Germany) and QuantiNova probe RT-PR kit (Quiagen Cat no:208352). Primer sequences are previous published [29, 30]. The following gene transcripts were measured: IL17a, IL23, IL4, IL5, IL6, IL10, IL21, IL33, IL1B, TGFβ, GATA, IL18, TLR4, IL1RL1, RORC, FOXp3, TBX21 and TNF.
Two-way ANCOVA models adjusted for sex, disease distribution and age were performed to compare cytokine levels between groups and estimate effect size. For cytokines with a P-value of < 0.05, the difference in ΔΔCT fold change were calculated and converted to fold difference indicating up or down regulated in LTR compared to HC and early relapse group. To evaluate predictors of surgery, biological free remission and remission without relapse, Cox regression analyses were performed and receiver operating characteristics (ROC) curves were constructed. All statistical analyses were carried out in IBM SPSS Statistics 24 (IBM Corporation, Armonk, New York, USA).