Treatment targets, real-world outcomes and optimisation of biologic therapies for the treatment of inflammatory bowel disease

BACKGROUND: Treatment targets in inflammatory bowel disease (IBD) have evolved from treating symptoms to mucosal healing (MH), with the goal of altering the natural history of the disease. The biomarker faecal calprotectin (FC) correlates well with MH and has been established as a treatment target in IBD. However, whether aiming for, early, normalisation of FC, results in improved long-term outcomes remains unknown. The anti-tumour necrosis factor alpha (anti-TNF) drugs, infliximab (IFX) and adalimumab (ADA), have revolutionised the treatment of IBD and allowed patients to achieve the objective end-points of normalisation of FC and MH. However, anti-TNF therapy is expensive. Now biosimilar versions of these drugs are available with significant cost reductions. Therefore, it is important to assess the suitability of switching patients on to biosimilar versions to allow for cost savings to be incurred. Furthermore, in the advent of new biologic therapies for IBD, like vedolizumab (VDZ) and ustekinumab (UST), which have different mechanisms of action, it is important to establish how these therapies perform in the real-world setting, especially at achieving objective endpoints. Although, many respond to these biologic therapies, a significant portion do not or lose response with time. This is partly mediated by low drug levels in the presence or absence of antibodies. The concept of performing therapeutic drug monitoring (TDM) to help optimise therapy is evolving. Establishing therapeutic thresholds for biologic drugs is therefore important to help guide therapy and improve outcomes. AIMS: This thesis aims to use prospective, cross-sectional, and retrospective cohorts of patients with IBD to investigate the following: 1. The prognostic ability of FC as an early treatment target in CD 2. The real-world outcomes of switching CD patients from originator to biosimilar CT-P13. 3. The real-world outcomes of the biologics VDZ and UST 4. The association of drug levels with objective outcomes in IBD patients receiving anti-TNF and VDZ therapy METHODS: FC study: A retrospective cohort study was performed utilising medical records from incident cases of CD (2005-2017). The last FC measurement within 12-months of diagnosis was used to determine early normalisation (cut-off <250 μg/g). The primary end-point was time to first disease progression (composite of progression in Montreal disease behaviour; CD-related surgery; or CD-related hospitalisation). CT-P13 switch study: A prospective switch study was performed at the Edinburgh IBD Unit. Clinical and biochemical parameters (Harvey Bradshaw Index [HBI], Creactive protein [CRP] and FC) plus serum for IFX levels were collected prior to patients final Remicade infusion and at 6- and 12-months after switching to CT-P13. All adverse events during follow up were recorded. Real-world outcome studies for VDZ and UST: Two large retrospective studies were performed across several NHS health boards in Scotland. Clinical outcomes were collected from patients receiving VDZ (UC and CD) and UST (CD only) by review of routine medical records. TDM studies: Two cross-sectional studies were performed. HBI, CRP, drug levels and FC were collected on patients receiving ADA (CD only) or VDZ therapy (CD and UC). Biologic remission was defined as: CRP <5 mg/L and FC <250 μg/g. For perianal CD, a retrospective review of medical records was performed to find anti-TNF TDM samples that were matched with assessments of fistula healing / closure. RESULTS: FC Study: A total of 375 patients out of 1368 incident cases were included (median follow up 5.3 years). Patients with normalised levels of FC had a significantly lower risk of composite disease progression (hazard ratio, 0.36; 95% CI, 0.24–0.53; p<.01). CT-P13 switch study: No significant difference was observed between HBI (p=0.07), CRP (p=0.13), FC (p=0.25) and trough IFX levels (p=0.47) comparing before and at 6- and 12-months after the switch to CT-P13 (n=110). Rate of serious adverse events was 13.5 per 100 patient years of follow-up (PYF). Real-world outcome studies in VDZ and UST: In UC (n=180), 12-month cumulative rates of clinical remission, MH, and deep remission on VDZ were 57.4%, 47.3% and 38.5% respectively. In CD (n=260), 12-month cumulative rates of clinical remission, MH, and deep remission on VDZ were 58.4%, 38.9% and 28.3% respectively. In the UST CD cohort (n=216 CD), 12-month cumulative rates of clinical remission, MH, and deep remission were 32.0%, 32.7%, and 19.3%, respectively. The serious adverse event rate was 15.6 per 100 PYF and 13.6 per 100 PYF for the VDZ and UST cohorts, respectively. TDM studies: For ADA (n=152 CD patients), patients in biologic remission had significantly higher ADA levels compared to others (12.0 vs 8.0 μg/mL, p<0.01). An optimum ADA level of >8.5 μg/mL was identified for predicting biologic remission. For VDZ, (n=30 UC; 43 CD), no difference was observed in median VDZ levels between patients in and not in biologic remission (10.6 vs. 9.8 μg/mL, p=0.35). For the anti-TNF levels in perianal disease (ADA, n=35; IFX, n=29), patients with fistula healing had higher levels of anti-TNF versus those without fistula healing (ADA: 12.6 vs 2.7 μg/mL, p<0.01; IFX: 8.1 vs 3.2 μg/mL, p<0.01). For ADA, an optimum level of >6.8 μg/mL and >9.8 μg/mL was identified for fistula healing and closure, respectively. For IFX, an optimum trough level of >7.1 μg/mL was identified for both fistula healing and closure. CONCLUSIONS: Normalisation of FC within 12-months of diagnosis is associated with a reduced risk of progression in CD, supporting its use as an early treatment target. Transition to CTP13 from originator for the treatment of CD has no negative effect on outcomes at 12- months. Furthermore, the newer biologics VDZ and UST are effective treatments for achieving both clinical remission and MH in IBD. Finally, TDM can be utilised to optimise anti-TNF therapy, whilst the role of TDM in the context of VDZ therapy is likely limited.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.