Introduction
Comparative effectiveness of biologics in preventing penetrating disease (PD) complications in Crohn’s disease (CD) is unknown. We compared the risk of developing luminal and perianal PD (LPD and PPD) between biologics used as first-line therapies.
Aims & Methods
We performed a propensity-score adjusted comparative effectiveness study. Adults (≥18 years) with CD (≥3 administrative claims for CD [K50.*]) who initiated their first biologic (anti-tumor necrosis factor [anti-TNF], ustekinumab ([UST]) or vedolizumab [VDZ]) were identified from Merative™ commercial database between 2006 and 2020. LPD was determined by ICD claims and PPD by a validated case definition consisting of ICD claims and procedural codes. We excluded pre-existing PD and prior biologic exposure using a minimum look-back period of one year. Patients were followed until a PD event and censored if a new biologic class was initiated, loss of medical coverage or a study end. Cohorts were balanced by inverse probability of treatment weighting (IPTW) based on age, sex, comorbidities, prior CD surgery, and CD severity. Pairwise comparisons were performed by cox-proportional hazard models, adjusted for immunomodulator exposure, and with biologic exposure treated as a time dependent variable based on a medication possession ratio of 0.8.
Results
Our analysis included 40,693 patients: (93%) anti-TNF, (3%) UST and (4%) VDZ. Our cohort had a median age of 39 years (IQR, 28-50) and 54% were female. Penetrating complications were identified during follow-up in 8567 (21%) patients overall: 2,204 (5%) LPD and 7,474 (18%) PPD. After IPTW all comparisons were well balanced (standardized difference <0.1 for each pairwise comparisons). PPD was more common than LPD overall (Log-rank p<0.0001). Anti-TNF was protective against LPD (HR, 0.66; 95% CI, 0.55–0.78; p<0.0001) and PPD (HR, 0.88; 95% CI, 0.80–0.96; p=0.0045;) compared with VDZ and LPD (HR, 0.37; 95% CI, 0.30–0.46; p<0.0001;) compared with UST (Table 1). There were no significant differences in the risk of LPD and PPD between VDZ and UST. The protective association between anti-TNF therapy and PD was more pronounced when our study period was limited to after 2016 when all three treatments were commercially available.
| Table 1: Pairwise comparisons of biologic classes and risk of composite, luminal and perianal penetrating disease. |
|---|
| Composite* PD
aHR (95% CI)
| P value
| Luminal PD
aHR (95% CI)
| P value
| Perianal PD
aHR (95% CI)
| P value
|
Overall cohort
|
Anti-TNF vs VDZ
| 0.87 (0.80, 0.95)
| 0.001
| 0.66 (0.55, 0.78)
| <0.0001
| 0.88 (0.80, 0.96)
| 0.0045
|
Anti-TNF vs UST
| 0.90 (0.80, 1.02)
| 0.113
| 0.37 (0.30, 0.46)
| <0.0001
| 1.04 (0.90, 1.19)
| 0.6075
|
UST vs VDZ
| 1.03 (0.67, 1.59)
| 0.885
| 1.62 (0.83, 3.18)
| 0.1603
| 0.89 (0.55, 1.42)
| 0.6166
|
Study cohort (2016-2020)
|
Anti-TNF vs VDZ
| 0.66 (0.56, 0.76)
| <0.0001
| 0.50 (0.36, 0.68)
| <0.0001
| 0.67 (0.58, 0.79)
| <0.0001
|
Anti-TNF vs UST
| 0.70 (0.59, 0.84)
| 0.0001
| 0.28 (0.20, 0.39)
| <0.0001
| 0.84 (0.70, 1.02)
| 0.0827
|
UST vs VDZ
| 1.01 (0.65, 1.58)
| 0.949
| 1.57 (0.80, 3.09)
| 0.1897
| 0.85 (0.52, 1.40)
| 0.5327
|
Conclusion
Anti-TNF therapy was associated with a lower risk of LPD and PPD compared with VDZ and LPD compared with UST. Further studies are needed to validate these findings and to determine potential reasons for these differences.
Disclosure
JM: Consultancy fees and/or Honoraria: Janssen, Abbvie, Takeda, Pfizer. Research funding: BMS, Pfizer and Leo Pharma, Honoraria: Bayer, Pfizer, Sanofi, BMS and Leo Pharma.
PD: has received research support under a sponsored research agreement unrelated to the data in the paper and/or consulting from AbbVie, Arena Pharmaceuticals, Boehringer Ingelheim, Bristol Myers Squibb, Janssen, Pfizer, Prometheus Biosciences, Takeda Pharmaceuticals, Roche Genentech, Scipher Medicine, Fresenius Kabi, Teva Pharmaceuticals, Landos Pharmaceuticals, Iterative scopes and CorEvitas, LLC.
The other authors have no conflicts