Introduction
Cyclical monoclonal antibody (mAB) therapy, defined as intermittent pausing of biological agents to reduce drug cost and risk of adverse events, has been proposed as a treatment strategy. There are few prospective discontinuation studies; most are uncontrolled, have small sample sizes, are retrospective, are short term, and do not directly compare multiple mABs. It is therefore unknown if treatment persistence is impaired following a drug holiday versus their initial trajectory. Whether persistence loss differs between mABs, is mitigated by immunomodulators or shorter drug holidays, is also unknown.
Aims & Methods
We aimed to evaluate the persistence of Crohn’s disease (CD) mABs with- and without drug holiday using a real world data registry. We interrogated the prospective population-based Persistence Australian National IBD Cohort (PANIC)5 registry of CD subjects up until Dec 2021. The PANIC registry includes the entire Australian population on advanced therapies. Drug holiday was defined as mAB discontinuation with recommencement of the same mAB after a >6-month dispensing gap. Non-persistence was defined as subsequent cessation of treatment for >6-months. Kaplan Meier survival curves were generated for persistence and compared using the log-rank test. Propensity score matched Cox proportional hazards regression models calculated hazard ratios (HR) and 95% confidence intervals (CI). P<0.05 was deemed statistically significant (SPSS, IBM, USA).
Results
In total 19,078 consecutive CD subjects (44.5% male, median age 39.0 years, 79,677 patient-years of prospective follow-up) were identified with 3,383 episodes of drug holiday (mean duration of 414 days) versus 28,584 episodes of continuous treatment as controls. Patients who resumed mAB therapy following a drug holiday had a statistically significant lower median persistence of 23 months versus 41 months for controls (-18 months; HR: 0.63, 95% CI: 0.60-0.66, P<0.001). TNF inhibitors (TNFi) and ustekinumab (UST) demonstrated significant loss of persistence following a drug holiday (both P<0.001) but not vedolizumab (P=0.63; Figure 1). Shorter versus longer median drug holiday duration did not impact persistence (P>0.05). Use of immunomodulator abrogated the loss of persistence of UST (P=0.15) but not TNFi (P<0.001). Propensity score matched independent predictors of non-persistence were drug holiday (HR: 0.72 [95%CI: 0:67-0.76]), TNFi, monotherapy, females, and bio-exposed status.
Conclusion
This study that recruited all CD subjects on advanced therapy in Australia showed pausing mAB therapy to be associated with a 37% decline in medication persistence for TNFi and UST, but not VED. Immunomodulator combination therapy mitigated the loss of persistence with UST but not TNFi. mAB drug holiday should be discouraged and closely monitored for disease relapse. These data do not support the cyclical use of mABs.
Disclosure
RL: advisory board: AbbVie, Aspen, BMS, Celgene, Celltrion, Chiesi, Ferring, Glutagen, Hospira, Janssen, Lilly, MSD, Novartis, Pfizer, Prometheus Biosciences, Takeda; research grants: Joanna Tiddy USYD, McCusker Charitable Foundation, Celltrion, Shire, Janssen, Takeda, Gastroenterological Society of Australia, NHMRC, Gutsy Group, Pfizer, MRFF
