Drug survival of ixekizumab, TNF inhibitors, and other IL‐17 inhibitors in real‐world patients with psoriasis: The Corrona Psoriasis Registry

Abstract To compare drug survival of ixekizumab to other IL‐17 inhibitors (IL‐17i) and TNF inhibitors (TNFi) among patients with psoriasis (PsO) in a real‐world setting. Participants included adult PsO patients enrolled in the Corrona Psoriasis Registry who initiated ixekizumab, TNFi, or other IL‐17i between 16 March 2016 to 10 August 2019 and completed ≥1 follow‐up visit. Multivariable adjusted hazard ratios (HR) were calculated to estimate the risk for drug discontinuation in the ixekizumab group relative to the other drugs. Among the 1604 drug initiations, 552 initiated ixekizumab, 450 initiated TNFi, and 602 initiated other IL‐17i. Mean age was 51 years, 49% were women, and 52% were obese (BMI > 30). Ixekizumab patients had a higher proportion of patients with PASI >12 at drug initiation (24%) than TNFi (15%) and other IL‐17i (19%). Over a median of 11 months of follow‐up, 723/1604 (45%) drug discontinuations occurred. Persistence of ixekizumab, TNFi, and other IL‐17i at 24‐months were 68%, 33%, and 46%, among biologic‐naïve patients (n = 543), and 46%, 23%, and 36%, for biologic‐experienced patients (n = 1061), respectively. Ixekizumab patients had a 64% lower risk of discontinuation vs TNFi (HR = 0.36; 95% CI 0.27‐0.47) and a 31% lower risk vs other IL‐17i (HR = 0.69, 95% CI 0.55‐0.87) after adjustment for biologic experience and other covariates. HRs were similar when limited to patients with moderate‐to‐severe PsO (BSA > 3, PASI > 3, and IGA > 1, n = 1076) at initiation. In our study of real‐world patients with PsO, initiators of ixekizumab had more prolonged drug survival than both initiators of TNFi and other IL‐17i up to 2 years of follow‐up.


Abstract
To compare drug survival of ixekizumab to other IL-17 inhibitors (IL-17i) and TNF inhibitors (TNFi) among patients with psoriasis (PsO) in a real-world setting. Participants included adult PsO patients enrolled in the Corrona Psoriasis Registry who initiated ixekizumab, TNFi, or other IL-17i between 16 March 2016 to 10 August 2019 and completed ≥1 follow-up visit. Multivariable adjusted hazard ratios (HR) were calculated to estimate the risk for drug discontinuation in the ixekizumab group relative to the other drugs. Among the 1604 drug initiations, 552 initiated ixekizumab, 450 initiated TNFi, and 602 initiated other IL-17i. Mean age was 51 years, 49% were women, and 52% were obese (BMI > 30). Ixekizumab patients had a higher proportion of patients with PASI >12 at drug initiation (24%) than TNFi (15%) and other IL-17i (19%). Over a median of 11 months of follow-up, 723/1604 (45%) drug discontinuations occurred. Persistence of ixekizumab, TNFi, and other IL-17i at 24-months were 68%, 33%, and 46%, among biologic-naïve patients (n = 543), and 46%, 23%, and 36%, for biologic-experienced patients (n = 1061), respectively. Ixekizumab patients had a 64% lower risk of discontinuation vs TNFi (HR = 0.36; 95% CI 0.27-0.47) and a 31% lower risk vs other IL-17i (HR = 0.69, 95% CI 0.55-0.87) after adjustment for biologic experience and other covariates. HRs were similar when limited to patients with moderate-to-severe PsO (BSA > 3, PASI > 3, and IGA > 1, n = 1076) at initiation. In our study of real-world patients with PsO, initiators of ixekizumab had more prolonged drug survival than both initiators of TNFi and other IL-17i up to 2 years of follow-up. Psoriasis negatively impacts quality of life and work productivity, [1][2][3] and is associated with several other comorbidities, including cardiovascular disease, Crohn's disease, depression, and anxiety. 4 Traditional therapies for PsO include topical corticosteroids, retinoids, ultraviolet B (UVB), narrowband UVB, and systemic therapies such as methotrexate and ciclosporin. The development of highly effective biologic treatments such as IL-17 antagonists and IL-12/23 antagonists has transformed the treatment of PsO, 5 offering dermatologists several therapeutic options for their patients with moderate-tosevere PsO in addition to the well-known tumor necrosis factor (TNF)-α inhibitors. Choice of optimal treatment for PsO can depend on several factors, and it is strongly driven by a drug's efficacy and safety profile. Randomized clinical trials (RCT) are still considered the best source of data when assessing such features, but they are mostly designed to address regulatory purposes under ideal and controlled circumstances in a selected patient population. The real-life setting is a much more complex environment, where the actual drug utilization can be biased by many factors depending on the treatment strategy adopted by the clinician, the drug, and the patient in a specific health care system. Therefore, describing the performance of any particular treatment in real-life is a more appropriate tactic to measure effectiveness. Drug survival has been extensively considered as a surrogate of drug effectiveness in real life. It is the duration of time a patient remains on the same drug. It can be impacted by several factors such as drug factors (ie, efficacy and safety profile, tolerability), health care system factors and access to the therapy, and factors related to the patient preference and adherence to the clinician prescription. 6 Ixekizumab, a high binding affinity monoclonal antibody that selectively targets IL-17A, approved for the treatment of moderateto-severe PsO in adults, has been recently approved for moderate-tosevere pediatric PsO patients (ages 6 to under 18). In rheumatology, ixekizumab has been approved for active psoriatic arthritis, ankylosing spondylitis, and nonradiographic axial spondyloarthritis (nr-axSpA). 7 Ixekizumab has demonstrated superior efficacy to placebo, etanercept, ustekinumab, and guselkumab in several large, randomized head-to-head clinical trials. [8][9][10] Furthermore, the safety profile of ixekizumab has demonstrated to be significantly consistent with up to 5 years of continuous treatment and across all PsO trials. [11][12][13][14][15][16][17][18][19][20] In a recent study of long-term safety data in 13 ixekizumab clinical trials, Genovese et al. 21

| Study population
The study population included 1604 patient drug initiations

| Outcome
The primary outcome was drug survival, defined as persistent use of drug during follow-up, allowing for a medication gap of ≤60 days. Drug survival time was defined from the date of drug initiation through the date of drug discontinuation. Patients persistent on drug at their last follow-up visit were censored on the date of their last follow-up visit.

| Statistical analysis
Baseline characteristics were summarized descriptively for the treatment groups (ixekizumab, TNFi, non-ixekizumab IL-17i). Pairwise standardized differences were used to assess differences in characteristics between the treatment groups. A standardized difference effect size is equivalent to a Z score of a standard normal distribution and a threshold greater than 0.1 was the criteria used to represent a relevant imbalance between groups. Additionally, thresholds of 0.2, 0.5, and 0.8 are often used to indicate small, medium, and large effect sizes. 27 Kaplan-Meier (KM) methods were used to estimate the probability of drug survival over time for each treatment group, separately for biologic-naïve and biologic-experienced patients. Cox proportional hazard models, with a shared frailty term assumed to follow a gamma distribution to account for within-patient correlation were used to estimate adjusted hazard ratios (HR) for drug discontinuation in the ixekizumab group relative to the TNFi and non-ixekizumab IL-17i groups. 28,29 Hazard ratios of less than 1 indicate a lower risk of discon- Note: Standardized differences >0.10 are suggestive of an imbalance between groups and thresholds of 0.2, 0.5, and 0.8 are often used to indicate small, medium, and large effect sizes (Cohen, 1988). Abbreviations: IXE, ixekizumab; DLQI, Dermatology Life Quality Index; TNFi (adalimumab, certolizumab, etanercept); non-ixekizumab IL-17i (non-IXE) (secukinumab, brodalumab); VAS, visual analogue scale; WPAI, Work Productivity and Activity Impairment Questionnaire (Absenteeism = % work hours missed, Presenteeism = % impairment while working, Work productivity loss = % impairment while working, and Activity impairment = % daily activities affected by PsO); EQ-5D-3L (EuroQol) consists of 2 pages: the EQ-5D descriptive system and the EQ visual analogue scale (EQ-VAS). a Not mutually exclusive. PASI > 3, and IGA > 1). The proportional hazards assumption was evaluated for the models by examining Schoenfeld residuals 30 and was met for drug group. However, there was evidence to suggest deviation from proportional hazards for some covariates (eg, weight, history of infection). Sensitivity analyses were conducted using timedependent coefficients for such covariates, which is one option for handling nonproportional hazards. 31
In unadjusted Cox regression analyses, the ixekizumab group had  Table 4, Tables S1 and S2).
In sensitivity analyses, estimates of drug survival and hazard ratios were similar within the subgroup of patients with moderate-to-severe disease activity at baseline (Tables 3 and 4 and a newer non-ixekizumab IL-17i such as secukinumab 25  measures to assess the drug performance in real patients and settings, and these should become important indicators when choosing the appropriate treatment for the best patient outcomes.
Our study is not without limitations. The data source is a North American Registry, and the results may not be generalizable to other regions. The data are subject to limitations inherent in all observational studies, such as the lack of randomization to patients for different biologics and the unknown patient factors that may be associated with access to care. Estimation of drug survival between treatments should also be cautiously interpreted when using nonrandomized data. Nevertheless, our study has important strengths. Data for the current study were collected across the United States and Canada from both academic and private practice dermatologists, and these patients are more likely to reflect the typical real-world patient population than those in clinical trials. The duration and 2-year follow up of the current study were reasonable and clinically meaningful. We were able to adjust for several potential confounders, including baseline disease severity, clinical outcomes, and treatment patterns, which are not captured in claims databases.

| CONCLUSIONS
The current study from a large cohort of North American PsO patients demonstrates that over 2 years ixekizumab has a better drug survival compared with TNFi and non-ixekizumab IL-17i. The findings were consistent in both biologic-naïve and biologic-experienced patients.
The current study data offer valuable evidence regarding drug performance and real-life differences, which may be taken into clinicians' consideration when choosing between biologics.

ACKNOWLEDGMENTS
This study was sponsored by Corrona, LLC and the analysis was funded by Eli Lilly and Company. Access to study data was limited to