Abstract
Objective Information on the persistence of tofacitinib (TOF) in psoriatic arthritis (PsA) is scarce in real-world conditions. Our objective was to analyze the persistence and safety of TOF under these conditions.
Methods This was a single-center retrospective longitudinal observational study of all patients with PsA who received at least 1 dose of TOF. The main focus was on adverse events (AEs) and drug survival. Drug survival was analyzed by Kaplan-Meier curves and persistence explanatory factors by multivariate Cox regression models. The hazard ratio (HR) was used to measure association.
Results Seventy-two patients were included, 54 women and 18 men, mean age 51.9 (SD 11.1) years, mean disease duration of 10.4 (SD 6.99) years. TOF was ≥ third line of therapy in > 70% of cases. The median survival was 13.0 (IQR 5.3-29.0) months. One-year retention rate was 52.7% (95% CI 42.4-65.6). TOF survival was not influenced by sex, disease duration, comorbidities, or line of treatment. Younger patients (HR 0.96, P = 0.01) and those with enthesitis (HR 0.37, P = 0.03) showed lower odds of drug discontinuation. The overall rate of AEs was 52.9 (95% CI 38.5-70.6)/100 person-years. Most AEs occurred during the first 6 months of exposure.
Conclusion In this real-world study, TOF showed a reasonably good retention rate in a PsA population that was mostly refractory to biologic and oral targeted synthetic disease-modifying antirheumatic drugs. There were no new causes for concern regarding safety. Patients with refractory PsA and enthesitis might be a specific target population for this drug.
- drug survival
- psoriatic arthritis
- safety
- targeted synthetic disease-modifying antirheumatic drugs
- tofacinitib
Psoriatic arthritis (PsA) is a chronic inflammatory joint condition that affects a nonnegligible proportion of the general population.1 Although the therapeutic options in PsA are wide-ranging, management strategies are not homogeneous, and treatment decisions and pathways for the use of the newest drugs are not always straightforward. The Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) has based its management recommendations on the most affected PsA domain, whereas the European Alliance of Associations for Rheumatology (EULAR) has opted for a more classic step-up scheme depending on whether the treatment objectives have been achieved.2,3
Tofacitinib (TOF) was the first Janus kinase inhibitor (JAKi) approved to treat PsA. Two well-designed phase III trials (Oral Psoriatic Arthritis Trial [OPAL] Broaden and OPAL Beyond) in patients with PsA with or without prior tumor necrosis factor inhibitor (TNFi) therapy showed that TOF 5 mg twice a day significantly improved the key clinical signs/symptoms and disability associated with PsA after 3 months of treatment, while also improving skin psoriasis, enthesitis, dactylitis, physical function, and fatigue.4,5 However, the conditions in which most randomized controlled trials (RCTs) are conducted bear little resemblance to the conditions that clinicians face on a daily basis when making therapeutic decisions. To date, there are very few studies that have explored the benefits and limitations of TOF in real-world conditions.6 Hence, it is essential to obtain information on the performance of TOF for the treatment of PsA in clinical practice. Our objective was to analyze the persistence and safety data of TOF in PsA under routine clinical conditions.
Ethics approval. This is a single-center retrospective longitudinal observational study. The ethics review board of Hospital Universitario Central de Asturias and Principality of Asturias (northern Spain) approved the study and exempted the participants from informed consent due to the retrospective nature of data collection (CEImPA no. 2023.246). This study was conducted in accordance with the Declaration of Helsinki, Good Clinical Practice Guidelines, and General Data Protection Regulation.
METHODS
Study population. We included all adult patients with PsA fulfilling the Classification Criteria for Psoriatic Arthritis (CASPAR) who received at least 1 dose of TOF from January 2020 to December 2022. TOF prescription was made in accordance with the local regulatory laws of the healthcare system of the Principality of Asturias. Briefly, in our setting, patients with PsA who require biologic therapy start with a biosimilar TNFi. If there is a contraindication for the use of these agents, or there is a lack of therapeutic response, a second line of treatment is opened, with a JAKi or interleukin (IL)-17 inhibitor as potential options, at the discretion of the prescribing physician.
Study outcomes. The primary outcomes were drug survival and safety. Safety was analyzed by reviewing the clinical charts from the date of initiation of TOF, as well as the hospital admission records. The following adverse events (AEs) were recorded: (1) infections (type, microorganism, location and whether it was accompanied by bacteremia); (2) neoplasms (type, location, and stage); and (3) events located or affecting any other organs or systems. Drug survival was defined as the time from the start of TOF to the last dose administered before discontinuation or loss to follow-up. The reason for and rate of suspension were also collected.
The following variables were collected: all previous therapies, age, weight, sex, disease duration, joint pattern, clinical enthesitis (for ultrasound confirmation, only the presence of thickened hypoechoic enthesis areas with clear power Doppler activity with/without erosions was accepted as a valid finding), dactylitis, smoking status, diabetes mellitus, hypertension, dyslipidemia, depression, chronic obstructive pulmonary disease, major adverse cardiovascular events (MACE), renal and hepatic insufficiency, biologic treatment line (1st line, 2nd, 3rd, 4th, etc.), previous AEs, previous disease-modifying antirheumatic drugs (DMARDs; yes/no and concomitant), and disease activity assessments (baseline and at 6 and 12 months).
Statistical analysis. The study sample was described in terms of the distribution of the descriptive variables by summary statistics. Drug survival was analyzed using Kaplan-Meier curves and the hazard ratio (HR) was used to measure association. Multivariate Cox regression was used to analyze the effect of the explanatory variables on drug survival. Potential confounding variables were age, disease duration, sex, previous DMARDs, smoking status, and depression. However, due to the relatively small sample size (n = 72), the number of potential correction variables (n = 6) could seem somewhat high. These variables were chosen based on their explanatory importance, collected from previous literature, and were introduced into the final regression model as long as their confidence intervals were reasonably narrow. The rate of AEs was estimated in total, by severity and by type of event. The denominator used was the total number of patients multiplied by years of follow-up. Level of statistical significance was set at P < 0.05. Statistical analysis was done using the software R 4.3.1 “Beagle Scouts” (R Foundation for Statistical Computing).
RESULTS
The study sample included 72 patients (54 women and 18 men), with a mean age of 51.9 (SD 11.1) years. The mean (SD) age at psoriasis onset was 36.2 (12.1) years, whereas the mean age at PsA onset was 45 (10.9) years. The Table shows the patient characteristics at the beginning of treatment with TOF and some variables collected during follow-up.
Disease characteristics of the study population.
In most patients (> 70%), TOF was the third or subsequent line of treatment (difficult-to-treat [D2T] PsA). The overall AE rate was 52.9 (95% CI 38.5-70.6)/100 person-years (PYs). Of the 45 patients who discontinued the drug, 24 (53.3%) did so due to loss or lack of effectiveness, 10 (22.2%) due to intolerance, and the rest (11, 24.4%) due to AEs during exposure. Most patients who discontinued due to AEs (66.7%) did so during the first 6 months of treatment. No MACE were detected during the exposure. AEs leading to discontinuation during drug exposure were Crohn disease (n = 1), cancer (n = 2), severe stomatitis (n = 1), herpes zoster (n = 2), severe headache (n = 2), severe abdominal pain (n = 2), and skin worsening (n = 1).
The median survival of TOF was 13.0 months (IQR 5.3-29.0). The overall retention rate at 1 and 2 years was 52.7% (95% CI 42.4-65.6) and 38.1% (95% CI 28-51.7), respectively (Figure 1). At 12 months, TOF survival was 66.7% (95% CI 11-48) in men and 48.1% (95% CI 6.8-36.5) in women. The resulting 18.6% difference in drug persistence between sexes was not statistically significant (log-rank P = 0.39). No differences in drug survival were found according to obesity (log-rank P = 0.34), smoking status (smokers vs nonsmokers, log-rank P = 0.29), line of treatment (log-rank P = 0.54), or cardiometabolic comorbidity (diabetes, hypertension, and dyslipidemia). Younger patients (HR 0.96, 95% CI 0.92-0.99, P = 0.01) and those with enthesitis (HR 0.37, 95% CI 0.15-0.92, P = 0.03) showed lower odds of TOF discontinuation (see overall Cox regression model in Supplementary Table S1, available with the online version of this article). The 1- and 2-year retention rates for patients with PsA-related enthesitis was the same: 66% (95% CI 45.7-95.4), whereas the 1- and 2-year retention rates for patients without enthesitis were 49.1% (95% CI 37.7-6) and 32.2% (95% CI 21.9-47.3), respectively (Figure 2). The TOF withdrawal rate in patients with enthesitis was 30.4 PYs (95% CI 11.1-66.1) vs 59.6 PYs (95% CI 42.4-81.5) in patients without enthesitis.
Kaplan-Meier curve of all-cause discontinuation of tofacitinib.
Kaplan-Meier survival curve stratified by enthesitis.
DISCUSSION
In this real-world evidence study, TOF demonstrated a reasonably good persistence rate in a population of PsA mostly refractory to various lines of biologic therapies and oral targeted synthetic drugs. A similar proportion of patients discontinued the drug due to lack or loss of effectiveness, as well as to intolerance or AEs during exposure. Most safety problems were detected during the first 6 months of treatment. The number of previous biologic DMARDs did not have any influence on drug survival. Neither sex nor the presence of cardiometabolic comorbidity (including obesity) or smoking affected drug survival. Interestingly, patients with clinical enthesitis (with ultrasound confirmation) showed the lowest risk of TOF discontinuation.
Despite the growing number of treatment options for PsA, the hierarchy of these therapies in treatment algorithms remains unclear.2,3,7,8 In general, the use of TNFi biosimilars seems to be a reasonable option in terms of first-line therapy for PsA. However, on many occasions, these agents present persistence and safety problems, which forces the treating clinician to choose among the different non-TNFi options available; this selection is not always easy.7,8 It is in this scenario that drug survival studies make sense since this type of study may respond to the uncertainties that arise in clinical routine and that are not contemplated in the majority of RCTs. For example, among the main factors related to reduced TNFi survival, different studies highlight female sex, shorter disease duration, number of previous biologic DMARDs, older age, current smoking, and comorbidities.9,10 Female sex, number of prior biologics, and depression have also been related to poorer survival of secukinumab, an IL-17A inhibitor, in real-world evidence studies of PsA and axial spondyloarthritis.11 Meanwhile, in a recent real-world persistence study of ixekizumab, another IL-17A inhibitor, prior methotrexate exposure and depression hampered the drug’s survival rate.12 However, the persistence rates of all these agents are quite different in real-world registries and observational studies.13-16 In a recent study taken from the Nordic country registries of biologic (b) and targeted synthetic (ts) therapies, uptake of newer b/ts DMARDs occurred mainly in biologic-experienced patients.16 Regardless of mode of action, only a minority of patients starting a second or later b/tsDMARD course remained on the drug and achieved treatment targets. Superior outcomes for adalimumab in this report indicate that the positioning of newer b/tsDMARDs in the PsA treatment algorithm remains to be established.16
Real-world survival studies of TOF are very scarce. The Spanish multicenter study by Galíndez-Agirregoikoa et al included 87 patients treated with TOF, the majority previously exposed to bDMARDs.6 Drug retention at 6 months was 77% (first-year retention rate was not reported) and few side effects were observed. Following TOF therapy, patients experienced rapid improvement and maintained joint function. The main outcomes showed significant improvement in the first month of TOF therapy that was maintained longer.6 Egeberg et al performed a Danish cohort study using the prospective nationwide registries DANBIO and DERMBIO, comprising all patients treated with biologics or novel small-molecule therapies for rheumatoid arthritis, axial spondyloarthritis, PsA, and psoriasis.17 Among patients receiving treatment for PsA (2251 patients with 3313 treatment series, 57.4% women), the most frequently used therapy was adalimumab (n = 897), followed by infliximab (n = 773) and etanercept (n = 572). For adalimumab, 63% of patients were biologic-naïve, whereas this was the case for only 8% of patients treated with TOF. Among patients with PsA, drug survival was generally highest among patients treated with either adalimumab, ixekizumab, or secukinumab. TOF (n = 75) and infliximab had the lowest drug survival rates compared to all other drugs.17
In our cohort, unlike in the 2 previous studies,6,17 the majority of patients were women (75%) and had been exposed to a greater number of bDMARDs, and a high proportion of cases presented concomitant depression (55.6%). These points are relevant, since female sex, number of previous biologic treatments, and depression are common factors that contribute to poor persistence of biologic therapies.9-12 Therefore, it is not surprising that our TOF retention rate has been significantly lower (55% and 52.7% at 6 months and 1 year, respectively) than those reported in the 2 aforementioned studies.6,17 However, when we analyzed the TOF persistence graphs based on sex, obesity, smoking, or treatment line, we did not find significant differences based on these factors. Our study is probably underpowered to detect difference between sexes or between other clinical factors. The survival curves were not different depending on the joint pattern or the presence of dactylitis. However, younger patients or those with enthesitis showed lower odds of TOF discontinuation. Thus, for each decade of life, the risk of TOF discontinuation increased by 4%, whereas patients with enthesitis reduced that risk by 63%. The former is an already classic finding regarding the persistence of DMARDs in PsA, but the second is an aspect of special consideration. Interestingly, the persistence rate of TOF in patients with enthesitis was exactly the same at the first and second year (66%) in our present study. Pooled data from 2 phase III studies (ClinicalTrials.gov: NCT01877668, NCT01882439) in patients with PsA receiving TOF 5 or 10 mg twice daily until month 6 or placebo until month 3 have demonstrated the efficacy of this agent on this disease domain.18 However, effectiveness data on enthesitis in real-world studies are limited.
More than 60% of our patients discontinued TOF after 2 years of exposure. It must be taken into account that the majority of our patients could be considered a D2T population, having failed treatment to 2 or more lines of therapies against different targets. Perhaps the most relevant aspect of our results is that in this D2T population, TOF seems to provide added value to the treatment of the entheseal component. However, we must be cautious with this finding, as it was based on a relatively small number of enthesitis cases.
This study has several limitations. First, this study comes from a single university center where a specific protocol for the use of b/ts DMARDS for PsA is applied, which means that its results cannot be generalized. Although the number of patients included may seem small a priori, it must be considered that multicenter studies, such as those previously mentioned from Spain (n = 87) and Denmark (n = 75), analyzed a similar number of patients to ours.6,17 On the other hand, although the correction variables were slightly higher than desirable for the number of events included in the regression model, the confidence intervals were reasonably narrow. Finally, this study provides valuable additional information of the benefits of TOF in patient profiles that are not usually considered in RCTs.
In conclusion, TOF, a pan-JAKi, showed a reasonably good persistence rate in a population of patients with refractory PsA. Factors that usually affect the survival of other drugs in PsA did not affect the persistence of TOF. Patients with refractory PsA and enthesitis might constitute a target group for this treatment.
Footnotes
This work was supported by Pfizer. Pfizer did not participate in design, writing, or intellectual content.
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication March 19, 2024.
- Copyright © 2024 by the Journal of Rheumatology
REFERENCES
ONLINE SUPPLEMENT
Supplementary material accompanies the online version of this article.
