FormalPara Key Summary Points

Why carry out this study?

There are few real-world studies of apremilast use in patients with psoriatic arthritis, and none in a Canadian population.

APPRAISE sought to assess clinical disease activity, patient-reported outcomes, and safety of apremilast use in routine clinical care in Canada.

What was learned from the study

Over half of all patients reached remission or low disease activity after 4, 8, and 12 months of apremilast treatment.

In real-world clinical practice in Canada, apremilast was safe, and significant improvements were observed in multiple physician- and patient-reported outcomes, including Clinical Disease Activity Index for Psoriatic Arthritis (cDAPSA), swollen joint count (SJC), tender joint count (TJC), pain visual analog scale (VAS), body surface area (BSA), and Health Assessment Questionnaire-Disability Index (HAQ-DI).

Introduction

Patients with psoriatic arthritis (PsA) experience substantial disease burden and reduced quality of life [1]. PsA is associated with comorbidities such as obesity, diabetes, and cardiovascular disease [2]. Joint damage occurs in many patients, impairing physical function [3, 4]. Even in patients with oligoarticular PsA (≤ 4 involved joints), many patients report impairment of quality of life despite having few active joints involved [5]. Several studies have indicated that beginning treatment early in disease course is important for achieving optimal outcomes [6,7,8,9]. The goal of PsA treatment is to improve the signs and symptoms of the disease by achieving remission (REM) or low disease activity (LDA), and improving quality of life [10].

Many biologic therapies used for the treatment of PsA directly target inflammatory cytokines like tumor necrosis factor (TNF), interleukin (IL)-12/IL-23, IL-23, and IL-17 [1]. However, these treatments must be administered by injection and can be associated with significant adverse events, particularly infections [11]. Apremilast is an oral phosphodiesterase 4 (PDE4) inhibitor with a unique immunomodulating mechanism of action. Inhibition of PDE4 elevates intracellular cyclic adenosine monophosphate (cAMP) levels, thereby modulating the inflammatory response by reducing the expression of TNF, IL-23, IL-17, and increasing the pro-inflammatory activity of IL-10 [12]. In phase 3 studies, treatment with apremilast was associated with achievement of Clinical Disease Activity Index for Psoriatic Arthritis (cDAPSA) treatment targets at week 52 in patients with PsA, particularly for patients with moderate disease activity [13, 14]. In addition, there is evidence for real-world effectiveness of apremilast in studies of patients with PsA and psoriasis (PsO) [15, 16].

Apremilast is approved in Canada for the treatment of adults with PsA who have had an inadequate response to, were intolerant to, or had a contraindication to a prior disease-modifying antirheumatic drug (DMARD) [12]. No prospective real-world studies with apremilast in the Canadian population for treatment of PsA have been reported. To better understand the 12-month effectiveness, tolerability, and sustainability of apremilast treatment in real-world settings, the APPRAISE study evaluated clinical and patient-reported outcomes (PROs) in patients with PsA receiving routine clinical care in Canada for up to 12 months.

Methods

Study Design

APPRAISE (NCT03608657) was a prospective, multicenter, observational study that enrolled adult patients with active PsA who were prescribed apremilast per routine care between July 2018 and March 2020 in Ontario, Quebec, British Columbia, Saskatchewan, and Nova Scotia. A maximum of 30 days was permitted from apremilast treatment initiation to the screening/baseline visit. Patients were treated as per regional policies and the product monograph. Patients were followed from treatment initiation to 12 months, with suggested follow-up assessment visits every 4 months and a follow-up call 12 months after the end of the study. Sites were selected on the basis of their expertise in treating patients with PsA, access to apremilast through local reimbursement options, and their ability to collect and report data for this study at the required quality standards. APPRAISE was conducted in accordance with the Declaration of Helsinki’s general ethical principles with the study protocol approved by either regional or institutional ethics boards. The study protocol was approved by the Canadian Shields Ethics Review Board (#18-05-00) and the University Health Network Research Ethics Board (18-6022). Informed written consent was obtained from each patient prior to any study-related procedure.

Eligibility Criteria

This study enrolled adult patients (≥ 18 years of age) with active PsA who satisfied the Classification Criteria for Psoriatic Arthritis (CASPAR). The decision to initiate apremilast was made by the prescribing physician prior to and independent of patient enrollment in the study. There were no limitations on concurrent treatment. Patients were excluded if they were pregnant/planning to become pregnant or breastfeeding.

Endpoints

The primary endpoint was achievement of REM or LDA at month 12, defined as a cDAPSA score ≤ 13. cDAPSA is calculated as the sum of swollen joint count (SJC [0–66]), tender joint count (TJC [0–68]), and Patient’s Global Assessment of PsA (PtGA); visual analog scale (VAS [0–10 cm]), and pain (VAS [0–10 cm]). cDAPSA categories are defined as high disease activity (HDA), > 27; moderate disease activity (ModDA), > 13 to ≤ 27; LDA, > 4 to ≤ 13; and REM, ≤ 4.

Clinical assessments at months 4, 8, and 12 included change from baseline in cDAPSA, SJC, TJC, patient assessment of pain (0–100 mm VAS), body surface area (BSA), and Physician Global Assessment of PsA (PhGA) (0–100 mm VAS); and presence of dactylitis (assessed by individual physician assessment) and enthesitis (assessed by the physician using the Leeds enthesitis index).

PROs at months 4, 8, and 12 included PtGA (0–100 mm VAS), achievement of Patient-Acceptable Symptom State (PASS), Treatment Satisfaction Questionnaire for Medication (TSQM) total score, Health Assessment Questionnaire-Disability Index (HAQ-DI), 36-item Short-Form Health Survey version 2 (SF-36v2) physical component summary score (PCS) and mental component summary score (MCS), and work productivity index scores. To determine PASS, patients responded to this single question: “If you were to remain for the next few months as you were during the last week, would this be acceptable or unacceptable to you?” with a binary (Yes or No) response [17]. The TSQM (version 1.4) is a validated 14-item self-administered patient questionnaire that assesses patient satisfaction with treatment; scores range from 0 to 100, with a higher TSQM score indicating greater patient satisfaction [18]. There are four domains that make up the TSQM: effectiveness, side effects, convenience, and overall satisfaction. A TSQM license was obtained for this study (Opportunity #2806168). For information regarding or permission to use the TSQM, please refer to IQVIA at www.iqvia.com/TSQM or TSQM@iqvia.com.

Common treatment-emergent adverse events (TEAEs) and AEs leading to discontinuation were also assessed.

Statistical Analysis

Baseline patient demographics were summarized descriptively. Outcomes at months 4, 8, and 12 were analyzed by paired Student’s t test or McNemar’s test. Data are presented as observed. SF-36 domain scores were transformed to T scores using a mean of 50 and standard deviation of 10.

Results

Patient Population

In total, 102 patients were enrolled across 18 rheumatology clinics. The majority of the population (96.1%) were white, women (56.9%), and had a mean age of 51.8 years (Table 1). Mean duration of PsA was 5.5 years. The most common PsA subtypes were oligoarticular arthritis (41.4%) and polyarticular arthritis (34.3%). Of the 92 patients who reported psoriasis, the most common subtype of psoriasis was plaque psoriasis (62.0%), followed by scalp (38.0%) and nail (33.7%).

Table 1 Demographics and baseline characteristics
Full size table

There was a high prevalence of comorbidities in the population; 74.5% had at least one comorbidity and 45.1% had more than two comorbidities. Cardiometabolic disease was present in half of all patients and anxiety and/or depression were reported by the investigator in a quarter of the patients. Three-quarters had moderate or high disease activity at baseline based on cDAPSA score. Prior PsA medication use included biologic DMARDs (bDMARDs) in 17.6% of patients and conventional DMARDs (cDMARDs) in 72.5%. One patient had previously used a Janus kinase inhibitor. Most (57.8%) of the population was concurrently taking cDMARDs for PsA, 12.7% were concurrently taking bDMARDs, and 17.6% were concurrently taking corticosteroids.

Of the 102 patients who enrolled, the number of patients reaching 4, 8, and 12 months follow-up were 93, 71, and 57, respectively; 45 (44.1%) patients discontinued the study by 12 months (Fig. 1). Common reasons for discontinuation included lack or loss of effectiveness (20 [19.6%]) and AEs (19 [18.6%]). Median time to discontinuation including the 24-month follow-up call was 13.6 months.

Fig. 1
figure 1

Patient disposition. *Two additional patients discontinued as a result of adverse events but completed a 12-month follow-up visit

Full size image

Primary Endpoint

The proportion of patients in cDAPSA REM or LDA increased from 24.5% at baseline to 63.7% at month 4, 67.2% at month 8, and 53.8% at month 12 with apremilast treatment (Fig. 2a).

Fig. 2
figure 2

Disease activity and treatment satisfaction at months 4, 8, and 12. a Proportions of patients in each cDAPSA disease category at each visit. cDAPSA categories are defined as HDA, > 27; ModDA, > 13 to ≤ 27; LDA, > 4 to ≤ 13; and REM, ≤ 4. Percentages were based on the number of patients with cDAPSA data available at each visit. b Mean TSQM scores. *Mean change from baseline significant at the level of P < 0.05 based on paired Student’s t test. Analysis is based on patients with available data. c Percent of patients achieving PASS. *P < 0.001 vs baseline based on McNemar’s test. Patients responded to this single question: “If you were to remain for the next few months as you were during the last week, would this be acceptable or unacceptable to you?” with a binary (Yes or No) response. Percentages were based on the number of patients with available data at each visit. d Mean cDAPSA scores. Based on patients with available data. cDAPSA Clinical Disease Activity Index for PsA, HDA high disease activity, LDA low disease activity, ModDA moderate disease activity, PASS Patient-Acceptable Symptom State, REM remission, SD standard deviation, TSQM Treatment Satisfaction Questionnaire for Medication

Full size image

Treatment Satisfaction

Treatment satisfaction, as measured by TSQM, increased over time with apremilast, although the changes were not significant (Fig. 2b). Overall satisfaction increased by a mean of 7.1 from baseline to month 12. Improvements from baseline were observed in all subdomains; at 12 months, satisfaction with effectiveness improved by a mean of 9.3, side effects by 8.3, and convenience by 4.4.

At baseline, 26.3% of patients reported PASS. After 12 months of apremilast treatment, the proportion of patients reporting PASS increased to 68.0% (P < 0.001) (Fig. 2c).

Improvements in Clinical Variables

From baseline to month 12, significant improvements were observed in cDAPSA, SJC, TJC, pain VAS, and BSA (Table 2, Fig. 2d). Significant improvements were seen in these parameters at the earliest time point assessed (month 4). Mean (SD) change from baseline in cDAPSA scores at month 12 was − 8.5 (14.2). Mean (SD) change from baseline to month 12 in SJC was − 3.4 (5.2) and that in TJC was − 3.2 (7.5). Mean (SD) change from baseline in pain VAS was − 10.2 (26.0) and that in BSA was − 3.0 (7.3) at month 12. Decreases were also seen in PtGA, PhGA, and the percentage of patients reporting enthesitis and dactylitis at month 12, although these were not significant (Table 2). Among patients with enthesitis at baseline and available enthesitis data during follow-up, resolution of enthesitis was reported by 55.6%, 70.6%, and 60.0% at months 4, 8, and 12, respectively. Among patients with dactylitis at baseline and available dactylitis data during follow-up, complete resolution of dactylitis was reported by 41.2%, 66.7%, and 100% at months 4, 8, and 12, respectively.

Table 2 Disease/clinical parameters and PROs over time
Full size table

Improvements in PROs

In addition to clinical parameters, significant improvements were observed in PROs. Mean HAQ-DI score significantly decreased at months 4 and 8 (Table 2), with a mean change from baseline of − 0.13 and − 0.15, respectively. A decrease from baseline of − 0.16 was seen at month 12, although this was not significant. Significant decreases were seen in SF-36 PCS score at all time points and in SF-36 MCS score at months 8 and 12 (Table 2). Mean changes in SF-36 MCS and PCS were clinically meaningful at months 8 and 12, as defined by a minimal clinically important difference of ≥ 2.5 [19, 20]. Improvements were also observed in SF-36 subdomains, particularly bodily pain, vitality, social functioning, and emotional role (Fig. 3).

Fig. 3
figure 3

SF-36 subscales at baseline and months 4, 8, and 12. SF-36 domain scores were transformed to T scores using a mean of 50 and standard deviation of 10. BP bodily pain, GH general health, MH mental health, PF physical functioning, RE role emotional, RP role physical, SF social functioning, SF-36 36-item Short-Form Health Survey, VT vitality

Full size image

Safety

Overall, more than 80% of patients experienced any TEAE (Table 3). Most TEAEs were mild to moderate in severity. Common TEAEs included diarrhea, headache, nausea, and upper respiratory tract infection. Diarrhea (9/102 [8.8%]), nausea (4/102 [3.9%]), and migraine (4/102 [3.9%]) were the most common AEs in patients who discontinued as a result of AEs. A total of five serious AEs (SAEs) occurred in four patients. Abdominal pain was reported by two patients, seizure and gastric cancer by one patient, and a fall by one patient. None were determined to be related to apremilast treatment and all but one resolved. There were no deaths, major adverse cardiac events (MACE), or COVID-19 cases during the study.

Table 3 Overview and common TEAEs
Full size table

Discussion

In this real-world study, apremilast demonstrated effectiveness for the treatment of PsA. After 12 months of treatment, significant improvements were observed in multiple measures of disease activity, including cDAPSA, SJC, TJC, pain VAS, and BSA. Of patients with enthesitis and dactylitis at baseline, 60% and 100% reported enthesitis and dactylitis to be resolved, respectively, at 12 months. Treatment satisfaction was high, with two-thirds of patients reporting overall satisfaction at month 12 based on the validated TSQM. This is consistent with results from the randomized, placebo-controlled UNVEIL study that demonstrated mean global satisfaction scores of approximately 60% with apremilast at 1 year in patients with chronic plaque PsO [21]. The proportion of patients who reported an acceptable disease state increased from 26.3% at baseline to 68.0% at month 12 with apremilast treatment. Improvements were also seen in physical function, as measured by HAQ-DI and SF-36. Changes in HAQ-DI were not, on average, clinically meaningful (as defined by a minimal clinically important difference of ≥ 0.35) [22]. However, it should be noted that the HAQ-DI is associated with several limitations in evaluating patients with PsA and establishing a standard for clinically meaningful improvement is difficult [23]. Despite this effectiveness and treatment satisfaction, approximately 40% of patients discontinued by month 12. Discontinuations were mainly due to lack or loss of effectiveness and AEs. Most TEAEs were mild to moderate in severity (> 95%) and the most common TEAEs were diarrhea, headache, and nausea, consistent with the known safety profile of apremilast [24,25,26,27]. Collectively, these results show many practical benefits of apremilast treatment in clinical practice.

APPRAISE complements phase 3 clinical trial data of apremilast [13, 28]. Pooled data from the PALACE 1–3 studies showed that 46.9% of patients with ModDA and 24.9% with HDA at baseline achieved cDAPSA treatment targets of REM or LDA at 52 weeks [13]. In APPRAISE, the proportions of patients achieving REM or LDA increased from a quarter (24.5%) at baseline to half (53.8%) at month 12. Achievement of cDAPSA treatment targets was associated with improvements in other PsA manifestations in PALACE 1–3 for up to 5 years as well [13, 28], in line with the improvements in SJC, TJC, enthesitis, dactylitis, and skin involvement seen at month 12 in APPRAISE. Notably, the proportions of patients in REM/LDA increased from week 16 to 52 in PALACE 1–3 [13] but decreased from month 4 to 12 in APPRAISE. The high rate of discontinuations in APPRAISE may have been a factor in this observation.

APPRAISE is one of few real-world evidence studies of apremilast, particularly apremilast treatment of PsA. The results of APPRAISE are aligned with the findings from the real-world APOLO study. APOLO was an observational, prospective study of apremilast in 106 patients with PsA in Belgium (NCT03096990) [15]. In APOLO, improvements were seen in enthesitis, dactylitis, SJC, and TJC over 12 months of treatment. Reported TEAEs were also similar between APOLO and APPRAISE; 45.3% of patients experienced any TEAE in APOLO and the most common TEAEs were diarrhea, nausea, and headache. The mean time since PsA diagnosis was 7.3 years in APOLO compared to 5.5 years PsA duration in APPRAISE. The shorter duration of PsA in APPRAISE, as well as the higher prevalence of oligoarticular arthritis, supports apremilast use earlier in disease progression. Despite the lower number of joints involved with oligoarthritis compared to polyarthritis, patients report impaired quality of life. The collective efficacy results from APPRAISE support the potential for apremilast to improve quality of life in these patients through the improvement of clinical outcome measures as well as patient satisfaction measures.

Additional real-world studies have supported the safety and effectiveness of apremilast for the treatment of psoriasis [29, 30]. High rates of comorbidities have been reported in these studies, including hypertension and other cardiovascular diseases, depression, diabetes, and cancer [29,30,31]. Similarly, comorbidities were high in the APPRAISE population at baseline, particularly cardiometabolic comorbidities, and approximately 20% of patients reported depression at baseline. Despite this, there were no new reports of TEAEs of MACE in APPRAISE. The real-world safety profile of apremilast in patients with psoriasis was similar to the profile seen in patients with PsA in APPRAISE, supporting consistent drug safety across patient populations [29]. In addition to safety and effectiveness, one study of apremilast use in Quebec found that treatment persistence was comparable between apremilast and biologics and use of apremilast resulted in significant healthcare savings [32].

Limitations

The study was limited by a lack of racial diversity. COVID-19 restrictions impacted in-office assessment visits, necessitating reliance on virtual visits. However, all sites were encouraged to have patients return to clinic for an in-person visit at month 12 for a full assessment. In addition to limiting clinical assessments, the COVID-19 pandemic may have impacted PROs. Real-world evidence studies do not have the rigor of randomized, placebo-controlled studies. Although reflective of a real-world treatment environment, the lack of limitations on concurrent treatment may confound the results. Reasons for apremilast initiation were not captured. Fatigue is an important aspect of PsA that was not directly captured in this study. Additionally, fibromyalgia and chronic pain syndrome were not evaluated, which may have impacted patient pain scores. Effectiveness assessments in this study were limited to 12 months, limiting the interpretation of long-term effectiveness and safety. The study also is limited by as-observed aggregate analyses and loss of patients over time. Apremilast was initiated at the clinical discretion of the treating physician, and there were no limitations for threshold of disease activity. As a result, some patients were in LDA at the beginning of the study. Because of this, some patients may not have shown a higher degree of disease improvement that is observed in phase 3 studies that include patients with moderate to high disease activity.

Conclusion

In this real-world analysis, patients with PsA who continued apremilast treatment in Canada achieved improvements in clinical and PRO parameters after 4 months, and the majority reported being satisfied with their achieved disease state. Consistent rates of improvement in meaningful clinical disease measures, such as pain, tender and swollen joints, and dactylitis resolution were observed at 8 and 12 months as well. Apremilast demonstrated effectiveness over 12 months in patients with PsA in Canadian clinical practices, with an efficacy and safety profile consistent with previously reported clinical data.