FormalPara Key Points

The S-flurbiprofen plaster (SFPP) with an innovative percutaneous absorption was developed for the treatment of osteoarthritis (OA) commonly seen in the elderly.

Long-term 52-week application of SFPP was well tolerated in the OA patients, who had a mean age of 66.3 years.

1 Introduction

Osteoarthritis (OA) is strongly linked to aging, and characterized by chronic pain, inflammation, and impaired overall functioning, significantly affecting quality of life [1, 2]. Non-steroidal anti-inflammatory drugs (NSAIDs) are widely used in the pharmacotherapy of OA [3]. Although oral NSAIDs in particular have been used as first-line therapy for many years, concerns about adverse reactions related to class effects of NSAIDs such as gastrointestinal injuries and related hemorrhage are increasingly seen in the elderly patient [4–6]

In recent years, the efficacy of topical NSAIDs is increasingly recognized and, currently, several clinical practice guidelines highly recommend topical NSAIDs over oral formulations [7–12].

S-flurbiprofen plaster (SFPP) is a tape-type patch that contains S-flurbiprofen (SFP)the active ingredient of flurbiprofen (FP), which has a potent cyclooxygenase (COX)-inhibitory effect—and has been developed to achieve improved percutaneous absorption of SFP and its penetration into deep tissues. A clinical pharmacology study of SFPP in knee OA patients showed high penetration of SFP into the synovial tissue [13]. In addition, 2-week randomized controlled studies demonstrated the superior clinical efficacy of SFPP over that of the placebo [14]) and an FP patch [15], which suggested that SFPP might be useful for the short-term treatment of OA.

Systemic exposure to SFP following the application of 80 mg/day of SFPP (two patches/day) for 7 days was estimated to be comparable to that of oral formulations of FP [14]. Adverse reactions considered as a class effect of NSAIDs include gastrointestinal, renal, and cardiac disorders [16]. These adverse reactions rarely occurred in the 2-week SFPP studies [14, 15]. However, in clinical practice, topical NSAIDs are commonly used as long-term treatments, including their intermittent use in several OA treatment paradigms. Thus, here, we evaluated the safety of SFPP applied to OA patients at doses up to 80 mg/day for 52 weeks.

2 Method

2.1 Study Design and Participant Selection

This was a phase III, multi-center, open-label, uncontrolled prospective study conducted between May 2012 and December 2013 at 11 study sites in Japan. The Declaration of Helsinki and Good Clinical Practice guidelines were followed throughout the study. The protocol and informed consent form were approved by the institutional review board at each participating study site. (Trial registration: JapicCTI-121840.)

Osteoarthritis patients with moderate or severe pain in a major joint, and who were ≥20 years of age and had provided written informed consent were included. The diagnosis of OA was based on both radiographic evidence and clinical symptoms determined by the clinician.

Patients who had three or more symptomatic OA sites were excluded. Patients were also excluded if they had diseases or required treatments that might affect the safety or efficacy assessment of NSAIDs, i.e., complication of gastrointestinal ulcers or other joint diseases such as rheumatoid arthritis. In principle, the concomitant use of oral NSAIDs and other analgesics as well as that of antiulcer and gastrointestinal agents was prohibited.

SFPP is a tape-type patch, 10 cm × 14 cm, containing 40 mg of SFP per patch (Tokuhon Corporation, Tokyo Japan). The condition of the OA sites from the whole body were assessed by a clinician, and one or two painful sites were selected for the study. SFPP was applied at a daily dose of one patch per site to be assessed for 52 weeks. The site to be assessed was not changed during the study.

2.2 Study Assessments

Drug safety was assessed by medical examination by the clinician; blood and urine tests (blood cell count, hepatic function, and renal function); and blood pressure and pulse rate at Day1 (before the application; baseline), 2 and 4 weeks after the application, and every 4 weeks thereafter until 52 weeks.

The relationship between SFPP and an adverse event (AE) was assessed using a 4-point scale (“related,” “probably related,” “possibly related,” and “not related”) by the clinician. All the “related,” “probably related,” and “possibly related” scores were defined as “Drug-related AE.” In addition, the severity of each AE was determined on a 3-point scale (mild = treatment not required or daily living not affected, moderate = some treatment required or daily living affected, and severe = particular emergency treatment required or daily living complicated). Since SFPP is a topical formulation, AEs were assessed separately for local AEs at the application sites (skin symptoms) and systemic AEs, except those at the application sites.

The efficacy of SFPP in the outpatients was assessed by the patient’s and clinician’s global assessments as well as clinical symptoms (CS) at 2 weeks before the application (screening), Day1 (before the application; baseline), 2 and 4 weeks after the application, and every 4 weeks thereafter until 52 weeks. The patient’s and clinician’s global assessments consisted of a 5-point scale (marked, moderate, mild, not changed, and worse). The severity of CS was assessed for seven parameters (exercise pain, rest pain, local tenderness, swelling, local heat sensation, limitation of range of motion, and disability of activities of daily living [ADL]) on a 4-point scale (0 = none, 1 = mild, 2 = moderate, and 3 = severe), which allowed for the calculation of the total clinical symptoms (tCS) score.

2.3 Statistical Analysis

The planned sample size was set to 100 patients in each treatment group to mainly evaluate drug safety during the long-term application of SFPP.

All analyses were carried out according to the pre-specified statistical analysis plan using SAS® 9.2. The significance level was set at 5 % (two-sided). Missing data at the end of the study period were imputed using the last observation carried forward (LOCF) method; missing data at other time-points were not imputed.

The safety analyses were based on the safety population that comprised all the patients who had applied SFPP at least once and of whom safety data had been obtained after the SFPP application. The number and percentage of patients who had an AE were summarized using the Medical Dictionary for Regulatory Activities (MedDRA/J ver.16.1) terminologies (system organ class and preferred term). Continuous outcomes in the laboratory tests and vital signs were analyzed using a paired t test.

Efficacy analyses were based on the full analysis set (FAS) that comprised all the patients who had applied SFPP at least once and for whom the efficacy data had been obtained after the SFPP application. For the efficacy outcomes, the descriptive statistics were summarized by the treatment group and each time-point, and the tCS was analyzed using a paired t test.

3 Results

3.1 Patient Disposition

The patient demographic characteristics are listed in Table 1. The study included 201 OA patients. The study population comprised 50 males and 151 females aged 66.3 ± 11.8 years (mean ± standard deviation) and a body mass index (BMI) of 24.76 ± 3.86 kg/m2. The sites assessed are given in Table 2. A total of 301 sites were assessed for the 201 patients since two sites had to be assessed for the 100 patients in SFPP 80-mg group. The site with the most assessments was the knee (n = 192, 63.8 %) followed by the lumbar spine (n = 66, 21.9 %). The combination of both knees was the most common site combination in the SFPP 80-mg group (n = 66, 66.0 %).

Table 1 Patient demographic characteristics
Full size table
Table 2 Sites assessed
Full size table

A total of 201 patients received the study drug (SFPP), of whom 161 completed the study (Fig. 1). Fourteen patients discontinued from the study as the result of an AE. The application periods of SFPP and the patient disposition per treatment group for each period are listed in Table 3. Of the 201 patients, 161 applied SFPP for 52 weeks or longer. Of these, 92.5 % (186/201 patients) had more than 80 % application rate (the actual total number of patches during the application period/the prescribed number of patches during the application period).

Fig. 1
figure 1

Patient disposition. SFPP S-flurbiprofen plaster

Full size image
Table 3 Application periods for the S-flurbiprofen plaster (SFPP)
Full size table

3.2 Safety Evaluations

Skin symptoms that occurred at least once during each application period were observed at 141 of the 301 application sites (Table 4). These included application site dermatitis, eczema, and erythema, and were mild and moderate in severity in 126 and 15 sites, respectively. None of the AEs at the sites were rated as severe. AEs at 128 sites were resolved by interruption of SFPP or by drug treatment, which allowed for study continuation until the end of the study period. The AEs at the remaining 13 sites necessitated discontinuation of the study; however, these AEs were resolved after the study discontinuation.

Table 4 Drug-related adverse events at the application sites (skin symptoms)
Full size table

Systemic drug-related AEs occurred in 18 of the 201 patients (Table 5). Gastrointestinal disorders, such as gastritis and abdominal discomfort, were observed in nine patients and laboratory abnormalities, such as increased blood urea, were observed in seven patients. In addition, hepatic function disorder, colon adenomatous polyp, and generalized dermatitis were observed in one patient each. Among the nine patients in whom gastrointestinal disorders were observed, severe gastric ulcer hemorrhage was observed in one patient, whereas the symptoms were mild in the other patients. The patient with the gastric ulcer hemorrhage complication underwent endoscopic hemostasis because of the occurrence of nausea and hematemesis at 129 days after the SFPP application. The patient discontinued the study at 22 days after symptom onset, and received treatment with proton pump inhibitors. The symptoms were resolved 81 days thereafter. Although gastric ulcer hemorrhage was observed in the patient suffering from atrophic gastritis due to a Helicobacter pylori infection, a possible relation between SFPP and gastric ulcer hemorrhage could not be excluded.

Table 5 Systemic adverse events
Full size table

3.3 Laboratory Tests and Vital Signs

The blood urea nitrogen (BUN) levels increased in the two treatment groups 2 weeks after the SFPP application compared with those at baseline; however, they stabilized thereafter. The maximum mean change from baseline was 1.91 mg/dL for the SFPP 40-mg group (baseline, 15.11 mg/dL) and 1.89 mg/dL for the SFPP 80-mg group (baseline, 16.12 mg/dL). The creatinine levels increased 44 weeks after the SFPP application compared with those at baseline. The maximum mean change from baseline was 0.019 mg/dL for the SFPP 40-mg group (baseline, 0.678 mg/dL) and 0.022 mg/dL for the SFPP 80-mg group (baseline, 0.678 mg/dL) (Table 6).

Table 6 Time-course changes in the laboratory tests
Full size table

Statistically significant changes were found in some of the laboratory parameters and vital signs assessed; however, these changes were not clinically significant.

3.4 Efficacy Evaluations

In the patient’s global assessment (Fig. 2a), the score “marked” appeared at 2 weeks after the SFPP application, and the percentage of “marked” increased continuously until 52 weeks after the SFPP application, reaching 44.2 % (72/163 patients; 45.8 and 42.5 % for the SFPP 40-mg and 80-mg groups, respectively). The percentage of the improvement (the sum of “marked” and “moderate”) was 72.4 % (118/163 patients; 73.5 % and 71.3 % for the SFPP 40- and 80-mg groups, respectively) at 52 weeks after the SFPP application.

Fig. 2
figure 2

a Patient’s global assessment; b Clinician’s global assessment; c time-course changes in the total clinical symptoms score (tCS, mean ± standard error). p values were calculated using a paired t test (two-sided, 5 % significance level). Significant improvements from baseline were observed at all assessment time-points except at the 48-week time-point of “other.” tCS total clinical symptoms

Full size image

Similarly, in the clinician’s global assessment (Fig. 2b), the score “marked” appeared at 2 weeks after the SFPP application, and the percentage of “marked” increased continuously until 52 week after the SFPP application, reaching 46.0 % (75/163 patients; 45.8 % and 46.3 % for the SFPP 40-mg and 80-mg groups, respectively). The percentage of the improvement was 74.8 % (122/163 patients; 74.7 and 75.0 % for the SFPP 40- and 80-mg groups, respectively) 52 weeks after the SFPP application.

Figure 2c shows the time-course changes in the tCS per site to be assessed. The mean ± standard error of the tCS for all sites was 6.2 ± 0.1, 4.5 ± 0.1, and 1.8 ± 0.1 at baseline, 2 weeks, and 52 weeks after the SFPP application, thus improving the tCS by 4.4 ± 0.1 points at 52 weeks. The tCS improved significantly from 2 weeks after the SFPP application, and improved continuously until 52 weeks after the application for all sites assessed. In addition, the CS improved for all seven parameters compared to that at baseline (Table 7).

Table 7 Time-course changes in the clinical symptoms score
Full size table

4 Discussion

In this study, the percentage of patients who continuously applied SFPP for 52 weeks (the completion rate) was high (80.1 %), which was much higher than the 50 % completion rate of previous long-term (52 weeks) studies of topical NSAIDs in OA patients [17, 18]. Although these patients were expected to be in the consistent systemic exposure (consistently high blood concentration of SFP) with the higher adherence of SFPP throughout the study. Only 7.0 % (14/201 patients) discontinued the study due to AEs. From these results, the present study suggested a relatively better safety profile for SFPP.

In the safety assessment of SFPP, the most common drug-related AEs were skin symptoms at the application sites, and most of the drug-related systemic AEs were not considered clinically significant.

The incidence of drug-related AEs causing gastrointestinal symptoms, which are a class effect of NSAIDs, was 3.0 % (3/101 patients) and 6.0 % (6/100 patients) in the SFPP 40-mg and 80-mg groups, respectively. Of the nine patients with gastrointestinal symptoms, gastrointestinal ulcer was found in two patients; however, the other seven patients experienced only mild symptoms such as gastritis, and no study discontinuations due to drug-related AEs occurred. Therefore, the risk of gastrointestinal disorders caused by SFPP was considered lower than with oral administration [19, 20]. The following two phenomena are reported in gastrointestinal disorders associated with oral NSAIDs: a decrease in endogenous prostaglandins (PGs) in the gastric mucosa induced by inhibition of plasma COX, and a direct effect of NSAIDs on the gastric mucosal epithelial cells [21–23]. SFP, which exhibits non-selective inhibition of COX-1 (IC50 = 8.97 nM) and COX-2 (IC50 = 2.94 nM) [24], induces gastrointestinal disorders by acting on the gastric mucosal cells. However, the transdermal administration of SFP did not induce gastric ulcer at exposures up to seven times the maximum plasma concentration (C max) and ten times the area under the concentration-time curve (AUC) after oral administration to rats [14]; therefore, the effects on the digestive organs might be avoided by the transdermal administration of SFP. Similarly, there were no drug-related AEs causing renal disorders. A significant but small increase in BUN levels was observed in previous 2-week SFPP application studies [14, 15]. Therefore, in the present study, we assessed the changes in the laboratory values of renal function-related parameters. Similar to these previous 2-week SFPP application studies, in the current study, a statistically significant increase in BUN levels was observed 2 weeks after the SFPP application, which did not increase further during the prolonged application periods. In addition, the creatinine levels significantly increased from 44 weeks after the SFPP application, although only slight changes in the mean values were observed and these values were within the normal range; therefore, we believe this finding does not suggest a clinically significant effect of SFPP on renal function.

Although patients with cardiovascular diseases, including a past history, were enrolled, no drug-related AEs classified as cardiovascular disorders are observed.

Statistically significant changes were found in some of the blood tests, urinalysis results, blood pressure, and pulse rate, although these changes were considered clinically insignificant.

In the present study, the major drug-related systemic AEs were also gastrointestinal disorders. Laboratory abnormalities were observed for a small number of patients, although these changes were observed for racemic flurbiprofen tablets [19, 20]; therefore, no new drug-related AEs were observed in the present study.

Some topical NSAIDs are known to cause serious photosensitive dermatitis [25, 26], while FP is known to pose no such risk. The benzophenone moiety in chemical structures has been shown to contribute to photoallergic reactions [27, 28]. SFP does not contain a benzophenone moiety. In the current study, SFPP was applied to various sites for 52 weeks without restrictions such as protection from sunlight, and no photosensitivity dermatitis-related AEs were observed. Therefore, SFPP may be used safely without concern for photosensitivity dermatitis.

As application of SFPP was started without any washout period of former NSAIDs, we expected that the apparent improvement of SFPP seemed to be very difficult. Supprisingly, both the patient’s and clinician’s global assessments showed apparent improvements as early as 2 weeks after the SFPP application, and the improvement rates continued to increase over the following weeks until the end of the study. However, our study had major limitations such as a lack of a control group and randomization. The result of efficacy only showed the possibility and further studies including control groups would be necessary to confirm the immediate and long-lasting effects of SFPP.

Since the number of elderly persons in Japan will continue to increase in the near future, prevention of “locomotive syndrome,” deterioration in locomotive function associated with locomotor disability, and extension of healthy life expectancy are challenges to be faced [29, 30]. OA, the most common joint disease, is an important disease underlying locomotive syndrome. Arthralgia due to OA worsens locomotive function. Hopefully, SFPP could open new avenues for preventing the progression of functional disorders in OA patients.

5 Conclusion

SFPP did not cause safety concerns during its continuous application for up to 52 weeks in OA patients. Therefore, SFPP could be an additional pharmacotherapy in the treatment of OA.