FormalPara Key Summary Points

Taiwan has one of the fastest growing aging populations in the world, and the number of people with dementia was projected to increase to up to 210,000 by 2020.

The 10 cm2 rivastigmine patch was approved in 2013 in Taiwan, yet its real-world treatment efficacy in Alzheimer’s disease (AD) patients from Taiwan is still limited.

The current study aimed to provide additional efficacy and safety data of the 10 cm2 rivastigmine patch in a real-world setting by primarily assessing cognitive assessment screening instrument score.

Treatment with the 10 cm2 rivastigmine patch was well tolerated and improved cognitive functioning, neuropsychiatric functioning, and treatment persistence in patients with mild-to-moderate AD.

The prospective study suggests that the 10 cm2 rivastigmine patch is a convenient treatment option in the management of mild-to-moderate AD in Taiwan.

Introduction

Alzheimer’s disease (AD) is a progressive neurodegenerative disease characterized by a steady decline in the patient’s cognition, function, and behavior. The World Alzheimer’s Report in 2019 estimated that there are over 50 million people living with dementia, and that this is set to increase to 152 million by 2050 [1]. Taiwan has one of the fastest growing aging populations in the world, and the percentage of people aged ≥ 65 years increased from 4.1% in 1980 to 10.7% in 2010 [2]. In a nationwide survey conducted from 2011 to 2012 in Taiwan, the prevalence of all-cause dementia in patients aged ≥ 65 years was 8.04% [3], and the number of people with dementia was projected to increase to up to 210,000 by 2020 [4].

Rivastigmine is a reversible cholinesterase inhibitor (ChEI), originally developed as an oral capsule and liquid formulation. In the United States, rivastigmine is indicated for the treatment of all stages of AD and mild-to-moderate Parkinson's disease.

In a large, 24-week, randomized, multicenter, placebo-controlled, double-blind study (IDEAL; Investigation of transDermal Exelon in ALzheimer’s disease), the 10 cm2 rivastigmine patch was shown to have comparable efficacy and improved tolerability than the 12 mg/day rivastigmine capsules. The patch also had fewer withdrawals due to gastrointestinal adverse events (AEs) and three-fold lower incidences of nausea and vomiting, allowing most patients to achieve the optimal dose compared to the 12 mg/day rivastigmine capsules (95.9% vs. 64.4%, respectively). The IDEAL study therefore established the 10 cm2 rivastigmine patch as the currently recommended target maintenance dose in the treatment of patients with mild-to-moderate AD [5, 6]. This patch, containing 18 mg of rivastigmine in line with a dosage of 9.5 mg/24 h, was approved in Taiwan in 2013.

Various structured neuropsychological tools have been used in AD clinical trials for different purposes [7, 8]. In Taiwan, annual cognitive changes are generally measured simultaneously by the mini-mental state examination (MMSE), cognitive assessment screening instrument (CASI) [9], and clinical dementia rating (CDR) [10].

Oral acetylcholinesterase inhibitors (AchEIs) have demonstrated efficacy in treating patients with AD [11]; however, many patients adhere to their treatment for a relatively short duration [12]. A number of factors may contribute to the non-adherence in AD. Decline in cognitive or functional abilities is inevitable in AD, and the non-adherence may be related to dissatisfaction of the treatment outcome. However, a greater proportion of non-adherence is related to AEs of AchEIs or forgetfulness about medication. Non-compliance with oral agents has been a common problem for the treatment of AD, mostly due to the gastrointestinal side effects associated with large fluctuations in acetylcholine levels [12]. The patch formulation allows smooth and continuous drug delivery. In addition, its favorable tolerability, efficacy, and convenience of use may increase treatment compliance.

The overall persistence and adherence to rivastigmine (oral and the 5 cm2 patch) versus donepezil in the Taiwanese population was recently reported using the national health dataset [13]. The real-world treatment efficacy of the 10 cm2 rivastigmine patch in patients with AD has still not been extensively reported. To our knowledge, the present study is the first prospective large-scale observational study of this patch in Taiwan. Here, we report the efficacy and safety results of a 48-week, observational study of the 10 cm2 rivastigmine patch. The study also evaluated the results of adherence to this patch and the most commonly accepted dosing regimen in Taiwan.

Methods

Study Design

This was a 48-week, single-arm, open-label, multicenter, prospective, non-interventional, observational, and post-marketing study of the 10 cm2 rivastigmine patch conducted in Taiwan between May 5, 2016 and July 10, 2017. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. The Independent Ethics Committee or Institutional Review Board reviewed the study protocol for each center (information provided within the supplementary material). All patients provided written informed consent before enrollment.

Study Workflow

Patients initiating treatment with the 10 cm2 rivastigmine patch were enrolled based on the physicians’ judgment and the Taiwan reimbursement criteria of rivastigmine.

The assignment of the patient to the rivastigmine patch was decided within the current practice and the medical indication. Patients were followed-up at outpatient clinics for 48 weeks to observe usage of the rivastigmine patch.

After informed consent, the study included one screening phase (Week 0) and two follow-up phases (Weeks 24 and 48). In the screening phase at baseline (Week 0), the inclusion and exclusion criteria were checked and the demographics data were collected. We also collected the neurobehavioral assessment data at baseline. At Week 24, there was a follow-up neurobehavioral assessment, which included collection of CASI, MMSE, CDR, and neuropsychiatric inventory (NPI) scores. AEs were also recorded and treatment persistency was calculated. At the second follow-up at Week 48, all effectiveness and safety assessments were performed and the persistency data were collected.

Eligibility Criteria

Patients aged 55–95 years with a diagnosis of mild-to-moderate AD based on the core clinical criteria proposed by the National Institute on Aging/Alzheimer’s Association workgroup were included in the study [14]. Eligible patients had to have received a new prescription of the 10 cm2 rivastigmine patch at the screening phase and had to provide a written informed consent. The prior treatments of eligible patients were the 5 cm2 rivastigmine patch, and oral rivastigmine 9 mg/day and oral rivastigmine 3 mg/day.

Patients were excluded if they had previously exhibited contraindications to rivastigmine or had contraindications to the rivastigmine as described on the drug label.

Study Endpoints and Assessments

The primary endpoint of the study was the change in the CASI scores [9] between baseline and Week 48. The CASI is used as a screening instrument for dementia, to monitor disease progression, and to provide a profile of impairment among various cognitive domains [9]. The maximum score of the CASI is 100, with higher scores indicating better cognitive ability.

Secondary endpoints were the changes in MMSE, CDR, and NPI scores from baseline to Week 48. The NPI is a questionnaire administrated to caregivers of AD patients to assess the 12 subdomains of neuropsychiatric behavioral symptoms in AD (including delusions, hallucinations, depression, anxiety, euphoria, and anomalous behavior) over the previous months by rating the frequency of the symptoms on a 4-point scale and their severity on a 3-point scale. NPI is quantified by calculation of the product of frequency (0–4 points) and severity (0–3 points) of each subdomain. A maximum score of 12 is given to each symptom, with an overall scale of 0–144 points; a higher score indicates more serious neuropsychiatric behavioral symptoms. The safety profile of rivastigmine patch was also assessed in this study.

We also calculated treatment persistency, which was the proportion of patients who continued using the 10 cm2 rivastigmine patch to the end of the study. The calculation was based on the following equation:

$${\text{Treatment }}\,{\text{persistency }}\left( \% \right)\, = \,\left( {{\text{full}}\,{\text{ analysis }}\,{\text{set }}\left[ {{\text{FAS}}} \right]\, - \,{\text{number}}\,{\text{ of }}\,{\text{withdrawals}}} \right)/({\text{FAS}}\, \times \,{1}00).$$

Statistical Analysis

In this study, there were two analysis sets: the full analysis set (FAS) and the per-protocol (PP) population. The FAS included all enrolled patients who met the eligibility criteria of the protocol. The PP population included patients who completed the 10 cm2 rivastigmine patch treatment for 48 weeks. Data collection from the patients was missed at certain time points. Hence, there was a difference in the number of patients analyzed from week to week.

The primary and secondary endpoints were presented as descriptive statistics for both absolute values and the change from baseline. A paired t test or a Wilcoxon signed rank test was performed, as appropriate, in calculating longitudinal changes. Summary statistics for continuous variables included number, mean, standard deviation (SD), minimum, median, maximum, and the 95% confidence interval. For discrete variables, summary statistics were presented in contingency tables with absolute and relative frequencies. If not otherwise specified, p values were presented as two-sided and the significance level was set at 0.05. AEs were coded using the Medical Dictionary for Regulatory Activities (v.21.1). The occurrence of each AE was counted and reported as a relative percentage.

Results

Between May 5, 2016 and July 10, 2017, 285 patients were enrolled across seven sites. Of the 285 patients (FAS), 181 (63.5%) were titrated from the 5 cm2 rivastigmine patch, and the rest were switched/titrated from various doses of the rivastigmine capsule. In total, 216 (75.8%) patients completed the study (patients were not considered as dropped out if they switched to another form/dose of rivastigmine after the 10 cm2 rivastigmine patch) and 180 (63.2%) patients remained on the patch for 48 weeks. The mean (SD) duration of the patch treatment was 284.3 (143.3) days (range 1.0–1161.0). Sixty-nine patients (24.2%) discontinued the study, due to AEs (26.1%, 18/69), withdrawal of consent (27.5%, 19/69), lost to follow-up (15.9%, 11/69), and death (15.9%, 11/69) (Fig. 1). None of the deaths were considered related to usage of the rivastigmine patch.

Fig. 1
figure 1

Patient flow chart. *180 patients persisted on the 10 cm2 rivastigmine patch for 48 weeks. AE adverse events, BID twice a day, QD once a day, SAE serious AE

Full size image

Baseline demographics and disease characteristics are summarized in Table 1. Female patients accounted for 57.2% of the FAS population. At baseline, the mean (SD) age of the FAS population was 78.1 (7.7) years and the mean (SD) body weight was 58.3 (10.8) kg. The majority of patients (n = 254; 89.8%) were diagnosed with mild AD with a CDR score of 0.5 or 1 at baseline. Most of the patients (n = 276; 96.8%) lived with family or caregivers.

Table 1 Baseline demographics and disease characteristics of the FAS population
Full size table

The primary endpoint was the change in CASI score after the 48-week treatment period. After 48 weeks of treatment, the CASI score (mean [SD]) was significantly reduced by 2.1 (9.3) points (p = 0.005) in the FAS population (Fig. 2). The reduction was from a baseline score of 64.1 (17.2) to a score of 62.5 (18.8) at Week 48 (Fig. 2).

Fig. 2
figure 2

Change in the CASI score. *Significant p value. CASI cognitive assessment screening instrument. n number of patients analyzed

Full size image

The FAS population showed stable MMSE scores with small mean (SD) changes from baseline of 0.2 (2.7) at Week 24 and − 0.4 (2.8) at Week 48 (Fig. 3). CDR scores were well sustained over the 48-week treatment period (Table 2). The proportion of patients from the FAS population with CDR score ≤ 1 at baseline dropped from 89.7 to 84.6% at Week 48. Most patients (78.6%) had no change or improvement in CDR score at Week 48. The FAS showed numerically stable NPI scores with small mean (SD) changes of 0.9 (12.4) at Week 24 and 0.4 (11.0) at Week 48 for the total score. Changes from baseline in scores of single NPI domains, including delusions, hallucinations, apathy, and depression, were also limited to between − 0.2 and 0.5 points at Week 48 (Table 2).

Fig. 3
figure 3

Change in the MMSE score. *Significant p value. MMSE mini-mental state examination

Full size image
Table 2 Secondary endpoints in the FAS population
Full size table

The overall treatment persistency was 63.2%. Treatment persistency was also calculated for patients based on their prior treatment. It was observed that the majority of patients in the study had used either oral rivastigmine 3 mg/day (n = 21, 7.4%), oral rivastigmine 9 mg/day (n = 55, 19.3%) or the 5 cm2 rivastigmine patch (n = 181, 63.5%) as the prior treatment. Treatment persistency over a 1-year period was similar in patients irrespective of prior treatment (oral rivastigmine 3 mg vs. oral rivastigmine 9 mg vs. 5 cm2 rivastigmine patch: 66.7% vs. 65.5% vs. 66.3%). Effects of persistency on CASI, MMSE, and CDR scores were also investigated. We observed that CASI score [mean (SD)] was reduced by 2.0 (9.3) (p = 0.010) in patients who persisted on the 10 cm2 rivastigmine patch treatment for 48 weeks (Table 3). The reduction was from a baseline score of 64.4 (16.3) to a score of 63.1 (17.9) at Week 48. MMSE scores were stable with small mean (SD) changes from baseline of 0.1 (2.6) at Week 24 and − 0.3 (2.7) at Week 48 (Table 3). CDR scores were well sustained over the 48-week treatment period. Most patients (79.2%) had no change or an improvement in CDR score at Week 48 (Table 3). Over the study period, 523 AEs were reported by 158 (55.4%) patients; among these, 102 were serious AEs (SAEs) occurring in 46 (16.1%) patients. The most common drug-related AEs were pruritus (11.2%), nausea (3.5%), rash (3.2%), and vomiting (2.8%) (Table 4). Forty-eight (16.8%) patients discontinued the 10 cm2 rivastigmine patch due to AEs, and these were most commonly pruritus (4.9%) or rash (2.1%). Most AEs were managed by symptomatic treatments (39.3%). The most common SAE (≥ 5.0%) was infection and infestations [n (%); 21 (7.4%)], of which pneumonia [11 (3.9%)] and urinary tract infection [7 (2.5%)] occurred in ≥ 2% patients. Although 58 of the 102 SAEs were graded as severe [occurring in 29 (10.2%) patients], eight as life threatening [3 (1.1%) patients], and seven as fatal in severity [7 (2.5%) patients], none were suspected to be related to the 10 cm2 rivastigmine patch. Of these SAEs, 53 were resolved in 30 (10.5%) patients. However, 11 SAEs caused 7 (2.5%) patients to discontinue the rivastigmine patch and 22 SAEs caused the death of 11 (3.9%) patients. However, none of the deaths were related to use of the rivastigmine patch. The cause of death included metastases to lung/colon cancer, septic shock, cardiac failure, metastases to liver, necrosis, renal failure, respiratory failure, colon cancer stage IV/hepatic cirrhosis, acute kidney injury, pneumonia, hemophagocytic lymphohistiocytosis, multiple organ dysfunction syndrome, cardiac arrest, sepsis, urinary tract infection, fall, and myocardial infraction.

Table 3 Effect of persistency on CASI, MMSE, and CDR scores
Full size table
Table 4 Drug-related adverse events
Full size table

Discussion

Our results demonstrated that the 10 cm2 rivastigmine patch provides clinical benefit in patients with mild-to-moderate AD. Over 80% of patients maintained a CDR score ≤ 1 in the 48 weeks of treatment with the patch. We observed a decrease of < 3 points in the CASI score, especially in patients with mild-to-moderate dementia at baseline, which was lower than that in previous research [9].

To the best of our knowledge, this is the first study to evaluate CASI outcomes after treatment with the 10 cm2 rivastigmine patch in a real-world setting. Our study demonstrated an increase of 0.5 points in CASI score at Week 24 compared with baseline, which compares favorably with the decrease of 1.5 points in CASI score in the 6-month pilot study of rivastigmine 4.5 mg capsules [15]. The disparity in this result could be due to varying disease severity and the distinct dose and drug formulation. Patients in the pilot study had a lower mean CASI score at baseline (47.5 vs. 64.1 in the present study), representing a population with a more advanced disease stage. Again, the 10 cm2 rivastigmine patch provides similar exposure to rivastigmine as the capsule, although the dose of rivastigmine in the capsule formulation is slightly higher (12 mg/day vs. 9.5 mg/day) than that in the patch [16]. Finally, unlike capsules, the patch formulation allows continuous and steady delivery of rivastigmine through the skin, thus avoiding the first-pass effects after oral administration [17]. This could mean that titrating to a higher dose earlier may allow patients to achieve an optimal therapeutic dose and also benefit from a longer duration of treatment. However, findings suggest that further research is required to determine which population of patients may benefit from titrating to a high dose [16, 17]. Clinical adherence of the 5 cm2 rivastigmine patch has been explored and the results have shown a significant negative correlation between subscapular skin fold thickness and serum metabolite levels [18].

A decline of 2.3 points per year in MMSE scores has been observed in those who progressed to dementia without treatment [19]. Per the National Health Insurance regulations in Taiwan, patients with AD using reimbursed rivastigmine should switch to another treatment if the MMSE score decreases by more than two points [13]. During the follow-up period, patients treated with the 10 cm2 rivastigmine patch maintained cognitive performance, and only one patient had to withdraw from the study due to a failure in re-submission for health insurance reimbursement; however, the reason for the failure was not specified. The results were as expected, supporting the beneficial role of the 10 cm2 rivastigmine patch in the maintenance of global cognition and disease severity. Based on the results for the CASI total score, MMSE scores, and CDR scores, we observed that a higher persistency rate with the 10 cm2 rivastigmine patch is associated with numerically better clinical outcomes. Moreover, even though the overall treatment persistency was 63.2%, the remaining ~ 37% of patients still adhered to the treatment for 274 days. Our effectiveness data were also comparable to prior observational studies. An 18-month observational Canadian study comprising of patients treated with the 5 or 10 cm2 rivastigmine patches showed a mean change in MMSE score of 0.5 at 6 months and 0.2 at 12 months [20]. In an observational study conducted by Minthon et al., the change in the MMSE score was 0.11 at 6 months and − 0.62 at 12 months [21]. A 6-month, observational study of switching from donepezil or rivastigmine capsules to the 5 or 10 cm2 rivastigmine transdermal patches also reported a stable MMSE outcome, with a minimal change of − 0.5 [22].

There are a number of AchEIs used in real-world practice for AD patients. From a statistical perspective, oral rivastigmine 3 mg/day, oral rivastigmine 9 mg/day, and the 5 cm2 rivastigmine patch are comparable in bridging to the 10 cm2 rivastigmine patch when the persistency rate for the 10 cm2 rivastigmine patch serves as the major clinical outcome. These results may indicate the relatively applicable usage of either 3-transformation formula, i.e., switching from either of the three doses of rivastigmine (oral rivastigmine 3 mg or 9 mg and the 5 cm2 rivastigmine patch) to the 10 cm2 rivastigmine patch is feasible. Of particular note is that the shift from oral rivastigmine 3 mg/day to the 5 cm2 rivastigmine patch is tolerable in the Taiwanese population [23].

In our study, the majority of AEs were mild, local skin tolerability was good, discontinuations due to drug-related AEs occurred in less than a fifth of the patients, and no unexpected safety issues arose. The IDEAL study [5] demonstrated that the 10 cm2 rivastigmine patch provided similar efficacy as the rivastigmine capsule (12 mg/day), but with a superior tolerability profile due to lower incidences of vomiting (6.2%) and nausea (7.2%) over 6 months; however, in our study, such incidences occurred in only 1.4% patients, each over a longer observational period of 48 weeks. In the current study, 63.2% of patients continued on the 10 cm2 rivastigmine patch treatment at Week 48, which was in line with the 55–65% of patients observed with earlier studies at Week 24 with an equivalent oral dose (obtained with 12 mg/day rivastigmine capsules) [21, 24, 25]. These findings were comparable to those in an open-label study, which reported a study completion rate of 74.5% after 24 weeks of treatment with the 10 cm2 rivastigmine patch, and the 6-month IDEAL study where 83.8% of participants stayed on the 10 cm2 rivastigmine patch for at least 8 weeks [25]. The results from the Real-world Evaluation of Compliance And Preference in Alzheimer's disease treatment (RECAP) [23] and Exploring and Managing Dementia in Black African and Caribbean Elders (EMBRACE) [20] studies also showed that 82.4% and 88.2% of caregivers of patients with AD preferred the rivastigmine transdermal patch over oral medication. Therefore, the 10 cm2 rivastigmine patch may allow patients easier access to higher doses compared with the 12 mg/day rivastigmine capsule, thereby enabling patients to stay on and benefit from long-term effective treatment.

The safety data in our study were also comparable to prior observational studies. In a Canadian study with the 5 or 10 cm2 rivastigmine patches, 18.3% of patients discontinued due to an AE, with pruritus (4.0%), erythema (2.9%), nausea (2.5%), rash (1.9%), skin reaction (1.7%), application site erythema (1.6%), vomiting (1.3%), decreased appetite (1.1%), and dizziness (1.0%) being the most common [20]. Similarly, in a 6-month observational study of switching from donepezil or rivastigmine capsules to the 5 or 10 cm2 rivastigmine patches, discontinuation due to AEs occurred in 18% of patients, with skin reactions and gastrointestinal disorders causing 9% and 3% of patients, respectively, to stop the treatment [22].

All AchEIs require titration from a low dose and the complex dosing regimen [26] may affect drug adherence and, therefore, the outcome. Rivastigmine is administered in four oral formulations (doses of 1.5 mg, 3 mg, 4.5 mg, and 6 mg) and two transdermal patch formulations (5 cm2 and 10 cm2) [27, 28]. Consequently, the dosing regimen may vary widely among practicing physicians as observed in this study and hence might interfere with compliance. The other two approved AchEIs in Taiwan are galantamine and donepezil, which are administered in two oral doses, and the dosing regimens are titrated if an AE occurs. In clinical practice, AD patients also fail to adhere to AchEI treatment due to a lack of efficacy. When one AchEI may fail to reach the therapeutic expectation and the dosing strategy may not improve efficacy, the physician may consider switching to another AchEI [29], using an add-on of memantine, or discontinuing the AchEI altogether [26, 30].

The efficacy and safety of the rivastigmine patch has been validated in patients with AD who failed to benefit from treatment with donepezil [31], and the switch from donepezil to the (5 cm2) rivastigmine patch and the gradual switch or a cross-tapering strategy both showed a high rate of adherence and low incidences of side effects [22, 32, 33]. The favorable safety and tolerability profile reported with the 10 cm2 rivastigmine patch and increased ease of use than an oral formulation may increase adherence to a higher dose therapy. This could encourage patients with AD to stay on treatment for a longer period, offering the possibility of enhanced outcomes in clinical practice. Although the transdermal patch may cause skin reactions, these events were manageable. In line with prior studies [5], erythema/rash and pruritus were the most commonly reported reactions in our study; however, importantly, no patient experienced a skin reaction that was reported as an SAE. Our data support a favorable skin tolerability profile for the 10 cm2 rivastigmine patch and therefore further reinforce the fact that its benefits should not be dismissed due to skin irritation problems.

Limitations

The open-label and real-life observational design of our study has some limitations. As there was no placebo or parallel control group, outcomes in the absence of the 10 cm2 rivastigmine patch are unknown. It is suggested that researchers in this field should consider including a control group in future studies. Physicians were not blinded to study treatment and were able to adjust the dosage freely as needed within the study, such as temporarily switching to the other formulation (i.e., rivastigmine capsule) or a lower dose (i.e., the 5 cm2 rivastigmine patch). These adjustments could be influential to treatment response. Therefore, the clinical outcome might not entirely represent the effects of the continuous 10 cm2 rivastigmine patch treatment. Also, it is important to note that randomized controlled trials, despite being conducted in a controlled setting, are associated with several limitations including incomplete understanding of AD pathophysiology that might have led to selection of the wrong targets, inappropriate patient selection, variable rates of progression, suboptimal dosing, drug exposure and/or target engagement, inappropriate time of intervention, inappropriate outcome measures, and low sensitivity of clinical scales. These variables are even more difficult to control in a real-world study.

The results may not be generalized to the entire Taiwan population because, in the real world, patients face the issue of failing the re-application of drug reimbursement and the effect of this situation is not captured in our study. However, the present study offers real-world insights into the persistency, treatment outcomes and unique treatment pattern in Taiwan.

Conclusions

The use of the 10 cm2 rivastigmine patch in a real-life setting was efficacious for patients with mild-to-moderate AD and was not associated with any significant safety concerns. Thus, the 10 cm2 rivastigmine patch represents an efficacious, tolerable, and convenient treatment option in the management of mild-to-moderate AD in Taiwan.