Patients meeting the inclusion criteria for this study were women or men aged 50-85 years with a diagnosis of dementia of the Alzheimer's type, according to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition Text Revision, and probable AD, according to the criteria of the National Institute of Neurological and Communicative Disorders and Stroke and the AD and Related Disorders Association13 Eligible patients had mild-to-moderate disease, which was confirmed by a score on the Korean version of the Mini-Mental State Examination (K-MMSE) of 10-26.14 Each patient underwent a comprehensive evaluation with a neurological examination and appropriate laboratory tests.15 Patients had to have received treatment with oral ChEIs (donepezil, galantamine, or rivastigmine capsules) for a minimum of 3 months prior to baseline, and in the investigator's clinical judgment must have been responding poorly to or been deteriorating on their current treatment.

The patients were assigned to either a poor-response group or an adverse-events group. The poor-response group comprised patients with a loss of at least 2 points on the K-MMSE within the previous 6 months or a decline in the activities of daily living (ADL) or global functioning, as determined by the investigator or caregiver. All patients were required to have a caregiver in contact with them for a minimum of 3 days/week and who was available to accompany them on all visits associated with the study. The adverse-event group was classified as patients who had experienced adverse reactions such as nausea or vomiting after the administration of oral ChEIs and who could no longer take oral ChEIs.

Exclusion criteria included an advanced, severe, or unstable medical condition of any type that could interfere with the evaluation and any condition other than AD that could explain the dementia. Patients were also excluded if they had experienced any cerebrovascular accidents within 6 months prior to baseline, a current diagnosis of active, uncontrolled seizure disorder, or any psychiatric diagnosis that might interfere with the response of the patient to a clinical trial.

Prior to participation in the study, patients provided written informed consent once they were determined by the investigator to be mentally competent. If the patient was not able to provide written informed consent, this was obtained from the caregiver or an authorized representative on the patient's behalf. The study was conducted according to the ethical principles of the Declaration of Helsinki, as revised in 2000.

This was a prospective, 24-week, open-label, multicenter study conducted in South Korea from June 2009 to June 2010. After assessments of eligibility for this study, patients underwent baseline efficacy and safety assessments. They were then assigned to either the poor-response or adverse-events groups. Screening was performed at baseline (visit 1) and at visits 2, 3, and 5, which occurred during weeks 4, 12, and 24, respectively. All patients were started on the 5 cm² rivastigmine transdermal patch, and the first patch was applied between 24 and 36 h after the last dose of oral ChEIs. If the patient had already discontinued oral ChEIs prior to the baseline visit, the time between the last dose of oral ChEI and the first application of the rivastigmine transdermal patch could not exceed 7 days. If the patient demonstrated good tolerability to the initial 5 cm² patch, the rivastigmine patch dose was increased to 10 cm² after 4 weeks. If the target dose was not achieved during the titration period, the investigator could resume titration during the maintenance period. Patients were maintained at their highest well-tolerated dose until the end of the study. The patch was applied by caregivers to clean, dry, and hairless skin on the patient's upper back every morning, and it was worn for 24 h. Normal activities, including bathing, were permitted. In order to minimize any skin irritation, the application site of the patch was alternated daily between the left and right sides.

An interview for the Clinical Global Impression of Change (CGIC) scale (the primary efficacy measure) was performed at baseline, and changes from baseline were assessed at weeks 12 and 24 (or at early termination).16 The CGIC was rated on a 7-point scale ranging from 1 (very much improved) to 7 (very much worse), with 4 indicating no change. The CGIC score was used to determine the primary efficacy variable, which was the percentage of patients receiving treatment with rivastigmine (i.e., who had achieved a score of ≤4) at 24 weeks. Responders to treatment were defined as patients with a CGIC score of 4 or less and poor responders were defined as those who had a CGIC score of 5-7. According to the protocol, each patient was assessed by the same clinician who interviewed them at their baseline visit. For all ratings of change from baseline on the CGIC, the clinician relied on information obtained from the patient at the baseline visit, as well as clinical information obtained throughout the study period. The clinician did not have access to any efficacy data collected during the current study visit. For this reason, the CGIC was rated prior to all other efficacy evaluations.

The K-MMSE, the Korean Instrumental Activities of Daily Living (K-IADL), and the Clinical Dementia Rating-Sum of Box (CDR-SB) were used as secondary outcome measures to assess the effects of the treatment with the rivastigmine transdermal patch on cognitive functioning and ADL.14,17,18 These assessments were performed at baseline, and weeks 12 and 24.

Tolerability and safety evaluations were assessed at each visit during the treatment period through the collection of information concerning adverse events, including serious adverse events. Vital signs and body weight were recorded at each visit. Routine laboratory tests were performed at baseline; postbaseline laboratory tests were not performed routinely unless any abnormal clinical laboratory findings developed and induced clinical signs or symptoms that were considered clinically significant or that required therapy; these were recorded as adverse events. Skin irritation was evaluated at every visit by the investigator based on inspection of the skin at the site of application. Skin irritation also was assessed by the caregiver, who provided a summary irritation rating.

All patients who took at least one dose of study medication were included in the safety analysis, which those patients who additionally provided a valid baseline and at least one postbaseline measurement were included in the intent-to-treat population. Statistical analyses of efficacy outcome measures were performed on the final group of enrolled cases of the study population. The primary analysis time point was week 24, and changes in efficacy measures from baseline were tested using a paired t-test; the level of statistical significance was set at p<0.05. The 95% confidence interval corresponds to the percentage of patients who demonstrated an improvement or no change from baseline (as assessed using CGIC scores).

Of the 164 patients who were enrolled in this study, 116 patients (70.5%) completed all 24 weeks (Fig. 1). The patients were divided into the poor-responder group (75/116, 64.7%) and the adverse-events group (41/116, 35.3%). The patients' baseline demographic and background characteristics are summarized in Table 1. There were no significant differences between the poor-responder group and the adverse-events group at baseline.

At week 24, approximately 82.8% (96/116) of the patients demonstrated either an improvement or no further deterioration in global functioning, as assessed by the CGIC rating of change from baseline (CGIC scores ≤4; Fig. 2). Similar results were observed for the two groups, with 82.7% (61/75) of the poor-responder group and 83% (34/41) of the adverse-events group demonstrating either an improvement or no further deterioration in global functioning as assessed by the CGIC rating of change from baseline (Fig. 3). Of the 116 patients with K-MMSE scores of 10-26, 64.3% showed an increase or a steady state from baseline at week 24, but those changes did not differ significantly from baseline (p>0.05). The change in K-MMSE score at week 24 relative to baseline was 0.42±0.8 (mean±SD). The K-IADL and CDR-SB scores of the entire cohort did not change significantly between baseline and week 24. There were no statistically significant differences in the subgroup analysis (poor-response and adverse-events groups) of the secondary efficacy measures. The changes from baseline for the secondary efficacy variables are given in Table 2.

Forty-eight patients (29.5%) left the study due to adverse events (n=20, 12.2%), follow-up loss (n=17, 10.5%), poor compliance (n=6, 3.7%), or protocol violation (n=5, 3%). The most common adverse event that resulted in discontinuation was skin lesions, such as erythema or an itching sensation, which occurred in 11% of the patients; there were no reported cases with serious skin problems. The next common adverse event was gastrointestinal problems, such as nausea, vomiting, or anorexia, which occurred in 1.2% of the patients. The most frequently reported adverse events are summarized in Table 3.