Abstract
The rivastigmine patch is the first transdermal treatment for Alzheimer disease (AD). By providing continuous delivery of drug into the bloodstream over 24 hours, transdermal delivery may offer benefits superior to those of oral administration. This study compared the efficacy, safety and tolerability of rivastigmine patches with capsules and placebo. IDEAL (Investigation of transDermal Exelon in ALzheimer’s disease) was a 24-week, double-blind, double-dummy, placebo- and active-controlled study. Patients with AD were randomized to placebo or one of three active treatment target dose groups: 10-cm2 rivastigmine patch (delivering 9.5 mg/24 hours); 20-cm2 rivastigmine patch (17.4 mg/24 hours); or 6-mg BID rivastigmine capsules. Primary efficacy measures were the Alzheimer’s Disease Assessment Scale–Cognitive subscale and Alzheimer’s Disease Cooperative Study–Clinical Global Impression of Change. Secondary outcome measures assessed a range of domains, including behavior, cognitive performance, attention, executive functions, and activities of daily living. A total of 1,195 AD patients participated. All rivastigmine treatment groups showed significant improvement relative to placebo. The 10-cm2 patch showed similar efficacy to capsules, with approximately two-thirds fewer reports of nausea (7.2% vs 23.1%) and vomiting (6.2% vs 17.0%), incidences statistically not significantly different from placebo (5.0% and 3.3% for nausea and vomiting, respectively). The 20-cm2 patch showed earlier improvement and numerically superior cognitive scores vs the 10-cm2 patch with similar tolerability to capsules. Local skin tolerability was good. The transdermal patch with rivastigmine may offer additional therapeutic benefits and may prove to be the best delivery system for this drug to treat AD.
Previous treatments approved for Alzheimer disease (AD) have been administered orally. Transdermal drug delivery offers many advantages over conventional oral administration. By providing continuous delivery of medication from the surface of the skin into the bloodstream, patches achieve sustained, more even drug concentrations in the plasma over 24 hours.1 The decreased fluctuation of drug concentration in the plasma may reduce side effects, making optimal therapeutic doses easier to achieve and potentially improving efficacy and treatment compliance.
Rivastigmine (Exelon®, Novartis) inhibits the two enzymes that break down acetylcholine (ACh) in the brain (acetylcholinesterase and butyrylcholinesterase). Administered orally, rivastigmine is widely approved for the symptomatic treatment of mild to moderate dementia of the Alzheimer type and mild to moderate dementia associated with Parkinson disease (PDD). The chemical properties of rivastigmine are well suited to transdermal delivery because it is a small lipophilic and hydrophilic molecule, in addition to being a potent cholinesterase inhibitor (ChE-I). A once-daily rivastigmine patch has recently been developed and tested in a large, randomized, double-blind, double-dummy, placebo- and active-controlled trial. The aim of the Investigation of transDermal Exelon in ALzheimer’s disease (IDEAL) study was to evaluate the efficacy, safety, and tolerability of the rivastigmine patch in patients with probable AD.
METHODS
Detailed methods of the IDEAL trial have been published.2 Briefly, patients (50–85 years; Mini-Mental State Examination [MMSE]3 scores of 10–20) with probable AD were randomized to one of four target treatment groups: 10-cm2 rivastigmine patch (delivering 9.5 mg/24 hours), 20-cm2 rivastigmine patch (delivering 17.4 mg/24 hours), rivastigmine capsules (6 mg BID), or placebo. The study was conducted in 21 countries. The protocol followed the ethical principles of the Declaration of Helsinki, as revised in 2000.
All patients were required throughout the study to be living with someone in the community or to be in daily contact with a responsible caregiver. Every morning a patch (rivastigmine or placebo) was applied by the caregiver to clean, dry, hairless, skin on patients’ upper back. Normal daily activities including bathing were permitted over the 24-hour period that patches were worn. Placement of the patch was alternated between the left and right sides of the back daily. Twice daily all patients also took a capsule (rivastigmine or placebo).
Patients were titrated, tolerability permitting, to their target dose in 4-week intervals over 16 weeks and then maintained at their highest well-tolerated dose (target patch size or capsule) for a further 8 weeks. Patients in the target 10-cm2 rivastigmine patch group received a 5-cm2 rivastigmine patch for Weeks 1–4 and then the target 10-cm2 rivastigmine patch for the remainder of the study. Patients in the target 20-cm2 rivastigmine patch group also started on a 5-cm2 patch and their dose was increased every 4 weeks to a 10-cm2, 15-cm2 and 20-cm2 patch. Patients in the rivastigmine capsule group started on 3 mg/day (1.5 mg BID) and were titrated every 4 weeks in steps of 3 mg/day to a maximum of 12 mg/day.
Outcomes.
Primary efficacy outcome measures were the Alzheimer’s Disease Assessment Scale–Cognitive subscale (ADAS-cog) and the Alzheimer’s Disease Cooperative Study–Clinical Global Impression of Change (ADCS-CGIC).4,5
Secondary efficacy outcome measures were the Alzheimer’s Disease Cooperative Study–Activities of Daily Living (ADCS-ADL), MMSE, Neuropsychiatric Inventory (NPI), Ten Point Clock-drawing Test, and Trail-making Test part A.6–9
Adverse events and vital signs were monitored and recorded specifically throughout the study. Signs of skin irritation at the site of patch application were assessed at every visit by the investigator using a skin irritation rating scale. Caregivers provided summary irritation ratings for the intervals between scheduled outpatient visits. Patch adhesion was also evaluated by caregivers throughout the study, with ratings provided in accordance with a patch adhesion scoring system (see table 3).
Table 3 Patch adhesion as evaluated by caregivers in the IDEAL study (safety population, adhesion assessed at each visit)
Statistical analysis.
Full statistical details of the IDEAL study and its hypotheses have been published.2 Study objectives were assessed using a hierarchical testing strategy. The main efficacy analysis was based on the Intent to Treat (ITT) population using a Last Observation Carried Forward (LOCF) imputation. Additional supportive analyses were also included and have been described previously.2 The safety analysis population comprised patients who received at least one dose of study medication and at least one safety evaluation after baseline.
Changes from baseline on ADAS-cog and secondary efficacy variables were assessed by analysis of covariance (ANCOVA), with baseline values as covariates and treatment groups and countries as factors, or by a Cochran-Mantel-Haenzel (CMH) test. The ADCS-CGIC analysis was based on a stratified Wilcoxon rank sum test using country as a blocking factor. Robustness analyses using a proportional odds model were prospectively planned.
RESULTS
Study population.
Conducted between November 2003 and January 2006, the IDEAL study randomized 1,195 patients: 293 to 10-cm2 rivastigmine patch, 303 to 20-cm2 rivastigmine patch, 297 to rivastigmine capsules, and 302 to placebo. Completion rates were similar in the three rivastigmine groups, 78.2%, 79.5%, and 78.8%, respectively, compared with 88.1% for the placebo group. In total, 970 (81.2%) patients completed the study. The safety population comprised 1,190 patients. The main efficacy population, ITT-LOCF, comprised 1,053 patients. Baseline demographic and background characteristics are summarized in table 1. Ethnic diversities in each treatment group were comparable. There were no significant differences in any of the background characteristics recorded at baseline, among any of the treatment groups.
Table 1 Baseline characteristics and demographics of patients
Efficacy.
At the end of the study (Weeks 20–24), the mean (SD) rivastigmine exposures of patients in the three active treatment groups were 9.8 (1.0) cm2, 16.5 (5.3) cm2, and 9.7 (3.4) mg/day for the 10-cm2 and 20-cm2 patch groups and the capsule group, respectively.
Patients in all rivastigmine treatment groups (patch and capsule) showed significant improvements compared with placebo at Week 24 with respect to ADAS-cog (ITT-LOCF; p < 0.05 vs placebo for all rivastigmine groups) (figure 1, table 2). The 10-cm2 patch provided efficacy similar to that of the capsule. The 20-cm2 patch was numerically superior to the capsule and the 10-cm2 patch.
Figure 1 Changes from baseline at 24 weeks on the ADAS-cog (ITT-LOCF population)
*p < 0.05 vs placebo.
Table 2 Mean changes in efficacy score relative to placebo at Week 24 (ITT-LOCF population)
At Week 24 there was global improvement of all three rivastigmine groups vs placebo on ADCS-CGIC. This was statistically significant (p < 0.05) for the 10-cm2 patch and capsule groups (figure 2; table 2). The treatment difference for the 20-cm2 patch vs placebo with respect to ADCS-CGIC marginally missed significance (p = 0.054) in the ITT-LOCF analysis. However, the 20-cm2 patch yielded consistent and statistically significant results (p < 0.05) across all other prespecified analysis populations (ITT+RDO and observed case [OC]) and statistical analysis methods (predefined proportional odds model). Therefore, significance on the ADCS-CGIC was attained in five of the six pre-planned analyses. This substantial evidence of effectiveness was considered sufficient to continue testing the remaining hypotheses in the hierarchical testing strategy.
Figure 2 Changes from baseline at 24 weeks on the ADCS-CGIC (ITT-LOCF population)
*p < 0.05 vs placebo, †p = 0.054 vs placebo.
Secondary efficacy outcome measure results are shown in table 2. Rivastigmine patches and capsules provided statistically significant benefits over placebo on the ADCS-ADL (figure 3; p < 0.05), with earlier statistically significant improvement vs placebo observed in the rivastigmine 20-cm2 patch group. Numerically superior ADCS-ADL scores were achieved with the rivastigmine 20-cm2 patch compared with the 10-cm2 patch. In all three rivastigmine treatment groups, statistically significant benefits vs placebo were also observed on the MMSE and Trail-making Test part A (figures 4 and 5, respectively; both p < 0.05). Statistically significant treatment effects were not attained on the NPI or Ten Point Clock-drawing test.
Figure 3 Changes from baseline at 24 weeks on the ADCS-ADL (ITT-LOCF population)
*p < 0.05 vs placebo.
Figure 4 Changes from baseline at 24 weeks on the MMSE (ITT-LOCF population)
*p < 0.05 vs placebo, †p < 0.05 vs capsule.
Figure 5 Changes from baseline at 24 weeks on the Trail-making Test part A (ITT-LOCF population)
*p < 0.05 vs placebo.
Tolerability and safety.
Adverse events (AEs) were mostly mild or moderate, predominantly cholinergic in nature, with nausea, vomiting, and diarrhea occurring most frequently. Notably, no statistical differences in AE rates between the 10-cm2 rivastigmine patch and placebo groups were observed, and rates of nausea and vomiting in the 10-cm2 patch group were approximately two-thirds lower than that of the capsule (figure 6). The incidence of AEs was comparable for the 20-cm2 rivastigmine patch and rivastigmine capsule groups, with a tendency to be higher than in the placebo group.
Figure 6 Most frequently reported adverse events occurring in at least 10% of patients in either group (safety population)
*p < 0.05 vs placebo.
The occurrence of serious AEs across the rivastigmine 10-cm2 patch, 20-cm2 patch, and the capsule and placebo groups was 7.9%, 11.9%, 7.1%, and 8.6%, respectively. AEs resulted in discontinuation in 10.7%, 10.2%, 8.5%, and 6.0% of patients in the rivastigmine 10-cm2 patch, 20-cm2 patch, capsule and placebo groups, respectively. The AEs responsible for study discontinuation were predominantly related to the gastrointestinal tract (1.0%, 3.3%, 4.4%, and 1.3%) and nervous system (3.1%, 3.0%, 2.7%, and 1.3%, for the 10-cm2 patch, 20-cm2 patch, capsule and placebo groups, respectively). Of 16 deaths that occurred during the study or within 30 days of treatment being discontinued, five, five, two, and four were in the rivastigmine 10-cm2 patch, 20-cm2 patch, capsule group, and placebo group, respectively. None of the deaths were considered to be treatment-related, with causes typical of an elderly AD population (cardiac and nervous system disorders being the most common causes).
Skin tolerability of the rivastigmine patch was good, with almost all (>90%) patients experiencing no, slight or mild skin irritation. Figure 7 summarizes the overall incidence of skin irritation observed in the 10-cm2 rivastigmine patch group. In the event that skin irritation was of moderate or severe intensity the most frequent symptoms were erythema or pruritus. Withdrawals due to skin irritation were low: 2.4%, 2.3%, 1.4%, and 0.3% in the 10-cm2 rivastigmine patch, 20-cm2 rivastigmine patch, rivastigmine capsule group, and placebo group, respectively.
Figure 7 Overall incidence of skin irritation based on investigator’s ratings in the 10-cm2 rivastigmine patch group (safety population)
For patients experiencing multiple occurrences of the same skin irritation, the most severe rating is shown.
Adhesion properties of the rivastigmine patch were very good, with the majority of patches remaining either completely on or detached only at the edges throughout the 24-hour application period (96% and 94% for 10-cm2 and 20-cm2 patches, respectively) (table 3).
DISCUSSION
Patients in all rivastigmine treatment groups experienced significant improvements over placebo in primary efficacy measures (cognition and global performance). Significant improvements over placebo were also observed in secondary measures of ability to perform activities of daily living, cognitive performance, attention, visual tracking, and motor processing speed. The 10-cm2 rivastigmine patch provided similar efficacy to highest doses of capsules. The 20-cm2 rivastigmine patch showed numerically superior scores on the ADAS-cog, compared with the 10-cm2 patch. Statistically significant benefits of the 20-cm2 patch vs placebo were obtained on five of the six pre-planned analyses of the ADCS-CGIC (all except the ITT-LOCF analysis). The superiority of all rivastigmine groups vs placebo on both co-primary outcome measures and on secondary outcome measures clearly supports the clinical benefits of the rivastigmine patch.
Statistically significant differences between the rivastigmine treatment groups and placebo were not observed on the NPI. There was a marked improvement of the placebo group over baseline on this measure of behavior and psychiatric symptoms. It is not uncommon for untreated patients to show improved behavior or less apathy or aggression when they receive additional attention, as they might do in a clinical study setting. Moreover, patients were allowed to take concomitant antipsychotic or anxiolytic medications during the study and the effects of these treatments might have masked any treatment differences.
It is well established that higher doses of rivastigmine are associated with greater efficacy.10–12 In the IDEAL study, this dose-dependent relationship was apparent on the ADAS-cog. The IDEAL study was not designed or powered to assess differences among the active treatment groups, and it may simply be due to the higher sensitivity of the ADAS-cog compared with other assessment scales13,14 that dose-dependent differences were more easily detected on this outcome measure compared with others. An important feature of IDEAL was the positive control arm with a target dose of 12 mg/day rivastigmine capsules, which revealed that treatment effect sizes in this study were attenuated relative to previous studies of rivastigmine in AD.11 This may also have masked an overall rivastigmine dose–response. The diminished treatment effects may be a consequence of the reluctance of caregivers and health-care professionals to enter patients into clinical trials where they might receive placebo, as proven effective treatments are available. This reticence may apply particularly to those patients who are declining rapidly and who usually manifest the best responses to rivastigmine treatment. It is feasible that patients entering more recent trials show less marked treatment effects due to slower placebo decline,15 compared with patients entering early studies.
Adverse events most frequently reported in the IDEAL study were typical of the cholinergic stimulation experienced with treatment by any ChE-I. These side effects with capsules have been attributed to high peak plasma concentrations and short times to peak plasma concentrations.16 Administration of rivastigmine capsules with food improves tolerability by slowing absorption, reducing peak plasma concentrations, and lengthening the time to peak concentrations. Consistent with transdermal patches providing a continuous, more even delivery of drug from the skin into the bloodstream over 24 hours, nausea and vomiting rates reported for the 10-cm2 rivastigmine patch were two-thirds lower than those of the capsule. Significantly improved tolerability with the rivastigmine 10-cm2 patch was also supported by the majority of patients in this group achieving the target dose. Based on modeling data (data on file, Novartis Pharma AG), transdermal delivery of rivastigmine via the 10-cm2 patch enabled patients to achieve levels of exposure associated with the highest (oral) daily dose of rivastigmine (12 mg), with a substantially lower incidence of the side effects associated with capsules. The 20-cm2 rivastigmine patch was associated with much higher levels of exposure than the highest (oral) daily dose of rivastigmine (equivalent to 18–20 mg per day), yet it was tolerated equally as well as the capsule.
The rivastigmine 10-cm2 patch was associated with two-thirds fewer reports of gastrointestinal (GI) side effects compared with capsules. Correspondingly, withdrawals due to GI adverse events were reported to be approximately 75% lower in the 10-cm2 rivastigmine patch group than in the capsule group. However, owing to very small (presumably random) increases in other adverse events that led to withdrawal, no overall difference in discontinuation rates for the rivastigmine patch and capsule groups was observed. The discontinuation rate for the rivastigmine capsule group (8.1%) was markedly lower than those seen in earlier trials in which discontinuation rates of up to 29% were reported.11,17 This suggests that the discontinuation rate in the rivastigmine capsule group of the current study is in fact low. This may be due to caregivers and investigators being more experienced with ChE-Is and the management of associated side effects. The adverse events of ChE-Is are well known and can usually be managed by halting dose titration or decreasing the dose. Therefore, patients might be less likely to withdraw from modern trials as a result of GI side effects. By providing similar efficacy to the highest doses of capsules with three times fewer reports of nausea and vomiting, the 10-cm2 rivastigmine patch is likely to enable patients to stay on treatment during upward dose increases and beyond.
The rivastigmine patch demonstrated a very good skin tolerability profile. In clinical practice, skin irritation may be further minimized by rotating the patch to different upper body sites (upper and lower back, upper arm, or chest) and applying it to skin that is clean and dry with no cuts, rashes, or other skin problems. The excellent skin tolerability of the patch is exemplified by the remarkably low incidence of withdrawals due to skin irritation. Adhesion rates of the rivastigmine patch were also good, with the vast majority of patches remaining attached to the patients’ skin at the end of the 24-hour application period. This is an excellent result considering that the study was conducted in 21 countries, including some where the climate is hot and perspiration might have been expected (e.g. Chile, Taiwan, Venezuela), and that patients were allowed to bathe and perform normal activities while wearing the patches.
Caregivers of patients who participated in the IDEAL study expressed a preference for the rivastigmine patch over conventional oral therapy, primarily based on ease of use and ease of following schedule.18 The patch also offers visual reassurance that the drug is being taken, and a patch puts the caregiver in control of treatment administration. These benefits might be expected to improve treatment compliance. However, it should be noted that compliance was not prospectively assessed in the IDEAL study, and with many incentives for patients to complete clinical trials, the potential of treatments to positively impact compliance can only truly be seen in routine clinical practice.
A 28-week open-label extension to the IDEAL study has also been completed, the results of which will be published separately. All patients who completed the double-blind study and chose to enter the extension trial (approximately 90%) received an initial daily dose of 10-cm2 patch and their dose was increased at 4-week intervals in 5-cm2 increments, to the highest well-tolerated dose of rivastigmine patch (up to 20 cm2). Along with the 24-week double-blind study presented here, this open-label extension will provide long-term data for up to 52 weeks of rivastigmine patch treatment.
The IDEAL study was the first successful controlled evaluation of a transdermal treatment for patients with AD. Both oral and transdermal administration of rivastigmine demonstrated superiority vs placebo. The once-daily rivastigmine patch provided benefits across a range of symptoms. The target dose of 10-cm2 patch provided similar efficacy to the highest doses of capsules (12 mg/day), with a markedly improved tolerability profile in the majority of patients after a 4-week dose-increase period. The rivastigmine 10-cm2 patch may further enhance the ability of AD patients to stay on and benefit from treatment. Effective and better tolerated medications that have the potential to enhance ease of use and compliance in patients with dementia comprise a significant medical need that is not fully met by current oral treatment options. The rivastigmine patch may offer additional therapeutic benefits and may prove to be the best way to deliver this drug to treat AD.
ACKNOWLEDGMENT
This study was supported by Novartis Pharma AG, Basel, Switzerland. We wish to thank the following principal investigators for their valuable contribution to the IDEAL study:
Chile: S. Gloger, G. Rohde.
Czech Republic: J. Raboch, I. Rektor, V. Pidrman, J. Svestka, R. Talab, P. Kanovsky.
Denmark: G. Waldemar, K. Sørensen, H. Elkjaer, A. Koerner.
Finland: H. Soininen.
Germany: L. Frölich, H. Benes, T. Müller, K. Jendroska, V. Schumann.
Guatemala: J. Ariel Ramirez, L. Salguero.
Israel: T. Treves, J. Aharon Perez, T. Dwolatzky, Z. Meiner, A. Korczyn.
Italy: M. Onofrj, U. Bonuccelli, G. Comi, C. Defanti, G. Bottini.
Korea: S.-H. Han, D.L. Na, S. Yun Kim, J.-H. Lee, D.-W. Yang, J.-W. Kim.
Mexico: J. Reyes, M. Macias, O. Ugalde, A. Macias.
Norway: S. Sparr, A. Banach, H. Davidsen, O. Sletvold, L. Bjørnson, G. Rognstad.
Peru: P. Mazzeti, R. Suarez, J. Altamirano, R. Salinas.
Poland: M. Barcikowska, L. Bidzan, K. Gustaw, A. Potemkowski, W. Kozubski.
Portugal: L. Cunha, A. Herrero Valverde, A. Bastos Lima.
Russia: S. Gavrilova, N. Yakhno, V. Kontsevoy, A. Guekht, N. Neznanov, A. Gnezdilov, V. Mikhailov, Y. Zarubin.
Slovak Republic: P. Turcani, E. Kolibas, I. Mateffy, L. Vircik.
Sweden: B.G. Winblad, J. Louhija, H. Thostrup, A. Börjesson Hanson.
Taiwan: H.-C. Liu, S.-T. Chen, L.-J. Chuo, C.-K. Liu.
United States: J. Winston, J. Bertoni, J. Goldstein, S. Potkin, M. Sauter, J. Ross, K.L. Wilks, M. Lossada, A. Marcadis, B.E. Saferstein, J Shua-Haim, G. Grossberg, C. Sadowsky, B. D’Souza, D. Munoz, J.S. Meyer, B. Arias, D. Linden, G. Figiel, D. Freidenberg, R. Bravo, A. Patel, M. Agronin, W. Petrie, J. Shi, P. Reyes.
Uruguay: A. Pintos.
Venezuela: C.I. Ramirez, I. Mosquera.
Footnotes
-
Disclosure: Prof. Winblad, Prof. Grossberg, and Prof. Frölich have received personal honoraria and grant support from the sponsor during their careers. Prof. Farlow has received personal honoraria (in excess of $10,000) and grant support from the sponsor during his career. Dr. Nagel, Dr. Zechner, and Dr. Lane are employees of the sponsor and have equity or ownership interest. This supplement has been made possible by an educational grant by Novartis Pharma AG, Basel, Switzerland.
REFERENCES
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.