Long-term response to galantamine in relation to short-term efficacy data: pooled analysis in patients with mild to moderate Alzheimer's disease

Abstract

Background: This analysis aimed to identify an operational, clinically relevant definition of response achieved in short-term clinical trials to support the identification of patients with Alzheimer's disease (AD) who would benefit most from long-term galantamine therapy.

Methods: Data were analyzed from 6 randomized placebo-controlled trials of up to 6 months' duration, which included patients with mild to moderate AD receiving maintenance doses of galantamine 16-24 mg/day, and from 12 open-label extensions (galantamine 24 mg/day maintenance therapy). Assessments included changes from baseline in the 11-item AD Assessment Scale-Cognitive subscale (ADAS-Cog 11).

Results: Pooled analysis of the 5-6 month trial data showed that at the trial endpoint (2-5 months after reaching maintenance doses), the proportions of galantamine- (n=1,173) versus placebo-treated patients (n=801) with probable AD categorized according to "improved", "stable" or "non-rapid decline" criteria, were 45.8% versus 27.2%, 59.5% versus 37.1%, and 87.6% versus 69.7%, respectively (observed cases analysis), whilst changes in ADAS-Cog 11 scores versus baseline were -4.9, -4.7 and -2.9 points, respectively, for "improved", "stable" and "non-rapid decline" galantamine-treated patients (-1.5 points for galantamine recipients overall). "Improved" or "stable" galantamine-treated patients displayed mean improvement in ADAS-Cog 11 scores over baseline until 18 months after starting treatment, and attenuated deterioration thereafter; for galantamine-treated patients exhibiting "non-rapid decline", mean ADAS-Cog 11 score returned to baseline after approximately 12 months.

Conclusions: Patients who demonstrate improvement, stability, or limited cognitive decline 2-5 months after reaching maintenance doses of galantamine are more likely to experience continued benefit from long-term galantamine therapy.

Fig. (1)

Overview of the 6 randomized, double-blind, placebo-controlled trials of galantamine (GAL) in patients with Alzheimer’s disease (shaded boxes) and the subsequent open-label extensions included in the analyses. IR, immediate-release formulation (twice daily); n, number of patients starting the trial; PL, placebo; PR, prolonged-release formulation (once daily). Note that 18 patients from GAL-USA-10 entered GAL-USA-12 without first completing GAL-USA-11. Also, 285 patients in GAL-INT-6 had Alzheimer’s disease with concomitant cerebrovascular disease and were included in the analyses; the remaining 307 patients in this trial had probable vascular dementia and were excluded from the analyses.

Fig. (2)

Responder rates (“Responder 1 — improved” definition: stable/improved cognition, and improved global assessment or function or behavior) for galantamine- and placebo-treated patients in individual trials of 5-6 months’ duration (LOCF analysis). LOCF analysis is defined as the intent-to-treat analysis with missing data for patients who discontinued, computed as the last observation carried forward.

Fig. (3)

Long-term changes in scores on the 11-item Alzheimer’s Disease Assessment Scale–Cognitive subscale (ADAS-Cog 11) for responders to galantamine (A), defined as “improved” (“Responder 1” criteria: stable/improved cognition, and improved global assessment or function or behavior) at 5-6 months, compared with the Stern prediction [17] of untreated decline (unadjusted and adjusted for the baseline characteristics of patients available at each assessment point). Non-responders are also shown (B).

Fig. (4)

Long-term changes in scores on the 11-item Alzheimer’s Disease Assessment Scale–Cognitive subscale (ADAS-Cog 11) for responders to galantamine (A), defined as “stable” (“Responder 2” criteria: stable/improved cognition, and stable/improved global assessment or function or behavior) at 5-6 months, compared with the Stern prediction [17] of untreated decline (unadjusted and adjusted for the baseline characteristics of patients available at each assessment point). Non-responders are also shown (B).

Fig. (5)

Long-term changes in scores on the 11-item Alzheimer’s Disease Assessment Scale–Cognitive subscale (ADAS-Cog 11) for responders to galantamine (A), defined as “non-rapid decline” (“Responder 3” criteria: improvement, stability, or minimal deterioration in cognition) at 5-6 months, compared with the Stern prediction [17] of untreated decline (unadjusted and adjusted for the baseline characteristics of patients available at each assessment point). Non-responders are also shown (B).