Effect of donepezil on brain acetylcholinesterase activity in patients with AD measured by PET

Materials and methods.

Three patients (Patient 1, a 58-year-old woman; Patient 2, a 63-year-old man; and Patient 3, a 71-year-old woman) with a clinical diagnosis of probable AD8 took part in this study. Duration of illness in years was 5, 5, and 1, Clinical Dementia Rating9 scores were 2, 2, and 1, and Mini-Mental State Examination (MMSE) scores were 7, 13, and 14 at first PET study in Patients 1, 2, and 3. None of the patients had a family history of dementia. MRI revealed mild atrophy but no focal lesion in the three patients. In Japan, donepezil was approved for marketing in October 1999, and a standard dose is 5 mg/d. PET studies were performed once before and once during donepezil treatment, with a 2-month interval in Patient 1 and a 7-month interval in Patients 2 and 3. The second PET study was performed 1 month after start of oral administration of donepezil (5 mg/d) in Patients 1 and 2. Patient 3 was started with 5 mg/d of donepezil, but the dose was reduced to 3 mg/d in 2 weeks because she showed restlessness. Patient 3 was well with 3 mg/d of donepezil for the next 6 weeks and then underwent the second PET scan. The patients were assessed with the cognitive subscale of the AD assessment scale (ADAS-cog)10 once within a week of the first PET study and once within a week of the second PET study. On the day of the second PET study, all patients with AD took donepezil at 8:00 am and underwent the PET scan at 3:00 pm. The venous blood was sampled to measure plasma concentration of donepezil just before tracer injection at the second PET study.

Fourteen age-matched healthy volunteers (67 ± 10 years of age; 9 men, 5 women; MMSE score 29.1 ± 1.2) participated in the study as normal controls. Five had repeat PET scans with a mean interval of 11 ± 7 months to assess reproducibility of PET scans without donepezil. This study was approved by the Ethics and Radiation Safety Committee of the National Institute of the Radiologic Sciences, Japan. Written informed consent was obtained from all subjects or their family members.

A sequence of 14 PET scans was acquired with an EXACT 47 scanner (Siemens/CTI, Knoxville, TN) covering 40 minutes after IV injection of [¹¹C]MP4A (approximately 740 MBq) in each subject. Arterial blood samples were collected 24 times in 15 minutes after IV tracer injection for measurement of total radioactivity and metabolite analysis.5 The region of interest analysis was carried out by delineating frontal, sensorimotor, temporal, parietal, occipital, and posterior cingulate cortices, hippocampus, amygdala, thalamus, striatum, and cerebellum in each hemisphere of PET images with reference to corresponding MRIs.4-6⇓⇓ A three-compartment model was used to yield estimates of K₁ (transport into tissue), k₂ (tissue clearance of unchanged tracer into blood), and k₃ (hydrolysis rate of [¹¹C]MP4A by AChE, i.e., AChE activity) using metabolite-corrected arterial plasma input function.4-6⇓⇓ Cortical k₃ values were calculated as the mean value of those in the frontal, sensorimotor, temporal, parietal, and occipital cortices. Values are expressed as mean ± SD.

Results.

The ADAS-cog subscale improved from 55.6 to 48.7, 30.0 to 25.6, and 30.0 to 26.0 in Patients 1, 2, and 3 during donepezil therapy. None of the patients had a peripheral side effect of donepezil. The plasma concentrations of donepezil at the second PET study were 27.8, 22.0, and 23.8 ng/mL in Patients 1, 2, and 3.

In the second PET study with donepezil, radioactivity in the plasma (unchanged tracer) and the brain (total radioactivity) after [¹¹C]MP4A injection increased from the baseline study, but decline of radioactivity after the peak in the cerebral cortex became faster than that in the baseline study in all patients with AD ( figure 1). The image contrast between striatal and nonstriatal areas increased ( figure 2), and the ratio of striatum to cortex from 22 to 40 minutes after tracer injection increased from 2.8 to 4.5 in Patient 1, 2.3 to 2.8 in Patient 2, and 2.3 to 2.7 in Patient 3 by effect of donepezil treatment.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Time courses of radioactivity in the plasma (unchanged tracer) and the brain (total radioactivity) after IV injection of N-[¹¹C]methylpiperidin-4-yl acetate in the baseline study (A) and the second study with donepezil (B) in a patient with AD (Patient 1). Radioactivity is expressed as percentage of injected dose per liter tissue.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. PET images of a healthy control participant (a 65-year-old man) and a patient with AD (Patient 1) representing radioactivity collected during the period of 22 to 40 minutes after tracer injection. The radioactivity in the cerebral cortex of the patient (B) is lower than that in the normal control (A). With donepezil (5 mg/d), there is a further relative reduction of radioactivity accumulation in the cerebral cortex and thalamus compared with that in the striatum in the patient (C). These images are scaled to radioactivity in the striatum; in other respects, the color scale is arbitrary.

In the pretreatment baseline study, cortical k₃ values were lower by 19 ± 7% in the patients with AD than the 14 normal controls ( figure 3). In the second PET study, cortical k₃ values were reduced by 44, 35, and 38% in Patients 1, 2, and 3 as compared with the baseline study. The cortical K₁ values in the baseline study were lower by 30 ± 11% in the patients with AD than the 14 normal controls. In the second PET study, cortical K₁ values were changed by −1 ± 9% in the patients with AD as compared with the baseline study.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. Effect of donepezil on cortical k₃ values in patients with AD. The open circle indicates mean and SD for the 14 normal subjects, closed circles indicate cortical k₃ values for the five normal subjects who underwent repeat PET scans without donepezil, and closed squares indicate cortical k₃ values for the patients who underwent repeat PET scans with donepezil.

In the assessment of reproducibility of this technique in five normal controls, cortical k₃ values were changed by −4 ± 10% in the second PET study without donepezil as compared with the baseline study, and the paired t-test results revealed no significant difference (see figure 3).

Discussion.

Plasma concentrations of donepezil in the patients with AD were in the medium range of steady-state donepezil plasma concentrations for 5 mg/d donepezil therapy,1,2⇓ although Patient 3 had taken only 3 mg/d. Tracer kinetics in plasma and brain were remarkably changed in the patients with AD with donepezil because AChE activity was inhibited by donepezil both in blood and brain. The image contrast between striatal and nonstriatal areas increased with donepezil because AChE activity is so high in the striatum that radioactivity accumu-lation is strongly dependent on delivery and is insensitive to AChE activity change, whereas it responds to AChE activity change in nonstriatal areas.4-6⇓⇓ Tracer kinetic analysis revealed a remarkable reduction of cortical k₃ values by 39 ± 5% without K₁ change in the patients with AD in the second PET study with donepezil, suggesting that the clinical doses of donepezil substantially inhibit AChE activity in the brain of patients with AD. All patients showed some degree of symptomatic improvement, and it was concluded that the symptomatic improvement was likely caused by improved cholinergic activity by inhibition of AChE in the brain.

In the current study, the patients with AD differed in doses of donepezil administered, duration oftreatment, disease stage, and age. Future studies are required to elucidate whether the degree of AChE inhibition by AChE inhibitors is dependent on these variables.