The trusted source for
healthcare information and
Source: Nitsch RM, et al. Ann Neurol 2000;48:913-918.
AF102B, an investigational cholinergic agonist that binds to the muscarinic M1 receptor, is reported to lower total cerebrospinal fluid beta amyloid levels in Alzheimer’s disease (AD) patients. AF102B was previously found to decrease amyloid secretion in cell cultures and to have cognitive enhancing properties in mice and aged rats.
Nitsch and colleagues administered doses up to 80 mg of AF102B three times a day to 19 AD patients of mild to moderate severity. The patients underwent lumbar puncture (LP) at baseline prior to treatment and during four weeks of maximal dosing of AF102B. A separate set of AD patients who were involved in clinical trials of hydrochloroquine and physostigmine also underwent LPs at comparable intervals to the first group. The spinal fluid obtained from all patients was subjected to analysis using an ELISA technique to determine total amyloid beta-protein levels as a surrogate marker of drug effect. Total amyloid beta protein levels are not reported to change substantially over time in untreated AD patients.
The most common side effects of AF102B treatment were gastrointestinal side effects, excessive sweating, and headaches. Among the 19 AD patients treated with AF102B, 14 experienced a decline in total amyloid beta levels, averaging 22% below baseline. In three cases, amyloid beta protein levels increased and two remained unchanged. Among 10 patients receiving hydrochloroquine, amyloid levels increased in five, decreased in two, and were unchanged in three. In the nine patients treated with physostigmine, amyloid levels increased in four, decreased in four, and remained the same in one. The change in average amyloid levels between baseline and maximal treatment were statistically significant in the AF102B-treated group only. Levels of the 42 peptide form of amyloid beta and truncated amyloid precursor protein derivatives did not change significantly in any group.
This interesting study forgoes the use of the standard cognitive and functional-based outcome measures usually used in AD clinical trials, and substitutes serial measurements of a surrogate biological marker in the cerebrospinal fluid as the primary outcome measure. Nitsch et al judiciously avoid making any claims of having found a disease-modifying therapy and acknowledge the limitations of using a single marker as a measure of therapeutic efficacy. They also point out that clinical and behavioral measures will be required to judge the value of amyloid-reducing therapies in the treatment of AD.
Unfortunately, the observation of a decrease in total amyloid beta protein levels with AF102B treatment neither proves or disproves that this compound has a positive effect on the AD brain. In the CSF of patients with probable AD, levels of the 42 peptide form of amyloid beta protein are lower than normal, perhaps reflecting sequestration of this peptide in plaques or other deposits in the brain. It is, therefore, unclear whether lowering total amyloid beta levels, while leaving the 42 peptide form unchanged, is a therapeutically positive, negative, or indifferent result. The lack of alteration in amyloid precursor protein derivatives further confounds interpretation of these findings, since these derivatives were altered by the compound in cell culture studies.
We are likely to see more and more use of surrogate biological markers in future studies of putative AD therapies. This study demonstrates the viability of involving patients receiving other investigational therapies as controls and suggests the importance of measuring multiple biological and clinical markers in future studies. It also highlights some of the difficulties ahead in establishing whether potential disease-modifying agents exert positive effects in the treatment of this chronic, complex neurodegenerative disorder. — Norman R. Relkin