Antibiotics for Alzheimer’s?

Abstract & Commentary

Source: Loeb MB, et al. A randomized, controlled trial of doxycycline and rifampin for patients with Alzheimer’s
disease. J Am Geriatr Soc. 2004;52:381-387.

Although infectious agents are not widely believed to be the primary cause of Alzheimer’s disease (AD), some evidence implicates certain neurotrophic viruses and bacteria as possible contributing factors. Canadian investigators Loeb and colleagues suggest that Chlamydia pneumonia infection might play a role in AD. They carried out a randomized, blinded, and placebo-controlled trial testing whether antichlamydial antibiotics (rifampin/doxycycline) could serve as a potential therapy for patients with mild-to-moderate AD.

In this multicenter study, 51 mild-to-moderate AD patients were randomized to receive 3 months of treatment with rifampin 300 mg daily plus doxycycline 200 mg daily. Fifty other age- and severity-matched AD patients received a placebo. The primary outcome measure was the standardized form of the ADAS-Cog, a test commonly used in AD drug trials. Additional measures of cognition, behavior, and daily function were obtained at 3 months, 6 months, and 12 months from the start of therapy. The majority of patients in the trial were on stable doses of cholinesterase inhibitors during the study period. Blood tests for C pneumonia, including PCR and immunoglobulins, were evaluated before and after antibiotic treatment.

Approximately 84% of the subjects completed the trial, with relatively equal numbers of dropouts from the treatment and placebo arms. There was no significant difference in adverse events between groups. The primary outcome variable measured was ADAS-Cog. There was no significant difference between antibiotic-treated patients and placebo controls at 3 or 12 months. However, there was a statistically significant effect favoring antibiotic treatment at 6 months. A significant difference favoring antibiotic treatment was found on the Mini-Mental State Examination at 12 months and measures of function, behavior, and mood at 3 months.

The PCR and immunoglobin tests did not indicate irradication of C pneumonia infection in those treated with antibiotics. Loeb et al suggested that the putative beneficial effects of these antibiotics observed in this trial were probably not mediated through any action against chlamydia. Instead, they postulate an anti-amyloid effect based on in vitro studies using antibiotics such as rifampin. Since amyloid levels were not measured in this study, this hypothesis could not be confirmed.


The successful treatment of some forms of peptic ulcer disease (PUD) with antibiotics after the realization that Helicobacter pylori played a role in the pathogenesis of PUD provides a rationale for examining whether other diseases once thought to be unrelated to infection can be treated more effectively through antimicrobial therapies. However, the evidence supporting bacterial infection as a cause of AD is considerably less convincing than the associations between H pylori and PUD. If chlamydial infection does promote AD, however, relatively little is known about the mechanism. This lack of knowledge undoubtedly handicapped Loeb et al, whose study was otherwise well designed and executed despite the uncertainties inherent in a first attempt to use antibiotics to treat a neurodegenerative dementia. An attractive feature of the experimental design was the assessment of outcome several months after completing a 3-month treatment period. This reduces the likelihood that the effects observed were mediated through treatment of acute infections that are common in dementia patients, such as UTIs and pneumonias.

In this study, the primary cognitive outcome measure at 12 months was negative, although results at an interim time point (6 months) suggested possible benefit with antibiotic treatment. Positive results in other domains, such as behavior and function, did not temporally correlate with improvement in cognition. This study, therefore, neither proves nor disproves a role for antimicrobial therapy in the treatment of AD. Before additional trials with antimicrobials are undertaken, more research should be carried out examining the possible association between infections and AD, including studies of the mechanisms involved and their relationship to known aspects of the AD pathogenesis. — Norman R. Relkin, MD, PhD, Associate Professor of Clinical Neurology and Neuroscience, New York Presbyterian Hospital-Cornell Campus and Assistant Editor of Neurology Alert.