Is the microbe-dementia hypothesis finally ready for a treatment trial?

Could infections really have something to do with disorders that cause dementia? Several chronic diseases, such as peptic ulcer, certain cancers, and autoimmune conditions, have established infectious etiologies, either bacterial or viral, but mainstream cognitive disorder research has not fully considered microbial involvement. Evidence is emerging, but the size of the problem— millions of patients with Alzheimer disease (AD) worldwide, and increasing—compels searching for clues in the dark, and not only below the streetlight, especially if established and safe treatments are available. Thirty years ago, on the basis of the similar topographic distribution of AD and herpes encephalitis— in the temporal and frontal lobes, but also in the hippocampus—Ball speculated about a connection between dementia and herpesviridae. Thereafter, professor Ruth Itzhaki’s group in the United Kingdom has been stubbornly pursuing the idea: they have repeatedly argued for associations between herpes simplex virus (HSV) and AD, especially in individuals with the APOE e4 genotype, known to make one more vulnerable to AD. A plethora of studies elegantly link microbial effects to AD pathology: inflammation, formation of amyloid plaques, genetic host responses, and so forth. These pathogenetic mechanisms have been recently exhaustively reviewed. There are also increasing numbers of epidemiologic reports, not only of HSV, but also of other viruses, especially cytomegalovirus, which seems to play a substantial role in immunosenescence. In this issue of Neurology®, Katan et al. present further epidemiologic evidence of an association between infectious burden and cognitive function in the multiethnic and population-based Northern Manhattan Study. Although the authors primarily used a composite serologic measure of exposure to both bacterial (Chlamydia pneumoniae and Helicobacter pylori) and viral (cytomegalovirus and HSV-1 and -2) pathogens, closer analyses implied that the association was primarily driven by viral infectious burden. These results strengthen the idea of an association between herpesviridae and cognitive decline. However, is the evidence sufficient to consider new types of prevention and treatment? Moreover, what causes damage in the brain is uncertain: a specific pathogen? a collection of pathogens? interactions among pathogens? In the study by Katan et al., the infectious burden variable assumed an additive effect and therefore does not elucidate pathogen interactions or the possibility of competitive antagonism between or among pathogens. Further research should focus on the roles of combinations of pathogens in a way that allows for testing of specific interactions in relation to cognitive outcomes. Undoubtedly, demonstrating that old-age cognitive disorders, including AD, are slowly progressing diseases of viral etiology would revolutionize the dementia research field (and be Nobel Prize worthy). However, great challenges remain. Most studies to date have been cross-sectional and associative, which do not prove cause and effect. There may be important differences in the ability of various microbes to provoke amyloid substance aggregation, the pathognomonic feature of AD. Herpesviridae may be a more attractive candidate in this respect, given their ability to get into the brain and their potential interactions with APOE. However, in case– control studies of brain tissue, chlamydia and spirochetes actually seemed more robustly associated with AD than herpesviridae. Furthermore, during the 1990s C pneumoniae infection was enthusiastically greeted as a treatable, microbial cause of coronary artery disease. Failures in large antibiotic trials tempered the enthusiasm, and most now consider C pneumoniae infection possibly important, but only a single factor contributing to the development of atherosclerosis. As always in evidence-based medicine, a randomized controlled treatment trial would provide final proof. Interesting findings are emerging from schizophrenia research. Although valacyclovir treatment did not alleviate schizophrenia symptoms in patients seropositive for cytomegalovirus, a new study seems more promising. In a small test-of-concept study in individuals with schizophrenia, a group from Pittsburgh and Harvard demonstrated a favorable effect of valacyclovir 3 g on