By Makoto Ishii, MD, PhD
Assistant Professor of Neuroscience and Neurology, Feil Family Brain and Mind Research Institute, Department
of Neurology, Weill Cornell Medical College
SYNOPSIS: Twenty-two pairs of viral exposures associated with an increased risk of neurodegenerative diseases were identified and replicated across two national European biobanks.
SOURCE: Levine KS, Leonard HL, Blauwendraat C, et al. Virus exposure and neurodegenerative disease risk across national biobanks. Neuron 2022; Jan 11. doi: 10.1016/j.neuron.2022.12.029. [Online ahead of print].
Viral infections have long been associated with the development of neurodegenerative diseases, from the discovery of a link between the 1918 flu pandemic with postencephalitic parkinsonism to recent studies finding an increased risk of multiple sclerosis (MS) after Epstein-Barr virus exposure. In this study, Levine and colleagues sought to systematically examine the potential associations between viral exposures and neurodegenerative disease risks in older adults using two large national biobanks.
FinnGen (Finnish biobank with genotyping data from more than 300,000 individuals) contained time series data and was used as a discovery phase cohort, and the UK Biobank (genotyping data from nearly 500,000 individuals from the United Kingdom) with cross-sectional data was used as a replication phase cohort. Six neurodegenerative diseases (Alzheimer’s disease [AD], amyotrophic lateral sclerosis [ALS], generalized dementia, MS, Parkinson’s disease [PD], and vascular dementia) and 32 common viral illnesses, including influenza, viral encephalitis, meningitis, varicella-zoster virus, and viral warts, were examined.
Using ICD-10 codes, 45 significant neurodegenerative disease-virus associations were identified in FinnGenn, with 22 of these associations replicated in the UK Biobank. All viral associations had an increased risk of neurodegenerative diseases, with none associated with a protective effect, and most, but not all, viral associations (81%) were with viruses considered neurotropic.
Of the 22 associations that were found in both cohorts, the highest hazard ratio (HR) was seen between viral encephalitis and AD (discovery phase HR, 30.72; confidence interval [CI], 11.84-79.68; uncorrected P value, 1.89E-12; replication phase odds ratio [OR], 22.06; CI 5.47-88.94; uncorrected P value, 1.37E-05). In FinnGenn, 24 of 406 (5.9%) viral encephalitis cases developed AD, which was significantly higher than the general prevalence of AD in this population (less than 3%).
The highest number of viral associations was in the generalized dementia category, with six (viral encephalitis, other viral diseases, viral warts, all influenza, influenza and pneumonia, and viral pneumonia) that were found in both FinnGenn and the UK Biobank. Pneumonia caused by influenza was associated with five of the six neurodegenerative diseases (AD, ALS, generalized dementia, PD, and vascular dementia) in both FinnGenn and the UK Biobank. As a positive control, the previously published finding of an association between a preceding Epstein-Barr virus exposure and increased MS risk was replicated in FinnGenn but not in the UK Biobank, which may have been the result of differences in the use of diagnostic codes and the design of the two biobanks.
The authors then examined FinnGen to investigate if the viral infection could predate the neurodegenerative disease diagnosis. In general, the neurodegenerative disease risk was highest in the first year after a viral infection and decreased over time. Notably, some associations, such as generalized dementia and pneumonia caused by influenza, remained significant 15 years after the viral exposure.
Finally, to address the possibility of reverse causality or neurodegenerative diseases increasing the risk of acquiring viral infections, the viral risk after a diagnosis of neurodegenerative disease was examined. Although there was a moderate association between the risk of certain viral infections after a diagnosis of neurodegenerative disease, in general this was found to be significantly less than the risk of developing neurodegenerative disease after the viral exposure.
COMMENTARY
This important study provides further evidence linking several viral infections to the development of neurodegenerative diseases. Using longitudinal data in FinnGen, the strongest associations were in viral infections that preceded the neurodegenerative disease, which seemingly would suggest that viral infections may be a causative agent. However, since many neurodegenerative diseases occur several years to decades prior to clinical diagnosis, it is possible that even those individuals with a viral exposure 15 years prior to neurodegenerative disease onset already were in the early stages of the neurodegenerative disease. Therefore, it would be important to examine viral exposures even earlier in life. Additionally, it is not known if those with a viral exposure had worse clinical severity of the neurodegenerative disease compared to those who never had the viral illness.
Finally, it is interesting to speculate the possible mechanisms linking viral infections and neurodegenerative diseases. While one possibility is that viral infections cause a general neuroinflammation, another intriguing possibility is that viral infections could trigger the core pathogenic factors of the neurodegenerative disease. Supporting this possibility, recent published studies have found that amyloid-beta has endogenous antimicrobial properties. Therefore, a viral infection could lead to increased amyloid-beta production, which would help fight the viral infection but would be detrimental by increasing the burden of Alzheimer’s disease pathology.
