Cross-Sectional and Longitudinal Associations Between Treatment for Herpes Virus Infection and the Dispensing of Antidementia Medicines: An Analysis of the Australian Pharmaceutical Benefits Scheme Database

INTRODUCTION

Dementia, a complex clinical syndrome characterized by various physiological and neuropathological changes, significantly impacts both higher cortical abilities and functional capacity [1]. The global prevalence of dementia is projected to surge from 57.4 million cases worldwide in 2019 to 152.8 million cases by 2050 [2]. This trajectory brings forth substantial associated costs and health burdens, and highlights the urgent need for effective action in the coming years [3].

Despite comprehensive research into potential medications for dementia, existing symptomatic treatments are limited in both scope and effectiveness [4]. This underscores the importance of targeting risk-modifying interventions to reduce the prevalence of dementia worldwide. The 2020 report for Dementia Prevention, Intervention and Care emphasized various risk factors for dementia including: lower educational achievement, hearing loss, hypertension, alcohol use, obesity, social isolation, physical inactivity, diabetes, air pollution and depression [1]. Promisingly, studies have indicated the potential efficacy of such interventions [5].

Herpes simplex virus (HSV) types 1 and 2 are neurotropic viruses shown to have associations with dementia [6, 7]. HSV-1, in particular, has been strongly linked to Alzheimer’s disease (AD) [8]. Similarly, varicella zoster virus (VZV), another neurotropic member within the Herpesviridae family, has also been implicated in dementia development [9]. While a recent meta-analysis failed to confirm the association between VZV and the incidence of AD [10], population based cohort studies have suggested a 10% increase in the risk of dementia with VZV infection [9]. Additionally, herpes zoster ophthalmicus (HZO), affecting the trigeminal nerve, has been associated with a 3-fold increased risk of dementia [10].

Notably, the treatment of HSV shows promise in reducing dementia risk [11, 12]. Registry-based studies in Taiwan, Sweden and Korea have reported reduced dementia risk associated with antiviral treatment for both symptomatic HSV and herpes zoster infections [13, 14]. However, this association has not been consistent across studies [15, 16], possibly because of sampling issues, confounding and differences in study design. Emerging research has also suggested that shingles vaccination against Herpes Zoster may reduce the incidence of dementia [17, 18]. Drugs currently approved for the treatment of HSV/VZV-associated diseases in Australia include acyclovir, its oral pro-drug valacyclovir, and famciclovir.

Due to the high prevalence of HSV/VZV infections in the adult population [19, 20], emphasizing antiviral management and vaccination may play a significant role in decreasing the risk of AD. We used data from the Pharmaceutical Benefits Scheme (PBS) to investigate if antiviral treatment of HSV/VZV is associated with a reduction in the prevalence and incidence of dementia in the Australian Population as measured by the dispensing of anti-dementia medicines.

METHODS

RESULTS

The 2013–2022 10% PBS sample contained 4,891,920 records related to 723,151 men and women aged 60 years or over. In 2022 6,868 (1.2%) of 559,561 individuals were dispensed an antidementia medicine. Table 1 shows their distribution by age, sex, and the dispensing of medicines for the management of diabetes, hyperlipidemia, hypertension, ischemic heart disease, and herpes. The odds ratio of being dispensed antidementia medicines among those dispensed pharmacological treatment for herpes (OR 0.73, 99% CI = 0.56–0.95) after adjustments were made for age group, sex, and the dispensing of medicines to treat diabetes, hyperlipidemia, hypertension, and ischemic heart disease. Individuals who were dispensed antiherpetic medications were more likely to be female, and have medical comorbidities such as diabetes, hyperlipidemia, and hypertension (Supplementary Table 1).

A multilevel logit model using all data available between 2013 and 2022 for the 723,151 individuals aged ≥60 years in the 10% PBS sample showed that the dispensing of HSV medicines was associated with decreased odds of dispensing of medicines for dementia (OR = 0.87, 99% CI = 0.75–1.00), adjusted for age group, sex, and the dispensing of medicines to treat diabetes, hyperlipidemia, hypertension, and ischemic heart disease. Longitudinal analysis of the data using split time span series showed an adjusted HR of 0.98 (99% CI = 0.89–1.07), which remained largely unchanged when we excluded participants dispensed antidementia medicines within the first 3 and 5 years of follow up (Table 2). Among participants dispensed antidementia agents between 2014–2022, those who had been dispensed HSV treatment beforehand were older when they were first dispensed antidementia medicines than those not dispensed HSV treatment (80.5±6.8 versus 79.8±7.4, p < 0.01).

DISCUSSION

Our study found suggestive, but inconclusive evidence that use of antiviral medications for HSV and VZV is associated with a decreased risk of dementia. Time-series analyses spanning from 2013 to 2022 indicated that older Australians aged over 60 years, who were dispensed antiherpetic medications, were 13% less likely to be dispensed antidementia medications. Based on cross-sectional data, individuals dispensed antiherpetic medications were 27% less likely to also receive antidementia drugs compared to those not dispensed antiherpetic medicines, although the direction of the association was unclear. A time-series analysis covering a 10-year period showed that older individuals exposed to HSV treatment were 13% less likely to be dispensed medicines for the management of AD. The findings of our longitudinal analyses could not dismiss the null hypothesis regarding the association between exposure to HSV medicines and the dispensing of anti-dementia agents.

Our findings are generally consistent with those of previous population-based studies conducted across various administrative databases, investigating the impact of antiherpetic medications on dementia risk. Tzeng and colleagues, using a Taiwanese administrative database, reported a 90% reduction in dementia risk with the use of anti-herpetic medications [14]. Other population studies utilizing administrative databases from South Korea and Sweden described similar results, with individuals treated at least once with antiherpetic medications having 10% –24% lower risk of dementia compared to untreated individuals [13, 24]. A Swedish nested case-control study of people infected with HSV-1 (confirmed by positive IgG), found that individuals treated with antiherpetic agents had 71% lower odds of AD compared with controls [25]. Ongoing research in Sweden, involving a clinical trial of valacyclovir for early AD, is expected to provide further insights into dementia treatment and prevention [26].

The underlying mechanisms of the association between HSV infections and dementia remain unclear. Early research by Libikova et al. in 1975 provided initial insight, after noting an increased incidence of HSV 1 serum antibodies in cerebrospinal fluid of individuals with dementia [27]. Subsequent efforts were made to establish a link between herpes infection and AD using neuronal and glial cell cultures. Wozniak et al observed a direct induction of amyloid-β production by HSV1 in both cell cultures and mice, a characteristic feature of AD [28]. Subsequently, in 2008, Zambrano et al. demonstrated that HSV-1 infection of mouse neuronal cultures led to tau hyperphosphorylation, another hallmark feature of AD [29]. Expanding on these findings, in 2009, Wozniak et al. revealed the presence of HSV-1 DNA in 90% of amyloid-β plaques in postmortem brains from AD patients [30]. Moreover, they found that 72% of HSV1 DNA in the brain was concentrated within these plaques. Despite this compelling evidence, the mechanistic links between viral infection, AD genetic risk factors such as APOE4, and the etiology of AD remains elusive.

AUTHOR CONTRIBUTIONS

Stephanie Tan (Conceptualization; Writing – original draft; Writing – review & editing); Erin Kelty (Conceptualization; Data curation; Writing – review & editing); Amy Page (Data curation; Project administration; Writing – review & editing); Christopher Etherton-Beer (Data curation; Writing – review & editing); Frank Sanfilippo (Data curation; Writing – review & editing); Osvaldo P. Almeida (Conceptualization; Data curation; Formal analysis; Writing – review & editing).