We describe herein two young patients who underwent epilepsy surgery following remote infectious encephalitis, and whose subsequent neuropathologic findings demonstrated phosphorylated tau accumulation in the context of chronic HSV encephalitis.
Case 1: A 15-year-old girl underwent surgical resection of the right fronto-parietal-insular region for treatment of drug-resistant epilepsy with progressive epileptic encephalopathy six years after a diagnosis of Herpes Simplex Virus-1 encephalitis (HSVE). The HSVE had initially been confirmed via HSV-1-positive cerebrospinal fluid polymerase chain reaction and she underwent a course of acyclovir intravenously. Post-operatively, an additional course of acyclovir was prophylactically administered. The resected tissue revealed a chronic, granulomatous lymphohistiocytic meningoencephalitis on microscopic examination. Real-time polymerase chain reaction (PCR) detected HSV-1 viral DNA in the formalin-fixed paraffin-embedded sample of parenchyma, whereas immunohistochemical staining did not show evidence of HSV-1 or HSV-2 antigen. Several pyramidal cortical neurons contained basophilic cytoplasmic fibrillary inclusions, and phosphorylated tau immunohistochemistry (AT8 – Thermo Scientific; 1:200) confirmed the focal presence of neurofibrillary tangles, pre-tangles, and fairly dense neuropil threads (Fig. 1).
Figure 1: Resected cortical specimen samples from Case 1. Specimen consisted of right posterior frontal cerebral cortex and underlying subcortical white matter. Foci of parenchymal lymphohistiocytic granulomatous inflammation (a, b and c [CD68 immunohistochemistry]) and perivascular lymphocytic cuffing (d), as well as dystrophic parenchymal mineralization (closed arrow in B) are demonstrated. (e &f) Phosphorylated tau immunohistochemical staining demonstrates intracytoplasmic tau (open arrow in F) and neuropil thread staining (closed triangle in F). Phosphorylated tau staining was completed using AT8, 1:200; phospho-epitope Ser202.
Case 2: A 21-year-old woman, previously reported by Arnold et al., had undergone surgical resection of the left temporal lobe for treatment of drug-resistant focal epilepsy, five years after experiencing a reported non-specified viral encephalitis in her country of origin. Post-operative prophylactic acyclovir was not administered, and her course was complicated by HSVE reactivation and significant morbidity. Reference Arnold, Parkins, Hamilton, Hader and Federico1 The microscopic examination of resected tissue was compatible with chronic encephalitis due to HSVE. Real-time PCR detected HSV-1 DNA; there was no histological or immunohistochemical evidence of HSV-1/-2 inclusions or antigens. Considering the findings described in Case 1, phosphorylated tau staining was performed (ad hoc; immunohistochemistry protocol as in Case 1) and demonstrated neuronal cytoplasmic tau and neuropil threads in the cortex and to a lesser extent within the hippocampus, both of which were in close proximity to foci of parenchymal reaction and inflammation.
Discussion
These cases, with contrasting post-operative courses, shared pathologic findings of chronic inflammation and parenchymal HSV-1 DNA detectable by PCR only. Of note, the identification of HSV-1 DNA on PCR does not distinguish between stages of the virus life cycle, and the absence of viral antigen detection by immunohistochemistry likely indicates a non-active stage of HSV-1 infection. Reference Love, Perry, Ironside and Budka2 Remarkably, foci of abnormal phosphorylated tau accumulation, associated with neuropathologic chronic granulomatous HSVE, were demonstrated within regions of inflammation as well as seemingly uninvolved areas. These foci of accumulation appeared as cytoplasmic phosphorylated tau-immunopositive components, neurofibrillary tangles, pre-tangles, and neuropil threads.
To our knowledge, these represent the youngest cases reported of chronic HSVE associated with tau-related neuropathologic findings. Danics et al. recently reported a series of 13 autopsied cases surviving between 9 days and 6 years after HSVE. Reference Danics3 Eight cases demonstrated neuronal tau-immunoreactivity and neuropil threads, and in three of these, the phosphorylated tau was found in inflammation-associated regions. Reference Danics3 However, the youngest case found to have tau-related pathology was 41 years old at the time of death. Reference Danics3
HSV-1 primary infection and reactivation have been associated with abnormal phosphorylated tau accumulation in in vivo murine models and in vitro murine, monkey, and human models. Reference Harris and Harris4–Reference Álvarez, Aldudo, Alonso, Santana and Valdivieso6 HSV-1 in vitro infection models exposed to acyclovir treatment demonstrate dose-dependent reductions in phosphorylated tau accumulation. Reference Harris and Harris4,Reference Wozniak, Frost, Preston and Itzhaki5 Further investigation into the possibility that tau accumulation contributes to long-term neurologic sequelae among HSVE survivors may be warranted. Lastly, for patients with a past or suspected medical history of HSVE, prophylactic use of acyclovir during neurosurgical interventions should be considered.
Funding
None.
Competing interests
Dr JPA wishes to declare research grants from UBC Pharmaceutical and Canadian Institutes of Health Research, reimbursement from Jazz Pharmaceutical and EnlitenAI Inc., and leadership positions with the Canadian Society of Clinical Neurophysiologists and the Medical Therapeutics Committee of the Canadian League Against Epilepsy. None of the other authors have conflicts to declare.
Statement of authorship
All authors contributed equally in reviewing the final version of the manuscript and provided feedback. Dr KS, Dr JJ, Dr WH, Dr QX, Dr JPA, and Dr PF provided direct clinical care to these patients and provided details of the clinical history to complete the manuscript. Dr KS, Dr KL, and Dr JPA had the conceptual idea for the manuscript. Dr KS collected data and wrote the manuscript. Dr KL and Dr JPA revised the preliminary versions of the manuscript and provided valuable feedback to improve it.
Open access
