Glioblastoma (GBM) is a malignant tumor with a median survival of only 10–18 months even with standard treatment.
1 We recently encountered a patient with GBM who experienced a transient “remission” on acyclovir treatment.
CASE
A 67-year-old man with no past medical illness presented with fever and myalgia, followed by a single brief episode of focal to bilateral tonic-clonic seizure. The neurological exam revealed no focal neurological deficit apart from anterograde amnesia noted from the Mini-Mental State Examination (28/30). Electroencephalogram revealed a moderate amount of generalized slowing with no epileptiform discharge. On the fluid-attenuated inversion recovery (FLAIR) images, homogenous high signal intensities were observed mainly in the left medial temporal lobe (
Fig. 1A). The lesions were partially accompanied by diffusion restriction and extended to the left parieto-occipital lobe (
Fig. 1B,
C). Contrast enhancement was faintly present on the adjacent leptomeninges (
Fig. 1D). Cerebrospinal fluid (CSF) exam revealed 0 white blood cells/μL, protein 71.
6 mg/dL, and glucose 74 mg/dL. Polymerase chain reaction (PCR) was negative for herpes simplex virus (HSV) 1, HSV 2, varicella zoster virus,
Mycobacterium Tuberculosis, and cytomegalovirus (CMV) on two consecutive CSF samples obtained 6 days apart.
Although the clinical manifestation closely resembled that of HSE, negative CSF HSV PCR led to another possibility, a glioma. However, the lesions were mostly confined to the medial temporal lobe and were too dangerous to biopsy. Moreover, his symptoms were too minor. Therefore, the patient was first started on intravenous acyclovir treatment for 2 weeks, which surprisingly improved his symptoms and magnetic resonance imaging (MRI). MRI obtained 3 months after the treatment revealed markedly reduced FLAIR lesions, diffusion restriction, and leptomeningeal enhancement (
Fig. 1E-H). No new focus of disease was noted. Corticosteroids had not been administered.
He visited Boramae Medical Center 1 year later for headaches and memory loss that had developed 2 weeks prior. Recurrent episodes of nonconvulsive seizures were suspected. MRI showed extensive heterogenous FLAIR hyperintensities accompanied by a mass effect (
Fig. 1I). Contrast enhancement showed the breakdown of blood-brain barrier (
Fig. 1J). High values on apparent diffusion coefficient suggested a vasogenic edema (
Fig. 1K). In addition, parenchymal hemorrhage was newly noted on the susceptibility-weighted image (
Fig. 1L). Stereotactic biopsy performed from the lateral temporal lobe revealed anaplastic cells with microvascular proliferation and necrosis (
Fig. 2A-C). On immunohistochemistry (IHC), positive glial fibrillary acidic protein and negative isocitrate dehydrogenase-1 (IDH-1) confirmed the diagnosis of a glioblastoma, IDH-1 wild-type. IHC for p53 was negative and for ATP-dependent helicase (ATRX) was positive (no mutation). Nuclear viral cytopathic changes indicating HSV infection were not observed. In addition, IHC using antibodies to HSV1, HSV2, and CMV antigens were negative (
Fig. 2D).
DISCUSSION
Several clues suggest that temporal lobe lesions were not HSE but a glioma from the beginning. First, two consecutive CSF HSV PCR tests, the widely accepted test of choice for HSE with 98% sensitivity, were negative.
2 Second, initial MRI lesions outside the limbic regions are associated with lower odds of HSE.
3 Third, the absence of brain atrophy following antiviral treatment is rare in HSE.
4 Finally, there was no histopathological evidence of HSE in the biopsied brain. This raised a low possibility of coincident HSE. Altogether, it was assumed that a glioma was transiently suppressed by acyclovir. Based on the MRI features and the 1-year latency before aggravation, its grade would have been II or III, and later progressed to glioblastoma multiforme.
Theoretically, acyclovir could suppress regulatory T cell function and therefore amplify the immune reaction toward the tumor tissue.
5,6 However, clinical experience is largely limited to case reports, in which GBM patients were misdiagnosed and treated as HSE.
7-9 Moreover, some patients among them had received corticosteroids concomitantly; this made it difficult to judge the effect of acyclovir.
9 In addition, HSE occasionally coincides with gliomas; in these cases, acyclovir would act by reducing the viral load but not the size of the glioma.
10 In contrast, we suggest a clear-cut example in which we administered acyclovir without corticosteroid and objectified the improvement on a follow-up MRI.
Physicians should be cautious when diagnosing HSE, a typical clinical picture and a clear response to acyclovir does not guarantee the diagnosis. Paraclinical features should be reviewed carefully. As long as those findings are atypical for HSE, physicians should consider a brain biopsy. If it is not feasible, a long-term surveillance with frequent MRI check-ups is recommended.