Mosunetuzumab-associated fatal HHV-6 encephalitis in a patient with follicular lymphoma

Introduction

Follicular lymphoma is the most common indolent B-cell lymphoma and typically follows a relapsing-remitting course. Grade 3A follicular lymphoma represents a biologically and clinically distinct subtype characterized by increased proliferation and the need for a more aggressive therapy. ¹ While most patients respond to initial chemoimmunotherapy, many patients experience relapse and become treatment-refractory. ² Mosunetuzumab, a bispecific monoclonal antibody that engages CD3 on T-cells and CD20 on B-cells, has demonstrated high efficacy in patients with relapsed/refractory follicular lymphoma. We present herein a case of HHV-6 encephalitis in a patient treated with mosunetuzumab.

Case presentation

A 76-year-old woman with grade 3A follicular lymphoma diagnosed in July 2015 (Ann Arbor stage IVA, with bone marrow and pleural involvement) was treated initially with six cycles of bendamustine and rituximab, followed by rituximab maintenance therapy until November 2017. She experienced a relapse in May 2020, confirmed by biopsy (Ki-67 of 67%; BCL2 positive, BCL6 and MYC negative), and was treated with six cycles of rituximab, cyclophosphamide, etoposide, vincristine, and prednisone (without doxorubicin due to the patient's alcoholic cardiomyopathy and reduced ejection fraction). She achieved a near-complete response by January 2021 but relapsed again in July 2022. Restaging positron emission tomography/computed tomography (PET/CT) in October 2022 showed extensive adenopathy and splenic involvement. She was re-treated with bendamustine-rituximab from December 2022 through February 2023, and achieved a complete response by March 2023. However, progression in retroperitoneal and iliac lymph nodes was documented in October 2024.

She began IV mosunetuzumab monotherapy in November 2024, administered every three weeks per institutional step-up protocol (Cycle 1 Day 1: 1 mg, Cycle 1 Day 8: 2 mg; Cycle 1 Day 15: 60 mg, Cycle 2 Day 1: 60 mg, Cycle 3+ Day 1: 30 mg). Improvement was noted on a CT scan of chest, abdomen and pelvis in February 2025. IgG levels at that time were 552 mg/dL [reference range: 610–1616 mg/dL].

In March 2025, she presented to the emergency department with progressive confusion, anorexia, and generalized weakness. Her family noted last intact mentation ten days prior, and voiced that her symptoms started the day after receiving her 4th treatment cycle with mosunetuzumab and gradually worsened afterwards. On admission, she was oriented only to self and place, with reduced attention and intermittent agitation. Her vitals were within normal limits; examination revealed no focal neurologic deficits or meningeal signs. Initial laboratory studies revealed leukocytosis of 11.7 K/μL [reference range: 3.8–10.8 K/μL], baseline anemia at 10.8 g/dL [reference range: 12–16 g/dL], normal platelets at 165 K/μL [reference range: 150–450 K/μL], lymphopenia at 0.4 K/uL [reference range: 1.2–3.7 K/uL], elevated LDH at 348 U/L [reference range: 140–271 U/L], elevated beta-2 microglobulin at 0.353 mg/dL [reference range: 0.097–0.184 mg/dL] and baseline chronic kidney disease with GFR at 31.0 mL/min/1.73 m2 [reference range: >60 mL/min/1.73 m2]. Liver functions were unremarkable. Non-contrast CT scan of the head was negative. Initial plan of care involved psychiatry evaluation for assessment of delirium. Given her history of alcohol use and malnutrition, thiamine 500 mg IV three times daily was initiated for presumed Wernicke's encephalopathy.

Due to persistent encephalopathy and new onset low-grade fever (37.8°C), a hematology-oncology consult was requested and a brain MRI with and without contrast was obtained on hospital day 4. Imaging revealed new T2/FLAIR hyperintensity in bilateral mesial temporal lobes, left parahippocampal gyrus, bilateral hypothalami, thalami, basal ganglia, caudate nuclei, putamina, and pons, with restricted diffusion and heterogeneous enhancement in the left parahippocampal region (Figure 1(a) and (b)). These findings were concerning for viral or autoimmune encephalitis. ICANS was considered as well. Dexamethasone at 10 mg IV daily and empirical acyclovir 10 mg/kg IV were commenced.

OPEN IN VIEWER

Figure 1.

a, b legend: sequential axial T2-weighted MRI images in the index patient demonstrating diffuse symmetric hyperintensity involving the bilateral mesial temporal lobes (arrows), thalami, caudate nuclei, hypothalami, and brainstem.

A lumbar puncture was performed on hospital day 6. Opening pressure was 12 cm H₂O [reference range: 6–20 cm H₂O]. Cerebrospinal fluid (CSF) analysis revealed 28 nucleated cells/μL [reference range: 0–5 cells/μL], 22% neutrophils [reference range: 0–6%], 71% lymphocytes [reference range: 40–80%], 7% monocytes [reference range: 15–45%], protein 211 mg/dL [reference range: 15–45 mg/dL], IgG of 8.8 [reference range: 0.8–7.7 mg/dL] and glucose 82 mg/dL [reference range: 40–70 mg/dL], while serum glucose was 189 mg/dL [reference range: 70–105 mg/dl]. CSF meningoencephalitis PCR panel was positive for HHV-6 DNA; all other pathogens were negative (including Herpes simplex virus 1 [HSV1], Herpes simplex virus 2 [HSV2], Cytomegalovirus [CMV], Varicella zoster virus [VZV], enterovirus, human parechovirus, Cryptococcus neoformans, Streptococcus pneumoniae, Streptococcus agalactiae, Neisseria meningitidis, Listeria monocytogenes, Haemophilus influenzae, and Escherichia coli K1). CSF Gram stain, Eastern equine encephalitis antibody, venereal disease research laboratory (VDRL), West Nile virus antibody and paraneoplastic antibodies panel were negative as well. An electroencephalogram (EEG) done on hospital day 7 demonstrated moderate diffuse cerebral dysfunction without seizures.

Management & outcome

After onset of symptoms, mosunetuzumab was permanently discontinued and given the patient's CSF meningoencephalitis PCR panel being positive for HHV-6 DNA, acyclovir was switched for intravenous ganciclovir at 1.25 mg/kg daily (renally dosed). Infectious disease team was consulted and initiated broad-spectrum antibiotics, which were later de-escalated to monotherapy with cefepime once all CSF studies resulted negative for bacterial infection. On hospital day 7, the patient experienced an episode of pulseless electrical activity. She received 5 min of cardiopulmonary resuscitation, was intubated and transferred to the intensive care unit. Post-resuscitation, she remained comatose, with loss of brainstem reflexes. After interdisciplinary discussion with her family, care was transitioned to comfort-focused measures. She was transitioned to inpatient hospice and died on hospital day 12.

Discussion

This case illustrates the diagnostic dilemma of differentiating ICANS from viral encephalitis in a patient receiving T-cell–redirecting therapy and underlying severe lymphopenia.

A modern BiTE therapy, mosunetuzumab showed an objective response rate of 80% with 60% complete responses in heavily pretreated follicular lymphoma patients in a pivotal phase 2 trial. ³ However, this immune-based approach causes significant immunosuppression and carries unique toxicities. Cytokine release syndrome (CRS) and ICANS are well-recognized complications, particularly in the setting of CAR T-cell therapy. Although mosunetuzumab is associated with a lower rate of neurotoxicity, rare cases of ICANS have been described with its use as well. In clinical trials, grade ≥3 neurotoxicity has been reported in 1–3% of patients on mosunetuzumab, which is significantly lower than the 28% incidence reported with CAR T-cell therapies.^3–5

ICANS manifests as a diffuse encephalopathy and may feature confusion, aphasia, seizures, and cerebral edema, typically presenting within days of immune effector therapy, often in conjunction with CRS. These features may overlap with the ones of infectious encephalitis. ⁶ Differentiating ICANS from viral encephalitis is vital, as management strategies diverge significantly: corticosteroids and immunosuppression are the backbone of ICANS therapy, whereas viral encephalitis requires urgent antiviral treatment. Imaging in ICANS is frequently normal; EEG may show diffuse slowing. CSF is typically unremarkable, although mild protein elevation and/or lymphocytosis may occur.^4,5,7

In contrast, HHV-6 encephalitis is caused by a neurotropic virus that lies latent in most adults but can reactivate under immunosuppressive conditions and preferentially affects limbic structures.^8,9 In a prospective cohort, HHV-6 encephalitis reactivation, defined as new detection of viral DNA in blood or CSF via PCR within 12 weeks following CAR T-cell infusion was observed in approximately 6% of recipients, all of whom had presumed prior latency from childhood exposure. Among these, new-onset HHV-6 encephalitis was rare, with an incidence of only 0.17%^7,8

Our patient with bilateral temporal lobe encephalitis secondary to HHV-6 presented with time-based association of 12 weeks of mosunetuzumab therapy and initial symptoms compatible with ICANS, making this syndrome a reasonable early diagnostic consideration. However, the MRI changes, EEG findings, lack of preceding high-grade CRS, and significant CSF pleocytosis raised suspicion for an alternative etiology. The positive CSF HHV-6 PCR confirmed viral encephalitis. Naranjo score of 6 suggests a probable mosunetuzumab-associated encephalitis, which in this case was secondary to an infectious etiology (Table 1).

OPEN IN VIEWER

Table 1.

The Naranjo adverse drug reaction (ADR) probability scale questionnaire.

To assess the adverse drug reaction, please answer the following questionnaire and give the pertinent score	Yes	No	Do not know	Score
1. Are there previous conclusive reports on this reaction?	+1	0	0	+1
2. Did the adverse event appear after the suspected drug was given?	+2	−1	0	+2
3. Did the adverse reaction improve when the drug was discontinued or a specific antagonist was given?	+1	0	0	0
4. Did the adverse event reappear when the drug was re-administered?	+2	−1	0	0
5. Are there alternative causes that could have caused the reaction?	−1	+2	0	+2
6. Did the reaction reappear when a placebo was given?	−1	+1	0	0
7. Was the drug detected in blood (or other fluids) at toxic concentrations?	+1	0	0	0
8. Was the reaction worsened when the dose was increased or lessened when dose was decreased?	+1	0	0	0
9. Did the patient have a similar reaction to the same or similar drugs in any previous exposure?	+1	0	0	0
10. Was the adverse event confirmed by any objective evidence?	+1	0	0	+1
			Total	6

The Naranjo Adverse Drug Reaction Probability Scale is a validated, questionnaire-based algorithm designed to estimate the likelihood of a causal relationship between a drug and an observed adverse event. It incorporates weighted criteria including temporal correlation, alternative etiologies, drug concentration data, dose-response relationships, and prior patient experience. Based on the cumulative score, the adverse event is classified as definite (≥9), probable (5–8), possible (1–4), or doubtful (0). The assessment is performed on an individual drug basis and does not account for pharmacologic interactions; the presence of confounding factors may reduce the assigned probability of causality.

Standard treatment options for HHV-6 encephalitis includes parenteral foscarnet or ganciclovir, with variable outcomes.^8–10 In our case, ganciclovir was initiated promptly, but neurologic recovery did not occur, possibly due to delayed diagnosis, immunosuppression, and rapid disease progression. ¹¹ Though false positives may occur due to chromosomally integrated HHV-6, the patient's clinico-radiologic features supported active infection. ¹²

Importantly, this case demonstrates how overlapping clinical syndromes (ICANS, HHV-6, Wernicke's, other metabolic encephalopathy) can converge in complex immuno-oncology patients, especially in heavily pre-treated patients with underlying lymphopenia. While HSV-1 remains the most common cause of viral encephalitis in the elderly population, rare cases of HHV-6 reactivation have been reported following CAR T-cell therapy and allogeneic stem cell transplantation.^7,11 To date, no cases of viral encephalitis due to HHV-6 have been documented in the literature in association with bispecific T-cell engager (BiTE) therapy. Furthermore, given the scarcity of HHV-6 encephalitis post-CAR T-cell therapy and lack of documented cases post-BiTE therapy, routine prophylactic antivirals are not currently recommended. ⁸

Our case reinforces the importance of early lumbar puncture, brain MRI studies and EEG in any patient on BiTE therapy presenting with new encephalopathy as fatal outcomes can result from diagnostic delays, inappropriate steroid use or inconspicuous adverse drug reactions.