Abstract
Axicabtagene ciloleucel (AC) is an FDA-approved anti-CD19 autologous chimeric antigen receptor T-cell (CAR-T) therapy for refractory diffuse large B cell lymphoma (DLBCL). While its efficacy in DLBCL has been promising, neurotoxicity remains a significant concern. We present a case of a 22-year-old woman with chemotherapy-refractory DLBCL who exhibited Grade IV neurotoxicity in the setting of sepsis, after undergoing AC infusion. Despite prophylactic levetiracetam given per guidelines,1,2 she experienced a precipitous mental status decline on post-infusion day 8 (D8) followed by hypoxic respiratory failure in the setting of clinical status epilepticus on D11 and nonconvulsive status epilepticus (NCSE) on D18. While neuroimaging was unremarkable, EEG demonstrated diffuse slowing and 2.5–3 Hz generalized periodic discharges consistent with NCSE. Seizures were initially refractory to lorazepam, increasing doses of levetiracetam, and phenobarbital, requiring a midazolam drip titrated to 50–70% burst suppression for resolution. Methylprednisolone and tocilizumab were used to treat neurotoxicity and cytokine release syndrome, respectively. Empiric antibiotics were used for sepsis. After cessation of sedatives on D19, mental status improved to near baseline. PET/CT just prior to discharge showed a complete response of the DLBCL (Deauville 3). She was discharged on D37 with no further seizure activity. Unfortunately, a 3-month interval PET/CT demonstrated disease progression which continued through salvage pembrolizumab eventually leading to death 1.2 years post-CAR-T infusion. This case illustrates the clinical management challenges of a complex and rare neurotoxic side effect of CAR-T cell therapy, namely NCSE following status epilepticus.
Keywords: seizures, epilepsy, status epilepticus, epilepsy, neurotoxicity, syndromes, neuroimmunology, clinical specialty, neurohospitalist, clinical specialty
Introduction
Despite the promising results of axicabtagene ciloleucel (AC), 3 the first FDA approved anti-CD19 autologous chimeric antigen receptor (CAR)-T cell therapy against refractory diffuse large B-cell lymphoma (DLBCL), the potential for neurotoxicity, which can in rare instances be severe, continues to present challenges. Immune effector cell associated neurotoxicity syndrome (ICANS) 4 includes acute delirium, tremor, global aphasia, and more severe neurotoxic effects (e.g., seizures, status epilepticus, cerebral edema). AC-associated neurotoxicities occur in 64% of patients within the first 8 weeks following treatment and persist for a median duration of 11-17 days. While 11-28% of these patients experience Grade ≥3 toxicities, approximately 10% have non-convulsive status epileptics (NCSE).1,5 Neurotoxicities are less predictable compared to the more common cytokine release syndrome (CRS)1,2 that includes fever (T ≥ 38°C), hypoxia, hypotension, chills, diarrhea, and myalgias.
While grading scales and protocols exist to help clinicians manage CRS, there are limited data to predict and manage the neurological complications. Further, it is often clinically challenging to distinguish encephalopathy secondary to a high CRS fever from the acute confusional state of ICANS. This case illustrates the clinical management challenges of a rare neurotoxic side effect – NCSE following status epilepticus – in DLBCL not previously described in the literature. It highlights the importance of prompt seizure identification and control, identifying concomitant seizure triggers including sepsis, and uses a multi-disciplinary approach to patient care.
Case Description
A 22-year-old woman with history of chemotherapy-refractory primary mediastinal B cell lymphoma (rPMBCL) presenting with fatigue, weight loss, and night sweats was diagnosed with refractory DLBCL after failing to respond to multiple previous therapies (EPOCH-R, 6 R-DHAP, 7 and mediastinal radiation). Following apheresis, she received bridging therapy with methylprednisolone, high-dose cyclophosphamide, and brentuximab vedotin. As the lymphoma continued to progress, she received full-dose cyclophosphamide and fludarabine lymphodepletion followed by AC infusion (Day 0 [D0], 2×106 CAR-T cells/kg). In the week prior to lymphodepletion, she exhibited septic symptoms (Tmax 39°C, 88/65 mmHg, SpO2 88%) and progressive, non-malignant, aseptic, bilateral exudative pleural effusions. Urinalysis and procalcitonin were unremarkable while CRP (mg/dL)/lactate (mmol/L)/LDH (U/L)/ferritin (ng/mL) were elevated to 17.8/2.9/864/2,995 respectively. She was managed with vancomycin, cefepime, daily thoracentesis, and eventually indwelling bilateral pleural catheters for effusion control. Blood and pleural fluid cultures showed no growth of microorganisms.
Following CAR-T infusion, she remained neurologically intact (9-10 out of 10 on immune effector cell-associated encephalopathy [ICE] assessment 4 ) for the first 7 days. On D8, she had a marked neurological decline. She was tachycardic (∼160 beats/min) without acute metabolic derangements. Neurological exam revealed disorientation, naming difficulty, and involuntary fast frequency low amplitude involuntary asynchronous movement of the face and bilateral upper extremities present at rest and increased with extension of the upper extremities. Her ICE score decreased from 9 to 3 within 4 hours. Based on these symptoms, she met criteria for Grade III ICANS with concurrent Grade I CRS, 4 which was managed per CAR-T treatment guidelines.1,2 Saline boluses, a dose of tocilizumab (IL-6 receptor antagonist) 8 mg/kg IV, dexamethasone 10 mg IV q6 h, and continuation of prophylactic levetiracetam 500 mg b.i.d improved her ICE score to 8 within 4 hours. However, persistent deficits in attention, word finding, and writing continued over the next 1.5 days. Head computed tomography (CT) and brain magnetic resonance imaging (MRI) were unremarkable, and lumbar puncture was attempted unsuccessfully. A continuous video electroencephalogram (cEEG) demonstrated moderate diffuse slowing with rare generalized rhythmic delta activity (GRDA) suggesting generalized encephalopathy without seizures (Figure 1). The cEEG, without seizures or epileptiform discharges, was discontinued after 72 hours.
Figure 1.
Summary of 42-day hospital course. Diffuse large B-cell lymphoma (DLBCL), drip (gtt), immune effector cell-associated encephalopathy (ICE) score, intensive care unit (ICU), non-convulsive status epilepticus (NCSE), spontaneous breathing trial (SBT), unable to assess (UTA).
On D11 shortly after cEEG discontinuation, she developed multiple bilateral tonic-clonic seizures described as repeated bilateral limb contractions with perioral cyanosis and generalized mottled skin. A clinical diagnosis of convulsive status epilepticus was made following her inability to regain awareness between episodes 8 with subsequent hypoxic respiratory failure requiring rapid sequence intubation. In the ICU, convulsions resumed, and persisted despite intravenous administration of 4 mg of lorazepam, 1,500 mg of levetiracetam, and a 20 mg/kg phenobarbital load; however, they ceased after initiation of a midazolam infusion drip that was quickly escalated to 10 mg/hr ( Table 1 ). Initial cEEG on midazolam 10mg/hr demonstrated a burst suppression pattern without epileptiform or seizure activity. The midazolam drip was titrated to 50–70% burst suppression and discontinued on D12. There were no signs of seizure or epileptiform activity on cEEG after midazolam discontinuation. Phenobarbital levels were also within therapeutic limits, and a repeat head CT was unremarkable. Empiric antimicrobials were broadened for possible meningitis and, per protocol for Grade IV ICANS, she received methylprednisolone 1,000 mg/day ×3 days. Although she was more alert and able to again follow simple commands on D13, her mental status fluctuated. From D13–16, she remained off sedation but intubated due to tachypnea and unsuccessful weaning trials in the setting of ongoing non-malignant, aseptic, bilateral exudative pleural effusions. She was maintained on levetiracetam 1,500 mg b.i.d. and phenobarbital 3 mg/kg/day.1,2
Table 1.
Immune Effector Cell Associated Neurotoxicity Syndrome (ICANS), Cytokine Release Syndrome (CRS), Medication, and Anti-Seizure Drug Levels
| Days post CAR T-cell infusion | Medication | Drug levels (µg/mL) |
|---|---|---|
| D0 | Levetiracetam 500 mg b.i.d | |
| D8 (ICANS Grade III, CRS Grade I) | Levetiracetam 500 mg b.i.d | |
| Tocilizumab 8 mg/kg IV Q4–6 h PRN (single dose) | ||
| Dexamethasone 10 mg IV Q6h | ||
| D9–10 | Levetiracetam 500 mg b.i.d | |
| Dexamethasone 10 mg IV Q6h | ||
| D11 (ICANS Grade IV) convulsive status epilepticus | Levetiracetam 500 mg b.i.d + 1,500 mg loading | |
| Levetiracetam increased to 1,500 mg b.i.d | ||
| Lorazepam 4 mg | ||
| Midazolam 2 mg during rapid sequence intubation | ||
| Phenobarbital 20 mg/kg loading + 3 mg/kg/day t.i.d | ||
| Midazolam gtt 10 mg/hr | ||
| Methylprednisolone 1 g daily × 3 days | ||
| D12 | Levetiracetam 1,500 mg b.i.d | P: 29.2 |
| Phenobarbital 3 mg/kg/day t.i.d | ||
| Midazolam weaned to 4 mg/hr gtt then discontinued | ||
| Methylprednisolone 1,000 mg daily | ||
| D13–16 | Levetiracetam 1,500 mg b.i.d | P: 30.1 |
| Phenobarbital 3 mg/kg/day t.i.d | P: 27 | |
| Methylprednisolone 1,000 mg daily, tapered to 125 mg daily | P: 26.2 | |
| D17 | Levetiracetam increased to 2,000 mg b.i.d | |
| Phenobarbital 3 mg/kg/day t.i.d | ||
| Methylprednisolone 250 mg for mild AMS | ||
| D18 (ICANS Grade IV) NCSE | Levetiracetam 2,000 mg b.i.d | P: 21.1a |
| Lorazepam 4 mg | L: > 100+ | |
| Phenobarbital 10 mg/kg loading + 3mg/kg/day t.i.d | ||
| Midazolam gtt 5–10 mg/hr | ||
| Tocilizumab 8 mg/kg IV Q4–6 h PRN (single dose) | ||
| Methylprednisolone 1,000 mg daily × 3 days | ||
| Cefepime discontinued | ||
| D19–22 | Levetiracetam 2,000 mg b.i.d | P: 22.1-31.3 |
| Phenobarbital 3 mg/kg/day t.i.d | L: 12.8 | |
| Methylprednisolone 1,000 mg daily, tapered to 500 mg daily | ||
| D23 | Levetiracetam 2,000 mg b.i.d | P: 20.4 |
| Phenobarbital 80 mg b.i.d | ||
| D33–41 | Phenobarbital taper and discontinuation | |
| Levetiracetam 2,000 mg b.i.d | ||
| D86–450 (death) | Levetiracetam tapered to 500 mg b.i.d |
a Level abnormally elevated. +Drug level trough. Altered mental status (AMS), bis in die/2 times a day (b.i.d), day number post-CAR-T infusion (D), drip (gtt), hour (hr), levetiracetam (L), non-convulsive status epilepticus (NCSE), phenobarbital (P), within normal limits (wnl).
On D17 the patient’s mental status began to decline. She became unresponsive to verbal commands, no longer made eye contact, and had periodic eyelid fluttering movements. Levetiracetam was increased to 2,000 mg b.i.d. due to concerns for subclinical seizures. Further neurological decline over the next 12 hours prompted a repeat cEEG which demonstrated sharply contoured 2.5–3 Hz continuous generalized periodic discharges (GPDs, Figure 2) consistent with generalized NCSE according to the Salzburg Consensus Criteria. 9 The patient was treated with lorazepam 4 mg and re-loaded with a single dose of phenobarbital 10 mg/kg (Table 1). As NCSE persisted, she was then placed back on a midazolam drip, with resolution of NCSE 60 minutes after cEEG was initiated. At this time, 1,000 mg methylprednisolone and 8 mg/kg tocilizumab were administered per grade 4 neurotoxicity guidelines1,2 and cefepime discontinued due to concerns of potential contribution to encephalopathy, depressed level of consciousness, and lowered seizure threshold. The midazolam was discontinued on D19, and her mental status improved immediately thereafter. She reached a full ICE score of 10 by D24 with no further clinical or electrographic seizures.
Figure 2.
Day 8, 17–18 post-CAR-T infusion continuous video electroencephalography. (A) Day 8: generalized delta frequency slowing (longitudinal bipolar-double banana montage). (B) Day 18: sharply contoured 2.5–3 Hz generalized periodic discharges with frontal predominance consistent with non-convulsive status epilepticus (longitudinal bipolar-double banana montage).
[18F]FDG PET/CT scan on D30 showed complete therapeutic response of the DLBCL (Figure 1). She was discharged on D37 on levetiracetam 2,000 mg b.i.d. and phenobarbital 80 mg b.i.d. Phenobarbital was gradually tapered and discontinued on D41. DLBCL recurrence was noted 3 months following discharge. Biopsy confirmed that the tumor no longer expressed CD19. She had a partial response to pembrolizumab and was then prescribed brentuximab. She remained seizure-free on levetiracetam 500 mg b.i.d until her death due to progressive DLBCL, 1.2 years after the AC infusion.
Discussion
This case illustrates that ICANS may result in protracted clinical seizure and NCSE in a somewhat delayed fashion (convulsive status epilepticus on D11; NCSE on D17) after a period of improving neurologic function. Importantly, NCSE was effectively managed and produced sustained clinical improvement, a scenario not previously described in the CAR-T literature. The seizure risk was increased by numerous factors, most notably ongoing sepsis without bacteremia or electrolyte derangements. As part of pre-CAR-T infusion chemotherapy, the patient also received fludarabine and cyclophosphamide, with the former known to be associated with delayed onset neurological symptoms and potentially seizures. 10 However, the clinical context of ongoing CRS and encephalopathy make ICANS the most likely cause for her medically refractory status epilepticus.
The incidence and severity of ICANS increases with higher peak T-cell expansion, increased tumor burden, and malignancy subtype, with variations among CAR-T cell products. The proposed pathophysiology of CAR-T neurotoxicity relates to the direct and indirect systemic inflammatory responses leading to increased blood brain-barrier permeability and elevated CNS levels of pro-inflammatory cytokines, a mechanism similarly described in sepsis. 11 The development of seizures and GPDs after CAR-T therapy is likely due to elevated cytokine levels leading to thalamocortical hyperexcitability. 12
Septic encephalopathy is estimated to affect more than half of sepsis patients with symptoms ranging from delirium to seizures to coma. 13 In septic ICU patients, the incidence of electrographic seizures is 15–19%, and EEG frequently demonstrates GPDs on EEG, a pattern arguably associated with a high risk of seizures. 14 In ICANS, the most frequently encountered EEG pattern was diffuse slowing consistent with encephalopathy. 1 Approximately 10% of patients exhibited NCSE with <5% developing NCSE after convulsive status epilepticus as seen in this patient. 1 Extended exposure to cefepime and hypoxemia secondary to recurrent bilateral pleural effusions may have contributed, as they are also independently known to increase seizure risk. Although the incidence of severe neurotoxicity is quite low in patients receiving CAR-T cell therapy, it is reasonable to pursue cEEG monitoring in those with prolonged encephalopathy or clinical worsening after an initial period of improving ICANS, particularly with concurrent hypoxemia and/or sepsis.
Seizure prophylaxis with levetiracetam is based on the CAR-T cell therapy-associated TOXicity (CARTOX) Working Group recommendation. 2 Levetiracetam is used as a first-line therapy because it does not interact with other medications and does not affect cytokine levels. 1 There is no evidence indicating that levetiracetam prophylaxis impacts the incidence of seizures or patient outcomes in CAR-T therapy. While the utilization of phenobarbital for second-line seizure suppression was in-line with management guidelines, 1 it is not otherwise favored as the therapy of choice due its side-effects and potential for interactions with other medications. Because our patient had severe and prolonged ICANS, she received both high-dose corticosteroids and 2 doses of tocilizumab. There is a mechanistic concern that immunosuppressive interventions may reduce the long-term efficacy of CAR-T therapy, though this has not been demonstrated in early intervention cohorts receiving corticosteroids and tocilizumab following anti-CD-19 CAR-T therapy. 15 Our patient achieved a complete response on follow-up PET imaging, and had substantial neurological recovery following hospitalization. At the time of DLBCL recurrence, lymph noted biopsy showed no expression of CD19, indicating that recurrence occurred via an escape mechanism such as antigen loss or modulation rather than ineffectiveness of AC therapy. 16
Declaration of Conflicting Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
Search Terms: Neurohospitalist, Neuroimmunology, Neurotoxicity Syndromes, Seizures, Status Epilepticus.
ORCID iD
Rosyli Reveron-Thornton, MS
https://orcid.org/0000-0002-4943-581X
References
- 1.Neelapu SS, Tummala S, Kebriaei P, et al. Chimeric antigen receptor T-cell therapy—assessment and management of toxicities. Nat Rev Clin Oncol. 2018;15(1):47–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Neelapu SS, Tummala S, Wierda W, et al. Chimeric antigen receptor (car) cell therapy toxicity assessment and management—Adult. 2017; https://www.mdanderson.org/documents/for-physicians/algorithms/clinical-management/clin-management-cytokine-release-web-algorithm.pdf. Accessed September 12, 2019.
- 3.Neelapu SS, Locke FL, Bartlett NL, et al. Axicabtagene ciloleucel CAR T-cell therapy in refractory large b-cell lymphoma. N Engl J Med. 2017;377(26):2531–2544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lee DW, Santomasso BD, Locke FL, et al. ASTCT Consensus Grading for Cytokine Release Syndrome and Neurologic Toxicity Associated with Immune Effector Cells. Biol Blood Marrow Transplant. 2019;25(4):625–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chavez JC, Bachmeier C, Kharfan-Dabaja MA. CAR T-cell therapy for B-cell lymphomas: clinical trial results of available products. Ther Adv Hematol. 2019;10:2040620719841581–2040620719841581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Giulino-Roth L, O’Donohue T, Chen Z, et al. Outcomes of adults and children with primary mediastinal B-cell lymphoma treated with dose-adjusted EPOCH-R. Br J Haematol. 2017;179(5):739–747. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Schirmbeck NG, Mey UJ, Olivieri A, et al. Salvage chemotherapy with R-DHAP in patients with Relapsed or refractory non-hodgkin lymphoma. Cancer Invest. 2016;34(8):361–372. [DOI] [PubMed] [Google Scholar]
- 8.Trinka E, Cock H, Hesdorffer D, et al. A definition and classification of status epilepticus--report of the ILAE task force on classification of status epilepticus. Epilepsia. 2015;56(10):1515–1523. [DOI] [PubMed] [Google Scholar]
- 9.Beniczky S, Hirsch LJ, Kaplan PW, et al. Unified EEG terminology and criteria for nonconvulsive status epilepticus. Epilepsia. 2013;54(suppl 6):28–29. [DOI] [PubMed] [Google Scholar]
- 10.Lowe KL, Mackall CL, Norry E, Amado R, Jakobsen BK, Binder G. Fludarabine and neurotoxicity in engineered T-cell therapy. Gene Ther. 2018;25(3):176–191. [DOI] [PubMed] [Google Scholar]
- 11.Gust J, Hay KA, Hanafi LA, et al. Endothelial activation and blood-brain barrier disruption in neurotoxicity after adoptive immunotherapy with CD19 CAR-T Cells. Cancer Discov. 2017;7(12):1404–1419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Herlopian A, Dietrich J, Abramson JS, Cole AJ, Westover MB. EEG findings in CAR T-cell therapy-related encephalopathy. Neurology. 2018;91(5):227–229. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Helbing D-L, Böhm L, Witte OW. Sepsis-associated encephalopathy. CMAJ. 2018;190(36):E1083–E1083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Azabou E, Magalhaes E, Braconnier A, et al. Early standard electroencephalogram abnormalities predict mortality in septic intensive care unit patients. PLoS One. 2015;10(10):e0139969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gardner RA, Ceppi F, Rivers J, et al. Preemptive mitigation of CD19 CAR T-cell cytokine release syndrome without attenuation of antileukemic efficacy. Blood. 2019;134(24):2149–2158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Shah NN, Fry TJ. Mechanisms of resistance to CAR T cell therapy. Nat Rev Clin Oncol. 2019;16(6):372–385. [DOI] [PMC free article] [PubMed] [Google Scholar]


