Oyae 186

The Oncologist, 2024, XX, 1–13
https://doi.org/10.1093/oncolo/oyae186
Advance access publication 26 July 2024
Original Article
Immune-related encephalitis after immune checkpoint

inhibitor therapy
Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

Monica W. Buckley1,2, Aanika Balaji Warner3,4, , Julie Brahmer3,4, Laura C. Cappelli3,5,
William H. Sharfman3,4, Ephraim Fuchs3, Hyunseok Kang3,6, Patrick M. Forde3,4,
Douglas E. Gladstone3,7, Richard Ambinder3, Ronan J. Kelly3,8, Evan J. Lipson3,4, , Ivana Gojo3,
Edward J. Lee9, Tory P. Johnson1, Shiv Saidha1, Rafael Llinas1, Lyle W. Ostrow1,10,
Jarushka Naidoo3,4,11,12, John C. Probasco*,1,
1
Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, United States,
2
Department of Neurology, University of Virginia School of Medicine, Charlottesville, VA 22903, United States,
3
Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, United States,
4
Bloomberg-Kimmel Institute for Cancer Immunotherapy, Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, MD
21287, United States,
5
Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD 21287, United States,
6
Department of Medicine, University of California, San Francisco, San Francisco, CA 94143, United States,
7
R.J. Zuckerberg Cancer Center at Hofstra/Northwell Health, Lake Success, NY 11042, United States,
8
Charles A. Sammons Cancer Center, Baylor University Medical Center, Dallas, TX 75246, United States,
9
Maryland Oncology Hematology, Columbia, MD 21044, United States,
10
Department of Neurology, Lewis Katz School of Medicine at Temple University, Philadelphia, PA 19140, United States,
11
Department of Oncology, Johns Hopkins Bayview Medical Center, Baltimore, MD 21224, United States,
12
Department of Medicine, Beaumont Hospital Dublin and RCSI University of Health Sciences, Dublin, 9, Ireland
*
Corresponding author: John Probasco, MD, Johns Hopkins Hospital, 600 N. Wolfe Street, Meyer 6-113, Baltimore, MD 21287, USA ([email protected]).
Abstract
Background: Immune checkpoint inhibitors (ICI) have revolutionized cancer treatment but can trigger immune-related encephalitis. We report
one of the largest case series of patients with immune-related encephalitis and review of the literature.
Methods: Retrospective series of patients with immune-related encephalitis and literature review.
Results: Fourteen patients with cancer treated with ICI (50% combination therapy) developed immune-related encephalitis. Diagnostic testing
revealed cerebral spinal fluid (CSF) lymphocytic pleocytosis (85%) and elevated protein (69%), abnormal brain magnetic resonance imaging(MRI)
(33%) or brain FDG-PET (25%), electroencephalogram (EEG) abnormalities (30%), and autoantibodies (31%). Encephalitis treatment included:
corticosteroids (86%), intravenous immunoglobulin (IVIg) (36%), plasmapheresis (7%), and rituximab (29%). There were no deaths and 12
patients had significant recovery, although long-term complications were observed. All patients discontinued ICI. Longitudinal follow-up demon-
strated anti-cancer response to ICI at 3 months (85%) and 6 months post-ICI initiation (77%). A literature review identified 132 patients with
immune-related encephalitis. Most were treated with PD-1 inhibitors (18% combination). Common abnormalities included elevated CSF protein
(84%) or pleocytosis (77%), abnormal brain MRI (65%), or autoantibodies (47%). Nearly all were treated with corticosteroids, many required
additional therapy with IVIg (26%) or rituximab (12%). Most patients had clinical improvement (81%) but a minority (10%) had a clinical relapse
after completing corticosteroid taper. ICIs were resumed in 7 patients (5%), with relapse in 3.
Conclusions and relevance: Immune-related encephalitis is treatable and improves with corticosteroids in most cases but may require addi-
tional immunosuppression. Re-emergence of encephalitis is rare and does not typically result in adverse outcomes, and this should be consid-
ered in neurological immune-related adverse event management guidelines.
Key words: encephalitis; autoimmune; immune checkpoint; cancer.
Implications for practice

Encephalitis due to immune checkpoint inhibitors (ICE) is treatable and improves with corticosteroids in most cases but may require
additional immunosuppression. Re-emergence of this phenomenon is rare and does not typically result in adverse outcomes. While
current guidelines recommend halting ICE after irEncephalitis, continued research is needed to evaluate whether patients may resume
therapy.
Received: 29 January 2024; Accepted: 28 June 2024.

© The Author(s) 2024. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/),
which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
2 The Oncologist, 2024, Vol. XX, No. XX
Introduction included the subacute onset of memory deficits, altered men-

tal status, and/or psychiatric symptoms accompanied by at
Immune checkpoint inhibitors (ICI) are monoclonal antibod-
least one of the following: new focal neurological findings,
ies targeting pathways involved in maintenance of immune tol-
seizures, CSF pleocytosis, and/or brain MRI suggestive of
erance. These include antibodies against programmed death-1
encephalitis with exclusion of alternative causes.14 Patients
(PD-1), PD ligand-1, (PD-L1), and cytotoxic T-lymphocyte
who met criteria for possible autoimmune encephalitis were
antigen-4 (CTLA-4).1 ICI improve outcomes in patients
further evaluated to determine whether they met criteria for
with a variety of cancers, including melanoma, non-small
probable or definite autoimmune encephalitis. Patients with
cell lung cancer (NSCLC), and renal cell carcinoma (RCC),
neuronal cell-surface or onconeural autoantibodies detected
both in metastatic and early-stage settings.1-8 Immune-related
in the serum and/or CSF or typical findings of limbic enceph-

adverse events (irAEs) can affect any organ with higher inci-
alitis with either CSF pleocytosis or epileptiform activity
dences after combination.1,9,10 Neurological irAEs (nirAEs)
involving the temporal lobes were classified as having definite
affect 1.5% of patients treated with ICI with serious events
autoimmune encephalitis. Probable autoimmune encephalitis
occurring in 0.2%-0.8%. Serious nirAEs include encephalitis,
included patients who did not fulfill criteria for definite auto-
meningitis, meningoencephalitis, myasthenia, myositis, and
immune encephalitis but with at least 2 of the 3 supportive
vasculitis.11-13
findings: MRI suggestive of autoimmune encephalitis, CSF
Recently, a definition of immune-related encephalitis as
pleocytosis, presence of CSF-specific oligoclonal bands and/
a nirAE was proposed including diagnosis of autoimmune
or elevated IgG-index, or brain biopsy showing inflammatory
encephalitis and clinical improvement/stabilization with
infiltrates.14 Board-certified neurologist independently adjudi-
immunomodulation or discontinuation of ICI.13 In 2016,
cated the diagnosis (J.P.). Patient demographics, presentation,
experts proposed criteria for the diagnosis of autoimmune
diagnostic results, immunosuppressive treatments, neurolog-
encephalitis, incorporating subacute memory deficits with
ical outcomes, and response to ICI were collected by chart
supporting findings such as magnetic resonance imaging
review (M.B.; A.B., J.P.). First-line immunosuppressive treat-
(MRI) brain changes, new onset seizures, and/or cerebral spi-
ments for autoimmune encephalitis included intravenous (IV)
nal fluid (CSF) pleocytosis.14 Autoimmune encephalitis varies
methylprednisolone or oral prednisone, intravenous immuno-
clinically, but most commonly presents with subacute mem-
globulin (IVIg), and plasmapheresis (PLEX), while rituximab
ory deficits, confusion, mood or behavioral changes, seizures,
and cyclophosphamide were considered second-line ther-
and/or movement disorders and imaging is highly variable.14
apy.12 Antibody Prevalence in Epilepsy and Encephalopathy
MRI brain can be normal, but may evolve with emergence of
(APE2) and Responsive to Immunotherapy in Epilepsy and
fluid-attenuated inversion recovery (FLAIR) or T2 sequence
Encephalopathy (RITE2) scores were determined retro-
hyperintensities in the medial temporal lobes (typically bilat-
spectively (A.B., J.P.).17 Tumor response and durable clini-
eral in limbic encephalitis), gray matter, and white matter.
cal benefit to ICI were defined as a radiologic reduction in
Fluorodeoxyglucose (FDG)-PET may show hypermetabolism
size or stable disease on contrast-enhanced CT imaging at
in the mesiotemporal regions (such as in limbic encephalitis)
3 and 6 months post-ICI-initiation respectively and verified
or hypometabolism such as in the visual cortex in anti-N-
by an oncologist (J.N.). The severity of encephalitis for each
methyl-d-aspartate receptor (anti-NMDAR) encephalitis.15
patient was graded retrospectively per Common Terminology
Electroencephalogram (EEG) abnormalities include focal or
Criteria for Adverse Events (CTCAE) version 5.0.
generalized slow activity, epileptiform activity, fast super-
A literature search for immune-related encephalitis was per-
imposed on slow activity (delta brush observed in anti-
formed using PubMed, Embase, the Cochrane Library, Web
NMDAR encephalitis), or seizures. CSF abnormalities include
of Science, Scopus, and ClinicalTrials.gov from respective
pleocytosis, elevated protein, elevated IgG relative to serum,
database inception through October 2020. A search was con-
and neuronal autoantibodies.14 Autoimmune encephalitis can
structed for encephalitis and ICI therapy (supplement). The
be associated with malignancy in the absence of ICI treatment.14
search terms included controlled vocabulary, index terms, and
Here, we present a retrospective cohort of patients with
additional keywords. Non-English language publications were
immune-related encephalitis and review of the literature.
excluded. There were 2601 records identified, 1134 duplicates
We aim to determine whether patients fulfilled consensus
removed and 1467 records screened. All abstracts were eval-
criteria for autoimmune encephalitis, annotate management
uated by independent reviewers (M.B., J.P.) using Covidence
and response, and investigate long-term cancer and nirAE
systematic review software (Veritas Health Innovation,
outcomes.
Melbourne, Australia; www.covidence.org). Candidate studies
were evaluated by full text (Supplementary Figure S1). Careful
Methods assessment ensured patients reported in different studies were
We retrospectively identified patients hospitalized at the excluded from duplicate analysis. A total of 85 individual pub-
Johns Hopkins Hospital and Johns Hopkins Bayview lications and 132 patients were identified (Supplementary File
Medical Center between June 1, 2014 and July 31, 2020 S1). Each publication was reviewed to extract patient demo-
through a review of records and an institutional immune- graphics and clinical data. Using available data, the patients
related toxicity team consultation service.16 Patients and the were categorized as having a diagnosis of possible, probable,
public were not involved in the design, conduct, or report- or definite autoimmune encephalitis as defined above.14
ing of this study. Included patients fulfilled 2016 consensus
clinical criteria for diagnosis of possible, probable, or definite
Results
autoimmune encephalitis and had received at least 1 dose of
ICI as their most recent cancer therapy with stabilization or Patients
clinical improvement after immunomodulation or discontin- We identified 14 patients with immune-related encephali-
uation of ICI.14 Criteria for possible autoimmune encephalitis tis (Table 1).14 The median age was 57.5 years (IQR: 27.3
Table 1. Patients with immune-related encephalitis after treatment with immune checkpoint inhibitors.
Pt. Cancer ICI Symptoms on admission Other adverse events APE2 score/ Intervention Response Oncologic status at last follow
no. regimen and CTCAE grade for RITE2 score Time to recovery up, total follow up after ICI-
Time to ICI-related encephalitis (weeks) related encephalitis (weeks)
onset
1 Melanoma, Nivo/ • Headache, N/V, 8/10 • IV MP 1g x 3 d • Resolved • No Clinical Response:

Brain metas- ipi, 6 fevers • IV MP 1g x 5 d • 3 weeks Deceased due to progres-
tasis weeks • Gait impairment • Antibiotics, acyclovir sive cancer
• Confusion, AMS, • Levetiracetam • Total follow up ~15 weeks
agitation requiring • Steroid taper
intubation
• R arm twitching
• CTCAE 3
The Oncologist, 2024, Vol. XX, No. XX
2 SCC of skin Pembro, • Headache • Nephritis (simulta- 4/6 • IV MP 1g x 5d • Resolved • Alive

7 weeks • Cognitive changes, neous) • Steroid taper • 6 weeks • Clinical Response: Stable
AMS disease at 3 months post
• Fatigue, sleepiness ICI
• CTCAE 3 • No Durable Clinical
Benefit: Progressive disease
at 6 months post-ICI
• Total follow up ~216
weeks
3 Adnexal Nivo/ • Headache, N/V, • Thyroiditis 7/9 • IV MP 1 g x 5d • Resolved • Clinical Response: Stable
carcinoma ipi, 22 fevers (antecedent) • Antibiotics • 12 weeks disease at 3 months post-
of skin weeks • Confusion, memory • Hepatitis (anteced- • Long term steroids ICI
impairment, person- ent) • Durable Clinical Benefit:
ality changes • Myasthenia Stable disease at 6 months
• Double vision, (simultaneous) post-ICI
fatigue, dysarthria, • Deceased due to progres-
• CTCAE 3 sive cancer
• Total follow up ~75 weeks
4 HNSCC Nivo/ • Headache, photo- 7/9 • IV MP 1 g x 1d • Resolved • Clinical Response: Clinical

ipi, 35 phobia, N/V, chills, • Steroid taper • 24 hours remission at 3 months
weeks neck pain • Acyclovir post-ICI
• Mild confusion • Relapse: IV MP 1 g x 3 days, steroid taper • Durable Clinical Benefit:
• CTCAE 3 Clinical remission at 6
months post-ICI
• Clinical remission at 32
months post-AE
weeks
3

4
Table 1. Continued
onset
5 NSCLC Pembro, • Cognitive decline, • Skin rash (subse- 5/7 • Antibiotics, acyclovir • Resolved • Clinical Response: Clinical
53 weeks confusion quent) • 1 mg/kg IV MP, increased to 1g x 5 d • 4 weeks response at 3 months post-
• Hallucinations, • Hypophysitis • Long term steroids ICI
hyperactive (subsequent) • Durable Clinical Benefit:
delirium, picking • PMR Clinical response at 6
behavior months post-ICI
• Gait impairment, • Alive with clinical
falls response
• CTCAE 3 • Total follow up ~154
weeks
6 Papillary Thy- Pembro, • Mild confusion • Myasthenia 5/5 • IVIG x 5 d • Resolved • Clinical Response: Stable
roid Ca 13 weeks • Ptosis, fatigue, (simultaneous) • Long term IVIG • 5 days disease at 3 months post-
diplopia ICI
• CTCAE 3 • Durable Clinical Benefit:
Stable disease at 6 months
post-ICI
• Alive with stable disease
weeks
7 NSCLC Atezo, • Cognitive decline • Immune Thrombo- 6/8 • 80 mg prednisone Minimal • Lost to follow-up
2 days • Gait impairment cytopenia (simulta- • IV MP 1 g x 5 d improvement post-hospitalization
• Bradykinesia, dys- neous) • IVIG x 5 d while hospi- • Total follow up ~1 week
metria • Steroid taper talized
• Emotional lability
• CTCAE 3
8 Hodgkin Lym- Nivo/ • Hallucinations, 6/8 • IV MP 1 g x 5 d • Resolved • Clinical Response: Clinical
phoma anti- delusions, psychosis, • PLEX x 5 d • 4 weeks response at 3 months post-
LAG3, pressured speech, • Rituximab ICI
42 weeks paranoia, grandeur • No Durable Clinical
• CTCAE 4 Benefit: Clinical progres-
sion at 6 months post-ICI
• Alive following treatment
for progression
weeks

Table 1. Continued
onset
9 Hodgkin/ Nivo, • Seizures • Thyroiditis 12/14 • IV MP 1g x 5 d • Resolved • Clinical Response: Clinical

Grayzone 27 weeks • Headaches, nausea (antecedent) • Acyclovir • 6 weeks response at 3 months post-
Lymphoma • Memory loss • Levetiracetam, lacosamide, clobazam ICI
• Difficulty with • IVIG x 5 d • Durable Clinical Benefit:
concentration • Rituximab Clinical response at 6
• Aphasia, dysarthria • Corticosteroid taper months post ICI
• Personality changes, • Relapse: IV MP 1g x 5 d • Alive with clinical remis-
irritable, easy to sion

anger • Total follow up ~171
• CTCAE 3 weeks
10 SCLC, Nivo/ipi, • Memory loss • SIADH (simulta- 7/9 • Prednisone 60 mg • Resolved • Clinical Response: Clinical
Brain metas- 3 days • Gait impairment neous) • Steroid taper • 4 weeks response at 3 months post-
tasis • Tremor, ataxia ICI
• CTCAE 3 • Durable Clinical Benefit:
Clinical response at 6
months post-ICI
• Deceased due to progres-
sive disease
11 Melanoma, Nivo/ipi, • Fevers, N/V 10/14 • IV MP 1 g x 5 d • Resolved • Clinical Response: Clinical

Brain 15 days • Memory loss, AMS • IVIG • 12 weeks response at 3 months post-
metastasis • Inappropriate laugh- • Rituximab ICI
ter • Levetiracetam, phenytoin, lacosamide • Durable Clinical Benefit:
• Gait impairment Clinical response at 6
• Dysautonomia months post-ICI
• Seizures • Deceased due to progres-
• CTCAE 3 sive disease
weeks
12 SCLC Nivo/ipi • Confusion • SIADH 7/9 • IV MP 0.5 g x 4 d • Resolved • Clinical Response Clinical
10 days • Headache, N/V, • Steroid taper • 4 weeks response at 3 months post-
chills ICI
• Gait impairment, • Durable Clinical Benefit:
falls Clinical response at 6
• Ataxia months post-ICI
• CTCAE 3 • Alive with sustained
response
weeks
5

6
Table 1. Continued
onset
13 AML Pembro, • AMS • Dermatitis 9/9 • None • Resolved • Clinical response at 3

9 days • CTCAE 3 • 6 weeks months post-ICI
• Durable Clinical Benefit:
Clinical response at 6
months post-ICI
• Deceased due to progres-
sive disease
14 NSCLC, sus- Pembro, • AMS, seizures, 7/9 • IV MP 1 g x 5 d No improvement • Clinical response at 3

pected brain 71 weeks aphasia, vertigo • IVIG at 3 months months post-ICI
metastasis • CTCAE 4 • Rituximab post-AE onset • Durable Clinical Benefit:
• Prednisone taper Clinical response at 6
months post-ICI
• Lost to follow-up after
entering hospice
Abbreviations: AML, acute myelogenous leukemia; AMS, altered mental status; APE2, antibody prevalence in epilepsy and encephalopathy; Atezo, atezolizumab; CTCAE, common terminology criteria for adverse
events; d, days; F, female; HNSCC, head and neck squamous cell carcinoma; IVIG, intravenous immunoglobulins; IV, intravenous; M, male; MP, methylprednisolone; N/V, nausea and vomiting; nivo, nivolumab;
ipi, ipilimumab; NSCLC, non-small cell lung cancer; No., number; Pt., patient; pembro, pembrolizumab; RITE2, response to immunotherapy in epilepsy and encephalopathy; SCC, squamous cell carcinoma; SCLC,
small cell lung cancer; SIADH, syndrome of inappropriate antidiuretic hormone secretion.

The Oncologist, 2024, Vol. XX, No. XX 7
years). Cancer diagnosis included: NSCLC (21%), melanoma findings typical of autoimmune encephalitis and one patient
(14%), small cell lung cancer (SCLC; 14%), Hodgkin’s lym- had a single contrast-enhancing lesion (Table 2; Figure 1A,
phoma (14%), and other cancers (36%) (Table 1). Only 29% B). On brain FDG-PET/CT imaging, 1 patient had findings
of patients had a history or suspicion of brain metastasis at consistent with autoimmune encephalitis with hypermetab-
presentation. Patients were treated with nivolumab (7%), olism in the medial temporal lobe (Figure 1C). CSF analy-
pembrolizumab (36%), atezolizumab (7%), and combination sis frequently revealed lymphocytic pleocytosis (85%) and/
therapy (50%) with primarily ipilimumab/nivolumab (Table or elevated protein levels (69%). All patients had negative
1). The time of onset of immune-related encephalitis after ICI infectious CSF evaluations. Oligoclonal bands were evalu-
initiation varied from 2 days to 71 weeks (median: 70 days, ated in 10 patients with 60% having matched oligoclonal
IQR 219.8 days) (Table 1). bands in their serum and CSF and 2 patients with unique

The literature review identified 132 patients with immune- oligoclonal bands in the CSF alone, consistent with intrathe-
related encephalitis. Median age was 61.5 years (IQR: 17 cal immunoglobulin synthesis. Only 3 patients (30%) had
years). Patients had melanoma (32%), NSCLC (26%), RCC epileptiform activity or seizures on EEG. All 14 patients had
(8%), and hematologic malignancies (5%). Brain metastasis an APE2 score ≥4, suggestive of an associated neuronal auto-
was present in 20% of patients. Most patients were treated antibody. Paraneoplastic/autoimmune encephalopathy auto-
with PD-1 inhibitors (63%) and approximately 18% of antibody testing was completed in 12 patients (n = 3 serum,
patients were treated with combination ICI (Supplementary n = 5 CSF, n = 4 both); 25% had positive testing on commer-
Table S1). The onset of symptoms from initiation of ICI ranged cially available tests (anti-NMDAR, AGNA, VGKC complex
from less than 24 hours to more than 1 year with a median of without further specification) while one patient had a pos-
62 days (IQR 110.75 days; Supplementary Table S1). itive autoantibody on research testing (anti-neurofilament
light chain; Table 2). Five patients (36%) met the criteria for
Clinical presentation definite autoimmune encephalitis based on the presence of
Patients in the institutional cohort presented with confusion, autoantibodies or typical imaging findings.
memory impairment, and/or altered mental status 93%; gait Among patients identified in the systematic review, evalu-
imbalance (43%); definite or suspected seizures (29%); fevers/ ations typically included brain MRI, lumbar puncture, and
chills (21%); meningeal signs reflecting meningeal inflamma- EEG. MRI brain imaging was unremarkable or stable in 35%
tion or meningitis including headache (50%) and personality of patients. Typical findings such as bilateral temporal or mul-
changes (29%; Table 1). Thirteen patients presented with a tiple T2/FLAIR lesions were frequently observed (21% and
combination of symptoms, most commonly headache with 19%, respectively). Leptomeningeal or dural enhancement
altered mental status (43%) reflective of likely meningoen- was seen primarily in patients presenting with CSF pleocytosis
cephalitis. Most patients (64%) were diagnosed with addi- and meningeal symptoms suggestive of meningoencephalitis
tional irAEs (Table 1) including 2 patients with myasthenia. (Table 3). Of note, several patients initially had unremarkable
Due to the presence of concurrent irAEs (14%) and treat- brain MRI, but had abnormal follow-up imaging (n = 11)
ment for brain metastasis (7%), 3 patients were on corticoste- with findings of autoimmune or limbic encephalitis, dural/
roids at the time of developing symptoms of immune-related leptomeningeal enhancement, or demyelination.
encephalitis. The severity of encephalitis was as follows: grade CSF analysis included cell count, protein, glucose, infec-
3 (86%) and grade 4 (14%; Table 1). tious studies, cytopathology or flow cytometry, and auto-
Among the patients identified in the literature review, 92% antibody testing. CSF pleocytosis was common, although
met the criteria for autoimmune encephalitis based on avail- 23% of patients had normal CSF white blood count (WBC).
able data: possible (51%), probable 4%, and definite (37%; Lymphocytic pleocytosis was predominant and observed
Supplementary Table S1). Typical symptoms included: con- in 93% of patients with pleocytosis. However, neutrophilic
fusion, altered mental status, or memory impairment (87%) pleocytosis was seen (7%). Most patients had elevated CSF
and psychiatric symptoms (25%). Signs of meningeal inflam- protein (84%, Table 3). Infectious workup and additional
mation were common and included fevers (28%), headaches screening for malignancy with CSF cytopathology or flow
(21%), and nausea/vomiting (15%) (Supplementary Table cytometry was negative in all cases tested.
S2). Approximately 27% of patients presented with sei- A total of 96 patients were tested for specific neuronal
zures. Interestingly, concurrent irAEs were common (33%), autoantibodies, of these 45 patients (47%; Table 3) had
and included dermatitis (10%), hepatitis (6%), hypophysitis positive results. Detected autoantibodies primarily targeted
(2%), and thyroiditis (5%). Concurrent nirAEs occurred in intracellular antigens and included: anti-Hu (16%), anti-Ma2
17 patients (13%) and included sensory neuronopathy (3%), (22%), anti-GAD65 (11%), anti-GFAP (9%), and a neuron
optic neuritis (2%), Guillain-Barre syndrome (1.5%) and specific autoantibody of unknown specificity (18%). Anti-
myopathy (1.5%) (Supplementary Table S2). NMDAR encephalitis was diagnosed in 4 patients (80%).18
Seven patients (16%) had more than 1 autoantibody iden-
Diagnostic evaluation tified. Retrospective analysis of pre-ICI serum revealed the
Among the patients in the institutional cohort, the differ- presence of autoantibodies (n = 5); however, pre-treatment
ential diagnoses included brain metastasis, leptomeningeal samples were rarely tested (numbers unavailable). In some
disease, infectious meningitis, or infectious encephalitis. The cases, patients had other classical paraneoplastic syndromes
full evaluation included testing for neuronal autoantibod- associated with the detected autoantibody; for example, sen-
ies (n = 13), brain MRI (n = 14), lumbar puncture (n = 13), sory neuronopathy was diagnosed in a patient with anti-Hu
brain FDG-PET/CT (n = 4), EEG (n = 10), and electromyo- autoantibody.19,20
gram and nerve conduction studies (EMG/NCS; n = 2) EEG results were reported in 47 patients (36%) and
(Table 2). Brain imaging was unremarkable or non-specific included slowing or consistent with diffuse encephalopa-
in the majority of patients (64%). Two patients had image thy (64%) and slow activity or temporal lobe epileptiform
Table 2. Evaluation of cohort patients with immune-related encephalitis.
MRI brain Lumbar Autoantibodies EEG

puncture
Cell count Protein Oligoclonal bands

(cells/µL) (mg/dL) (OCBs)
1 Hemorrhagic metastasis 166 80 ND Negative Generalized periodic

discharges
2 WNL 0 19.7 Pattern 4, OCBs in CSF Negative WNL

identical to those in
serum
3 WNL 12 53.4 None Negative WNL
4 WNL 75 68 ND ND ND
5 Right Parietal infarction 12 39 ND Negative Slow
6 T2/FLAIR in left corona radi- ND ND ND AChR Ab+ ND
ata with contrast enhance-
ment
7 WNL 1 53.2 None Negative WNL
8 WNL 20 113.1 Pattern 4, OCBs in CSF Negative WNL
identical to those in
serum
9 T2/FLAIR hyperintensities of 22 37.7 Pattern 4, OCBs in CSF Negative Left temporal spikes and
left medial temporal lobe, identical to those in slowing
left lateral frontal lobe and serum
posterior cingulate
10 Right hippocampus T2/FLAIR 18 98 Pattern 4, OCBs in CSF Anti-glial nuclear Ab ND
hyperintensity identical to those in
serum
11 Metastatic disease 8 24 Pattern 4, OCBs in CSF NMDAR Ab Left temporal spikes and
identical to those in seizures
serum
12 Nonspecific subcortical T2/ 30 166 Pattern 4, OCBs in CSF Novel unclassified Ab WNL
FLAIR lesions identical to those in subsequently iden-
serum tified as directed to
neurofilament light
chain
13 Multiple cortical and subcorti- 8 64.6 Pattern 2 with ≥2 OCBs Negative ND
cal T2/FLAIR hyperintensi- in CSF only
ties, some of which enhance
14 Punctate enhancing T2/FLAIR 6 68 Pattern 3, ≥2 OCBs in VGKC complex Diffuse, symmetric irreg-
hyperintensities concerning CSF with ≥1 separate antibody, LGI1 and ular occipital rhythm
for metastases as well as bands in CSF also in CASPR2 negative and background
nonspecific subcortical T2/ serum slowing without epilep-
FLAIR hyperintensities tiform activity
Abbreviations: Ab, antibody; AChR, Acetylcholine Receptor; CASPR2, contactin-associated protein-like 2; EEG, electroencephalogram; FLAIR, fluid-
attenuated inversion recovery; LGI1, leucine-rich glioma inactivated 1; MRI, magnetic resonance imaging; ND, not done; NMDA-R, N-methyl-D aspartate
receptor; OCBs, oligoclonal bands; VGKC, voltage gated potassium channel; WNL, within normal limits.
activity (13%) (Table 3). Confirmed seizures were captured (n = 1 as initial therapy, n = 4 following IVMP) and plasma-
in 4 patients and 2 patients presented with status epilepticus. pheresis in one patient (7%). Patients who did not respond to
first-line immunosuppressive therapy were treated with ritux-
Treatment of immune-related encephalitis imab as second-line therapy (4/14, 29%; Table 1).
Among the institutional cohort, clinical suspicion for Most patients recovered from immune-related enceph-
immune-related encephalitis resulted in discontinuation of alitis over several weeks, although one patient was lost
ICI and initiation of immunosuppression in 13 cases. Given to follow up. Among the 13 with follow-up, the tempo of
subacute presentation, 1 patient received additional ICI recovery was variable, and 1 patient (patient 4) with primar-
after initial symptoms (patient 5). RITE2 score was ≥7 in 12 ily meningeal symptoms with headaches and mild cognitive
patients (86%), predictive of a favorable response to immu- changes recovered immediately after corticosteroid initiation.
nosuppression. Intravenous corticosteroids (500 or 1000 mg Unfortunately, 1 patient did not demonstrate improvement
IV methylprednisolone) for 3-5 days with corticosteroid taper after immunosuppression with rituximab (patient 14). As
was the typical initial therapy, and 86% received corticoste- expected, patients with longer recovery time had more severe
roids. Treatment with IVIg was initiated in 36% of patients immune-related encephalitis characterized by seizures and

Figure 1. MRI and brain FDG-PET imaging of patients with immune-related encephalitis. (A) T2/FLAIR MRI brain images at presentation and follow
up after treatment in a patient with typical imaging findings of autoimmune encephalitis (patient 9). Note T2/FLAIR hyperintensities of the left
hippocampus, left frontal lobe, and posterior cingulate. The left hippocampal lesion evolves to medial temporal sclerosis while the left frontal and
posterior cingulate lesions resolve. (B) T2 FLAIR and T1-post-gadolinium MRI brain images at presentation and after treatment in patient with lesion of
the right corona radiate demonstrating persistence of T2/FLAIR hyperintensity with resolution of gadolinium enhancement on follow-up (patient 6). (C)
FDG-PET/CT brain imaging 112 days prior to and 99 days after immunosuppressive treatment initiation for immune-related encephalitis demonstrating
hypermetabolism of the left medial temporal lobe that was not evident on the pre-encephalitis FDG-PET performed through the course of oncologic
care (patient 9).
cognitive impairment. Although most patients with follow-up rapidly fatal or resolving symptoms. Patients with suspected
had a marked clinical recovery (85%) and there were no cases or confirmed seizures were treated with anti-epileptics.
of mortality due to ICI-related encephalitis, some patients Most patients had partial or complete recovery (81%).
had residual cognitive impairment (31%), recurrent seizures Typical findings of limbic encephalitis were more com-
(8%), or required continued treatment with anti-seizure med- monly found in patients with no recovery or death (82%)
ications (23%) (Table 1). Two patients (15%) had a clinical as compared to patients with partial or full recovery (25%).
relapse during corticosteroid taper which responded to repeat Additionally, patients with partial or full recovery were more
treatment with IV methylprednisolone. likely to have a normal or stable MRI (54%). Stabilization
Of patients in the literature review, ICI was discontinued of symptoms but minimal or no clinical improvement was
in all patients with immune-related encephalitis, although 7 associated with hippocampal atrophy secondary to inflamma-
patients had repeat exposure after recovery for further cancer tory damage. Mortality without any recovery from immune-
treatment. Patients received corticosteroids as first-line treat- related encephalitis (12%) was attributed to encephalitis;
ment, typically either IV methylprednisolone (500-1000 mg/ however, progressive cancer, comorbid infections, and respi-
day) or IV dexamethasone (1-2 mg/kg/day) for 3-5 days ratory failure also contributed. Among patients who recov-
(Table 4). While improvement was seen with corticosteroids ered, clinically significant improvement occurred on average
alone, patients frequently received additional first-line immu- 18 days after treatment and improvement continued over
nosuppressive therapy with IVIg (26%) and plasmapheresis weeks. However, some patients with predominately menin-
(10%). The most common second-line therapy was rituximab geal symptoms had rapid recovery within 24 hours.
(12%). No immunosuppressive medications were admin- Relapse of immune-related encephalitis was infre-
istered to 5% of patients due to unrecognized diagnoses or quent (10%) and occurred during corticosteroid taper or
re-exposure to ICI. Of the 7 patients re-exposed to ICI, 3

Table 3. Summary of evaluation of patients with immune-related encephalitis as a neurological immune related adverse event following immune checkpoint inhibitor therapy identified by systematic literature
had relapse of encephalitis. Treatment for relapse was similar
(46.9%)
Positive
to initial treatment; corticosteroids were administered to all
Antibodies (96/132)
patients. However, 2 patients required treatment with ritux-
45
imab and 1 patient died.
Response to ICI in patients with immune-related

Negative
(56.1%)
encephalitis
Auto-
51
Of patients in the institutional cohort with longitudinal
follow-up (n = 13), 85% demonstrated anti-tumor response

Non-specific
to ICI at 3-month post-ICI initiation, and 77% at 6-month

follow-up (Table 1). Of those with a RITE2 score greater
(7.9%)
than or equal to 7 (n = 13), 85% had favorable responses to

9
immunosuppression. At last follow-up, 38% of patients died

of progressive disease, and 7 had stable disease or continued
Enhancement
(lepto/dural)
partial response (median follow up 1098 days, IQR 1290

(11.5%)
days). Chronic corticosteroid use was in 38% for continued

treatment for encephalitis, metastatic cancer, or hypophysitis.
13
No patients were re-exposed to ICIs.

The patients identified in the systematic review were fol-
Demye-llinating
lowed an average of 274 days after diagnosis of immune-

related encephalitis (Table 4). After removing patients
without available data, a total of 41 patients were reported to
(5.3%)
have died (33%). Cause of death included progressive malig-

Abbreviations: AE, autoimmune encephalitis; LE, limbic encephalitis; MRI, magnetic resonance imaging; WNL, within normal limits.
6
nancy (44%), immune-related encephalitis (37%), other irAEs

such as autoimmune colitis (5%), or other comorbidities (12%).
Typical AE
(18.6%)
Discussion
21
Immune-related encephalitis is a well-recognized neurological

irAE following ICI.11,12 Here, we present a retrospective 14
Typical LE
patient case series, one of the largest to date, accompanied by

(21.2%)
a review of the literature.

Melanoma and NSCLC were the most common underly-
24
ing malignancies, likely reflecting the frequency of adminis-

tration and length of time since FDA approval of ICI rather
MRI brain
(113/132)
(35.4%)
than intrinsic cancer factors as these malignancies are not

Stable
WNL
frequently associated with paraneoplastic encephalitis.1 Most

40
previously reported patients were treated with PD-1 inhibi-

tors alone. However, our case series had a higher frequency of
patients treated with combination ICI (44% as compared to
Abnormal
18% in the literature review). Additionally, brain metastasis

(83.1%)
(74/89)
(251.7)
were found in the minority of patients both in the case series

183.8
as well as literature review (20%-30%). The presence of brain

74
Protein (mg/dl)
metastasis and treatment with radiation or radiosurgery may

contribute to blood brain barrier breakdown and altered
(89/132)
immune responses in the brain. Interestingly, although most

(16.9%)
Normal
(15/89)
irAE occur in the first 3 months of treatment, immune-related

15
encephalitis could occur more than 1 year after first exposure,

as is observed in pneumonitis.21 However, the onset is highly
variable and symptoms can occur within 24 hours after the
Abnormal
start of ICI.
(79/103)
(76.7%)
(115.3)
Patients with immune-related encephalitis typically pres-

Cell Count (cells/ul)
69.2
ent with subacute cognitive changes, meningeal signs, and

Lumbar Puncture
79
personality changes, similar to what is seen in autoimmune

encephalitis cases not associated with ICI. These symptoms
24 (23.3%)
(103/132)
(103/132)
(24/103)
are also present in leptomeningeal carcinomatosis as well as

Normal
infectious meningitis/encephalitis, highlighting the impor-

tance of excluding alternative diagnoses. Many patients had
other co-existing irAEs, including nirAEs, reminding us that
Mean (SD)
patients are at risk for multiple as well as isolated irAEs. Our

Total no.
review.
series included 2 cases of encephalitis and myasthenia gravis,

(%)
which to our knowledge has not been previously reported.

Table 4. Treatments and outcomes of reported patients with immune-related encephalitis as a neurological immune related adverse event following
immune checkpoint inhibitor therapy identified by systematic literature review.
PDL-1 inhibitors PD-1 inhibitors (nivolumab/ CTLA-4 inhibitors Combination All

(atezolizumab) pembrolizumab) (ipilimumab)
Treatment
IV or PO steroids 15 76 8 21 120 (96%)
IVIG 6 18 2 7 33 (26.4%)
Plasmapheresis 0 10 1 1 12 (9.6%)

Rituximab 3 8 0 4 15 (12%)
Other IS 3 8 2 1 14 (11.2%)
No IS 0 4 0 3 7 (5.6%)
Outcome
Deceased from AE 0 11 0 3
Time to recovery, days, mean (SD) 11.3(16.6) 18.7 (16.0) 31.8 (39.6) 18.2 (18.1) 18 (18.6)
Time to recovery, days, median (IQR) 5 (6.5) 14 (21.8) 17.5 (27.3) 8 (21) 9 (23)
Total deceased at follow up 4 24 0 9 37 (29.6%)
Time to follow up, days, mean (SD) 107.7 (94.6) 311.6 (391) 202.5 (51.2) 283.7 (361) 274.4 (354.7)
Time to follow up, days, median (IQR) 90 (45.3) 180 (305) 195 (52.5) 150 (127.5) 150 (245.5)
Abbreviations: AE, autoimmune encephalitis; IS, immunosuppressive treatment; IV, intravenous; IVIG, IV immunoglobulins; PO, per oral
Interestingly, all patients within the institutional cohort corticosteroids while other patients died or had no recov-
had APE2 scores of ≥4 and 86% of the patients had a RITE2 ery from their autoimmune encephalitis. Associated factors
score ≥7, predicting presence of a neuronal autoantibody included abnormal MRI brain, especially typical findings of
and response of symptoms of encephalitis to immunother- limbic encephalitis which was found in 82% of patients with
apy. These scores may prove helpful in the identification poor recovery or death and only 25% of patients with partial/
of patients with immune-related encephalitis, guide con- full recovery. Although CSF pleocytosis was not predictive of
siderations for systemic immunosuppression, and identify recovery in our literature review cohort, neuronal autoan-
patients likely to have a novel neuronal autoantibody. Future tibodies were found in 62% of patients with poor recovery
prospective evaluation is warranted to assess the utility of or death and only 38% of patients with partial/full recov-
these scores in the identification and response to immuno- ery. In both cohorts, the relapse rate was lower than typical
suppression among patients suspected of having immune- autoimmune encephalitis, suggesting that long-term immu-
related encephalitis. nosuppressive therapy may not be required. Close clinical
The typical workup of patients in both our institutional monitoring should be continued during corticosteroid taper
series and those identified through the literature review as some patients relapsed during corticosteroid discontinu-
included an MRI brain, lumbar puncture, and EEG. One ation, although corticosteroid taper should be decided on a
discrepancy included the frequency of patients with MRI case by case basis.23,24 Considering that patients in our cohort
brain findings typical of autoimmune encephalitis; 28% of responded well to immunosuppression raises the question of
our case series compared to over 50% in a literature review. whether ICI could be continued if clinically indicated. It may
This potentially reflects increased awareness allowing for be that patients can be treated through the acute encephalitis
diagnosis prior to MRI brain changes as untreated patients and further ICI may be administered with careful neurologi-
can develop MRI abnormalities over time. Lumbar puncture cal monitoring and rapid re-initiation of immunosuppressive
most commonly showed lymphocytic pleocytosis or elevated treatments with recurrence.
protein. Workup included evaluation for infectious etiologies, Further research is required in this burgeoning area of
and none of our patients with concern for immune-related nirAEs. It is not known whether ICI therapy could unmask
encephalitis had. Nearly half of patients in the literature latent disease. Some patients had neuronal autoantibodies
review who were tested for neuronal autoantibodies had pos- identified in pre-treatment serum without a neurological syn-
itive titers. Although this may reflect publication bias, a recent drome. Although this was infrequently tested. It is difficult
study revealed a high frequency of neuronal autoantibodies to determine how many cases of immune-related encephali-
in patients with immune-related encephalitis.22 Importantly, tis represent pre-existing paraneoplastic encephalitis. Recent
patients with neuronal autoantibodies often had normal MRI studies have shown that pre-existing paraneoplastic disorders
brain and lumbar puncture, highlighting that normal studies have worsened symptoms in up to 50% of patients after ICI
do not rule out immune-related encephalitis and clinical sus- therapy.25 Identification of biomarkers will be crucial for diag-
picion is paramount. nosis of pre-existing paraneoplastic neurological syndromes,
In our case series, patients commonly responded to first-line diagnosis of nirAEs, and monitoring response of nirAEs to
immunosuppressive therapy defined as corticosteroids, IVIg, immunosuppressive therapy.11 Finally, most patients with
or plasmapheresis and had lower mortality than the literature immune-related encephalitis in this case series and litera-
cohort. However, recovery was variable with some patients ture review responded well to immunotherapy. Continued
having nearly immediate improvement after initiation of research is needed to evaluate whether the current guidelines
that recommend halting ICI therapy after immune-related Instil Bio, Merck, Merck KGaA, Natera, Nektar, Novartis,
encephalitis should be revised. OncoSec, Pfizer, Rain Therapeutics, Regeneron, Replimune,
Sanofi-Aventis, Sun Pharma, and Syneos Health, receiving
institutional research funding from Bristol-Myers Squibb,
Acknowledgments Haystack Oncology, Merck, Regeneron, and Sanofi, and
The authors acknowledge Carrie Price, MLS and Stella Seal, stock ownership in Iovance. Dr. Naidoo reports consul-
MLS of the Johns Hopkins University Welch Library for pre- tant work for AstraZeneca, Bristol Myers Squibb, Daiichi
paring the literature search for the systematic review. We also Sankyo, Roche/Genentech, Amgen, Arcus Biosciences, NGM
acknowledge cancer center grant, P30CA006973 (Nelson). Pharmaceuticals, Bayer, Regeneron, Takeda, Pfizer, Elevation
Oncology, Abbvie, Kaleido Biosciences; research funding

from AstraZeneca, Bristol Myers Squibb, Roche/Genentech,
Author contributions Amgen, Arcus Biosciences, Mirati, Novartis, Takeda, Pfizer;
Monica W. Buckley (Conceptualization, Data curation, Data safety monitoring board work with AstraZeneca, Bristol
Investigation, Methodology, Visualization, Writing—origi- Myers Squibb, Daiichi Sankyo; honoraria from AstraZeneca,
nal draft, Writing—review & editing), Aanika Balaji Warner Bristol Myers Squibb, Daiichi Sankyo, Roche/Genentech, and
(Data curation, Formal Analysis, Investigation, Validation, Mirati. Dr. Probasco reports being a site investigator of a trial
Writing—original draft, Writing—review & editing), Julie sponsored by Genentech as well as compensation as Editor-in
Brahmer (Writing—review & editing), Laura C. Cappelli Chief of NEJM Journal Watch Neurology. Dr. Saidha has
(Conceptualization, Writing—review & editing), William received consulting fees from Medical Logix for the devel-
H. Sharfman (Writing—review & editing), Ephraim Fuchs opment of CME programs in neurology and has served on
(Writing—review & editing), Hyunseok Kang (Writing— scientific advisory boards for Biogen, Novartis, Genentech
review & editing), Patrick M. Forde (Writing—review & Corporation, TG therapeutics, Clene Pharmaceuticals, Amgen
editing), Douglas E. Gladstone (Writing—review & editing), & ReWind therapeutics. He has performed consulting for
Richard Ambinder (Writing—review & editing), Ronan J. Novartis, Genentech Corporation, JuneBrain LLC, Innocare
Kelly (Writing—review & editing), Evan J. Lipson (Writing— pharma, Kiniksa pharmaceuticals, Setpoint Medical and
review & editing), Ivana Gojo (Writing—review & editing), Lapix therapeutics. He is the PI of investigator-initiated stud-
Edward J. Lee (Writing—review & editing), Tory P. Johnson ies funded by Genentech Corporation, Biogen, and Novartis.
(Writing—review & editing), Shiv Saidha (Writing—review He previously received support from the Race to Erase MS
& editing), Rafael Llinas (Writing—review & editing), Lyle foundation. He has received equity compensation for con-
W. Ostrow (Writing—review & editing),Jarushka Naidoo sulting from JuneBrain LLC and Lapix therapeutics. He
(Conceptualization, Data curation, Formal Analysis, was also the site investigator of trials sponsored by MedDay
Investigation, Methodology, Validation, Visualization, Pharmaceuticals, Clene Pharmaceuticals, and is the site inves-
Writing—original draft), John C. Probasco (Data curation, tigator of trials sponsored by Novartis, as well as Lapix ther-
Formal Analysis, Methodology, Resources, Writing—original apeutics. Dr. Sharfman reports consultant work for Bristol
draft, Writing—review & editing) Myers Squibb, Merck; research support from Bristol Myers
Squib, Merck, Novartis, Genentech, and AstraZeneca. Drs.
Ambinder, Buckley, Fuchs, Gladstone, Johnson, Lee, Llinas,
Funding Ostrow, and Warner report no relevant disclosures.
Dr. Brahmer reports grant funding from AstraZeneca,
Bristol Myers Squibb; advisory board membership with
AstraZeneca, Bristol Myers Squibb, Genentech, Merck. Dr. Data availability
Cappelli reports research support from Bristol Myers Squibb; The data underlying this article will be shared on reasonable
consultant work for Amgen. Dr. Forde reports research sup- request to the corresponding author.
port from AstraZeneca, BioNTech, BMS, Kyowa, Novartis,
Regeneron; consultant work for Amgen, AstraZeneca,
Bristol Myers Squibb, Daiichi, F-Star, G1, Genentech, Supplementary material
Janssen, Iteos, Merck, Sanofi, Novartis, Surface; and data Supplementary material is available at The Oncologist online.
safety monitoring board work for Polaris. Dr. Gojo reports
research support from Merck, Amgen, Amphivena, Gilead,
Genentech, Incyte and Celgene; advisory board work with References
Amgen, Immunogen, Gilead, Bristol Myers Squibb, Certara, 1. Martins F, Sofiya L, Sykiotis GP, et al. Adverse effects of
ClearView, Ono Pharmaceutical. Dr. Kang reports research immune-checkpoint inhibitors: epidemiology, management and
support from Eli-Lilly, Exelixis, Kura, Elevar therapeutics, surveillance. Nat Rev Clin Oncol. 2019;16(9):563-580. https://doi.
PDS Biotechnology, and NeoImmuneTech; Consultant work org/10.1038/s41571-019-0218-0
for Bayer, PIN therapeutics, Bertis, MitoImmune; Advisory 2. Antonia SJ, López-Martin JA, Bendell J, et al. Nivolumab alone
board work for Coherus, Bluedot Bio, Exelixis, LG Chem, and nivolumab plus ipilimumab in recurrent small-cell lung cancer
Bayer. Dr. Kelly research support and advisory work from (CheckMate 032): a multicentre, open-label, phase 1/2 trial. Lan-
cet Oncol. 2016;17(7):883-895. https://doi.org/10.1016/S1470-
Bristol Myers Squibb and Eli Lilly and advisory board/con-
2045(16)30098-5
sultant work for Merck, Novartis, Daiichi Sankyo, Novocure, 3. Schadendorf D, Hodi FS, Robert C, et al. Pooled analysis
Astra Zeneca, EMD Serono, Grail, Astellas, Ipsen, Takeda. Dr. of long-term survival data from phase II and phase III tri-
Lipson reports performing consultant work for and/or receiv- als of ipilimumab in unresectable or metastatic melanoma. J
ing honoraria from Bristol Myers Squibb, CareDx, Eisai, Clin Oncol. 2015;33(17):1889-1894. https://doi.org/10.1200/
Genentech, HUYA Bioscience International, Immunocore, jco.2014.56.2736
4. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab NXI.0000000000000413 http://www.ncbi.nlm.nih.gov/

and ipilimumab or monotherapy in untreated Melanoma. N Engl J pubmed/29159205. cited [2018 Dec 5].
Med. 2015;373(1):23-34. 16. Naidoo J, Zhang J, Lipson EJ, et al. A multidisciplinary tox-
5. Motzer RJ, Tannir NM, McDermott DF, et al. Nivolumab plus icity team for cancer immunotherapy-related adverse events.
Ipilimumab versus Sunitinib in advanced renal-cell carcinoma. N J Natl Compr Cancer Netw. 2019;17(6):712-720. https://doi.
Engl J Med. 2018;378(14):1277-1290. https://doi.org/10.1056/nej- org/10.6004/jnccn.2018.7268. Harborside Press
moa1712126 17. Dubey D, Kothapalli N, McKeon A, et al. Predictors of neural-
6. Hellmann MD, Paz Ares L, Bernabe Caro R, et al. Nivolumab specific autoantibodies and immunotherapy response in patients
plus ipilimumab in advanced non-small-cell lung cancer. N Engl with cognitive dysfunction. J Neuroimmunol. 2018;323:62-72.
J Med. 2019;381(21):2020-2031. https://doi.org/10.1056/NEJ- https://doi.org/10.1016/j.jneuroim.2018.07.009

Moa1910231 18. Dalmau J, Gleichman AJ, Hughes EG, et al. Anti-NMDA-receptor
7. Faivre-Finn C, Vicente D, Kurata T, et al. Four-year survival with encephalitis: case series and analysis of the effects of antibodies.
durvalumab after chemoradiotherapy in stage III NSCLC-an Lancet Neurol. 2008;7(12):1091-1098. https://doi.org/10.1016/
Update From the PACIFIC Trial. J Thorac Oncol. 2021;16(5):860- s1474-4422(08)70224-2
867. https://doi.org/10.1016/j.jtho.2020.12.015 19. Graus F, Delattre JY, Antoine JC, et al. Recommended diagnostic
8. Ascierto PA, Del Vecchio M, Mandalá M, et al. Adjuvant criteria for paraneoplastic neurological syndromes. J Neurol Neu-
nivolumab versus ipilimumab in resected stage IIIB-C and stage rosurg Psychiatry. 2004;75(8):1135-1140. https://doi.org/10.1136/
IV melanoma (CheckMate 238): 4-year results from a multi- jnnp.2003.034447
centre, double-blind, randomised, controlled, phase 3 trial. Lancet 20. Graus F, Dalmau J, Reñé R, et al. Anti-Hu antibodies in patients
Oncol. 2020;21(11):1465-1477. https://doi.org/10.1016/S1470- with small-cell lung cancer: Association with complete response to
2045(20)30494-0 therapy and improved survival. J Clin Oncol. 1997;15(8):2866-
9. Hassel JC. Ipilimumab plus nivolumab for advanced melanoma. 2872. https://doi.org/10.1200/JCO.1997.15.8.2866
Lancet Oncol. 2016;17(11):1471-1472. https://doi.org/10.1016/ 21. Tiu BC, Zubiri L, Iheke J, et al. Real-world incidence and impact
s1470-2045(16)30409-0 of pneumonitis in patients with lung cancer treated with immune
10. Jain P, Jain C, Velcheti V. Role of immune-checkpoint inhibi- checkpoint inhibitors: a multi-institutional cohort study. J Immu-
tors in lung cancer. Ther Adv Respir Dis. 2018;12:1. https://doi. noTher Cancer. 2022;10(6):e004670. https://doi.org/10.1136/jitc-
org/10.1177/1753465817750075 2022-004670
11. Hottinger AF, Hegi ME, Baumert BG. Neurologic complications of 22. Velasco R, Villagrán M, Jové M, et al. Encephalitis induced
immune checkpoint inhibitors. Curr Opin Neurol. 2016;29(6):782- by immune checkpoint inhibitors: a systematic review. JAMA
788. https://doi.org/10.1097/WCO.0000000000000390 Neurol. 2021;78(7):864-873. https://doi.org/10.1001/jamaneu-
12. Astaras C, de Micheli R, Moura B, Hundsberger T, Hottinger AF. rol.2021.0249
Neurological adverse events associated with immune checkpoint 23. Schneider BJ, Naidoo J, Santomasso BD, et al. Management
inhibitors: diagnosis and management. Curr Neurol Neurosci Rep. of immune-related adverse events in patients treated with
2018;18(1):3. https://doi.org/10.1007/s11910-018-0810-1 immune checkpoint inhibitor therapy: ASCO guideline update.
13. Guidon AC, Burton LB, Chwalisz BK, et al. Consensus disease defi- J Clin Oncol. 2021;39(36):4073-4126. https://doi.org/10.1200/
nitions for neurologic immune-related adverse events of immune jco.21.01440
checkpoint inhibitors. J ImmunoTher Cancer. 2021;9(7):e002890. 24. Thompson JA, Schneider BJ, Brahmer J, et al. Management of
https://doi.org/10.1136/jitc-2021-002890 immunotherapy-related toxicities, version 1.2019. JNCCN J
14. Graus F, Titulaer MJ, Balu R, et al. A clinical approach to diagnosis Natl Compr Cancer Netw. 2019;17(3):255-289. https://doi.
of autoimmune encephalitis. Lancet Neurol. 2016;15(4):391-404. org/10.6004/jnccn.2019.0013
https://doi.org/10.1016/S1474-4422(15)00401-9 25. Manson G, Maria ATJ, Poizeau F, et al. Worsening and newly
15. Probasco JC, Solnes L, Nalluri A, et al. Decreased occipital lobe diagnosed paraneoplastic syndromes following anti-PD-1 or anti-
metabolism by FDG-PET/CT. Neurol. Neuroimmunol. Neuroin- PD-L1 immunotherapies, a descriptive study. J ImmunoTher Can-
flamm. 2018;5(1):e413. Available from: https://doi.org/10.1212/ cer. 2019;7(1):337. https://doi.org/10.1186/s40425-019-0821-8

Oyae 186

Uploaded by

Copyright:

Available Formats

Oyae 186

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Oyae 186

Uploaded by

Copyright:

Available Formats

The Oncologist, 2024, XX, 1–13

Immune-related encephalitis after immune checkpoint

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

Implications for practice

Received: 29 January 2024; Accepted: 28 June 2024.

Introduction included the subacute onset of memory deficits, altered men-

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

1 Melanoma, Nivo/ • Headache, N/V, 8/10 • IV MP 1g x 3 d • Resolved • No Clinical Response:

2 SCC of skin Pembro, • Headache • Nephritis (simulta- 4/6 • IV MP 1g x 5d • Resolved • Alive

4 HNSCC Nivo/ • Headache, photo- 7/9 • IV MP 1 g x 1d • Resolved • Clinical Response: Clinical

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

9 Hodgkin/ Nivo, • Seizures • Thyroiditis 12/14 • IV MP 1g x 5 d • Resolved • Clinical Response: Clinical

irritable, easy to sion

11 Melanoma, Nivo/ipi, • Fevers, N/V 10/14 • IV MP 1 g x 5 d • Resolved • Clinical Response: Clinical

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

13 AML Pembro, • AMS • Dermatitis 9/9 • None • Resolved • Clinical response at 3

14 NSCLC, sus- Pembro, • AMS, seizures, 7/9 • IV MP 1 g x 5 d No improvement • Clinical response at 3

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

Table 2. Evaluation of cohort patients with immune-related encephalitis.

MRI brain Lumbar Autoantibodies EEG

Cell count Protein Oligoclonal bands

1 Hemorrhagic metastasis 166 80 ND Negative Generalized periodic

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

re-exposure to ICI. Of the 7 patients re-exposed to ICI, 3

had relapse of encephalitis. Treatment for relapse was similar

Response to ICI in patients with immune-related

Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae186/7721714 by guest on 22 October 2024

to ICI at 3-month post-ICI initiation, and 77% at 6-month

than or equal to 7 (n = 13), 85% had favorable responses to

immunosuppression. At last follow-up, 38% of patients died

partial response (median follow up 1098 days, IQR 1290

days). Chronic corticosteroid use was in 38% for continued

No patients were re-exposed to ICIs.

lowed an average of 274 days after diagnosis of immune-

have died (33%). Cause of death included progressive malig-

nancy (44%), immune-related encephalitis (37%), other irAEs

Immune-related encephalitis is a well-recognized neurological

patient case series, one of the largest to date, accompanied by

a review of the literature.

ing malignancies, likely reflecting the frequency of adminis-

than intrinsic cancer factors as these malignancies are not

frequently associated with paraneoplastic encephalitis.1 Most

previously reported patients were treated with PD-1 inhibi-

18% in the literature review). Additionally, brain metastasis

were found in the minority of patients both in the case series

as well as literature review (20%-30%). The presence of brain

metastasis and treatment with radiation or radiosurgery may

immune responses in the brain. Interestingly, although most

irAE occur in the first 3 months of treatment, immune-related

encephalitis could occur more than 1 year after first exposure,

Patients with immune-related encephalitis typically pres-

ent with subacute cognitive changes, meningeal signs, and

personality changes, similar to what is seen in autoimmune

are also present in leptomeningeal carcinomatosis as well as

infectious meningitis/encephalitis, highlighting the impor-

patients are at risk for multiple as well as isolated irAEs. Our