Pembrolizumab for treatment of progressive multifocal leukoencephalopathy in primary immunodeficiency and/or hematologic malignancy: a case series of five patients

Abstract

Progressive multifocal leukoencephalopathy is a rare opportunistic infection of the brain by John Cunningham polyomavirus in immune-compromised patients. In cases where no overt option for immune reconstitution is available [e.g., in patients with primary immunodeficiency (PID)], the disease is lethal in the majority of patients. Immune checkpoint inhibition has been applied in recent years with mixed outcomes. We present four novel patients and the follow-up of a previously published patient suffering from progressive multifocal leukoencephalopathy (PML) due to PID and/or hematologic malignancy who were treated with the immune checkpoint inhibitor pembrolizumab. In two patients with PID, symptoms improved and stabilized. One patient died because of worsening PML another of intracranial hemorrhage which was unrelated to PML or its treatment with pembrolizumab. The fifth patient suffered from PID and died of a pre-existing immune dysregulation, possibly exacerbated by pembrolizumab. The long-term follow-up of the first patient provides support for therapeutic decisions during this therapy and is the longest published clinical course of a patient with checkpoint inhibition for PML. We conclude that pembrolizumab can control PML symptoms long term in a subgroup of patients with PID, in our cases for 21 and 36 months. However, therapy must be started early because symptoms are only partially reversible. In light of severe adverse events, application of pembrolizumab is only justified if the prognosis for the individual patient is very poor.

Full text

Vol.:(0123456789)
1 3
Journal of Neurology (2022) 269:973–981
https://doi.org/10.1007/s00415-021-10682-8
ORIGINAL COMMUNICATION
Pembrolizumab fortreatment ofprogressive multifocal
leukoencephalopathy inprimary immunodeficiency and/
orhematologic malignancy: acase series offive patients
TimoVolk1 · KlausWarnatz2,3· ReinhardMarks4· HorstUrbach5· GiselaSchluh1· ValentinaStrohmeier2,3,6·
JessicaRojas‑Restrepo3,6,7,8· BodoGrimbacher3,7,8,9,10· SebastianRauer1
Received: 13 May 2021 / Accepted: 21 June 2021 / Published online: 1 July 2021
© The Author(s) 2021
Abstract
Progressive multifocal leukoencephalopathy is a rare opportunistic infection of the brain by John Cunningham polyomavirus
in immune-compromised patients. In cases where no overt option for immune reconstitution is available [e.g., in patients
with primary immunodeficiency (PID)], the disease is lethal in the majority of patients. Immune checkpoint inhibition has
been applied in recent years with mixed outcomes. We present four novel patients and the follow-up of a previously pub-
lished patient suffering from progressive multifocal leukoencephalopathy (PML) due to PID and/or hematologic malignancy
who were treated with the immune checkpoint inhibitor pembrolizumab. In two patients with PID, symptoms improved and
stabilized. One patient died because of worsening PML another of intracranial hemorrhage which was unrelated to PML
or its treatment with pembrolizumab. The fifth patient suffered from PID and died of a pre-existing immune dysregulation,
possibly exacerbated by pembrolizumab. The long-term follow-up of the first patient provides support for therapeutic deci-
sions during this therapy and is the longest published clinical course of a patient with checkpoint inhibition for PML. We
conclude that pembrolizumab can control PML symptoms long term in a subgroup of patients with PID, in our cases for 21
and 36months. However, therapy must be started early because symptoms are only partially reversible. In light of severe
adverse events, application of pembrolizumab is only justified if the prognosis for the individual patient is very poor.
Keywords PML· Pembrolizumab· PID· Hematologic malignancy· Autoimmunity
* Sebastian Rauer
sebastian.rauer@uniklinik-freiburg.de
Timo Volk
timo.volk@uniklinik-freiburg.de
1 Department ofNeurology, Medical Center – University
ofFreiburg, Faculty ofMedicine, University ofFreiburg,
Breisacherstr. 64, 79106Freiburg, Germany
2 Department ofRheumatology andClinical Immunology,
University Medical Center Freiburg, Freiburg, Germany
3 Center forChronic Immunodeficiency (CCI), University
Medical Center Freiburg, Freiburg, Germany
4 Department ofHematology andOncology, University
Medical Center, Freiburg, Germany
5 Department ofNeuroradiology, University Hospital Freiburg,
Freiburg, Germany
6 Faculty ofBiology, University ofFreiburg, Freiburg,
Germany
7 Institute forImmunodeficiency, University Medical Center,
Medical Faculty, Albert-Ludwigs-University ofFreiburg,
Freiburg, Germany
8 RESIST – Cluster ofExcellence 2155 toHanover Medical
School, Satellite Center, Freiburg, Germany
9 DZIF – German Center forInfection Research, Satellite
Center Freiburg, Freiburg, Germany
10 CIBSS – Centre forIntegrative Biological Signalling Studies,
Albert-Ludwigs University, Freiburg, Germany
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

974 Journal of Neurology (2022) 269:973–981
1 3
Introduction
Progressive multifocal leukoencephalopathy (PML) is a rare,
life threatening, opportunistic infection of the central nervous
system caused by the reactivation of John Cunningham poly-
omavirus (JCV) almost exclusively in immunocompromised
patients. Treatment focusses on restoring immunity of the
affected patients since no effective antiviral therapies for JCV
are available. To this end, e.g. combined antiretroviral therapy
is given in HIV-infected patients or immunosuppressive drugs
such as natalizumab are withdrawn in e.g., MS patients. In
contrast, PML in the context of hematologic malignancy or
primary immunodeficiency (PID) is lethal in the majority of
cases as there are no therapeutic options for immune recon-
stitution [1].
The checkpoint inhibitor pembrolizumab is a monoclonal
antibody disrupting interactions of programmed cell death pro-
tein 1 (PD-1) on T cells and its ligands on antigen-presenting
cells. By blocking this receptor involved in negative regulation
of T-cell activation, they boost immune responses. Originally,
checkpoint inhibitors were developed to foster the antitumor
immune response and were able to significantly improve sur-
vival in patients with metastatic melanoma [2] and others. In
line with this mechanism, an association between reduced
T-cell function and high PD-1 expression has also been shown
in chronic viral infections such as HIV [3, 4]. Moreover,
blockade of PD-1 expression in monkeys with chronic simian
immunodeficiency virus infection resulted in an increase in
activated virus-specific T cells, a reduced viral load, and pro-
longed survival of the animals [5, 6], ultimately contributing
to the establishment and maintenance of HIV-1 latency [7].
Laboratory evidence of PD-1 upregulation on T cells in
PML patients [8] led to the off-label use of checkpoint inhibi-
tors in the therapy of PML patients where no other option was
available to reinvigorate antiviral immunity. To our knowledge,
31 published patients with PML have so far been treated with
either pembrolizumab or nivolumab. Treatment outcomes
range from moderate improvement of symptoms with stabili-
zation to death from disease progression [9–25]. To date, there
are no clear prognostic factors to identify patients in whom
checkpoint inhibition will be able to ameliorate PML [26].
Here, we give an update on the clinical course of one pre-
viously published patient [21] and describe four additional
unpublished patients treated with pembrolizumab for PML in
the context of PID and/or hematologic malignancy.
Methods
Patients received pembrolizumab at a dose of 2mg per
kilogram body weight (unless otherwise stated) on a com-
passionate-use basis after informed consent was obtained.
Data of patients presented in this report were retrospectively
collected. For flow cytometric analyses of PD-1 expression,
peripheral blood mononuclear cells were isolated from
EDTA blood of patients and healthy controls by Ficoll den-
sity gradient centrifugation following Standard protocols.
Staining for PD-1 was performed on freshly isolated PBMCs
or whole blood. Staining was performed for 15min at 4°.
In case of whole blood staining, red blood cell lysis was
performed, following staining, using OptiLyse B (Beck-
man Coulter). Data were acquired using an LSR Fortessa
(BD Biosciences) or FACS CANTO II (BD Biosciences).
Data were analyzed using FlowJo Software (Treestar). Cells
were stained with antibodies against CD45RA (PE-Cy7,
HI100), CD3 (APC-H7, SK7), CD 4 (FITC, RPA-T4, all
above from BD Pharmingen, Heidelberg, Germany), CD8
(PerCP-Cy5.5, RPA-T8, Invitrogen, Carlsbad, USA), CD3
(PerCp-Cy5.5, SK7), CD4 (PE-Cy7, RPA-T4), CD8 (PE,
SK1), CD45 (BV421, HI30), CD45RA (APC-Cy7, HI100)
and PD-1 (APC, EH12.2H7, all above from Biolegend, San
Diego, USA). Additional laboratory analysis and magnetic
resonance imaging (MRI) were performed as part of stand-
ard clinical care.
Results
Patient 1: a 21-year-old male patient with the diagnosis of
CD40-ligand deficiency (MIM #308,230) was referred to us
with a diagnosis of PML after progressive visual impairment
over the past two months. MRI showed bi-occipital lesions
on FLAIR (fluid-attenuated inversion recovery) sequences
and a lesion in the globus pallidus on the right. JCV-PCR
from cerebrospinal fluid (CSF) was positive with 471 cop-
ies per milliliter. One week after the first dose of pembroli-
zumab, he developed a maculopapular rash, which dissolved
without further specific treatment and was considered an
immune‐related adverse event. After another week, the
patient was in a stable clinical condition. MRI showed mild
contrast enhancement (Fig.1) as sign of possible immune
reconstitution inflammatory syndrome (IRIS), the JCV-PCR
was still positive (but < 500 copies/ml). Another 2weeks
later the patient developed short term memory impairment
and mild disorientation. Contrast enhancement in MRI
remained stable while FLAIR lesions showed progression.
The patient received a second dose of pembrolizumab 5
weeks after the first one. Subsequently the disorientation
and memory impairment abated, JCV-PCR was negative
and MRI showed stable lesions with persistent contrast
enhancement. A third dose of pembrolizumab was adminis-
tered 8weeks after the first infusion. In follow-up visits the
patient reported minimal improvement of visual impairment
but showed no signs of cognitive impairment. He was able
to resume his studies. MRI showed constant FLAIR lesions
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

975Journal of Neurology (2022) 269:973–981
1 3
without contrast enhancement and JCV-PCR remained nega-
tive at follow-up 15months after the first dose of pembroli-
zumab. A graphical overview of the clinical course is given
in Figure S1 in this article’s Online Repository.
Patient 2: A 45-year-old woman presented with fluctuat-
ing hypesthesia of the left leg for a few weeks. The patient
was on immunoglobulin replacement therapy for common
variable immunodeficiency (CVID) with immune dysregu-
lation including enteropathy, autoimmune cytopenia and
interstitial lung disease. For enteropathy, she received bude-
sonide. Rituximab had been administered 15years before
presentation for immune thrombocytopenia and hemolytic
anemia. MRI showed a parietooccipital FLAIR lesion on
the right hemisphere without contrast enhancement. A ste-
reotactic biopsy was performed after extensive testing of
CSF (including negative JCV-PCR). PML was diagnosed
based on a strongly positive JCV-PCR from the biopsy and
pembrolizumab was administered. Within days, symptoms
progressed with mild hemiparesis on the right. MRI showed
mild contrast enhancement as sign of IRIS. After a complete
stabilization with only mild persistent hypesthesia and no
signs of worsening enteropathy, the patient received a sec-
ond dose of pembrolizumab after 3 weeks. Further treatment
was withheld because of stabilization after the second dose
and the potential reactivation of the pre-existing autoim-
mune condition. After intermittent diarrhea and leukopenia,
a third dose was given 4months after the first dose because
of mild progression of FLAIR lesions on MRI without
contrast enhancement concomitant with increase in PD-1
expression on T cells after an initial decrease (Fig.2). Sub-
sequently, the patient developed severe leukopenia requiring
corticoid therapy and administration of granulocyte colony-
stimulating factor and the diarrhea worsened. Additionally,
there was laboratory evidence of autoimmune hepatitis.
Because of newly developing cognitive deficits with neglect
and progression of the parietooccipital FLAIR lesion and
after discussion with the patient and relatives a fourth dose
of pembrolizumab was administered with a reduced dose of
0.5mg per kilogram body weight after leukopenia, diarrhea
and hepatitis had abated over a period for approximately 1
week. PD-1 expression on T cells decreased again after the
fourth dose but leukopenia reoccurred and she developed
a pneumonia. In light of the clinical deterioration and the
desperate situation the patient opted for palliative care and
subsequently died from bleeding due to severe thrombocyto-
penia 6months after the beginning of pembrolizumab treat-
ment. A graphical overview of the clinical course is given
in Figure S2 in this article’s Online Repository.
Patient 3: A 78-year-old man presented with slowly pro-
gressing hemianopia over a few weeks. He had received
chemotherapy including rituximab for the treatment of a
diffuse large B-cell lymphoma (DLBCL) 1month prior to
Fig. 1 Transiently contrast-enhancing lesion in Patient 1. A Two
months before treatment, bi-occipital lesions with minimal con-
trast enhancemend on the right as well as a lesion in the globus pal-
lidus are visible. B Lesions progress after treatment, a new lesion
in the left frontal lobe, as well as bi-thalamic lesions are visible.
Contrast enhancement is increased in the right occipital lesion. C
Lesion in the thalamus is less intense, otherwise stable MRI. D + E
Declining intensities of lesions, no contrast enhancement is detect-
able15 months after initial treatment. Top row: FLAIR (Fluid-atten-
uated inversion recovery) MRI sequences, bottom row: MP RAGE
(magnetization-prepared 180 degrees radio-frequency pulses and
rapid gradient-echo sequences) MRI sequences post gadolinium
application. m month, w week
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

976 Journal of Neurology (2022) 269:973–981
1 3
development of the symptom. MRI showed a large right
occipital FLAIR lesion without contrast enhancement. JCV-
PCR of CSF was positive (< 500 copies/ml). Pembrolizumab
was administered. Three weeks later, the patient reported
worsening vision of the left eye and a tumor of the iris was
identified. Additionally, he showed several tumors of the arm
and leg suspicious of enlarged lymph nodes. A biopsy of one
lymph node showed a relapse of the lymphoma. The occipi-
tal FLAIR lesion had progressed without contrast enhance-
ment. However, a left sided contrast-enhancing lesion had
developed, suspicious of lymphoma. While being evaluated
for further therapeutic options regarding the lymphoma,
he developed a coma due to an extensive intracranial hem-
orrhage on the left hemisphere under full anticoagulation
therapy for atrial fibrillation. He died a few days later under
palliative care. A graphical overview of the clinical course is
given in Figure S3 in this article’s Online Repository.
Patient 4: A 45-year-old man presented with progress-
ing dysarthria, impaired vision and attention deficit for 2
weeks. He suffered from a combined immunodeficiency
due to a deleterious compound heterozygous mutation
in DOCK8 (dedicator of cytokinesis 8, MIM # 243,700),
symptomatic with recurrent tumors of the skin, two epi-
sodes of meningitis in childhood, and mild lymphopenia
(including at presentation to our clinic). An MRI showed
bi-occipital FLAIR lesions without contrast enhancement
and after positive JCV-PCR in the CSF, the diagnosis of
PML was made. Pembrolizumab was administered for the
first time. Relatives described decreasing dysarthria after
the first dose. After another 2weeks the patient developed
non-fluent aphasia, attention deficit, further deterioration of
his vision and right-sided hemiparesis. JCV-PCR from the
CSF rose to 10.000 copies/ml and FLAIR lesions progressed
with contrast enhancement at the borders. A second dose of
pembrolizumab was administered. After being discharged to
a rehabilitation clinic with stable deficits, the patient devel-
oped abdominal pain without diarrhea. No clear etiology for
this pain (including autoimmune enteropathy, pancreatitis or
hepatitis) could be established, and the pain subsided after
antibiotic treatment. Symptoms of PML had progressed to
severe spastic hemiparesis, JCV-PCR from the CSF was
again positive with 68.500 copies/ml and the MRI showed
a progression of lesions to the thalamus. Since PML lesions
progressed with only mild maculopapular rash as sign of an
autoimmune adverse reaction a third dose of pembrolizumab
was administered. One week later the patient had two gen-
eralized epileptic seizures and fell into a persistent coma.
Palliative care was initiated and the patient died a few days
later. A graphical overview of the clinical course is given in
Figure S4 in this article’s Online Repository.
Patient 5: A 49-year-old patient presented to us with
aphasia, ataxia, apraxia and difficulties in memory. This
patient’s initial clinical course has already been pub-
lished [21]. In short, he was diagnosed with PML by
positive JCV-PCR (1,150 copies/ml) in CSF and left-
hemispheric FLAIR lesions on MRI. He had previously
been diagnosed with CVID and had received Rituximab
for DLBCL. Because of worsening of symptoms (mut-
ism) and progressing lesions on MRI pembrolizumab was
administered. The patient was subsequently able to speak
again and JCV-PCR was negative except for 1 analysis
when intervals of dosing were extended from 2 to 4 weeks.
Contrast enhancement was detected on MRI after the 6th
infusion compatible with IRIS, but the patient showed
Fig. 2 Patients show reduced expression of PD-1 on T cells after
Pembrolizumab treatment. PD-1 expression is within normal range on
CD4 + (A) and CD8 + (B) T cells in Patient 1 and 2 before treatment.
Patient 3 and 4 show normal expression on CD8 + (B) and high nor-
mal expression on CD4 + T cells before treatment. All patients show
a reduction of PD-1 expression after treatment. An increase is seen on
T cells from Patient 2 four and five months after the first infusion dur-
ing severe autoimmune adverse reactions. Cells from Patient 5 were
analyzed only 16 and 18months after commencement of therapy and
show low to low normal expression of PD-1
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

977Journal of Neurology (2022) 269:973–981
1 3
no corresponding symptoms. After the 8th infusion the
patient had self-limiting diarrhea for a few days. Because
of stable symptoms and MRI lesions dosing intervals
were extended to monthly intervals after the 7th infusion.
Before the 10th infusion, JCV-PCR was positive, there-
fore pembrolizumab was then given every 2 weeks and
after negative PCR every 3 weeks. After the 24th infusion
Transaminases were elevated and therefore pembrolizumab
was halted. JCV-PCR was positive for 4months (2months
after pembrolizumab was halted), but was negative again
without further administration of pembrolizumab. Lym-
phopenia had abated without specific treatment. Thirty-
nine months after onset of symptoms and 36months after
the first infusion, the patient is stable with severe aphasia,
cognitive impairment and is dependent on care from his
wife. However, during treatment he showed incremental
improvement of aphasia and interaction (ability to utter
needs with incomplete sentences and comprehension and
execution of simple commands) and ataxia (ability to
walk on his own and recently successful attempts to ride a
bike with support). MRI shows progressive tissue atrophy
without signs of active PML (Fig.3). JCV-PCR remained
negative. A graphical overview of the clinical course is
given in Figure S5 in this article’s Online Repository.
Discussion
Without efficient antiviral therapy, the mainstay of PML
therapy is immune reconstitution. In conditions where rever-
sal of immune suppression is not achieved mortality is high.
This is demonstrated by a median survival of 2months in
a case series of patients with in hematologic malignancies
and stem-cell transplantation [27]. Checkpoint inhibition has
been applied so far in 11 cases with nivolumab therapy and
20 cases with pembrolizumab therapy (including our previ-
ous report) with varying results [9–25]. Here were report
on a series of 4 novel patients with PML in the context of
PID or hematologic malignancy treated with pembrolizumab
and follow-up data on one previously reported patient [21].
Two patients in this cohort showed long-term stabiliza-
tion judged by regression of MRI lesions and ultimately neg-
ative JCV-PCR from CSF. Symptoms improved moderately
but patients still suffer from significant sequelae from PML.
Fig. 3 Shrinking PML lesion during therapy and progressive atrophy
in Patient 5. FLAIR images showing the largest lesion extension (A)
after two months. B Shrinkage of the lesion four months later (B),
previously published in [21]. (D) and (E) showing progressive tissue
atrophy between 30 and 35 months after commencement of pem-
brolizumab. C is an overlay of A and B, (F) is an overlay of (D) and
(E) with 50% transparency each. m indicating months after the first
administration of pembrolizumab, over overlap
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

978 Journal of Neurology (2022) 269:973–981
1 3
This is representative of other published patients with good
clinical outcomes, in whom deficits were stable or showed
moderate improvement [9, 10, 15, 23, 25]. Only one patient
with frontal lobe syndrome due to PML showed complete
independence in daily life after treatment with IL-2 and
pembrolizumab [17]. The death of Patient 3 due to intrac-
ranial hemorrhage—most likely in a lesion of his recurrent
lymphoma under anticoagulation therapy—rather demon-
strates the aggressive disease causing the underlying immu-
nodeficiency than failure of pembrolizumab. Death from
other causes (mainly progressive cancer) is also reported
in other PML patients treated with nivolumab or pembroli-
zumab [13, 17].
In Patient 4 disease progression could not be stopped,
demonstrated by rising copy numbers of JCV-DNA in CSF
and increasing lesions on MRI. Similar catastrophic clini-
cal courses have been published previously, including a
young patient who was diagnosed early after mild symp-
toms developed and had low copy numbers of JCV-DNA in
CSF at diagnosis [10, 12, 18, 20, 22]. This challenges the
concept that the disease can be controlled if treatment was
commenced early and raises the question, if the underly-
ing immunodeficiency itself is a prognostic factor. Medrano
etal. hypothesize that the fatal outcome in all of their three
patients was due to the cyclosporine given to prevent kidney
transplant rejection [18]. However, a patient who responded
to nivolumab also received cyclosporine one month prior to
developing PML [23]. In our case the patient suffered from
a combined immunodeficiency due to mutations in DOCK8
known for difficulties in the control of certain viral infec-
tions, so that even the blocking of PD1-mediated inhibi-
tion may have not been sufficient to overcome the T-cell
deficiency which are also known to have a migratory defect
in DOCK8-deficient patients, uncurable by checkpoint
blockade.
Therefore, a careful evaluation of the cellular immune
status may be informative of the potential outcome. Severe
and especially inborn lymphopenia or functional T-cell defi-
ciency might dampen the therapeutic success of checkpoint
inhibition. Along this line, two groups applied IL-2 before
administration of pembrolizumab in three patients. One
patient had a good outcome, the other two died of other
diseases (lymphoma and pneumonia) [11, 17]. Patients with
a desirable outcome in our report did not show lymphope-
nia before therapy (Patient 1) or showed rising lymphocyte
counts during therapy (Patient 5). Regarding lymphocyte
subsets, B-cell lymphopenia was most prominent in our
cohort (Table1) and in a recently published series of six
cases [25]. However, B-cell counts did not serve as a prog-
nostic marker exemplified by Patient 4 having normal B-cell
counts and no response to therapy. T-cell subset analyses
showed an expansion of activated memory T cells while
naïve populations were reduced – often seen in patients
with PID—and did not differ between patients with regard
to their outcome (TableS1). Pawlitzki etal. propose analyz-
ing lymphocyte subsets since they found higher frequencies
of progenitor-exhausted T cells in a patient with good clini-
cal response to pembrolizumab and higher frequencies of
terminally exhausted T cells in a patient without response to
pembrolizumab [20]. Progenitor and terminally exhausted T
cells have not been analyzed in out cohort.
PD-1 expression on T cells from peripheral blood from
our patients was within the range of healthy controls. Pem-
brolizumab administration abrogated PD-1 expression suc-
cessfully on circulating T cells in all patients and therefore
was not predictive for outcome or negativity of JCV-PCR
(Fig.2). Patient 1 showed an increase in PD-1 expression
14months after the last infusion. JCV-PCR was negative
13months after the last infusion, indicating that viral control
is possible even after the effect of pembrolizumab wears off.
In Patient 5 PD-1 expression was still decreased 4months
after the last infusion. JCV-PCR was transiently positive at
that point. Patient 2 showed clinical and MRI progression
of the disease after an initial response. While PML was pro-
gressing, PD-1 expression first increased and then decreased
again. This indicates that a reduction in PD-1 expression
might be prerequisite for viral control at beginning of ther-
apy but does not strictly correlate with viral control or symp-
toms of PML.
In line with these findings, a clinical response to pem-
brolizumab was not correlated to PD-1 suppression in a
cohort of eight patients published by Cortese etal. [10].
However, invitro T-cell reactivity to JCV peptides was only
detected in patients who also responded to therapy. This
indicates that pembrolizumab might only be able to support
a pre-existing immune response to JCV.
Although pre-existing autoimmune diseases not necessar-
ily exclude treatment with checkpoint inhibitors in patients
with cancer, clinical experience regarding exacerbation of
already established autoimmune conditions after applica-
tion of checkpoint inhibitors is rare. Nevertheless, autoim-
mune phenomena (e.g., diarrhea, pneumonitis) as a result of
checkpoint inhibition is frequently being observed in cancer
patients. Autoimmune adverse reactions, such as myositis,
rashes and worsening of a pre-existing digestive tract granu-
lomatosis, have been reported in PML patients [14, 23, 25].
None of the previously published PML patients experienced
autoimmune cytopenia after treatment with pembrolizumab
or nivolumab, despite one of the patients suffered from auto-
immune thrombocytopenia before development of PML [10,
16, 20, 24]. In contrast, Patient 2 showed marked exacerba-
tion of pre-existing cytopenia with only transient response
to granulocyte colony-stimulating factor and corticoster-
oid therapy and enteropathy, which was not prevented by
increased budesonide prophylaxis. She developed a pneu-
monia during leukopenia and ultimately died from bleeding
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

979Journal of Neurology (2022) 269:973–981
1 3
Table 1 Patient characteristics, laboratory data and clinical course
JCV John Cunningham polyomavirus, CSF Cerebrospinal fluid, IRIS immune reconstitution inflammatory syndrome, CE contrast enhancement on MRI, CVID common variable immunodefi-
ciency, Leuko Leukocytes, Lympho Lymphocytes, DLBCL diffuse large B-cell lymphoma, CID combined immunodeficiency
Bold numbers indicate values above (↑) or below (↓) reference values [30, 31]
† Table depicting values of subsets and subset defining markers in TableS1 in this article’s Online Repository
Patient Sex and age at
presentation underly-
ing condition
Cell counts before first dose
[/µl]
T-cell subsets before
first dose†
JCV PCR in CSF
at diagnosis [cop-
ies/ml]
JCV PCR in CSF
during treatment
Start of pembroli-
zumab after onset of
PML symptoms
IRIS Possible autoimmune
adverse reactions
Outcome
Patient 1
Male, 21years
CD40-ligand
deficiency
Leuko
Lympho
CD3
CD19
CD4
CD8
6810
3070 ↑
2442 ↑
479
457
1577↑
Expansion of
activated CD8 + T
cells and
early and late
CD8 + T cells
471 Negative after 2 infu-
sions 8weeks after
first infusion
10weeks Possibly, CE after
first infusion,
confusion 4weeks
later
Maculopapular rash Stable, significant
sequelae (cortical
blindness)
Patient 2
Female, 45years
CVID
Leuko
Lympho
CD3
CD19
CD4
CD8
4150
200 ↓
183 ↓
0 ↓
142 ↓
37 ↓
Reduction of naive
CD4 + T cells
Expansion of
memory CD8 + T
cells
Negative in CSF,
PCR from brain
biopsy positive
Not applicable
(negative in CSF at
diagnosis)
Approx. 3months Likely, CE 4days
after first infusion,
concomitant wors-
ening of symptoms
Recurrent diarrhea,
severe pancyto-
penia
Death, possibly due to
autoimmune com-
plications (bleeding,
severe thrombocy-
topenia), initially
improved regarding
PML symptoms
Patient 3
Male, 78years
DLBCL
Leuko
Lympho
CD3
CD19
CD4
CD8
7610
864 ↓
778
0 ↓
500
265
Not done Positive (< 500) Negative after 1 infu-
sion 4weeks after
first infusion
< 4weeks (precise
onset unknown)
No None Death, due to unrelated
disease, stable
regarding PML
symptoms
Patient 4
Male, 45years CID
due to DOCK8
deficiency
Leuko Lympho
CD3
CD19
CD4
CD8
7580
861 ↓
555 ↓
195
261 ↓
282
Reduction of naive
CD4 + T cells and
naive CD8 + T cells
500 Rising to 68.500
after 2 infusions,
6weeks after first
infusion
2weeks No, CE at borders
of PML lesion
concomitant with
rising JCV in
CSF suggestive of
advancing PML
maculopapular rash Death due to rapidly
worsening PML
Patient 5
Male, 49years
CVID, DLBCL
Leuko
Lympho
CD3
CD19
CD4
CD8
5000
234 ↓
111 ↓
0 ↓
40 ↓
62 ↓
Reduction of naive
CD4 + and CD8 + T
cells
Expansion of acti-
vated T cells, anti-
gen experienced
CD4 + T cells and
early CD8 + effec-
tor cells
1150 Rising to 252.500
after first infusion
Negative after 5
infusions
9weeks after first
infusion
Subsequently Tran-
siently positive
with low copy
numbers§
4months Unlikely, CE
9weeks after first
infusion without
corresponding
symptoms
Transient mild diar-
rhea, Transaminitis
Stable, initial improve-
ment of attention
and speech, severe
sequelae
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

980 Journal of Neurology (2022) 269:973–981
1 3
due to thrombocytopenia likely triggered by pembrolizumab
after wishing for no further treatment. Given her significant
initial improvement of symptoms caused by PML, our pro-
phylactic options for secondary immune exacerbation must
be improved and in the absence of suitable prophylaxis
benefits of checkpoint inhibition must be strictly weighed
against possible risks in patients with pre-existing autoim-
mune conditions and patients need to be plainly informed
about the potential risk.
IRIS after checkpoint inhibition has rarely been reported
[13, 21, 22, 24]. Four of our five patients showed contrast
enhancement on MRI after initiation of pembrolizumab, but
only Patient 2 had concomitant worsening of symptoms. Par-
alleled to his clinical deterioration patient 4 showed marked
contrast enhancement on the borders of the lesions on MRI
also in line with progressing PML. In HIV-infected patients
signs of IRIS have been proposed to indicate better progno-
sis [28, 29], results from our patients and previously pub-
lished patients are ambiguous.
Given the short history of pembrolizumab therapy in
PML, many therapeutic decisions have to be made with-
out precedence. Observations from patients in this report
including a unique 38-month follow-up add substantially to
previously published data: (i) Patients with certain pre-exist-
ing T-cell deficiencies may not respond to pembrolizumab
therapy for PML. (ii) The number of infusions required for
disease control may vary between patients and (iii) patients
with poorly controlled autoimmune disorders may not be
able to receive the minimal effective dose of pembrolizumab
due to severe life-threatening exacerbation of autoimmune
adverse reactions. (iv) Fluctuating results from JCV-PCR
may not necessitate reinstallation of pembrolizumab.
Conclusion
For a subset of patients with immunodeficiency due to PID
and/or hematologic malignancy pembrolizumab treatment
can stop disease progression while symptoms may improve
only moderately. Further studies must carefully define sub-
set of patients, in whom checkpoint inhibition is a valuable
treatment option. Laboratory analyses of invitro T-cell reac-
tivity to JCV [10] or detection of progenitor-exhausted T
cells are candidate surrogates for an immune state which
can be enhanced by checkpoint inhibition. Pre-existing
autoimmunity is a severe contraindication which requires a
careful and consented decision if therapy is justified on an
individual basis. Clinical courses of treated patients must be
closely monitored and analyzed to reach optimal outcome.
Supplementary Information The online version contains supplemen-
tary material available at https:// doi. org/ 10. 1007/ s00415- 021- 10682-8.
Author contributions TV analyzed and interpreted data and wrote the
manuscript. KW, BG, and SR recruited patients, designed the research
and edited the manuscript. RM, HU, VS, and JR-R analyzed and inter-
preted data.
Funding Open Access funding enabled and organized by Projekt
DEAL. B.G. receives support by the Deutsche Forschungsgemeinschaft
(GR1617/14-1/iPAD; SFB1160/2_B5; RESIST–EXC 2155–Project
ID 390874280; and CIBSS–EXC-2189–Project ID 390939984) and
the BMBF (GAIN 01GM1910A). He declares no conflict of interest
wrt the content of this manuscript. V.S. is funded by the Deutsche
Forschungsgemeinschaft (grant to K.W. SFB1160 TP A04). T.V., R.M.,
H.U., G.S., J.R.-R. and S.R. received no funding for this study.
Data availability Not applicable.
Code availability Not applicable.
Declarations
Conflicts of interest Declare no conflict of interest wrt the content of
this manuscript.
Ethics approval Not applicable.
Consent to participate All patients received pembrolizumab on a com-
passionate-use basis after informed consent was obtained.
Consent for publication All patients, guardians or next of kin have gave
written informed consent for publication.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
References
1. Cortese I, Reich DS, Nath A (2021) Progressive multifocal leu-
koencephalopathy and the spectrum of JC virus-related disease.
Nat Rev Neurol 17(1):37–51
2. Curry WT, Lim M (2015) Immunomodulation: checkpoint block-
ade etc. Neuro Oncol 17(Suppl 7):vii26–vii31
3. Grabmeier-Pfistershammer K etal (2011) Identification of PD-1
as a unique marker for failing immune reconstitution in HIV-
1-infected patients on treatment. J Acquir Immune Defic Syndr
56(2):118–124
4. Zhang JY etal (2007) PD-1 up-regulation is correlated with HIV-
specific memory CD8+ T-cell exhaustion in typical progressors
but not in long-term nonprogressors. Blood 109(11):4671–4678
5. Barber DL etal (2006) Restoring function in exhausted CD8 T
cells during chronic viral infection. Nature 439(7077):682–687
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

981Journal of Neurology (2022) 269:973–981
1 3
6. Velu V etal (2009) Enhancing SIV-specific immunity invivo by
PD-1 blockade. Nature 458(7235):206–210
7. Evans VA etal (2018) Programmed cell death-1 contributes
to the establishment and maintenance of HIV-1 latency. AIDS
32(11):1491–1497
8. Tan CS etal (2012) Increased program cell death-1 expression on
T lymphocytes of patients with progressive multifocal leukoen-
cephalopathy. J Acquir Immune Defic Syndr 60(3):244–248
9. Audemard-Verger A etal (2019) Sustained response and rationale
of programmed Cell Death-1-targeting for progressive multifocal
leukoencephalopathy. Open Forum Infect Dis 6(10):ofz374
10. Cortese I etal (2019) Pembrolizumab treatment for pro-
gressive multifocal leukoencephalopathy. N Engl J Med
380(17):1597–1605
11. Goereci Y etal (2020) Clearance of JC polyomavirus from cer-
ebrospinal fluid following treatment with interleukin-2 and pem-
brolizumab in an individual with progressive multifocal leukoen-
cephalopathy and no underlying immune deficiency syndrome.
Eur J Neurol 27(11):2375–2377
12. Grassl N etal (2020) Nivolumab for treatment of progressive mul-
tifocal leukoencephalopathy in Sezary syndrome. Eur J Neurol
27(11):2373–2374
13. Hoang E etal (2019) Progressive multifocal leukoencephalopathy
treated with nivolumab. J Neurovirol 25(2):284–287
14. Holmes A etal (2020) Progressive multifocal leukoencephalopa-
thy associated with a lymphoproliferative disorder treated with
pembrolizumab. J Neurovirol 26(6):961–963
15. Kapadia RK, Ney D (2020) Stabilization of progressive multifocal
leukoencephalopathy after pembrolizumab treatment. Neurohos-
pitalist 10(3):238–239
16. Kupper C etal (2019) Pembrolizumab for progressive multifocal
leukoencephalopathy due to primary immunodeficiency. Neurol
Neuroimmunol Neuroinflamm 6(6):e628
17. Mahler C, etal. (2020) Sequential interleukin 2 and pembroli-
zumab use in progressive multifocal leukoencephalopathy. Neurol
Neuroimmunol Neuroinflamm 7(4).
18. Medrano C etal (2019) Effectiveness of immune checkpoint
inhibitors in transplant recipients with progressive multifocal
leukoencephalopathy. Emerg Infect Dis 25(11):2145–2147
19. Mozo Ruiz M etal (2020) Pembrolizumab treatment for progres-
sive multifocal leukoencephalopathy in a patient with Human
Immunodeficiency Virus infection. Enferm Infecc Microbiol Clin
38(8):396–397
20. Pawlitzki M etal (2019) Ineffective treatment of PML with pem-
brolizumab: exhausted memory T-cell subsets as a clue? Neurol
Neuroimmunol Neuroinflamm 6(6):e627
21. Rauer S et al (2019) Treatment of progressive multifocal
leukoencephalopathy with pembrolizumab. N Engl J Med
380(17):1676–1677
22. Stogbauer J etal (2021) Clinical and magnetic resonance imaging
monitoring in progressive multifocal leukoencephalopathy treated
with pembrolizumab: a case report. Neurol Sci 42(1):357–359
23. Uzunov M etal (2020) Postallogeneic transplantation progres-
sive multifocal leukoencephalopathy successfully treated by
nivolumab. Br J Haematol 188(6):e82–e84
24. Walter O etal (2019) Treatment of progressive multifocal leukoen-
cephalopathy with nivolumab. N Engl J Med 380(17):1674–1676
25. Roos-Weil D, etal. (2021) Immune checkpoint inhibitors for pro-
gressive multifocal leukoencephalopathy: a new gold standard? J
Neurol.
26. Du Pasquier RA (2019) Pembrolizumab as a treatment for
PML? Waiting for Godot. Neurol Neuroimmunol Neuroinflamm
6(6):e629
27. Adrianzen Herrera D etal (2019) Characteristics and outcomes
of progressive multifocal leukoencephalopathy in hematologic
malignancies and stem cell transplant - a case series. Leuk Lym-
phoma 60(2):395–401
28. Berger JR etal (1998) Predictive factors for prolonged survival
in acquired immunodeficiency syndrome-associated progressive
multifocal leukoencephalopathy. Ann Neurol 44(3):341–349
29. Sainz-de-la-Maza S etal (2016) Incidence and prognosis of
immune reconstitution inflammatory syndrome in HIV-associ-
ated progressive multifocal leucoencephalopathy. Eur J Neurol
23(5):919–925
30. Hulstaert F etal (1994) Age-related changes in human blood
lymphocyte. II. Subpopulations varying kinetics of percentage
and absolute count measurements. Clin Immunol Immunopathol
70(2):152–158
31. Schatorje EJ etal (2011) Age-matched reference values for B-lym-
phocyte subpopulations and CVID classifications in children.
Scand J Immunol 74(5):502–510
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com