Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer
` (2018) 6:56
`https://doi.org/10.1186/s40425-018-0343-9
`
`R E V I E W
`Cytokine release syndrome
`Alexander Shimabukuro-Vornhagen1,2,3,4*†
`, Philipp Gödel1,2,3†, Marion Subklewe5,6,7,10, Hans Joachim Stemmler5,10,
`Hans Anton Schlößer1,8, Max Schlaak9, Matthias Kochanek2,3,4, Boris Böll2,3,4 and Michael S. von Bergwelt-Baildon1,4,5,7,10
`
`Open Access
`
`Abstract
`
`During the last decade the field of cancer immunotherapy has witnessed impressive progress. Highly effective
`immunotherapies such as immune checkpoint inhibition, and T-cell engaging therapies like bispecific T-cell
`engaging (BiTE) single-chain antibody constructs and chimeric antigen receptor (CAR) T cells have shown
`remarkable efficacy in clinical trials and some of these agents have already received regulatory approval. However,
`along with growing experience in the clinical application of these potent immunotherapeutic agents comes the
`increasing awareness of their inherent and potentially fatal adverse effects, most notably the cytokine release
`syndrome (CRS). This review provides a comprehensive overview of the mechanisms underlying CRS
`pathophysiology, risk factors, clinical presentation, differential diagnoses, and prognostic factors. In addition, based
`on the current evidence we give practical guidance to the management of the cytokine release syndrome.
`Keywords: Cytokine release syndrome, Immunotherapy, CAR T cells, T cell-engaging therapies, Cytokine storm
`
`Background
`Cytokine release syndrome (CRS) is a systemic inflam-
`matory response that can be triggered by a variety of fac-
`tors such as infections and certain drugs. The term
`“cytokine release syndrome” was first coined in the early
`‘90s, when the anti-T-cell antibody muromonab-CD3
`(OKT3) [1, 2] was introduced into the clinic as an im-
`munosuppressive treatment for solid organ transplant-
`ation. Subsequently, CRS has been described after
`infusion of several antibody-based therapies such as
`anti-thymocyte globulin (ATG) [3], the CD28 superago-
`nist TGN1412 [4], rituximab [5], obinutuzumab [6],
`alemtuzumab [7], brentuximab [8], dacetuzumab [9],
`and nivolumab [10]. CRS has also been observed follow-
`ing administration of non-protein-based cancer drugs
`such as oxaliplatin [11] and lenalidomide [12]. Further-
`more, CRS was reported in the setting of haploidentical
`donor stem cell transplantation, and graft-versus-host
`disease [13, 14]. Cytokine storm due to massive T cell
`
`* Correspondence: shima@uk-koeln.de
`†Alexander Shimabukuro-Vornhagen and Philipp Gödel contributed equally
`to this work.
`1Cologne Interventional Immunology, University Hospital of Cologne,
`Cologne, Germany
`2Intensive Care Program, Department I of Internal Medicine, University
`Hospital of Cologne, Cologne, Germany
`Full list of author information is available at the end of the article
`
`stimulation is also a proposed pathomechanism of severe
`viral infections such as influenza [15, 16].
`Lately, with the success of the newer T cell-engaging
`immunotherapeutic agents there has been a growing
`interest in CRS since it represents one of the most fre-
`quent serious adverse effects of these therapies. T cell-
`engaging immunotherapies include bispecific antibody
`constructs and chimeric antigen receptor (CAR) T cell
`therapies. Both these immunotherapeutic strategies have
`recently been carried forward into clinical application
`and have shown impressive therapeutic activity in sev-
`eral hematologic malignancies, such as acute lympho-
`blastic B cell
`leukemia (B-ALL), chronic lymphocytic
`leukemia (CLL), and diffuse large B cell
`lymphoma
`(DLBCL).
`In 2014, the CD19-directed CD3 BiTE blinatumomab
`was approved for Philadelphia chromosome-negative re-
`lapsed or refractory B-cell precursor ALL under the
`FDA’s accelerated approval program [17]. Recently, the
`first two CAR T cell therapies tisagenlecleucel and axi-
`cabtagene ciloleucel received FDA approval for refrac-
`tory CD19-positive B-ALL [18]
`and relapsed or
`refractory large B-cell
`lymphoma [19]. Multiple other
`bispecific antibody and CAR T cell constructs that target
`a variety of antigens are currently in clinical develop-
`ment. Furthermore, there are a number of related T cell-
`engaging
`immunotherapeutic
`approaches
`in earlier
`
`© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
`International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
`reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
`the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
`(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
`CUREVAC EX2035
`Page 1
`
`

`

`Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer (2018) 6:56
`
`Page 2 of 14
`
`clinical development. These include dual-affinity re-
`targeting antibodies (DART), immune-mobilising mono-
`clonal TCRs against cancer (ImmTAC), and other TCR-
`based strategies [20, 21].
`Studies of the first T cell-engaging therapies, i.e. blina-
`tumomab [22] and CD19-targeted CAR T cells [23–25]
`revealed that CRS is the most important adverse event
`of these therapies. Thus, most of the current CRS data is
`derived from CAR T cell and blinatumomab studies in
`hematologic malignancies where CRS has been reported
`in frequencies of up to 100% in CD19-targeted CAR T
`cell trials, sometimes with fatal outcome (Table 1). As in
`the future, T cell-engaging immunotherapeutic agents
`will increasingly be used outside of clinical studies and
`academic cancer centers it becomes paramount that on-
`cologists and intensive care specialists are familiar with
`this complication and its clinical management.
`
`Review
`Clinical presentation
`CRS can present with a variety of symptoms ranging
`from mild, flu-like symptoms to severe life-threatening
`manifestations of
`the overshooting inflammatory re-
`sponse (Fig. 1). Mild symptoms of CRS include fever, fa-
`tigue, headache, rash, arthralgia, and myalgia. More
`severe cases are characterized by hypotension as well as
`high fever and can progress to an uncontrolled systemic
`inflammatory response with vasopressor-requiring circu-
`latory shock, vascular leakage, disseminated intravascu-
`lar
`coagulation,
`and multi-organ
`system failure.
`Laboratory abnormalities that are common in patients
`with CRS include cytopenias, elevated creatinine and
`liver enzymes, deranged coagulation parameters, and a
`high CRP.
`Respiratory symptoms are common in patients with
`CRS. Mild cases may display cough and tachypnea but
`can progress to acute respiratory distress syndrome
`(ARDS) with dyspnea, hypoxemia, and bilateral opacities
`on chest X-ray. ARDS may sometimes require mechan-
`ical ventilation. Of note, in patients with CRS the need
`for mechanical ventilation is oftentimes not due to re-
`spiratory distress but instead a consequence of the in-
`ability to protect the airway secondary to neurotoxicity
`[26]. Patients with severe CRS can also develop renal
`failure or signs of cardiac dysfunction with reduced ejec-
`tion fraction on ultrasound. In addition, patients with se-
`vere CRS frequently display vascular
`leakage with
`peripheral and pulmonary edema.
`In severe cases CRS can be accompanied by clinical
`signs and laboratory abnormalities that resemble hemo-
`phagocytic lymphohistiocytosis (HLH) or macrophage ac-
`tivation syndrome (MAS). Patients with CRS-associated
`HLH display the typical clinical and laboratory findings of
`HLH/MAS such as high fevers, highly elevated ferritin
`
`levels, and hypertriglyeridemia. In a phase III study of bli-
`natumomab in B-ALL four out of 13 CRS patients showed
`signs of HLH [27].
`Some patients develop neurotoxicity after administra-
`tion of T cell-engaging therapies. Neurologic symptoms
`might span from mild confusion with word-finding diffi-
`culty, headaches and hallucinations to aphasia, hemipar-
`esis, cranial nerve palsies, seizures and somnolence. In
`the case of CAR T cell therapy, neurotoxicity represents
`the second most common serious adverse event and
`therefore the term “CAR T cell-related encephalopathy
`syndrome” (CRES) has been introduced [28]. The neuro-
`toxicity of CAR T cell therapy does not seem to be dir-
`ectly related to CRS since neurologic symptoms do not
`always coincide with CRS onset and neurotoxicity can
`occur prior to CRS or after CRS has resolved [29]. The
`pathophysiology of the neurologic symptoms is poorly
`understood, but the lack of a strict temporal association
`with CRS indicates that it might, at least in part, be in-
`dependent from CRS. In addition, experience from clin-
`ical
`trials suggests that
`treatment of
`the neurologic
`symptoms is different from that of CRS.
`
`Epidemiology
`The incidence of CRS in patients receiving cancer im-
`munotherapy varies widely depending on the type of
`immunotherapeutic agent. The onset of CRS can
`occur within a few days, and in the case of CAR T
`cell therapy, up to several weeks after infusion of the
`drug. With most conventional monoclonal antibodies
`the incidence of CRS is relatively low, whereas T cell-
`engaging cancer immunotherapies carry a particularly
`high risk of CRS.
`Although most responding patients experience at least
`some degree of CRS there seems to be no direct associ-
`ation between the severity of CRS and clinical response.
`CRS does not seem to be a prerequisite for response to
`T cell-engaging therapies. Some patients show complete
`remission without obvious signs of CRS, while other pa-
`tients display severe symptoms and laboratory abnor-
`malities but no clinical response.
`Clinical studies identified a number of predictors of
`CRS severity. The risk of CRS is influenced by factors re-
`lated to the type of therapy, the underlying disease, and
`characteristics of the patients. Several clinical factors are
`associated with the severity of CRS following CAR T-cell
`therapy. Many CRS-inducing agents display a “first-dose
`effect”, i.e. the most severe symptoms only occur after
`the first administered dose and do not recur after the
`subsequent administrations [30]. This “first-dose ef-
`fect” is thought to be due to the higher disease bur-
`den at
`initiation of
`treatment. Disease burden is
`among the most important predictors of severe CRS
`after CAR T cell therapy or bispecific T cell-engager
`
`CUREVAC EX2035
`Page 2
`
`

`

`Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer (2018) 6:56
`
`Page 3 of 14
`
`NR
`
`NR
`
`y
`
`NR
`
`NR
`
`NR
`
`NR
`
`y
`
`310
`
`14
`
`0,8
`
`NR
`
`36
`
`NR
`
`NR
`
`2/3CR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`19
`
`11
`
`3
`
`NR
`
`189
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`68
`
`5
`
`4
`
`11
`
`70
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`37
`
`9,4
`
`4,9
`
`14,2
`
`267
`
`B-ALL
`
`B-ALL
`
`B-ALL
`
`B-ALL
`
`blinatumomabblinatumomabblinatumomabblinatumomab
`
`9centers
`
`37centers
`
`26centers
`
`2014[92]6]
`Toppetal.,
`
`2015[68]7]
`Toppetal.,
`
`[91]5]
`etal.,2016
`Stackelberg
`
`[27]4]
`etal.,2017
`Kantarjian
`
`y
`
`y
`
`NR
`
`0
`
`y
`
`13
`
`44
`
`NR
`
`16
`
`B-ALL
`
`(CD28)
`CAR
`CD19
`
`MSKCC101centers
`[33]9]
`2014
`etal.,
`Davila
`
`n*/n*
`
`y*/NR
`
`y*/y*
`
`y*11/y*
`
`y*/y*
`
`y*/y*
`
`3
`
`0
`
`y
`
`27
`
`100
`
`30
`
`1
`
`0
`
`y
`
`32
`
`66
`
`20
`
`B-ALL
`
`B-ALL
`
`MM
`
`(4-1BB)
`CAR
`CD19
`
`UPhil
`UPenn/
`[32]8]
`2015
`etal.,
`Maude
`
`(CD28)
`CAR
`CD19
`
`(4-1BB)
`CAR
`CD19
`
`(CD28)
`CAR
`BCMA
`
`[39]5]
`2015
`etal.,
`Lee
`
`UPennNCI
`[90]3]
`2015
`etal.,
`Garfall
`
`y
`
`possible13
`
`possible13n
`
`NR
`
`NR
`
`y
`
`100%
`
`50%12
`
`y
`
`y
`
`n
`
`n
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`0
`
`NR
`
`9
`
`18
`
`11
`
`NR
`
`n
`
`NR
`
`NR
`
`100
`
`NR
`
`NR
`
`NR
`
`0
`
`0
`
`171
`501
`
`12
`
`n
`
`n
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`5/6
`
`5/6
`
`y/NR
`
`NR
`
`n
`
`y*/y*
`
`y*/y*
`
`y
`
`0
`
`21
`
`23
`
`93
`
`45
`
`16
`
`25
`
`8
`
`83
`
`24
`
`NR
`
`n
`
`NR
`
`0/0
`
`1/1
`
`NR
`
`NR
`
`NR
`
`15
`
`11
`
`18
`
`57
`
`28
`
`28
`
`93
`
`101
`
`MM
`
`B-ALL
`
`CLL
`
`(4-1BB)
`CAR
`CD19
`
`(4-1BB)
`CAR
`CD19
`
`TFL
`DLBCL/
`
`(4-1BB)
`CAR
`CD19
`
`(CD28)
`CAR
`CD19
`
`NCI
`[79]7]
`2016
`etal.,
`Ali
`
`FHCRCSCHRI
`[80]8]
`2017
`etal.,
`Gardner
`
`2017
`etal.,
`Turle
`
`22centersUPenn
`[89]00
`[44]00
`2017
`2017
`etal.,
`etal.,
`Schuster
`Neelapu
`
`25centersMSKCC
`[88]00
`[87]00
`2018
`2018
`etal.,
`etal.,
`Park
`Maude
`
`Author/Year
`Table1CRSreportedinrecentclinicaltrials
`
`n
`
`n
`
`NR
`
`NR
`
`NR
`
`y/n
`
`NR
`
`NR
`
`34
`
`13
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`y*
`
`13
`
`42
`
`26
`
`85
`
`53
`
`NR
`
`NR
`
`NR
`
`NR
`
`NR
`
`y/y
`
`n/y
`
`NR
`
`12
`
`13
`
`46
`
`77
`
`ORR?
`toreduced
`steroidsrelated
`
`reducedORR?
`relatedto
`tocilizumab
`
`toORR?
`CRSrelated
`
`Prognosis
`
`response
`steroid
`
`response
`tocilizumab
`
`Therapy
`
`IL-6/IFNg
`
`CRP/ferritin
`
`tumorburden
`
`sCRScorrelates
`
`deaths
`related
`treatment
`
`(>°II)
`%sNeurotox
`
`(>°II)
`%sCRS
`
`%CRS
`
`Incidences
`
`Numberofpatients75
`
`CUREVAC EX2035
`Page 3
`
`PMBCL
`TFL/
`DLBCL/
`
`B-ALL
`
`(CD28)
`CAR
`CD19
`
`B-ALL
`
`(4-1BB)
`CAR
`CD19
`
`Disease
`
`Appliedtherapy
`
`Institution
`
`

`

`Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer (2018) 6:56
`
`Page 4 of 14
`
`2014[92]6]
`Toppetal.,
`
`2015[68]7]
`Toppetal.,
`
`[91]5]
`etal.,2016
`Stackelberg
`
`[27]4]
`etal.,2017
`Kantarjian
`
`[33]9]
`2014
`etal.,
`Davila
`
`[32]8]
`2015
`etal.,
`Maude
`
`[39]5]
`2015
`etal.,
`Lee
`
`[90]3]
`2015
`etal.,
`Garfall
`
`[79]7]
`2016
`etal.,
`Ali
`
`[80]8]
`2017
`etal.,
`Gardner
`
`2017
`etal.,
`Turle
`
`[89]00
`2017
`etal.,
`Schuster
`
`CUREVAC EX2035
`Page 4
`
`ductaladenocarcinoma,PMBCLprimarymediastinallargeB-celllymphoma,TFLtransformedfollicularlymphoma,sNeurotoxsevereneurotoxicity
`B-ALLacutelymphoblasticBcellleukemia,CLLchroniclymphocyticleukemia,DLBCLdiffuselargeBcelllymphoma,MMMultiplemyeloma,MPMmalignantpleuralmesotheliomas,NRnotreported,PDApancreatic
`statisticallysignificant
`steroidsadditionaltotocilizumabfornotreportedreasons.13:2of9patientstreatedwithimmunosuppression(notfurtherspecifiedifrelatedtotocilizumaborglucocorticoids)relapsed.14:48patientsevaluable.*:
`aspossiblyrelatedtoblinatumomab.10:3deathsduetosepsisandcandidainfectionclassifiedaspossiblyrelatedtoblinatumomab.11:CRP>20mg/dl:positivepredictivevalueonly50%.12:2patientsreceived
`beforereductionofCARTcelldose.7:notfurtherspecified.8:6fatalAEsattributedtoblinatumomab,onedueto°IVCRSwith°Vrespiratoryfailure.9:1deathduetofungalinfectionofthebrainafterHSCTclassified
`adjustmenttodiseaseburden.4:2deathsattributedtoCRS,1topulmonaryembolism.5:1deathduetoneurotoxicity.6:1deathto°VCRSwith°Vcerebraledemarefractorytotocilizumab,siltuximab,dexamethasone
`1:nodefinitionofCRSsupplied,butpatientsshowedsignsofCRS.2:deathduetocerebralhemorrhageinthecontextofcoagulopathyandresolvingcytokinereleasesyndrome.3:1deathdueto°VCRSbeforedose
`
`69(61)
`
`75
`
`43(32)
`
`52
`
`39(20)
`
`45
`
`44
`
`NR
`
`88(75)
`
`NR
`
`66(62)90
`
`10
`
`33
`
`NR
`
`NR
`
`8
`
`25
`
`93(93)
`
`100
`
`17
`
`71
`
`57
`
`64
`
`83(6714)55
`82
`83
`
`81(81)
`
`%CR(MRD-)
`
`81
`
`%ORR
`
`Response
`
`Author/Year
`Table1CRSreportedinrecentclinicaltrials(Continued)
`
`[44]00
`2017
`etal.,
`Neelapu
`
`[88]00
`2018
`etal.,
`Park
`
`[87]00
`2018
`etal.,
`Maude
`
`

`

`Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer (2018) 6:56
`
`Page 5 of 14
`
`Fig. 1 Clinical presentation of CRS. Beginning with fever and unspecific symptoms CRS might impact most organ systems. Mild cases can present
`as flu-like illness. Grade °III to IV shows signs of life threatening cardiovascular, pulmonary and renal involvement. Neurotoxicity can occur concurrent
`or with delay. Abbreviations: DIC: disseminated intravascular coagulation; INR: international normalized ratio; PTT: partial thromboplastin time
`
`administration [5, 31–35]. For instance,
`in patients
`with ALL the burden of disease was associated with
`the severity of CRS [36]. Similar observations have
`been made in a murine lymphoma model, were injec-
`tion of CAR T cells into mice with high tumor bur-
`den resulted in lethal CRS, whereas mice with low
`tumor burden did not show signs of CRS [37, 38].
`The administered dose of the active agent is another
`factor that affects the risk of CRS [33, 35, 39]. Further-
`more, the strength of T cell activation and the degree of
`T cell expansion seem to correlate with the severity of
`CRS [40]. Children seem to be at a higher risk of devel-
`oping CRS than adults. In pediatric patients with ALL
`the incidence of CRS following infusion of CD19-
`targeted CAR T cells was 30/30 (100%) and 16/21 (76%)
`in two clinical trials testing distinct 19-targeted CAR
`constructs [39, 41]. The causes of the higher incidence
`of CRS in pediatric patients are unknown but may be
`
`related to higher cell dose used or the more immature
`immune system of children.
`The type of T cell-engaging agent affects the overall
`risk as well as the onset of CRS. Even though the activa-
`tion of T cells is the common underlying trigger in all
`types of T cell-engaging therapies, there are also import-
`ant differences between the different therapeutic agents
`that affect the incidence, time course, and clinical man-
`agement of CRS. Since CAR T cells can persist in the
`circulation for more than 1 year, the risk for CRS ex-
`tends for a longer period of time but generally is highest
`up to 2 weeks after infusion. In CAR T cell therapy the
`nature of the CAR construct influences the likelihood,
`severity and time to clinical manifestation of CRS.
`Whereas CRS was rarely observed in studies of first gen-
`eration CAR T cell constructs that lacked additional
`costimulatory signaling domains, CRS is much more
`commonly
`reported with second generation CAR
`
`CUREVAC EX2035
`Page 5
`
`

`

`Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer (2018) 6:56
`
`Page 6 of 14
`
`constructs [42]. Even among the different second gener-
`ation CARs there are differences in the rate of CRS.
`CARs with the CD28 costimulatory domain induce a
`brisk but self-limited CAR T cell expansion whereas the
`4-1BB costimulatory domain promotes longer persist-
`ence [43]. CARs that incorporate a CD28 costimulatory
`domain seem to be associated with a higher risk of CRS.
`In two randomized trials of CAR T cells in patients with
`NHL the incidence of CRS was 93% with a CD28-
`containing CAR and 57% with a 4-1BB-containing CAR
`[44, 45]. However, due to differences in the patient pop-
`ulations and differences in the definition of CRS, no de-
`finitive conclusions can be drawn with regard to CAR
`design and the associated risk of CRS. Finally, the type
`of lymphodepletion that was used prior to CAR T cell
`infusion affected the risk of CRS. A higher incidence of
`CRS was observed after lymphodepletion with cyclo-
`phosphamide or fludarabine [26]. This was most likely a
`consequence of the higher expansion rates secondary to
`the more pronounced lymphodepletion achieved by
`combination therapy.
`
`Differential diagnoses
`Clinically, CRS patients present with unspecific syn-
`dromes making the diagnosis challenging. It is important
`to distinguish CRS from other inflammatory disorders
`that present with similar clinical signs and symptoms
`but require different treatment.
`Tumor lysis syndrome (TLS) can mimic CRS and pre-
`sents with symptoms such as fever, acute renal failure,
`cardiac arrhythmia, and seizures. Although tumor lysis
`syndrome usually can be readily discriminated from CRS
`on the basis of characteristic laboratory abnormalities
`such as hyperuricemia, hyperkalemia, hyperphosphate-
`mia and hypocalcemia, it can sometimes be difficult to
`determine if CRS and tumor lysis syndrome occur
`concurrently [46].
`It is important to distinguish patients with CRS from
`those with sepsis since the treatment for CRS could be
`detrimental
`if used in patients with sepsis. Unfortu-
`nately, it is extremely difficult to distinguish sepsis from
`CRS. In fact, according to the most recent definition a
`large percentage of patients with severe CRS will fulfill
`the clinical criteria of sepsis, i.e. suspected infection with
`organ dysfunction defined as an increase of 2 points or
`more in the Sequential Organ Failure Assessment
`(SOFA) score [47]. Furthermore, a significant proportion
`of these patients will also fulfill the criteria for septic
`shock since they have an elevated lactate and require
`vasopressors.
`Patients with CRS are at a high risk of infection and
`the immunosuppressive treatment that is administered
`for the treatment of CRS can mask some of the signs of
`infection thereby delaying diagnosis and treatment of
`
`infection. In one study of 133 patients receiving CD19-
`targeted CAR T cell therapy, 23% of patients developed
`infection within the first 4 weeks after CAR T cell infu-
`sion [48]. The infections typically began after the onset
`of CRS. Among the 93 patients with CRS, 28 (30%) de-
`veloped an infection.
`Infections that occur in patients with CRS are pre-
`dominantly of bacterial origin, followed by viral infec-
`tions that primarily involve the respiratory tract. Fungal
`infections are rare and were primarily observed in pa-
`tients that had previously undergone autologous or allo-
`geneic stem cell transplantation and were suffering from
`severe CRS [48]. The majority of infections occurred
`early after CAR T cell infusion and CRS severity was the
`most important risk factor for infection. It therefore is
`crucial to maintain a high degree of vigilance for infec-
`tion and appropriate empiric antimicrobial
`therapy
`should be rapidly initiated if infection is suspected. All
`patients with CRS should receive an extensive diagnostic
`work-up to exclude infections, including a chest X-ray
`and blood cultures. Furthermore, before start of
`im-
`munotherapy patients should be carefully checked for
`any signs of infection [49]. The mechanism that is re-
`sponsible for the increased incidence of infection in pa-
`tients with CRS is unknown. The CRS-associated
`propensity for infections resembles the severe immuno-
`suppression in patients HLH/MAS, which also are at a
`high risk of serious infectious complications. A plausible
`explanation could be that the massive release of cyto-
`kines in CRS induces a form of immune paralysis, which
`predisposes the patients to an increased risk of infection.
`This hypothesis is consistent with the observation that
`the incidence of infections is higher in patients with
`more severe CRS [48].
`As already mentioned, a HLH/MAS-like syndrome
`can develop as part of the CRS and usually is a manifest-
`ation of severe CRS. CRS-related HLH is difficult to dis-
`tinguish from primary HLH or other conditions that can
`mimic HLH such as sepsis. Table 2 summarizes some of
`the factors that help to distinguish CRS-related HLH
`from other conditions
`that present
`similarly. Even
`though in most cases HLH/MAS that develops concur-
`rently with CRS is triggered by CRS, other causes of
`HLH/MAS, such as genetic defects (in pediatric pa-
`tients), autoimmune disease, infection, or the underlying
`malignancy itself should be taken into account. Patients
`with severe neurotoxicity require a thorough neurologic
`work-up which should include a careful neurologic exam
`and, if appropriate, brain imaging, a spinal tab and an
`electroencephalogram.
`Since most T cell-engaging agents contain non-human
`protein sequences there is a risk of allergic drug reac-
`tions. Hypersensitivity reactions can also present with
`rash and urticaria,
`fever, dyspnea, hypotension and
`
`CUREVAC EX2035
`Page 6
`
`

`

`Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer (2018) 6:56
`
`Page 7 of 14
`
`Table 2 Differential diagnoses of CRS-related HLH/MAS
`Familial HLH
`Secondary HLH/MAS
`Homozygous mutations
`Heterozygous mutations in some patients
`
`Genetic Predisposition
`
`Age group
`
`Biomarkers
`
`IL-10
`IFN-γ
`IL-6
`
`Ferritin
`
`Young children
`
`All ages
`
`↑↑↑
`
`↑↑↑
`
`↑
`
`↑↑↑
`
`↑↑↑
`
`↑↑↑
`
`↑
`
`↑↑↑
`
`↑
`
`↑↑↑
`
`↑↑↑
`
`↑↑↑
`
`CRS-related HLH/MAS
`unknown
`
`All ages
`
`Sepsis
`unknown
`
`All ages
`
`↑
`
`←→
`
`↑↑↑
`
`↑
`
`↑
`
`NDA
`CD163
`CRS cytokine release syndrome, HLH hemophagocytic lymphohistiocytosis, MAS macrophage activation syndrome, Sepsis. NDA no data available
`
`↑↑↑
`
`↑↑↑
`
`gastrointestinal symptoms culminating in cardiorespira-
`tory failure. Unlike in CRS symptoms of true type I reac-
`tions occur after repeated exposure to the causative
`agent [50, 51]. Physicians should consider allergic reac-
`tions as a cause for the patients’ symptoms, in particular,
`after repeat infusion of the immunotherapeutic agent.
`However, so far only few cases of severe allergic reac-
`tions or anaphylactic shock related to immunotherapeu-
`tics have been described in the literature [52].
`If
`anaphylactic shock is suspected epinephrine and antihis-
`tamines should be administered immediately [53].
`Given that all these differential diagnoses have a clin-
`ical presentation that is very similar to CRS, making a
`definitive diagnosis of CRS is very challenging. Since
`some of the therapies given for conditions other than
`CRS can mitigate the effectiveness of immunotherapy,
`the development of reliable diagnostic test that help to
`make the diagnosis of CRS are a high priority for future
`research. Such tests could greatly improve the effective-
`ness and safety of CAR T cell therapy.
`
`Pathophysiology of CRS
`The pathophysiology of CRS is only incompletely under-
`stood. CRS is usually due to on-target effects induced by
`binding of the bispecific antibody or CAR T cell receptor
`to its antigen and subsequent activation of bystander im-
`mune cells and non-immune cells, such as endothelial
`cells. Activation of the bystander cells results in the
`massive release of a range of cytokines. We know little
`about how the initial activation of CAR T cells results in
`the distortion of the cytokine network that drives the in-
`flammatory process in CRS. Depending on a number of
`characteristics of the host, the tumor, and the thera-
`peutic agent the administration of T cell-engaging ther-
`apies
`can set off
`an inflammatory
`circuit
`that
`overwhelms counter-regulatory homeostatic mechanisms
`and results in a cytokine storm that can have detrimen-
`tal effects on the patient. Figure 2 summarizes our
`current understanding of the pathophysiology of CRS.
`IL-6, IL-10, and interferon (IFN)-Υ are among the core
`cytokines that are consistently found to be elevated in
`
`serum of patients with CRS. In the setting of T cell-
`engaging therapies, CRS is triggered by the massive re-
`lease of IFN-γ by activated T cells or the tumor cells
`themselves. IFN-γ causes fever, chills, headache, dizzi-
`ness, and fatigue. Secreted IFN-γ induces activation of
`other immune cells, most importantly macrophages [54].
`The activated macrophages produce excessive amounts
`of additional cytokines such as IL-6, TNF-α, and IL-10.
`TNF-α elicits flu-like symptoms similar to IFN-γ with
`fever, general malaise, and fatigue but furthermore is re-
`sponsible for watery diarrhea, vascular leakage, cardio-
`myopathy, lung injury, and the synthesis of acute phase
`proteins.
`Interleukin 6 (IL-6) seems to hold a key role in CRS
`pathophysiology since highly elevated IL-6 levels are
`seen in patients with CRS [5, 55–57] and in murine
`models of the disease [58]. IL-6 can signal via two differ-
`ent modes. Classical IL-6 signaling involves binding of
`IL-6 to membrane-bound IL-6 receptor. Of note, the IL-
`6 receptor does not possess intracellular signaling do-
`mains.
`Instead, upon binding of
`soluble
`IL-6 to
`membrane-bound IL-6 receptors, the IL-6/IL-6 receptor
`complex binds to membrane-bound gp130, which initi-
`ates signaling through its intracellular domain. In trans-
`signaling, IL-6 binds to the soluble form of the IL-6 re-
`ceptor, which has been cleaved from the cell surface by
`metalloproteinases. This soluble IL-6/IL-6 receptor com-
`plex binds to gp130 and therefore can also induce sig-
`naling in cell
`types that do not express membrane
`bound IL-6 receptors [59].
`IL-6 contributes to many of the key symptoms of CRS.
`Via trans-signaling IL-6 leads to characteristic symptoms
`of severe CRS, i.e. vascular leakage, and activation of the
`complement and coagulation cascade inducing dissemi-
`nated intravascular coagulation (DIC)
`[57, 60].
`In
`addition, IL-6 likely contributes to cardiomyopathy that
`is often observed in patients with CRS by promoting
`myocardial dysfunction [61].
`Recently, Teachey et al. performed a screen for bio-
`markers in patients after CAR T cell therapy for ALL
`and found that peak levels of IL-6, soluble IL-6 receptor,
`
`CUREVAC EX2035
`Page 7
`
`

`

`Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer (2018) 6:56
`
`Page 8 of 14
`
`Target cell lysis
`
`T cell activation
`
`Nivolumab
`
`P D - L 1
`
`P D - 1
`
`Tumor
`cell
`
`T cell
`
`TCR
`
`T A g
`
`MHC-I
`
`TGN1412
`
`CD28
`
`T cell
`
`TCR-
`transgenic T
`cell
`
`T Ag
`
`TCR
`
`MHC-I
`
`Tumor
`cell
`
`TNF-
`
`IL-6
`
`B cell
`
`Blinatumomab
`
`CAR
`
`BiTE
`
`CAR T cell
`
`
`
`T cell
`
`NK cell
`
` Fc-Receptor
`
`Rituximab/Obinutuzumab
`
`B cell
`
`IFN- /TNF-
`
`B/T cell
`
` CD30
`
`Brentuximab
`
`T cell
`
`CCL2/MCP-1
`
`OKT3
`
`T cell
`
`Alemtuzumab
`
`T cell
`
`C D 4
`
`CD3
`
`Anti-thymocyte globulin
`
`T cell
`
`Macrophage
`
`Endothelial
`cell
`
`DC
`
`IL-6
`
`IL-6
`Ang-2/vWF
`
`IL-8
`
`Fig. 2 Reported inducers of CRS. CRS can be induced by direct target cell lysis with consecutive release of cytokines like interferon gamma (IFN-γ) or
`tumor necrosis factor alpha (TNF-α) or by activation of T cells due to therapeutic stimuli with subsequent cytokine release. These cytokines trigger a
`chain reaction due to the activation of innate immune cells like macrophages and endothelial cells with further cytokine release. Abbreviations: Ang-2:
`Angiopoetin 2; CAR: chimeric antigen receptor; DC: dendritic cell; IFN-γ: interferon gamma; MHC-I: major histocompatibility complex I;NK cell: natural
`killer cell; PD-(L)1: programmed cell death protein (ligand) 1; TCR: T cell receptor.; TNF-α: tumor necrosis factor alpha; vWF: von Willebrand factor
`
`IFN-γ, and sgp130 correlated with the risk of severe CRS
`in a cohort of 35 pediatric and adult B-ALL patients re-
`ceiving CD19-CAR T cell therapy. They subsequently
`validated these findings in 12 pediatric patients [62].
`While limited due to the relatively small number of pa-
`tients experiencing sCRS and limited availability of on-
`site cytokine measurement these tools might help in
`identifying patients that need a more intense monitoring
`and treatment.
`
`A hallmark of severe CRS seems to be the activation
`of endothelial cells. Typical marker of endothelial activation
`such as Ang-2 and von Willebrand factor are often elevated
`in the serum of patients with CRS [26]. This indicates that
`the endothelium plays an important role in the pathophysi-
`ology of CRS both by amplifying the inflammatory response
`and as a target organ. The crucial contribution of endothe-
`lial dysfunction in the pathogenesis of CRS provides an ex-
`planation for some of the hallmarks of severe CRS, i.e.
`
`CUREVAC EX2035
`Page 8
`
`

`

`Shimabukuro-Vornhagen et al. Journal for ImmunoTherapy of Cancer (2018) 6:56
`
`Page 9 of 14
`
`capillary leakage, hypotension, and coagulopathy [26]. As
`shown by a recent post mortem study in a patient who died
`of CRS after CD19-targeted CAR T cell therapy, endothelial
`cells seem to be an important source of IL-6 in severe CRS
`[63]. Importantly, endothelial activation and the ensuing
`vascular dysfunction might be the mechanistic factor link-
`ing CRS with neurotoxicity. A recent study found that
`neurotoxicity after immunotherapy with CD19-targeted
`CAR T cells was accompanied by findings consistent with
`endothelial activation [64].
`In patients with CRS who develop a HLH/MAS-like
`syndrome additional cytokines such as IL-18, IL8, IP10,
`MCP1, MIG, and MIP1β are also elevated [62]. These
`cytokines also have been reported to be elevated in clas-
`sical HLH and MAS. Why some patients develop HLH/
`MAS and others do not is poorly understood. Some pa-
`tients may harbor genetic variants that predispose them
`to developing HLH/MAS. In addition, IL-6 may also
`promote the development of HLH/MAS in the setting of
`
`CRS by inducing dysfunction of cytotoxic activity in T and
`NK cells, which is a hallmark of HLH and MAS [65].
`However, a link to genetic aberrations involved in the re-
`lease of cytotoxic molecules (perforin, syntaxin) related to
`familial HLH (PRF1, STX11, STXBP2, and MUNC13–4)
`could not be established in a recent CAR T cell trial [55].
`
`Clinical management of CRS
`The management of the toxicities of cancer immunother-
`apy is challenging clinical problem. Since T cell-engaging
`therapies are a relatively recent development there are still
`many unanswered questions regarding the optimal clinical
`management of CRS. The recommendations for the man-
`agement of CRS are thus still evolving constantly. Current
`treatment algorithms for CRS are based on expert opinion
`and represent the experience of the pioneers in the field of
`T cell-engaging immunotherapies [28, 29]. The most widely
`used grading scheme for the severity of CRS was developed
`by the National Cancer Institute (NCI) (Fig. 3) [29].
`
`Fig. 3 Proposed pathomechanism of CRS. Activation of manly T cells or lysis of immune cells induces a release of interferon gamma (IFN-γ) or
`tumor necrosis factor alpha (TNF-α). This leads to the activation of macrophages, dendritic cells, other immune cells and endothelial cells. These
`cells further release proinflammatory cytokines. I