Recognizing, defining, and managing CAR-T hematologic toxicities

Chimeric antigen receptor (CAR) T-cell therapy has altered the treatment landscape for an ever-increasing number of relapsed and refractory B-cell malignancies,^1-7  but it requires specialized attention to recognize and manage a unique toxicity profile. Next to CRS and ICANS as prototypical CAR-T side effects, additional hematologic toxicities are both frequent and clinically relevant.⁸  Cytopenia following CAR-T, recently termed immune effector cell- associated hematological toxicity (abbreviated as ICAHT and pronounced “eye-cat”) is perhaps the most common noncanonical CAR-T toxicity.⁹  A less frequently observed hematologic complication outside of CRS is a distinct hemophagocytic syndrome, recently termed immune effector cell-associated hemophagocytic lymphohistiocytosis- like syndrome (IEC-HS), which often manifests following resolution of CRS. Coagulopathies are common after CAR-T with both bleeding and thrombotic events experienced in some patients. Tumor lysis syndrome can manifest, although clinical impact is rare.

Here, we present a summary of our current understanding of the manifestations and management of these complications following CAR-T. This information is intended to allow practitioners to recognize and manage these complications promptly, to inform translational investigators trying to elucidate their mechanisms and optimal management approaches, and to prompt experts to action in further standardization of definitions and management algorithms.

A 62-year-old man with relapsed/refractory mantle cell lymphoma (MCL) was referred for CD19-directed CAR-T therapy with brexucabtagene autoleucel (brexu-cel). His prior disease course included chemoimmunotherapy, autologous transplantation, and ibrutinib. Chemotherapy bridging was employed during manufacturing. At lymphodepletion, his laboratory studies were notable for an elevated serum LDH, increased serum inflammatory markers (CRP 4.1  mg/dL, ferritin 881  ng/mL), and bilineage cytopenia (hemoglobin 8.8  g/dL, platelets 122 G/L). Baseline risk assessment was performed using the CAR-HEMATOTOX score, a risk stratification tool comprised of markers of hematopoietic function and baseline inflammation, with this patient presenting with a high-risk score of 4. Bone marrow studies revealed ∼70% cellular involvement by blastoid MCL. On day 2 following CAR-T, the patient developed fever and progressive hemodynamic and respiratory insufficiency (maximal CRS grade 3), which resolved over the course of several days to grade 1 with administration of tocilizumab and corticosteroids. He also developed mild neurocognitive impairment on day 5 with a minimal ICE score of 8 (ICANS grade 1), which resolved on day 7. Following a rapid taper of steroids, on day 10, he presents with pronounced pancytopenia (ANC 0.1 G/L, Hb 7.1  g/dL, platelets 8 G/L).

Profound and/or prolonged cytopenias can predispose patients to significant infectious complications,¹⁰  result in extended hospital stays,¹¹  and prevent subsequent salvage therapies at relapse.¹²  However, the underlying pathophysiology remains enigmatic. While early cytopenia is expected after lymphodepleting chemotherapy, low counts can last weeks, months, or even years after CAR-T.¹³  Neutrophil recovery typically follows either a biphasic or an aplastic trajectory (Figure 1).^8,14,15  Understanding underlying risk factors for hematotoxicity is critical for the application of risk-adapted management strategies.⁹ 

Direct comparison of the incidence of post-CAR-T hematotoxicity, including cytopenias, across trials, and disease entities, is difficult due to differences in trial design, CAR construct, cohort size, and patient population. However, the observed degree and duration of hematotoxicity varies depending on the disease subtype (B-cell precursor acute lymphoblastic leukemia, B-NHL, multiple myeloma) and target antigen (CD19, BCMA) (Table 1). A recent meta-analysis demonstrated higher incidence of post-CAR-T cytopenia in BCP-ALL, likely related to extensive bone marrow infiltration or more intensive prior therapy.¹⁶  High rates of cytopenia have also been noted for MCL, in line with the generally high toxicity burden in these patients.^17-19  Conversely, CAR-T trials for follicular lymphoma demonstrated low rates of hematotoxicity (Table 1), although real-world evidence is needed to confirm. When considering the co-stimulatory domain, matched comparison has revealed increased cytopenias in patients receiving CAR products harboring a CD28z as opposed to 4-1BB co-stimulatory domain.^16,20 

Real-world studies have confirmed the high rate of grade 3 or higher hematological toxicity and especially prolonged cytopenias following both CD19- and BCMA-directed CAR T-cell therapy (Table 2). Detailed studies on the nature of CAR-T-related cytopenias have shed light on the biphasic pattern of neutropenia, with second or even multiple decreases.¹⁴  There are three distinct phenotypes of post-CAR-T neutrophil recovery (Figure 1).⁸  These range from transient lymphodepletion-associated cytopenia (“quick”) to the aforementioned biphasic course (“intermittent”) and the clinically challenging “aplastic” phenotype associated with high morbidity and mortality.^15,21  The relative distribution of these phenotypes after CAR-T therapy is approximately 40%, 40%, and 20% (quick vs intermittent vs aplastic). Patients with the aplastic phenotype are often refractory to G-CSF and can develop prolonged neutropenia (Figure 1 bottom right shows median duration of severe neutropenia is 26 days). Interestingly, biphasic neutrophil recovery is linked to favorable treatment outcomes and higher levels of CAR T-cell expansion and persistence.²¹  Of note, the thrombocytopenic nadir is commonly observed in the second month following CAR-T infusion.⁸ 

Cytopenias can persist long after lymphodepletion and resolution of CRS and ICANS. However, there is marked heterogeneity in the reporting and definitions of these long-term hematological side effects across studies. To address this, an expert panel recently developed a consensus grading system for early (day 0-30) and prolonged/late (after day +30 and day +90 respectively) ICAHT (Table 3). These clear definitions will ease reporting in trials, enable comparative studies of ICAHT severity across disease entities and CAR products, and provide a basis for severity-based management recommendations.

The case highlights a key decision point in managing early ICAHT: namely, when to initiate G-CSF and whether to defer G-CSF in case of coincident immunotoxicity (e.g., severe CRS or ICANS, typically during the first 2 weeks). Hesitation to administer G-CSF stems from preclinical data suggesting that GM-CSF may exacerbate toxicities.²²  However, retrospective analyses from real-word data sets have demonstrated an acceptable safety profile with early G-CSF without increases in the rate of high-grade (e.g., ASTCT grade 3 or higher) CRS or ICANS.^23-27  For example, Miller et al showed that patients receiving prophylactic G-CSF prior to CAR-T infusion (mostly pegylated G-CSF) displayed faster neutrophil recovery, comparable treatment outcomes, and similar rates of severe ICANS.²⁶  Importantly, patients presenting with low-grade toxicity did not exhibit worsening CRS severity with G-CSF. A separate study by Lievin et al found that early G-CSF administration (day +2) reduced febrile neutropenia without increased high-grade CRS or ICANS.²⁵  Notably, G-CSF did not impact CAR-T expansion or efficacy.^24,25  Taken together, these data support early G-CSF in high-risk patients to shorten severe neutropenia and prevent infections. Nonetheless, the optimal day of initiation and G-CSF protocol (prophylactic vs early; pegylated vs non-pegylated) in the context of CAR-T remains unclear. It must be noted that most CAR-T patients (>80%) will adequately respond to growth factor support with count recovery.^12,28 

On day 21, the patient had continued pancytopenia despite daily G-CSF and was transfusion-dependent for platelets and red cells. He developed hospital-acquired pneumonia and received broad anti-infective treatment. Bone marrow studies demonstrated aplasia and no MCL. Viral causes and substrate deficiency were ruled out. Myelotoxic co-medications were paused.

Cases of ICAHT in which patients do not respond to G-CSF can be challenging, and this typifies the aplastic neutrophil recovery phenotype.^8,12  For diagnostic workup, a judicious incremental approach is warranted (Table 4). A first diagnostic tier should rule out other causes of BM insufficiency, including medications, viral infections, substrate deficiency, and severe IEC-HS (further described later). In patients without prior prophylactic G-CSF, we advocate initiation of G-CSF support on day 5-7 in neutropenic patients, particularly those with a high baseline CAR-HEMATOTOX score and in case the first diagnostic tests are inconclusive. If counts have not recovered despite G-CSF support, BM aspiration and biopsy should be employed no later than day 28 to rule out persistent BM infiltration (e.g. progression), perform IEC-HS diagnostics, and evaluate for dysplasia indicative of underlying myeloid dyscrasias which can evolve rapidly following CAR-T infusion.²⁹  The longer cytopenia persists beyond CAR-T infusion, the greater the impetus to perform in-depth cytogenetic studies and/or next-generation sequencing (myeloid panel). This is particularly important for workup of prolonged (day 30-90) and late (beyond day 90) cases of marrow aplasia or new-onset cytopenias long after CAR-T infusion.

In G-CSF refractory cases with an available cryopreserved autologous or allogeneic stem cell product from a prior treatment line (Figure 2), hematopoietic cell boosts should be strongly considered given their favorable safety profile and encouraging engraftment rates.^30-34  Unfortunately, these are available in only a minority of cases, with myeloma patients sometimes having extra stem cells for a potential second consolidative transplant.³⁵  Other options include thrombopoietin receptor agonists (TPO-RA) such as eltrombopag or romiplostim, though the data are restricted to a few small case series.^36-38  The potential improvement of hematopoietic function would mirror the efficacy of TPO-RA in other cases of acquired BM failure.⁴⁰  If the underlying etiology is deemed associated with inflammation or is HLH-like in nature, anti-inflammatory measures such as pulse-dose steroids or anticytokine therapy with anakinra or tocilizumab can be pursued. If the options described above do not facilitate count recovery and grade 4 ICAHT persists, allogeneic hematopoietic cell transplantation (allo-HCT) represents the last resort. However, gradual count recovery can occur over weeks to months, and allo-HCT will inevitably eradicate CAR T-cells. Before commencing with allograft, it is imperative to carefully consider all factors including time from CAR-T, the possibility of spontaneous count recovery, the risk for fatal infections, the likelihood of disease progression, donor suitability/availability, and the patient's goals of care.^28,41,42  Regardless of the treatment strategy, mitigating the risk of severe infections with adequate and broad anti-infective therapy is critical. Fatal infections are possible and represent the main driver of non-relapse mortality.^11,15,43,44  Due to the broad spectrum of opportunistic pathogens, infectious disease consultation is recommended.

The management of CAR-T-related toxicities ideally should be tailored to the patient's individual risk profile. To this end, the CAR-HEMATOTOX (HT) score was developed and externally validated to predict severe hematotoxicity in relapsed and/or refractory (r/r) LBCL.⁸  It was then subsequently extended to patients receiving CAR-T therapy for r/r MCL and multiple myeloma.^17,45  Calculated prior to lymphodepletion (day −5), the score incorporates both factors that relate to the baseline inflammatory state (e.g., C-reactive protein [CRP], ferritin) and the patient's hematopoietic reserve (e.g., hemoglobin, absolute neutrophil count [ANC], platelet count) (Figure 3). High-risk patients (score ≥2) exhibited increased rates of severe infection (especially bacterial infections), higher non-relapse mortality, and inferior treatment outcomes compared to their low-risk counterparts (score 0-1).^11,45,46  We posit that the score may be used to restrict antibacterial prophylaxis and prophylactic G-CSF support to patients with a high-risk profile, as HT^high patients are more likely to benefit from these supportive measures because of their significantly higher rate of febrile neutropenia and infections.^11,26,45,47  On the other hand, HT^low patients may be suitable for avoidance of anti-infectives (i.e., fluoroquinolones, anti-mold agents), which may be beneficial due to the important role of an intact gut microbiome in the context of CAR-T therapy.^48,49  In addition, longitudinal measurements of serum procalcitonin (PCT) may be used to help rule out infections in the context of CRS (e.g., HT^low patients with non-elevated PCT at time of first fever).^46,50  Finally, the score may be useful to identify patients who require more extensive baseline diagnostic evaluations (e.g., pre-CAR-T BM biopsy). Prospective or retrospective data supporting or refuting these strategies should be generated. Limitations of the score relate to the lower specificity and positive predictive value and the fact that it has not been validated for use in BCP-ALL, follicular lymphoma, and pediatric patients (especially following relapse after hematopoietic stem cell transplantation).

The CAR-HEMATOTOX score underlines the importance of baseline hematopoietic function and systemic inflammation as risks for hematological toxicity. Baseline cytopenias likely reflect underlying impairment of the hematopoietic stem and progenitor cell (HSPC) compartment as a result of prior cytotoxic therapies.⁵¹  They are also more frequently observed in cases of disease infiltration of the bone marrow, which represents an independent risk factor for hematotoxicity due to local inflammatory stressors and effects on HSPCs. In line with this observation, patients with prolonged cytopenias following BCMA-targeting CAR-T were more likely to display increased concentrations of CAR T-cells in the marrow post-infusion.⁵²  The high degree of systemic inflammation and secondary inflammatory insults that are stimulated by CAR-T infusion play a relevant pathophysiologic role, with several reports linking high-grade CRS and the associated inflammatory markers to prolonged cytopenias.^28,53  This may in part explain the more extensive hematotoxicity observed with the CD28z-endodomain harboring CAR-T products, although lymphodepletion dosing may also contribute.⁵  In what manner the presence of preexisting clonal hematopoiesis of indeterminate potential (CHiP) may contribute to an underlying inflammatory state and subsequent development of prolonged cytopenias remains to be explored.^54,55 

Oligoclonal CAR-T-cell expansion and T-cell receptor restriction have been observed on the single-cell level in patients with protracted BM aplasia, together with an inflammatory micromilieu reminiscent of acquired aplastic anemia.⁵⁶  Strati et al employed single-cell RNA sequencing of bone marrow from patients with prolonged cytopenia to demonstrate that clonally expanded CXCR1^hi, IFN-γ expressing cytotoxic T cells were associated with hematopoietic stem cells (HSCs) that express IFN-γ response signatures.⁵⁷  This is consistent with data that IFN-γ can impair self-renewal and skew HSC differentiation.^58,59  Most importantly, these results expose potential therapeutic vulnerabilities, as IFN signaling can be targeted using IFN-neutralizing antibodies (e.g., emapalumab) or TPO-RA like eltrombopag.

Patients undergoing CAR-T cell therapy are at risk for a spectrum of coagulopathies, from asymptomatic laboratory abnormalities to disseminated intravascular coagulation (DIC). Petechiae and ecchymosis are the most common manifestations of these coagulopathies. We discuss hemorrhage and bleeding complications, which are infrequent but potentially life-threatening, in a later section. Although coagulation abnormalities have been associated with poor progression-free survival (PFS), these reports are severely limited in their interpretability given their correlations to CRS and ICANS and other inflammatory markers such as CRP and ferritin, which are known to be higher in the face of an immunosuppressive tumor microenvironment and elevation of suppressive myeloid cells in the tumor and periphery.^60,61 

The common toxicities of CAR-T, CRS and ICANS, are associated with a systemic inflammatory response and endothelial breakdown, respectively, both of which are likely contributors to the pathogenesis of consumptive coagulopathy and DIC. In a trial of primarily adult ALL patients treated with CAR-T, increased IL-6 and severe CRS corresponded to laboratory-defined coagulopathies, including D-dimer elevation, increased fibrin degradation products, and activated partial thromboplastin time (aPTT) elongation, while treatment of CRS aligned with their resolution.⁶²  Elevations of platelet endothelial cell adhesion molecule-1 (PECAM-1) and tissue factor (TF) were also seen, consistent with a disruption of endothelial integrity in the setting of CRS and ICANS.⁶²  The incidence of severe CRS has decreased with intervention using tocilizumab and corticosteroids at earlier grades, and the impact of these interventions on the incidence of coagulation abnormalities remains unclear.^1,63 

The frequency and degree of coagulation testing is not standardized across CAR-T trials or in real-world settings. With frequent evaluation, over 50% of patients may be discovered to have coagulation abnormalities within the first month after receiving CAR-T. Prothrombin and thrombin time prolongation, aPTT, and D-dimer elevation typically peak within the first 6-9 days, while fibrinogen nadir may be slightly later at 12-14 days.^60,64  Several attempts have been made to apply standardized DIC criteria scores to laboratory results in CAR-T patients; however, an association with bleeding has not been seen.^62,64  For example, Johnsrud et al applied the International Society on Thrombosis and Hemostasis DIC criteria to LBCL CAR-T-treated patients and found high rates of DIC yet no correlation with bleeding events. They postulated that thrombocytopenia after CAR-T is more likely related to lymphodepleting chemotherapy or immune-mediated suppression rather than platelet consumption seen in classical DIC.⁶⁵  While evaluation of laboratory tests of coagulation may help aid decision-making and management in the CAR-T patient, the diagnostic utility of DIC-scoring algorithms remains uncertain, and more evaluation is needed. There is wide variation across practitioners and institutions on which tests, including D-dimer, fibrin degradation products, thrombin time, prothrombin time, aPTT, and fibrinogen, are checked and when. Furthermore, there is little data to indicate that utilization of these parameters to guide interventions improves outcomes. Our general practice in adult patients is to evaluate these parameters at baseline and again after CAR-T only when a patient experiences severe or refractory CRS or bleeding events.

Fibrinogen concentrate or cryoprecipitate can effectively correct hypofibrinogenemia in the setting of CAR-T therapy. In pediatric and adolescent/young adult ALL patients, severe CRS and consumptive coagulopathies are more commonly observed than in the adult population. Tisagenlecleucel investigators summarized coagulopathies seen in BCP-ALL patients and developed practical guidelines for monitoring and managing coagulopathies, particularly fibrinogen replacement. Despite the relatively common occurrence of hypofinrinogenemia, severe bleeding events were relatively rare, occurring in 1.4% of cases. Buechner and colleagues recommend intervening with cryoprecipitate or fibrinogen only when grade 3-4 CRS occurs and fibrinogen levels are very low at <1 g/L, replacing to >1.5 g/L until CRS resolves to <G3.⁶⁶  In a series of adult LBCL patients undergoing CAR-T, severe hypofibrinogenemia was detected in 6% of patients between day 0 and 100, and it was not associated with bleeding events. However, the fibrinogen level was only checked in 20% of patients and was at the discretion of the treating physician, thus preventing extrapolation of incidence or bleeding risk.⁶⁴  Although grade 3-4 CRS in adult patients is rare, and the incidence of concomitant hypofibrinogenemia is uncertain, practitioner awareness of these guidelines is important when the situation may arise.⁶⁶ 

While severe bleeding events are rare, the risk of hemorrhage after CAR-T may be underappreciated. In pivotal trials of CAR-T cell therapy, severe bleeding events were rare. Bleeding events are most likely to occur within the first 30 days after CAR-T.⁶⁵  In one retrospective series, bleeding events were seen in 11% of LBCL CAR-T patients, a minority of which were severe, and they were seen more frequently in the elderly or those with prior bleeding events, those with baseline and concomitant thrombocytopenia, and patients with ICANS consistent with its known association with endothelial dysfunction.⁶⁵  Reports of cerebral bleeding events, particularly in the setting of CRS and ICANS, may be attributed to this endothelial dysfunction.²⁹ 

The incidence of thrombotic events after CAR-T are variably seen in 2-11% of cases, having associations with D-dimer elevation and ICANS.⁶⁷  In contrast to bleeding events, thrombosis events are more likely to occur up to day +90. Importantly, anticoagulation medications, either as prophylactic or treatment, have not been associated with bleeding events. In a series of 148 LBCL CAR-T patients, 11% of patients experienced a thrombotic event, and these were safely managed with anticoagulation. In this series, routine prophylaxis was not employed. However, therapeutic anticoagulation after thrombotic events was not associated with bleeding.⁶⁴  A more recent series, also in LBCL patients, reported only a 2% rate of thrombotic events when prophylactic anticoagulation was used in most patients.⁶⁸  Taken together, these data suggest that patients at moderate to high risk of thrombosis or with a prior history of thrombosis requiring anticoagulation can safely receive prophylactic or therapeutic anticoagulation, respectively, after CAR-T with cessation at platelet counts <50,000/µL, and such intervention may reduce the incidence of thrombotic events after CAR-T.

Secondary hemophagocytic lymphohistiocytosis can occur after CAR-T, typically with hemaphaogocytosis and other HLH-like features occurring after the resolution of acute CRS and, more commonly, with CD22-directed CAR-T.^69,70  Classical definitions of secondary HLH have diagnostic criteria that overlap with the common features of CRS, which itself is associated with the HLH adjacent macrophage activation syndrome. A need to further delineate CRS from this discrete CAR-T-related HLH entity led one panel of experts on HLH and cell therapy to define and name it IEC-HS, as well as create a severity grading algorithm and management recommendations.⁷¹ 

According to the American Society for Transplantation and Cellular Therapy (ASTCT) expert panel, IEC-HS refers to “the development of a pathological and biochemical hyperinflammatory syndrome independent from CRS and ICANS that (1) manifests with features of macrophage activation/HLH, (2) is attributable to immune effector cell-associated hemophagocytic lymphohistiocytosis-like syndrome therapy, and (3) is associated with progression or new onset of cytopenias, hyperferritinemia, coagulopathy with hypofibrinogenemia, and/or transaminitis.”⁷¹  A requirement for diagnosing the syndrome is an elevated and/or rapidly rising serum ferritin. As biphasic CRS can occur, classical recurrence of CRS as an alternative diagnosis should be considered.

Early recognition is critical, and alternative etiologies such as malignancy progression or infection should be ruled out. Initial therapy with the IL-1 receptor antagonist anakinra, with or without corticosteroids, is suggested. Following 48 hours for response assessment, additional agents such as the JAK1/2 inhibitor ruxolitinib, the anti-IFN-γ antibody emapalumab, or low-dose etoposide chemotherapy could be considered depending on the patient's trajectory.⁷¹ 

Although tumor lysis syndrome (TLS) following CAR-T cell therapy can occur, clinically significant manifestations of the syndrome are unusual in ALL, lymphoma, or myeloma patients and potentially more common in chronic lymphocytic leukemia. As with coagulopathies, evaluation of electrolyte abnormalities may suggest tumor lysis; however, clinical manifestations such as creatinine elevation may have multifactorial etiologies, including hypotension, in the setting of CRS or nephrotoxicity from concomitant co-medications.⁷²  In patients with a history of TLS or those with high circulating tumor burden, allopurinol prophylaxis can be considered.

Recognition of noncanonical hematologic toxicities after CAR-T cell therapy is critical for the practitioner. The development of uniform diagnostic criteria and severity grading systems will inform reporting on their incidence and help investigators elucidate their mechanisms. Significant strides have been made to define ICAHT and IEC-HS as distinct toxicity categories of CAR T-cell therapy. This now sets the stage for evaluating severity-based management recommendations and studying the applicability of these grading systems across disease entities and indications, including for solid tumors, pediatric patients, bispecific antibody therapies, and novel CAR constructs. Future studies will need to examine the potential of prophylactic stem cell collection in patients at ultra high risk for ICAHT, although this may incur an additional logistic burden and increase costs. More work must be done to quantify and define monitoring and management strategies for coagulopathies, bleeding and thrombosis events, and tumor lysis after CAR-T. Emerging areas of interest relate to the association between CHiP and the development of prolonged cytopenias, CRS, ICANS, and therapy-related myeloid neoplasms. Innovative single-cell and multi-omic approaches and modern imaging techniques may help to uncover the still vexing pathophysiology of ICAHT and IEC-HS. Finally, these efforts must ideally leverage the power of multicenter collaborations that span multiple CAR-T centers and countries to optimize resources and leverage diverse patient populations.

Kai Rejeski: Kite/Gilead: Research Funding, Consultancy, Honoraria and travel support; Novartis: Honoraria; BMS/Celgene: Consultancy, Honoraria; Pierre-Fabre: travel support.

Marion Subklewe: receives industry research support from Amgen, BMS/Celgene, Gilead, Janssen, Miltenyi Biotec, Novartis, Roche, Seattle Genetics and Takeda and serves as a consultant/advisor to AvenCell, CDR-Life, Ichnos Sciences, Incyte Biosciences, Janssen, Miltenyi Biotec, Molecular Partners, Novartis, Pfizer and Takeda. She serves on the speakers’ bureau at Amgen, AstraZeneca, BMS/Celgene, Gilead, GSK, Janssen, Novartis, Pfizer, Roche and Takeda.

Frederick L. Locke: scientific advisory role/consulting fees: Allogene, Amgen, bluebird bio, BMS/Celgene, Calibr, Caribou, Cellular Biomedicine Group, Cowen, Daiichi Sankyo, EcoR1, Emerging Therapy Solutions, GammaDelta Therapeutics, Gerson Lehrman Group, Iovance, Kite Pharma, Janssen, Legend Biotech, Novartis, Sana, Takeda, Wugen, and Umoja; research funding (institutional): Kite Pharma, Allogene, Novartis, bluebird bio, CERo Therapeutics, and BMS; patents, royalties, and other intellectual property, including several patents held by the institution in his name (unlicensed) in the field of cellular immunotherapy; education or editorial activity: Aptitude Health, ASH, BioPharma Communications CARE Education, Clinical Care Options Oncology, Imedex, and Society of Immunotherapy of Cancer.

Kai Rejeski: nothing to disclose.

Marion Subklewe: nothing to disclose.

Frederick L. Locke: nothing to disclose.