Introduction
With standard immunochemotherapy, e.g., R‑CHOP (rituximab+ cyclophosphamide, doxorubicin, vincristine, and prednisone), patients with diffuse large B‑cell lymphoma (DLBCL) can achieve an overall response rate (ORR) of 60% resulting in a long-term event-free survival of 50% [
1].
Unfortunately, 30–40% of patients eventually relapse and 10% are primary refractory. Despite intensive salvage immunochemotherapy and autologous stem cell transplantation (ASCT), outcome in these patients is poor with an ORR of 27–63% and long-term survival in up to 48% [
2]. With the introduction of (mostly autologous) chimeric antigen receptor T‑cell (CAR-T) therapy, very encouraging results with CR rates up to 50% in r/r DLBCL have been demonstrated leading to approval of two products by the US FDA (Food and Drug Administration) and EMA (European Medicines Agency).
Chimeric antigen receptor T-cells
A CAR is commonly composed of a specificity-conferring extracellular antibody single chain variable fragment (scFv), a hinge region transmembrane domain, one or more intracellular costimulatory domains (e.g., CD28 or 4‑1BB [CD137]), and a T-cell receptor signaling domain (CD3 ζ). Several generations of CARs can be distinguished, differing by co-stimulating signaling domains (CD28, 4–1 BB) being responsible for T‑cell activation and expansion [
3].
CAR T-cells in non-Hodgkin’s lymphoma
With the introduction of CAR T‑cells targeting CD19-positive lymphoid malignancies, encouraging response rates have been observed in heavily pretreated patients.
The ZUMA1 trial (NCT02348216) was the pivotal trial for axicabtagene ciloleucel incorporating a CD28 costimulatory domain (Axi-cel; Yescarta®). The trial included two cohorts with r/r lymphoma: cohort 1 which included 77 patients with r/r DLBCL and cohort 2 which included 24 patients with primarily mediastinal B‑cell lymphoma (PMBCL) or transformed follicular NHL (tFL). After lymphodepleting chemotherapy with fludarabine 30 mg/m
2/day and cyclophosphamide 500 mg/m
2/day (FC 30/500) for 3 days, 91% patients received a target dose of 2.0 × 10
6 CAR T‑cells/kg body weight. Patients had median of 3 pretreatments (range 1–7), including autologous stem cell transplantation (ASCT) in 21%. Bridging therapy between enrollment in the study and lymphodepleting chemotherapy was not allowed. Observed specific toxicity was a cytokine release syndrome (CRS) in 93% (grade ≥3: 13%) and neurological toxicity (NT) in 64% (grade ≥3: 28%). More recently, an up-date with median follow-up of 15.4 months was published, confirming durable ORR of 83% and a CR of 58%. The COO had no impact on clinical outcome. Median duration of response was 11.1 months (4.2–not estimable), OS was not reached, progression-free survival (PFS) was 5.9 months (95% confidence interval [CI] 3.3–15.0). [
4]. Furthermore, analysis of “real-world data” from different centers utilizing the approved, commercial available product, confirmed high activity in r/r DLBCL treated outside of clinical trial, while 50% of the patients did not meet inclusion criteria for different comorbidities [
5,
6]. Despite higher comorbidity, slightly higher median age and higher proportion of patients receiving Axci-cel after relapse ASCT (27% and 33 % for “real-world data” compared to 23% in the ZUMA1 trial), the ORR was similar (71% and 81% for “real-world data” and 83% in the ZUMA1 trial) [
6,
7].
Tisagenlecleucel (Tisa-cel) (Kymriah®), incorporating a 4-1BB costimulatory domain, was approved based on the JULIET trial (NCT02445248) for r/r DLBCL, including patients with a median of 3 prior therapies (range 1–8) including ASCT in 49%. After lymphodepleting chemotherapy with FC (25/250) or bendamustine 90 mg/m
2/day for 2 days, 93% of patients received a median single dose of 3.0 × 10
8 (range, 0.1–6.0 × 10
8) CAR-T cells. In 102/111 (92%) patients bridging therapy was given between leukocyte collection and CAR T‑cell infusion. There was no treatment-related death observed. Best ORR was 52% (95% CI, 41–62) including 40% CR. Those patients that had obtained CR at 3 months were also more likely to remain in remission at 6 months after CAR-T infusion. Median duration of response was not reached at the time of analysis. For patients achieving a CR, the 12-month relapse-free survival rate was 79% with an OS of 95%. The OS probability at 12-months for all infused patients was 49% [
8]. More recently, “real-word data” for tisagenlecleucel clearly support clinical activity with an ORR of 66% (CR 42%) [
9].
A third CD19-directed CAR T‑cell product, not yet approved, is currently investigated in the JCAR017-TRANSCEND trial (NCT02631044) using lisocabtagene maraleucel (Liso-cel) (incorporating a 4-1BB costimulatory domain). During manufacturing, T-cells are selected in CD4+ and CD8+ cells and then further processed to CAR-T separately. The product is prepared in two tubes consisting of CD4+ and CD8+ CAR-T cells in a precise 1:1 ratio. Patients were treated in two cohorts, including 73 patients DLBCL NOS and high grade B cell lymphoma and 102 patients with DLBCL NOS, PMBCL or tFL. Median of prior therapies was 3(range, 2–8, ASCT in 38%). A CRS occurred in 37% (grade ≥3: 1%) and NT in 23% (grade ≥3: 13%), the ORR was 80% including 55% CR [
10]. Based on encouraging data, a submission to FDA for approval is expected next year. More recently, a subset of patients with secondary central nervous system (CNS) manifestation showed response in 4/9 patients cases [
10,
11].
Assessment and management of adverse events in CAR T-cell therapy
CAR T‑cell therapy is associated with significant acute toxicities, which can be severe or even fatal [
12]. The following symptoms can be observed: cytokine release syndrome (CRS), neurotoxicity (CAR-T cell related encephalopathy syndrome [CRES] or immune effector cell associated neurotoxicity syndrome [iCANS]), cytopenia, prolonged B cell aplasia, hypo-gammaglobulinemia resulting in increased risk for infections and rarely hemophagocytic lymphohistiocytosis (HLH).
Cytokine release syndrome
Virtually all patients experience at least a mild cytokine release syndrome (CRS) presenting with fever >38.5 °C. The therapy of mild, i.e., grade I CRS, is symptomatic. Grade ≥2 CRS, which can include systolic blood pressure ≤90 mm Hg and/or hypoxia (FiO2 ≤40%) needs prompt interventions with IV fluids or non-invasive oxygen supply through breathing mask. If conditions deteriorate, CRS 3 is diagnosed and multiple pressor/resuscitator or respirator might become necessary; a CRS 4 means life-threating conditions. For monitoring of CRS proinflammatory parameters, e.g., CRP, ferritin or IL‑6 can be used. As the time point of CRS may vary between day 1 to day 14 (median 5), patients will be observed in hospital for 14 days in most centers. As a side effect of lymphodepletion and CAR T infusion, B‑cell aplasia and leukopenia can occur and therefore infection has to be ruled out in febrile patients, despite low incidence of infections during CAR T‑cell therapy [
13].
In moderate to severe symptoms, i.e., CRS ≥2 not responding to supportive therapy, tocilizumab, an interleukin‑6 receptor antibody, is recommended. Tocilizumab inhibits direct binding of IL‑6 or IL-6/soluble IL‑6 receptor complex to cell membranes. Tocilizumab can be given at a dose of 8 mg/kg IV (max. 800 mg), every 8 h with a maximum of three doses within 24 h and a total of four doses. In case of CRS grade ≥3 or patients not responding to tocilizumab within 24 h, corticosteroids (methylprednisolone 1 g/kg or dexamethasone 10 mg twice daily) should be considered. So far, corticosteroids were not recommended as first-line for CRS due a possible interaction with T‑cell expansion. However, a recently proposed regimen showed no negative impact on T‑cell expansion [
14]. Comparison of clinical data are difficult between trials, as different scoring systems and algorithms for CRS treatment were used. More recently, a simplified scoring system was introduced [
15].
Immune effector cell associated neurotoxicity syndrome (iCANS)
Immune effector cell associated neurotoxicity syndrome (iCANS) is another major complication of CAR T‑cell therapy which can occur with or independently of CRS, mostly within 28 days after CAR T‑cell infusion. As CAR T‑cells can cross the blood–brain barrier (BBB), endothelial cell activation might play a role in the development of iCANS and cerebral edema [
16]. Clinical symptoms, e.g., diminished attention, confusion, word-finding difficulties, disorientation, aphasia, somnolence, seizures or cerebral edema may occur. Tocilizumab is not expected to cross the BBB and could theoretically increase the amount of circulating IL‑6 in the brain. Dexamethasone 10 mg IV every 6 h or methylprednisolone 1 mg/kg IV every 12 h is considered as rescue therapy of iCANS in patients without signs of CRS. In addition to seizure prophylaxis or treatment, brain imaging (CT-scan or MRI) should be considered to rule out cerebral edema. A vigilant observation and close monitoring using a neurological assessment score is strongly recommended.
B-cell aplasia
As off target toxicity B‑cell aplasia can occur; therefore some patients are in need of immunoglobulin infusion in the case of recurrent infections or even prophylactically, depending to local guidelines.
Hemophagocytic lymphohistiocytosis/macrophage-activation syndrome
Hemophagocytic lymphohistiocytosis/macrophage-activation syndrome (HLH/MAS) rarely occurs and is characterized by similar clinical manifestations as CRS as high fever, multiorgan dysfunction or CNS disturbances.
Conclusion and perspective
CAR T‑cell therapy is “en vogue” due to very promising response data in relapsed and refractory DLBCL and 50–70% of patients will be alive after 12 month trials. This has been confirmed in “real-world” data including patients not eligible for trials due to several comorbidities [
5,
6,
9]. Despite the high efficacy of CAR T-cell therapy with CR rate up to 50%, which is in contrast to poor outcome after conventional salvage immune-chemotherapy, results have to be interpreted with caution [
17]. So far, only data of single-arm phase II trials with highly selected patients and short observation times are available. Even “real-world” data do not really improve the information quality, as data were provided by only a few highly experienced centers. No randomized data comparing ASCT or allo-HSCT are available so far. Therefore, no recommendation can be given in patients relapsing or refractory to first line therapy if transplant eligible. Clinical trials comparing ASCT with CAR T-cell therapy are currently being initiated.
As CAR T‑cell therapy is associated with specific toxicity as described above, has to be established a dedicated and well-trained team in specific centers. Finally, financial burden of CAR T‑cell therapy is significant. Therefore, new funding systems are necessary. Furthermore, a significant proportion of patients are refractory to CAR‑T cells. Pathomechims are only partially explained by a loss of the target structure (CD19) at the tumor cell or receptor mutations [
18] or expression of checkpoint proteins by the tumor [
19]. Administration of checkpoint inhibitors along with CAR-T cells are currently being tested: ZUMA‑6 trial, axicabtagen-ciloleucel +anti-PD-L1 antibody atezolizumab or PORTIA trial using tisagenlecleucel +pembrolizumab. Combining different epitopes to improve activity of CAR T-cell therapy was recently applied in r/r leukemia providing CR of 73% in ALL using an anti-CD19/CD22 CAR‑T [
20].
Novel CAR-T constructs are directed against CD79b alone or in combination with CD19 in cell line- and patient-derived xenograft models and should be clinically tested in B‑cell lymphoma [
21]. A CD30 CAR T-cell therapy demonstrated activity in patients with Hodgkin lymphoma and anaplastic large cell lymphoma [
22]. Ongoing concepts use CAR T-cell therapy in earlier lines, as randomized trials (ZUMA 7‑NCT03391466; BELINDA-NCT03570892) compare CAR T-cell therapy versus ASCT in first relapse.
CAR‑T cells have demonstrated significant clinical benefit in all studies published so far. Although Adverse Events (AEs) such as CRS, neurological toxicity, and B‑cell aplasia are common, the majority of events are manageable when treated by an appropriately trained multidisciplinary team.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.