Promising Immunotherapeutic Modalities for B-Cell Lymphoproliferative Disorders

Over the last few years, treatment principles have been changed towards more targeted therapy for many B-cell lymphoma subtypes and in chronic lymphocytic leukemia (CLL). Immunotherapeutic modalities, namely monoclonal antibodies (mAbs), bispecific antibodies (bsAbs), antibody-drug conjugates (ADCs), and chimeric antigen receptor T (CAR-T) cell therapy, commonly use B-cell-associated antigens (CD19, CD20, CD22, and CD79b) as one of their targets. T-cell engagers (TCEs), a subclass of bsAbs, work on a similar mechanism as CAR-T cell therapy without the need of previous T-cell manipulation. Currently, several anti-CD20xCD3 TCEs have demonstrated promising efficacy across different lymphoma subtypes with slightly better outcomes in the indolent subset. Anti-CD19xCD3 TCEs are being developed as well but only blinatumomab has been evaluated in clinical trials yet. The results are not so impressive as those with anti-CD19 CAR-T cell therapy. Antibody-drug conjugates targeting different B-cell antigens (CD30, CD79b, CD19) seem to be effective in combination with mAbs, standard chemoimmunotherapy, or immune checkpoint inhibitors. Further investigation will show whether immunotherapy alone or in combinatory regimens has potential to replace chemotherapeutic agents from the first line treatment.


Introduction
In the last decade, immunotherapy experienced a very rapid development and established itself as one of the fundamental parts of cancer treatment. Its main aim is to activate and encourage the body's own immune system to distinguish and destroy malignant cells [1]. In 1997, rituximab was the very first CD20-specific mAb to obtain regulatory approval, and rapidly became a key part of chemotherapeutic regimens (so-called chemoimmunotherapy) in the care for both B-cell non-Hodgkin lymphoma (B-NHL) and chronic lymphocytic leukaemia (CLL) [2]. Since then, antibody drug-conjugates (ADCs), immune checkpoint inhibitors, and both bispecific and trispecific antibodies for CAR-T cell therapy have been developed. Each class of these immunotherapeutic modalities has a unique mechanism of action, structure, and pharmacokinetic properties which are reflected in their specific treatment-related toxicities.
Mature B-cell neoplasms are a diverse group of lymphoid disorders with varying clinical manifestations, pathologic features, and outcomes. Diffuse large B-cell lymphoma (DLBCL) along with follicular lymphoma (FL) represent the most common aggressive and indolent histological subgroups of B-NHL with the incidence of~5.6/100.000/year and~2.7/100.000/year respectively (https://seer.cancer.gov/statfacts/html/dlbcl.html accessed on 19 October 2021). Together they comprise 50% of all B-NHL cases [3]. Chronic lymphocytic leukemia is another indolent B-cell malignancy with an incidence of 4.9/100.000/year (https://seer.cancer.gov/statfacts/html/dlbcl.html accessed on 19 October 2021), and accounts for 25-30% of newly diagnosed leukemias in the Western world [4]. Expression of tumor cell antigens varies according to diagnosis and may change during the treatment. Nevertheless, some markers such as B-cell-associated antigens (CD19, CD20, CD22, and CD79b) are commonly expressed on mature B-lymphocytes including neoplastic ones, and became targets for different immunotherapeutic modalities [5]. Despite that, molecules more selectively expressed on malignant cells and ideal targets on effector cells (e.g., T-cells, NK cells, APCs-antigen presenting cells, macrophages) are still being explored. Some of such antigens (CD30, CD47, CD37, CD70) will be described later in more detail.
Our review is focused on the current role of ADCs and bsAbs in B-NHL and CLL treatment. We discuss mechanism of action specific for each class, point to differences in their serious treatment-related toxicities and give an overview of the most important preclinical and clinical trials. Figure 1 shows examples of some immunotherapeutic agents used in B-NHLs. lymphocytic leukemia is another indolent B-cell malignancy with an incidence of ~4.9/100.000/year (https://seer.cancer.gov/statfacts/html/dlbcl.html accessed on 19 October 2021). ), and accounts for 25-30% of newly diagnosed leukemias in the Western world [4]. Expression of tumor cell antigens varies according to diagnosis and may change during the treatment. Nevertheless, some markers such as B-cell-associated antigens (CD19, CD20, CD22, and CD79b) are commonly expressed on mature B-lymphocytes including neoplastic ones, and became targets for different immunotherapeutic modalities [5]. Despite that, molecules more selectively expressed on malignant cells and ideal targets on effector cells (e.g., T-cells, NK cells, APCs-antigen presenting cells, macrophages) are still being explored. Some of such antigens (CD30, CD47, CD37, CD70) will be described later in more detail. Our review is focused on the current role of ADCs and bsAbs in B-NHL and CLL treatment. We discuss mechanism of action specific for each class, point to differences in their serious treatment-related toxicities and give an overview of the most important preclinical and clinical trials. Figure 1 shows examples of some immunotherapeutic agents used in B-NHLs.

Antibody-Drug Conjugates
Antibody-drug conjugates could be considered a bridge between conventional chemotherapy which eliminates all rapidly dividing cells and more targeted immunotherapy combining the best anti-tumor properties of each of them [6].
Antibody-drug conjugates consist of three main parts: the mAb, the cytotoxic payload, and the linker, which connects the two together. Monoclonal antibody serves as a transporter of the payload and binds to the specific antigen on the surface of tumor cells. The antigen then triggers the internalization of the whole complex via endocytosis, the linker is degraded by lysosomal enzymes and the payload is released. Upon release, the cytotoxic agent eliminates the tumor cell by various mechanisms, most often by DNA damage or interaction with microtubules [7,8]. Several ADCs (e.g., brentuximab vedotin) exert their anti-tumor toxicity also via bystander effect that is characterized as capability of the payload to permeate through cell membranes and kill neighboring cells, regardless of the presence of the target antigen [6].
Monoclonal antibody should be directed against antigen highly expressed on tumor cells and with limited or no expression on normal tissues to avoid excessive off target toxicity [7]. Payload is required to be highly toxic in subnanomolar concentrations, to be able to conjugate with the antibody, and to remain stable in physiological conditions [6,8]. Nevertheless, the essential components of ADC from the perspective of biochemical stability are linkers and there are two types of them (cleavable and non-cleavable) [9]. The first group has the cleavage site located between payload and monoclonal antibody and the "separation" process may occur within the endosome, lysosome or cytosol itself, while the second group needs complete proteolytic degradation of mAb backbone within the lysosome to release the payload [7][8][9]. Another important attribute of ADC in terms of efficacy and safety is drug-antibody ratio (DAR) which is defined by the number of cytotoxic molecules attached to mAb and relates to ADC's homogeneity and conjugation strategies [7,8,10].
As of 2021, eleven ADCs have been approved by the US Food and Drug Administration (FDA) in various hematological and oncological indications (Table 1) [8,10,11]. And many other ADCs are undergoing various stages of clinical testing. Results of phase I/II clinical trials, summary of currently ongoing phase III clinical trials and chemotherapy free clinical trials of ADCs are to be found in Tables 2 and 3.   Polatuzumab vedotin (PV) consists of a humanized IgG1 mAb against CD79b (B-cell receptor component) and a microtubule-disrupting agent MMAE (monomethyl auristatin E) conjugated by a protease-cleavable linker [28][29][30]. GO29365 (NCT02257567) was an open-label, international, multi-center, randomized, phase Ib/II clinical trial evaluating safety and efficacy of PV plus bendamustine and rituximab (BR) or bendamustine and obinutuzumab (BG) in relapsed/refractory (RR) DLBCL and FL setting compared to BR alone [12,30]. The results proved superiority of PV plus BR opposed to BR. Overall response rate (ORR) was 45% (18/40) versus 18% (7/40), complete response (CR) was 40% (16/40) versus 18% (7/40), duration of response (DOR) was 10.3 versus 4.1 months, median progression free survival (PFS) was 7.6 versus 2.0 months and overall survival (OS) was 12.4 versus 4.7 months in RR DLBCL, but there was no benefit in RR FL patients [12,29]. Based on the results of this study, PV in combination with BR was approved for the treatment of transplant ineligible RR DLBCL patients after at least one (Europe) or two (USA) prior lines of therapy in combination with BR in 2019 [31].
Polatuzumab vedotin in combination with CHP (cyclophosphamide, doxorubicin, prednisone) and rituximab (R) or obinutuzumab (G) were also investigated in treatment naïve DLBCL patients in a phase I/II clinical trial (NCT01992653) [13]. Sixty-six patients were included in the phase II part. Overall response rate was 89% (59/66) with CR of 77% (51/66). 12-month PFS and OS were 91% and 94%, respectively. There was no significant difference in efficacy between PV plus R-CHP or PV plus G-CHP cohorts. Based on these promising results, a phase III study comparing standard of care for newly diagnosed (ND) DLBCL, R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone), with PV-R-CHP (NCT03274492, Polarix) has been ongoing and the results are impatiently awaited.
Other PV combination investigated in RR DLBCL with available primary data is PV plus rituximab and lenalidomide [14]. DLBCL cohort in this phase Ib/II clinical trial (GO29834, NCT02600897) was characterized by the median age of 71 years (28-92) and median of 2 prior therapies. The investigator-assessed best overall response was 74% (36/49) with 35% (17/49) of CR. It is worth mentioning that CRs were durable with 82% (14/17) of patients remaining in remission at the cutoff date which is a promising sign in this difficult-to-treat population. The same clinical trial included also a RR FL arm investigating PV plus obinutuzumab and lenalidomide and the interim results were presented at British Society for Haematology 2020 Virtual meeting [15]. Efficacy population included 56 highly pretreated patients (median of three prior lines of therapy). At the median follow-up of 15.1 months, the data are still immature, but investigator-assessed ORR was 83% (38/46) with CR of 61% (28/46) by Modified Lugano 2014 criteria. Twelve-month PFS was 83.4%. This triplet appears to be a potent option for RR FL patients. However, longer follow-up is warranted for either arm.
Preclinical data suggest synergistic anti-tumor effect of PV and venetoclax (inhibitor of anti-apoptotic Bcl-2 protein) when PV promotes MCL-1 (myeloid leukemia cell 1) degradation, a known mechanism of resistance to venetoclax (Ven) [32]. That was the rationale of a phase Ib/II study (GO29833, NCT02611323) investigating the combination of PV with Ven and G for RR FL patients [33]. Study population was highly pretreated (74% ≥ 2 prior therapies) and 55% of patients were double refractory (to anti-CD20 mAb and alkylating agent). Primary analysis was presented at ASCO (American Society of Clinical Oncology) Meeting 2021. At the end of induction, PET (positron emission tomography)-CR was 57% (28/49). Median PFS was not reached and 12-month PFS was 73% with a median follow up of 14.4 months [33]. These results look encouraging, but data from maintenance treatment and beyond will be crucial to determine the benefit for this patient population.
Another promising PV combination for RR B-NHL patients is with mosunetuzumab (bispecific antibody targeting CD20 and CD3) and will be discussed further in text.
In 2021, Hamadani et al. published final results of a phase I study investigating this ADC as a monotherapy in RR B-NHL (NCT02669017) [17]. Patients had received median of three prior lines of systemic therapy. Overall response rate was 45.6% (82/180) in all evaluable patients with CR of 26.7% (48/180). Median DOR (duration of response) was 5.4 months for the whole cohort and was not reached for DLBCL patients who achieved CR. As this was primarily the dose-escalation and dose-expansion study, recommended dose for a phase II trial (LOTIS-2, NCT03589469) aiming at RR DLBCL was determined as 150µg/kg every three weeks for two doses followed by 75µg/kg every three weeks [16]. Interim results showed ORR 48.3% (70/145) with 24.8% CR (36/145). Median DOR for responders (CR + PR; PR = partial remission) was 12.6 months and not reached for patients with CR. Median PFS and median OS were 4.9 and 9.5 months, respectively. Lonca showed efficacy even in high risk groups such as double or triple-hit, transformed or refractory DLBCL. These encouraging results led to the FDA approval of lonca monotherapy for large B-cell lymphomas (DLBCL, transformed DLBCL, high grade B-cell lymphoma) after at least 2 prior lines of therapy in April 2021 [34].
Besides monotherapy, lonca is currently being tested in combination with ibrutinib in RR DLBCL and RR MCL in phase I/II trial (LOTIS-3, NCT03684694). Preliminary results for 36 evaluable patients from a phase I part were recently presented at ICML (International Conference on Malignant Lymphoma) 2021 and the efficacy data looked promising with ORR of 63.9% (23/36) and CR of 36.1% (13/36) [18]. Another investigated combination includes lonca plus rituximab versus R-GemOx (rituximab, gemcitabine, oxaliplatin) for RR DLBCL patients in a phase III study (LOTIS-5; NCT04384484), but the results are not available yet [35]. Loncastuximab tesirine is also being considered for the treatment of RR FL in a phase II clinical trial (LOTIS-6; NCT04699461) opposed to idelalisib [36]. Brentuximab vedotin (BV) consists of a microtubule-disrupting agent MMAE, but this time covalently attached by a protease-cleavable linker to the humanized mAb directed against CD30 [37,38]. Transmembrane receptor CD30 can be found on classical Hodgkin lymphoma's (cHL) Reed-Sternberg cells, on some T-cell malignancies and on a subset of DLBCL as well [28,38]. As of 2021, BV has been approved for the treatment of RR cHL, ND, and RR CD30+ systemic anaplastic large cell lymphoma and RR CD30+ cutaneous T-cell lymphoma [37,[39][40][41][42].
Primary mediastinal large B-cell lymphoma (PMBCL) is a specific subtype of DLBCL which shares some characteristics with cHL, especially high CD30 expression [43]. That is why similar efficacy was expected from BV monotherapy in RR PMBCL patients. A phase II study investigating BV in this setting (NCT02423291) did not confirm this hypothesis with ORR of only 13.3% (2/15) and with no CR achieved. Therefore, BV alone is considered inactive in RR PMCBL. Nevertheless, Zinzani et al. presented the data of a phase I/II combination trial of BV with nivolumab (immune checkpoint inhibitor) in RR PMBL at ICML 2021 and the results were just the opposite (NCT02581631, CheckMate436) [44]. Thirty highly pretreated patients were recruited with a median of two prior lines of therapy. Results from the extended median follow up of 33.7 months showed ORR of 73% (22/30) with CR of 37% (11/30). Median DOR was 31.6 months and median duration of CR and median OS have not been reached. As this combination proved its efficacy and long-term survival benefits, a phase II study of BV + nivolumab with R-CHP (rituximab, cyclophosphamide, doxorubicin, prednisone) for newly diagnosed ND PMBCL is currently underway (PACIFIC, NCT04745949).
Not only PMBCL, but CD30+ DLBCL and gray zone lymphoma (GZL) could be appropriate targets for BV combination therapy. In a phase 1/2 multicenter trial (NCT01994850), 29 patients with ND CD30+ B-cell lymphomas (22 PMBCL, 5 DLBCL, 2 GZL) were treated with 6 cycles of BV + R-CHP and 52% of patients followed consolidative radiotherapy [19]. Overall response rate was 100% with 86% CR and there was no difference between radiotherapy + or-cohort. At the median follow-up of 30 months, 2-year PFS and OS were 85% and 100%, respectively. These results are quite promising but should be regarded with caution due to the small study cohort.
At ICML 2021, preliminary results from a phase I combination trial of BV plus lenalidomide and rituximab for RR DLBCL were foreshadowed [45]. Among 37 evaluated patients, ORR was 56.7% (21/37) and median DOR was 13.2 months for the responding patients (CR+PR). The PFS and median OS reached 11.2 months and 14.3 months, respectively. It is worth mentioning that the results were similar between CD30+ and CD30 < 1% cohorts. Based on these promising outcomes, a phase III study comparing lenalidomide and rituximab plus BV/placebo in RR DLBCL setting was initiated (ECHELON-3, NCT04404283).

Naratuximab Emtansine, Debio1562, IMGN529 (Debiopharm)
Naratuximab emtansine (nara) represents a humanized anti-CD37 IgG1 antibody conjugated via a non-reducible thioether linker to a potent anti-mitotic agent-maytansinoid DM1 [7,46] Transmembrane protein CD37 has the highest abundance on mature B-cells and it is expressed only in a low level on T-cells and myeloid cells [7,47]. Under physiological condition, CD37 plays a role in normal B-cell activation and survival. On the other hand, it is highly expressed on most histological subtypes of B-NHL.
Naratuximab emtansine was evaluated in a phase I study (NCT01534715) in heavily pretreated RR NHL patients (median of three prior systemic regimens) and the results were encouraging with ORR 22% (4/18) in DLBCL cohort [46]. Based on these results, nara was selected for a phase II clinical trial in combination with rituximab for RR B-cell NHL patients (NCT02564744) [48]. Seventy-four RR DLBCL patients were evaluable for efficacy and ORR of 43.2% (32/74) with 32.4% (24/74) of CRs were observed in this group. Median DOR was not reached during median follow-up in responders of 13.7 months.

Bispecific Antibodies
Bispecific antibodies (bsAbs) are immunotherapeutic agents designed to recognize and bind two different antigens or two different epitopes on the same antigen. Many formats have been produced so far, but in general, bsAbs can be divided into two major classes, those bearing an Fc region and those lacking an Fc region. The former resembles IgG or IgG-like molecules with Fc-mediated effector functions (ADCC-antibody-dependent cell cytotoxicity, CDC-complement-dependent cytotoxicity, ADCP-antibody-dependent cell phagocytosis) and improved stability [57]. The latter is represented by small molecules composed of either a complete fragment antigen-binding (Fab) region or only some parts of Fab immunoglobulin fragments connected with a peptide linker [57]. Small molecules gain an advantage of better tissue penetration, but on the other hand, they have no carrier system protecting them from rapid intracellular degradation and renal elimination. Due to their properties, many of these bsAbs require frequent or even continual administration [57][58][59].
The mechanism of action depends on biological targets of particular bsAb. Currently, T-cell engagers (TCEs) that typically target tumor associated antigen (TAA) with one arm, and T-cell binding domain with another one, represent the most frequently used class of bsAbs. Protein CD3 is a part of T-cell receptor (TCR) signaling complex [60], and acts as the binding domain of many effector T-cells. In contrast to physiologically activated T-lymphocytes, immunological synapse mediated via TCEs induces polyclonal T-cell activation independent of TCR epitope specificity, independent of major histocompatibility complex (MHC), and co-stimulatory signals [60]. Once activated, cytotoxic lymphocytes release granules containing perforins and granzymes and mediate lysis of the target cells [58,61,62]. Afterwards, multiple cytokines such as interleukin (IL)-2, IL-6, IL-10, interferon-gamma (IFN-γ), or tumor necrosis factor-alpha (TNF-α) are secreted and, besides T-lymphocytes, other immune cells are activated (e.g., B-cells, macrophages, NK cells, etc.) [58,61].
Early clinical data for a phase I/II study (NCT03671018) of mosenutuzumab plus PV administrated i.v. (8-17 cycles of mosenutuzumab plus 6 cycles of PV) were presented at European Hematology Asociation (EHA) congress 2021. In a dose escalation cohort, 68% (15/22) of all RR B-NHL (DLBCL and FL with a median of three previous therapies) responded to treatment and 54.5% (12/22) achieved CR. [23]. The phase expansion cohort of mosenutuzumab plus PV for RR DLBCL (NCT03671018), and phase III clinical trial of mosenutuzumab plus lenalidomide versus R plus lenalidomide for RR FL (NCT04712097) are currently ongoing [66].

Epcoritamab, GEN3013, (Genmab, Abbvie)
Epcoritamab is a product of DuoBody technology that generates a full-length anti-CD20xCD3 IgG1 TCE. The bispecific construct is generated by the controlled Fab-arm exchange of two parental IgG1 mAbs (each with a matched single point mutation in the Fc region), and as a result, it can target two different epitopes as a single mAb [68].

IGM-2323 (IGM Biosciences)
IGM-2323 is a bispecific IgM antibody that has ten high-affinity binding domains for CD20 and one for CD3 [72]. Compared with other TCEs, it has a slightly different mechanism of action based on repeatable T-cell activation and high CDC. In preclinical in vitro studies, IGM-2323 irreversibly binds to CD20 positive cells and depletes them with limited cytokine secretion. Nevertheless, so far only limited preliminary data of efficacy and tolerability have been shown in the first-in-human clinical trial with eight RR B-NHL patients [72]. Blinatumomab was the first in class bispecific T-cell engager (BiTE) construct consisting of two flexibly linked single-chain variable fragments that bind to CD3 and CD19. Molecular weight of blinatumomab is approximately 55 kDa which is reflected in its high systemic clearance and short half-life (2.10 h) [59]. To minimize the risk of treatment-related AEs and maintain plasma concentration, blinatumomab is administrated as a dose-escalating continuous infusion (i.v.) usually lasting four to eight weeks [59,73].

Bispecific Antibodies
The clinical proof of blinatumomab efficacy was initially demonstrated in patients with RR B-NHL (median of four previous therapies) in the phase I study (NCT00274742), which defined 60 mg/m 2 /day as the maximum tolerated dose (ORR 69% (24/35), CR 37% (13/35)). In a DLBCL subgroup, 55% (6/11) of patients responded to treatment with 36% (4/11) achieving CR [74]. Responses were significantly better among patients who received the targeted dose with median PFS 1.5 years and median OS 5.8 years [75]. Afterward, several phase II clinical trials were designed for RR B-NHL subsets. One of those assesed blinatumomab as a second salvage in aggressive RR B-NHL patients who did not achieve CR with platinum-based regimen. The ORR, CR, and PR were 37% (15/41), 22% (9/41), and 15% (6/41), respectively. Median OS was 11.2 months among all patients. Among those who achieved CR or proceed to autologous stem cell transplantation (ASCT), median OS was not reached and median PFS was 8.4 months [76].
Combination of blinatumomab with lenalidomide was assessed in a phase I study with RR B-NHL (median od 2.5 previous therapies). At a median follow-up of 14.3 months, the ORR and CR were 83% (15/18) and 50% (9/18), and median PFS was 8.3 months. Three patients who achieved response underwent allogeneic stem cell transplantation and remained in remission for 14.2 to 22.3 months thereafter [77].
Blinatumomab was also tested as consolidation therapy post autologous stem cell transplantation (ASCT) in RR DLBCL or transformed FL [78], post standard chemoimmunotherapy in newly diagnosed high-risk DLBCL [79] or as maintenance therapy post allogeneic stem cell transplantation in RR B-NHL patients [80]. It probably has a potential to increase the response rate in some cases but bigger cohorts and longer follow-up are needed.

Other Bispecific Antibodies
TNB-486 (TeneoBio) is a fully human asymmetric IgG4 anti-CD19xCD3 TCE that was designed to bind CD3 antigen with low affinity. Given the modification, TNB-486 demonstrates efficient tumor cell cytotoxicity coupled with reduced cytokine release (IL-2, IFNγ, IL-6, IL-10 and TNF) without the decrease of secreted granzymes and perforins [81]. This might be an advantage if reduced CRS-related toxicity will be proved in clinical trials.
Different ImmuneOnco Biopharmaceuticals generates IMM0306, the first antibody receptor recombinant protein (mAb-Trap) targeting CD47 and CD20. Transmembrane protein CD47 is overexpressed by multiple malignant cells and gives macrophages and other phagocytes "do not eat me" signal. The clinical benefit is based on a hypothesis that binding CD47 may block inhibitory signal mediated via this antigen and proceed the process of phagocytosis [82]. IMM0306 will be soon evaluated in a phase I clinical trial in RR B-NHL patients (NCT04746131) and another, but in this case anti-CD47xCD19 bsAb, namely TG-1801 (TGTherapeutics), is currently evaluating as a single agent (NCT03804996) or in combination with umblituximab (anti-CD20 mAb) (NCT04806035) in RR B-NHL and RR CLL patients.
Finally, tebotelimab, MGD01 (MacroGenics), a dual-affinity re-targeting antibody (DART), was designed to bind PD-1 and LAG-3 to sustain or restore the function of exhausted T cells. There are only preliminary data from a small cohort of RR DLBCL (ORR 27% (3/11), CR 18% (2/11), PR 9% (1/11)) at the moment. Treatment-related AEs were reported in more than half of patients (64.7%) with one grade ≥3 pneumonia [83]. As this is the dual immune checkpoint inhibitor, autoimune treatment related toxicity should be closely monitored.
T-cell engagers have encountered hurdles in the clinical practice dominantly related to cytokine release syndrome (CRS) and immune effector cell-associated cytotoxicity syndrome (ICANS). Cytokine release syndrome results from T-cell over-stimulation likely arising from the high affinity engagement of CD3 [84,85]. Regardless of the subclass of TCE, approximately half of patients develops CRS of any grade during cycle one (mosunetuzumab: 23%, glofitamab: 50,3%, odronextamab: 57%, epcoritamab: 57%, blinatumomab: 44%) and only minimum during the next cycles [25,67,73,76]. Stepwise dose escalation, corticosteroid premedication as well as tocilizumab (anti-IL-6 mAb) administration in the early onset of CRS are effective and significantly reduce the number of grade ≥3 AEs [86]. It is still unclear whether ICANS is attributed to on-target toxicity or more likely to damage caused by excessive CRS with hyperinflammation and disruption of the blood-brain barrier [81,87]. Therefore, efforts to mitigate CRS may indirectly reduce the incidence of associated neurotoxicity. Undoubtedly, except blinatumomab which required treatment discontinuation in several studies [73,76], severe ICANS occur rarely, and frequency of grade 1 or 2 differs among TCEs (ICANS of any grade reported in mosunetuzumab: 45%, glofitamab: 5%, epcoritamab: 6%, blinatumomab 9-56%) [26,65,69,76]. Due to the risk of IL-6 increase in central nervous system after tocilizumab administration, corticosteroids are recommended for the treatment of ICANS while combination with tocilizumab is recommended for CRS and concurrent ICANS [86]. Other common severe AEs were hematologic toxicity, dominantly neutropenia, and lymphopenia along with infections [27,65,76]. Subcutaneous application reduces the number of severe CRS. However, more local injection-site reactions were reported (47%) [26].

Conclusions
From huge amount of ADCs, loncastuximab tesirine, anti-CD19 ADC, is the only one that has been approved for RR DLBCL as a single agent [34]. Other ADCs used for aggressive B-NHL subsets work more effectively in combination with standard chemoimmunotherapy (e.g., PV plus R-CHP, BV plus R-CHP, BV plus BR) [19,31,52]. Furthermore, long-lasting follow-up is being awaited for chemotherapy-free regimens such as BV plus nivolumab in RR PMBCL [88], triplet of PV plus rituximab and lenalidomide in RR DL-BCL [14], PV plus obinutuzumab and lenalidomide [15] or PV plus Ven [32] in RR FL.
Even though bsAbs, namely blinatumomab, has been tested in clinical trials for B-NHL since 2011, TCEs are still quite young molecules with no one commonly used in clinical practice for B-cell lymphomas and CLL treatment. Three anti-CD20xCD3 TCEs (epcoritamab, mosunetuzumab, glofitamab) have already entered phase 3 clinical trials (NCT04628494, NCT04712097, NCT04408638). The most promising seems to be epcoritamab, showing high response rates in both indolent and aggressive pretreated lymphoma patients [26]. Further investigation will reveal the potential of combinatory regimens such as glofitamab plus obinutuzumab [25], or mosenutuzumab plus PV [23].
In terms of anti-CD19 immunotherapeutic modalities, blinatumomab, loncastuximab tesirine, along with coltuximab ravtansine, are the only ADCs and TCEs clinically evaluating in B-NHL patients. This is quite the opposite to anti-CD19 CAR-T cells that have been widely used and gradually approved for aggressive and indolent B-NHL subtypes [89][90][91]. Furthermore, tafasitamab-cxix (MorphoSys US), which is a humanized Fc-modified cytolytic CD19 targeting mAb, has been approved by FDA in combination with lenalidomide for the treatment of RR DLBCL patient. It will be interesting to observe the development of anti-CD19 TCEs and ADCs in this wide field of immunotherapeutic agents, as they also have the advantage of so-called "off the shelf" products.
Despite the fact that anti-CD20xCD3 TCEs have been highly effective in indolent lymphomas, only limited data are available for TCEs used in CLL. At the moment, we can only speculate if this is because of primary impaired T-cell function [89,92,93] predicting failure of TCEs, too small CLL cohorts in clinical trials, or we just have to wait for studies with a dual or triple combination of TCE, Bcl-2 inhibitor, and/or B-cell receptor inhibitor. Funding: This work originated due to financial support from the project no. FV40145 Genetically modified donor T cells as a universal cell source for the immunotherapy of tumors using T lymphocytes equipped with a chimeric antigen receptor (CAR).