CD20 is a surface protein that exhibits ubiquitous expression in B cells with minimal occurrence in other tissues, rendering it an ideal target for immunotherapy against B cell-derived malignancies. CD20 expression initiates during the pre-B cell stage and persists until B cells undergo terminal differentiation into plasma cells ( Figure 1 ). Immunotherapy directed at CD20 is extensively employed for treating mature B cell-derived malignancies, such as chronic lymphocytic leukemia (CLL) and various B cell-derived non-Hodgkin lymphomas (B-NHL), including follicular lymphoma (FL), diffuse large B-cell lymphoma (DLBCL), and mantle cell lymphoma (MCL). CD20 is also present in multiple subtypes of B cell precursor acute lymphoblastic leukemia (B-ALL), albeit its expression at diagnosis is heterogeneous and frequently low (1–3). Notably, documented upregulation of CD20 after induction treatment suggests a potential expansion of CD20-directed immunotherapy applications for B-ALL (4, 5). CD20-specific therapies offer precise B cell targeting, minimizing impact on other cell types. These therapies efficiently deplete CD20-expressing B cells without hindering the replenishment of the B-cell compartment from early B cell precursors. Hence, upon cessation of anti-CD20 treatment, the B-cell population can recover (6). Notably, the absence of CD20 on fully mature plasma cells enables patients to maintain protective humoral immunity against previously encountered pathogens during treatment (6).

CD20-targeted immunotherapy encompasses diverse modalities administered at various treatment stages. Rituximab, the pioneering anti-CD20 monoclonal antibody (mAb) introduced in 1997, stands out as a well-studied, low-toxicity immunotherapy with manageable side effects. It is a crucial component of the common therapy regimens, such as BR (bendamustine + rituximab) or FCR (fludarabine + cyclophosphamide + rituximab), which are often used as a first-line treatment in specific groups of CLL and B-NHL patients. In addition, following positive phase 3 trial results, rituximab has been recently integrated into chemotherapy for adult B-ALL patients with at least 20% CD20-positive leukemic cells (7). Beyond rituximab, the engineered anti-CD20 mAb obinutuzumab is registered and employed in combination with chemotherapy as first-line therapy for defined cases of CLL and FL. In addition to mAbs, new immunotherapies targeting CD20 have been developed and successfully introduced into the clinic for patients refractory to first-line therapy or with relapsed disease (r/r). These include bispecific antibodies (BsAbs) targeting the CD20 molecule and simultaneously recruiting cytotoxic T cells, as well as adoptive therapies using autologous T cells modified with chimeric antigen receptors (CAR-T). Three BsAbs targeting CD20 have received FDA approval, while CD20-specific CAR-T cells are presently undergoing clinical trials. Notably, CAR-T cells simultaneously targeting CD19 and CD20 aim to address CD19-negative clones, with ongoing clinical trials in advanced r/r B-cell malignancies ( Table 1 ).

This comprehensive review explores various CD20-directed immunotherapies, including mAbs, radio-immunoconjugates, BsAbs, and CD20 CAR-T cells. The discussion encompasses both approved drugs and novel solutions undergoing investigation in preclinical and clinical trials ( Figures 2 , 3 ; Tables 1 – 3 ). Mechanisms of resistance to CD20-directed immunotherapies are presented ( Figure 4 ), and the potential for various combinations with immunotherapies is discussed.

CD20 is a transmembrane protein whose significance as a target for immunotherapy is well recognized, although its biological role remains elusive. Encoded by the MS4A1 gene, CD20 is part of the MS4A family, which consists of 18 proteins with similar structures. The CD20 protein spans the cell membrane with four transmembrane helices and features two extracellular loops, which are the main epitopes recognized by anti-CD20 mAbs. Notably, both the N-terminal and C-terminal ends of CD20 are situated inside the cell. The detailed structure and dimeric assembly of CD20 on the cell membrane have recently been extensively characterized (9). This structural research sheds light on the molecular architecture of CD20 and anti-CD20 mAbs biding modes, contributing to our understanding of its potential as a target for immunotherapy (9).

While no identified physiological ligand binds to CD20, it is known to form nanoclusters on the B cell membrane with proteins such as IgD or IgM-class B cell receptors (BCR), CD19, CXCR4, and CD40 (10). A recent study utilizing CRISPR/Cas9-mediated CD20 elimination from mature B cells revealed CD20’s role as a gatekeeper in maintaining the resting state. The knockout of CD20 resulted in the translocation of the BCR toward the CD19 coreceptor, transient B cell activation, and internalization of various B cell-specific proteins (10). Additionally, initial research proposed that CD20 may function as a calcium channel (11), however, subsequent findings suggested that calcium flux induction is mediated by the CD20-BCR complex rather than CD20 alone (12). Therefore, the role of CD20 as a regulator of B cell activity seems to be inherently linked to interactions of CD20 with other surface proteins, primarily with the BCR complex.

The regulation of the MS4A1 gene was attributed to several transcription factors, summarized in (13). The described positive regulators include transcription factors essential for B cell development and maturation, such as PU.1, PiP (IRF4), NFκB (14–16), as well as other factors, such as USF, TFE3.1 (16), OCT1, OCT2 (17), ELK1, ETS1 (18), SP1, CHD4 and MBD2 (19). Recently, another member of interferon regulatory factors (IRF) engaged in B cell development, IRF8, was shown to promote CD20 expression in DLBCL as well as in healthy B cells (20). Negative regulators of CD20 include FOXO1 (21), CREM (19), SMAD2/3 (22), and MYC (23, 24).

Additional regulation of CD20 expression occurs on the epigenetic level. Histone deacetylase (HDAC) family members HDAC1/2, HDAC1/4, and HDAC6 as well as methyltransferase enzyme EZH2 can repress CD20 in healthy and malignant B cells (25–27). Recently, the occurrence of four 5’-UTR variants of MS4A1 mRNA with differential translation efficacy was described (28).

The evolution of anti-CD20 mAbs marks a progression toward enhanced compatibility and reduced immunogenicity. The first therapeutic anti-CD20 mAb, rituximab, comprises a chimeric murine-human structure, contributing to the development of immune response and infusion-related reactions due to its limited resemblance to natural human antibodies (29, 30). Enhanced human content in subsequent mAbs correlates with decreased immunogenicity and improved binding affinity to human Fc receptors. The newer generations of anti-CD20 mAbs, exhibiting humanized (obinutuzumab) and fully human (ofatumumab) designs have reduced immunogenicity (31–33). Furthermore, heightened human sequence content enhances interactions with immune effector cells and FcRn receptors on hepatic and epithelial cells, thereby prolonging IgG antibodies’ half-life (34).

CD20-targeting mAbs elicit their cytotoxic function by at least four different mechanisms (35). Upon binding CD20 on target cells, they can activate complement-mediated cytotoxicity (CDC), engage immune effector cells to mediate antibody-dependent cytotoxicity (ADCC) and phagocytosis (ADCP), as well as directly induce cell death. The anti-CD20 mAbs currently employed in cancer treatment vary in their degree of activating specific mechanisms. These differences are the basis for a categorization of anti-CD20 mAbs into two types. The majority of mAbs are characterized as type I, which exhibit the ability to cluster CD20 into membrane lipid rafts, which is associated with potent induction of CDC. On the other hand, type I antibodies display a higher rate of internalization, which can limit their therapeutic efficacy (36). Type II mAbs do not stabilize CD20 in lipid rafts and are weak inducers of CDC, but they potently evoke direct cell death (37). Rituximab, ofatumumab, and ublituximab are classified as the type I anti-CD20 mAbs, whereas obinutuzumab represents a type II anti-CD20 mAb.

CD20-directed mAbs are also used in the form of immunoconjugates – Abs linked with drugs (ADCs), toxins (immunotoxins or engineered toxin bodies – ETBs) or radioactive isotopes (radio-immunoconjugates). Unlike conventional antibodies, which largely depend on the immune effector cells or the complement system for their cytotoxic effects, immunoconjugates can directly induce apoptosis of cancer cells. Due to the relatively poor internalization of CD20, few ADCs and immunotoxins were developed. A single-chain variable fragment-based targeting CD20 and conjugated with Shiga-like toxin A subunit, MT-3724, presented promising preclinical and early clinical results, but its development was ceased by the manufacturer (38). Another strategy that does not require CD20 internalization for direct and targeted cell killing is the use of radio-immunoconjugates. This approach has gained significant attention in targeting lymphoma cells, which are highly radiosensitive (39). Radio-immunoconjugates utilize ionizing radiation to induce cytotoxicity of the target cell. Concurrently, they can trigger classical effector mechanisms such as CDC, ADCC, and ADCP.

In this section, we describe anti-CD20 mAbs and conjugates that were approved for clinical use in lymphoid malignancies. Additionally, other anti-CD20 agents that displayed effectiveness in clinical trials of B-cell neoplasms are listed in Table 2 .

BsAbs represent one of the most promising classes of off-the-shelf immunotherapies for the treatment of r/r B cell malignancies (91, 92). Notably, several BsAbs targeting the CD20 antigen received clinical approvals in 2023, with numerous others showing promising results in ongoing clinical trials ( Figure 2 , Table 3 ) (93). These engineered proteins, featuring dual binding sites, can simultaneously target two different antigens or two epitopes of the same antigen. This dual specificity allows BsAbs to bridge immune cells, such as T cells, with target tumor cells, promoting their interaction and subsequent cytotoxicity against the tumor cells.

Over the past few years, there has been a rapid development of this technology, resulting in various molecular BsAb formats, including IgG-like and non-IgG-like platforms (91, 92). IgG-like BsAbs mimic the structure of IgG, featuring an Fc region for effector functions, like ADCC and CDC, and provide a larger molecular weight, which increases solubility, stability, serum half-life. This allows for a wider spectrum of dosing frequency from daily to weekly or even less frequent and its administration both intravenously and subcutaneously. In contrast, non-IgG-like BsAbs lack the Fc region, typically have a smaller molecular weight, and primarily exert therapeutic effects through direct antigen binding (92). This applies also to blinatumomab, the first BsAb approved for medical use. Blinatumomab is a CD3xCD19 bispecific T-cell engager that consists only of two scFvs connected by a linker, which contributes to a relatively short half-life and the necessity for frequent dosing in prolonged infusions (94).

Among the various formats, the anti-CD20xCD3 BsAb engaging cytotoxic T cells is the most popular format. Here we focus on CD20-targeting IgG-like BsAbs as a new therapeutic option for patients with B-cell malignancies who have already undergone several lines of mAbs and CD19 CAR-T therapy.

CD20 is also under exploration as a target for CAR-T cells in preclinical and clinical trials. CARs are synthetic constructs comprising extracellular antigen recognition domains, hinge and transmembrane regions, and intracellular signaling domains responsible for their activation and proliferation. Approved CAR T-cell therapy involves genetically engineered autologous products, utilizing the patient’s CAR T cells to target tumor cell antigens. Currently, four CD19-targeted CAR T-cell therapies are approved for treating r/r B-ALL and r/r B-NHL. Despite its efficacy, around 60% of patients experience disease relapse post-CD19 CAR-T treatment, often due to mechanisms like CD19 antigen loss. Also, in some patients, life-threatening toxicities occur, including severe CRS and ICANS (144, 145). Ongoing clinical trials suggest that CD20 CAR T-cell therapy could be a promising treatment for r/r NHL, even in cases of CD19-negative disease post-CD19 CAR-T cell relapse ( Table 1 ). The structure of the clinically tested CD20 CAR T-cells is presented in Figure 3 .

Phase 1/2 clinical trials utilizing second- and third-generation CAR constructs have confirmed the feasibility and efficacy of autologous anti-CD20 CAR-T cells in r/r CD20⁺ B-NHL (146). Particularly noteworthy is the efficacy of CD20 CAR T-cell therapy in treating r/r B-NHL patients who had previously failed chemotherapy, including R-CHOP. Studies indicate that CD20-targeted CAR T cells exhibit effectiveness even in cases of low antigen expression, proposing their potential utility for patients with CD20-downregulated B-NHL refractory to CD20 mAb therapy (28, 147). A comprehensive overview of ongoing and completed clinical trials for single CD20 CAR T-cell therapy in hematologic malignancies can be found in Table 2 of a recent review (148).

CD20 is also a pivotal target in CAR-T cell immunotherapies designed to mitigate antigen escape risks. Strategies targeting both CD19 and CD20 include bispecific/tandem CARs, co-administration of CD19 and CD20-directed CAR-T cells as well as sequential treatment with CD19 and CD20-directed CAR-T cells. Tan CAR7 T cells are bispecific CAR T cells composed of tandem extracellular domains targeting CD20 and CD19 tumor antigens linked in frame to the tisa-cel backbone, capable of activation via binding to either CD19 or CD20 tumor antigens, or both (149). Long-term remissions were observed following the use of Tan CAR7 T cells in r/r B-NHL with a safety profile that included CRS but few cases of high-grade CRS (150, 151). In a recent phase 1 dose-escalation trial, autologous CD19/CD20 bispecific CAR-T cells derived from naïve and memory T cells demonstrated safety and strong efficacy (90% ORR, 70% CR rate) in patients with r/r B-NHL (152). Beyond bispecific CD19/CD20 CAR T-cells, ongoing clinical trials explore sequential CD20 CAR-T after CD19 CAR-T infusion and combined infusion of CD19 and CD20-specific CAR-T cells for r/r B-ALL or DLBCL. However, a phase 2 trial combining anti-CD19 and anti-CD20 CAR-T cells in r/r DLBCL showed limited long-term responses (153). Recently, a combinatorial CAR-T cell approach targeting three antigens, CD19, CD20, and CD22, demonstrated efficacy in preclinical models, including leukemic cells that do not express CD19, thereby showcasing the promising potential for treating CD19-negative relapses (154). This approach is now undergoing testing in a clinical trial (NCT05418088).

Despite substantial progress in CD20-targeting immunotherapies, the issue of resistance and post-treatment relapse remains prominent. Resistance to CD20-targeted therapies encompasses a spectrum of mechanisms, ranging from alterations in CD20 antigen levels to compromised immune system effector functions, and extending to diverse mechanisms of immune evasion ( Figure 4 ). One of the main causes of resistance is the loss of the CD20 antigen on the surface of the target cell, which can be caused by changes in the expression of the MS4A1 gene, including silenced expression and alternative splicing (28, 155–158). A recent study has shown that the gene encoding CD20 in both healthy and malignant B cells is alternatively spliced into four 5’-UTRs variants, of which especially variants V3 and V4 support robust translation. It has also been demonstrated that resistance to the BsAbs therapy targeting CD20 results from the V3-to-V1 shift (28). A potential strategy to combat this resistance through the use of phosphorodiamidate morpholino oligomers or antisense oligonucleotides was presented in preclinical studies (28). Other mechanisms of antigen loss which may also cause resistance include the internalization of the CD20-mAb complex by cancer cells through endocytosis as well as the transfer of membrane fragments containing CD20 from a cancer cell to an effector cell called trogocytosis (159–161).

Resistance to CD20-directed immunotherapies may also be caused by impaired effector functions of the immune system, such as CDC and ADCC. Therapies targeting CD20 may cause complement depletion and overexpression of its inhibitors CD55 and CD59 (44, 162). Additionally, downregulation of the complement component C1qA was associated with the resistance of DLBCL cells to rituximab in vitro (163). Potential strategies to overcome these mechanisms may include the use of inhibitors of complement regulatory proteins as well as the use of the new asymmetric CD55-binding bispecific antibodies (164). Moreover, rituximab-coated tumor cells were shown to significantly downregulate CD16 (FcγRIII), leading to impaired ADCC (165). Mutations that modify the binding of the Fc fragment of antibodies to FcγR can be used to increase the effector functions of antibodies (166, 167). However, as previously mentioned, this approach may not always be optimal when utilizing BsAbs, as it carries an increased risk of premature activation of T cells, cytokine release, and tissue damage.

Additionally, genetic alterations within signaling pathways can also contribute to resistance to CD20-directed therapies, especially in the context of T cell activation, which is crucial for the activity of BsAbs and CAR-T cells (168, 169). Moreover, the tumor microenvironment can play a role in resistance by creating an immunosuppressive milieu. Tumor cells and immunosuppressive cells in the tumor microenvironment, e.g. myeloid-derived suppressive cells (MDSCs), tumor-associated macrophages (TAMs), and regulatory T cells (Tregs), can secrete suppressive cytokines that inhibit the activity of effector cells (T cells, NK cells, phagocytes), thereby reducing the effectiveness of immunotherapy (170, 171). It has also been shown that some proteins secreted by the tumor may have a suppressive effect, including galectin-1, which inhibited CD20 mAb-induced phagocytosis in the lymphoma microenvironment (172). Moreover, overexpression of PD-L1 by tumor cells can contribute to resistance to CD20-targeting therapies by dampening the activity of effector T cells induced by these therapies. Tumor cells may increase PD-L1 expression in response to treatment, leading to T cell exhaustion and reduced efficacy of CD20-targeting therapies (173). Combining CD20-targeting therapies with immune checkpoint inhibitors is a potential strategy to overcome resistance and is currently tested in clinical trials, as discussed in more detail in section Combination therapies with CD20 immunotherapeutics.

Due to the multitude of resistance mechanisms, it is crucial to actively search for new methods that can increase the effectiveness of immunotherapy. Research efforts encompass the identification of novel therapeutic targets beyond CD20, the refinement of patient stratification, and the incorporation of combination therapeutic strategies. A recent review summarizes potential solutions to overcome resistance to CAR-T therapy (174).

To enhance their efficacy, anti-CD20 mAbs are commonly administered in combination with other drugs. One notable combination is the R-CHOP regimen, which integrates rituximab with cyclophosphamide, doxorubicin, vincristine, and prednisone, and has been extensively employed in treating patients with DLBCL and MCL. Similarly, R-pola-CHP (rituximab, polatuzumab, cyclophosphamide, doxorubicin) regimen is an approved treatment for advanced-stage DLBCL. Other regimens include a combination of rituximab, dexamethasone, high-dose cytarabine and a platinum-based agent (R-DHAP) used in the treatment of MCL, and the addition of lenalidomide to rituximab (R-lenalidomide) which has shown promising results, particularly in patients with r/r FL. Combinations such as bendamustine and rituximab (BR) and fludarabine, cyclophosphamide, and rituximab (FCR) have demonstrated efficacy in the treatment of CLL. Furthermore, novel combinations of rituximab with targeted agents have shown significant potential. Rituximab in combination with venetoclax, a BCL-2 inhibitor, as well as with idelalisib, a PI3K inhibitor, has been approved for the treatment of CLL. Furthermore, the R-GemOx regimen, which combines rituximab with gemcitabine and oxaliplatin, has exhibited notable efficacy in r/r B-NHL in phase 2 clinical trial (175), and is currently being compared to a similar regimen using glofitamab instead of rituximab (glofit-GemOx) in a phase 3 clinical trial (176). In the treatment of CD20⁺ B-ALL, rituximab is added to standard chemotherapy regimens in patients with a Philadelphia chromosome-negative (Ph-) B-ALL. In patients with Philadelphia chromosome-positive (Ph+) CD20⁺ B-ALL, rituximab is combined with chemotherapy and BCR-ABL1 tyrosine kinase inhibitors, such as imatinib and dasatinib.

Several clinical trials have investigated the efficacy of other combination therapies involving CD20-targeting in patients with B-NHLs. Among these trials, the combination of R-DHAP regimen with temsirolimus, an mTOR inhibitor, has shown encouraging results, demonstrating improved outcomes in patients with r/r DLBCL (47). Temsirolimus has also demonstrated effectiveness in combination with rituximab alone in patients with r/r MCL in phase 2 clinical trial (177). Moreover, checkpoint inhibitors are also being tested in phase 1 and 2 clinical trials in combination with anti-CD20 therapies. The addition of atezolizumab to an R-CHOP regimen in previously untreated DLBCL patients resulted in 77,5% CR (178). It is also tested in combination with BsAbs glofitamab and mosunetuzumab in phase 1/2 clinical trials in patients with NHL (NCT02500407, NCT03533283) (179). Additionally, pembrolizumab (anti-PD1 mAb) was evaluated in combination with rituximab in a single-arm phase 2 clinical trial and resulted in 67% ORR with 50% CR among patients with r/r FL (180).

Notably, drugs that are used in combinations with CD20-targeting immunotherapies may have a bidirectional impact on CD20 antigen expression and thus the effectiveness of these therapies. Prednisolone, a glucocorticosteroid present in many chemotherapeutic regimens, was shown to upregulate CD20 on some primary B-ALL samples in vitro (4). On the other hand, some drugs that are used together with anti-CD20 mAbs, such as BTK inhibitor ibrutinib, PI3Kδ inhibitor idelalisib, or SYK inhibitor dasatinib, were shown to downregulate CD20 and demonstrated inhibitory effects on cytotoxic effector cells (181–184). These drugs decreased the efficacy of anti-CD20 mAbs in vitro (183, 184). It may also be one of the reasons for the failure of an attempt to improve ibrutinib efficacy in CLL by the addition of rituximab, as demonstrated in a randomized clinical trial showing no improvement in PFS in the rituximab+ibrutinib group versus ibrutinib alone (185). This highlights the need for further research on the drug-induced changes in cellular signaling and related CD20 regulation. Understanding these relationships may be important for selecting the most effective therapies and improving therapeutic results. Interestingly, several classes of CD20-upregulating drugs were described, including aurora kinase inhibitors, FOXO1 inhibitors, and chromatin modulators, enabling the increase in anti-CD20-mAbs efficacy in preclinical settings (13). Combining these drugs with CD20-targeting therapies could be a potentially valuable strategy to overcome resistance, however, it requires further evaluation in a clinical setting.

A breakthrough in the treatment of B cell malignancies is evident with recent approvals of CD19 CAR-T cells and BsAbs, particularly those targeting CD20xCD3, offering effective treatment and potential cure for r/r patients. Over the past 25 years, CD20, an early target in immunotherapy, has demonstrated remarkable effectiveness. However, the widespread use of cytotoxic T cell-based therapies appeared with new challenges such as treatment-related complications and side effects. Effective management requires the accumulation of comprehensive knowledge and experience, including identifying risk factors for CRS, ICAN, and refining treatment guidelines. These improvements are crucial for the widespread use of these innovative drugs.

With diverse treatment modalities emerging, from naked mAbs to BsAbs and CAR-T cells, understanding determinants of activity and resistance mechanisms for the specific types of treatment are crucial for their optimal selection and clinical efficacy. Decent levels of CD20 are essential for the efficacy of all types of CD20-directed immunotherapies, however, recent preclinical reports emphasize that different types of anti-CD20 therapies require different amounts of CD20 protein on the cell surface to be effective. While a certain level of reduction in CD20 compromises the activity of anti-CD20 mAbs and BsAbs, it may still be adequate for the effectiveness of CD20 CAR-T cells (28). Although the CD20 CAR-T constructs currently being tested in the clinic show great efficacy, further refinements to the CD20 CAR constructs, including changes around the scFv sequence, have shown significant superiority in preclinical models and offer the prospect of even better outcomes for patients (186).

Key directions for CD20 immunotherapy improvement also include combination strategies with small molecule drugs and simultaneous targeting of multiple immunotherapy targets to enhance precision and minimize relapse risks. Simultaneous targeting of CD20 with other antigens like CD19 and CD22 demonstrates efficacy in preclinical models (154) and ongoing clinical trials. Noteworthy, a better understanding of the determinants of response and resistance will be critical for patient selection and future rational combinations.

AD: Conceptualization, Writing – original draft, Writing – review & editing, Visualization, Investigation. KD: Conceptualization, Writing – original draft, Writing – review & editing, Investigation. MF: Conceptualization, Investigation, Writing – original draft, Writing – review & editing, Funding acquisition, Supervision, Visualization.