Intended for healthcare professionals

Clinical Review State of the Art Review

Novel treatments in B cell non-Hodgkin’s lymphomas

BMJ 2022; 377 doi: (Published 20 April 2022) Cite this as: BMJ 2022;377:e063439
  1. Yazeed Sawalha, assistant professor,
  2. Kami Maddocks, professor
  1. Internal Medicine – Division of Hematology, Ohio State University Comprehensive Cancer Centre, Columbus, OH, USA
  1. Correspondence to: Y Sawalha yazeed.sawalha{at}


The improved understanding of lymphoma biology and recent advances in the field of cancer immunology have paved the way for the development of many effective small molecule inhibitors and immunotherapies in B cell non-Hodgkin’s lymphomas. This article reviews novel treatments that have been approved recently by the US Food and Drug Administration and are now routinely used in clinical practice. It discusses their mechanisms of action, efficacy and safety, current therapeutic roles, and future directions in the treatment paradigm of different types of B cell non-Hodgkin’s lymphoma. It also reviews other exciting novel treatments that are not yet approved but have unique mechanisms of action and have shown encouraging early results.


Non-Hodgkin’s lymphomas (NHLs) are a heterogeneous group of cancers of the immune system that can involve any organ in the body, have broad range presentations that span the spectrum of indolent to very aggressive clinical behavior, and can be seen by physicians from most specialties. In this review, we discuss novel treatments approved recently by the US Food and Drug Administration (FDA) in B cell NHL (fig 1). We also highlight novel treatments under investigation that have shown promising results and are anticipated to have a future role in lymphoma therapy. We discuss these treatments in two major groups based on their mechanisms of action: immune based cellular or antibody therapies and small molecule inhibitors (fig 2 and fig 3).

Fig 1
Fig 1

Timeline of recent approvals by the US Food and Drug Administration of novel small molecular inhibitors and immunotherapies in diffuse large B cell lymphoma (DLBCL), mantle cell lymphoma (MCL), follicular lymphoma (FL), marginal zone lymphoma (MZL), and lymphoplasmacytic lymphoma (LPL). Approvals before 2017 are not shown. Axi-cel=axicabtagene ciloleucel; BR=bendamustine plus rituximab; Liso-cel=lisocabtagene maraleucel; Tisa-cel=tisagenlecleucel

Fig 2
Fig 2

Simplified schematic of mechanisms of action of recent immune based antibody and cellular therapies in B cell lymphoma. ADC=antibody-drug conjugate; CAR=chimeric antigen receptor; mAb=monoclonal antibody

Fig 3
Fig 3

Mechanisms of action of selected small molecule inhibitors approved in B cell lymphoma. AKT=protein kinase B; BCR=B cell receptor; BTK=bruton tyrosine kinase; LYN=Lck/Yes kinase; MAPK=mitogen activated protein kinase; mTOR=mammalian target of rapamycin; NFAT=nuclear factor of activated T cells; NF-κB=nuclear factor-κB; PI3K=phosphatidylinositol-3 kinase; PIP2/PIP3=phosphatidylinositol bi/triphosphate; PRC2=polycomb repressive complexes 2; SYK=spleen tyrosine kinase; XPO1=exportin 1

Despite it being an aggressive lymphoma, most patients with diffuse large B cell lymphoma (DLBCL) respond to chemoimmunotherapy with R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone) or a similar regimen. However, approximately 40% of patients will have refractory disease or relapse, most of whom will ultimately die from their lymphoma.1 For more than 20 years, the standard of care treatment for eligible patients with relapsed or refractory DLBCL has been second line or “salvage” chemotherapy followed by consolidation with high dose chemotherapy and autologous hematopoietic stem cell transplantation (AHCT). Despite the success of AHCT, many patients are ineligible for it, and the long term disease-free survival is still only about 50% even for those who are eligible.23 Overall, only a minority of patients with relapsed or refractory DLBCL are cured with AHCT.

Mantle cell lymphoma (MCL) is a B cell NHL characterized by an aggressive clinical course in most patients. A small subset (10-15%) of patients with MCL have indolent disease and can be observed, but most have rapidly progressing disease and need treatment at the time of diagnosis.4 Although most patients with MCL respond to frontline chemotherapy, relapses are inevitable. Patients with relapsed MCL generally have poor outcomes, with a median progression-free survival (PFS) of less than two years when treated with currently available therapies.4 Follicular lymphoma, marginal zone lymphoma (MZL), and lymphoplasmacytic lymphoma are indolent B cell NHLs typically characterized by slow growth and relapsing and remitting course. Survival outcomes for patients with these indolent NHLs have improved significantly over the past few decades, but these are generally considered incurable lymphomas with progressively shortening remissions with each subsequent treatment. Patients with refractory disease or those who relapse early after first line therapy generally have more aggressive disease and poor outcomes.5

Sources and selection criteria

We searched PubMed and Medline from 2010 to 2021 to identify references for this review article. We selected peer reviewed articles in the English language on the basis of the following search terms: lymphoma, CAR T cell, bispecific, antibody-drug conjugate, BTK, BCL2, EZH2, XPO1, and PI3K. We prioritized references from clinical trials whenever available and emphasized trials that led to FDA approvals of new treatments. We also identified references from relevant review articles. As this review article discusses novel treatments and results of ongoing studies, we included preliminary results of clinical trials presented at major national or international meetings but not yet published. We excluded articles published in non-peer reviewed journals.


Of the estimated 81 560 new cases of NHL in the USA in 2021, approximately 85-90% will originate from B cells and the rest from T cells or natural killer cells.6 DLBCL and follicular lymphoma are the two most common types of lymphoma in the USA, with an estimated 27 650 and 13 960 new cases, respectively, in 2016.7 MZL, MCL, and lymphoplasmacytic lymphoma are relatively uncommon, with an estimated 7460, 3320, and 2330 new cases, respectively, in 2016.7

Immune based therapies

Chimeric antigen receptor T cells

Chimeric antigen receptor (CAR) T cells are T cells transfected by a viral vector to introduce a construct coding for a CAR into the T cell’s DNA.8 The CAR is a synthetic receptor composed of a single chain antibody specific to a tumor antigen fused to T cell activation (CD3ζ chain) and co-stimulatory domains (4-1bb or CD28). When expressed on the surface of the T cell, the CAR mediates binding to the target tumor antigen, independently of major histocompatibility complex recognition, and subsequently activates the T cell and induces target cell killing. Examples of B cell surface antigens that have been targeted by CAR T cells in lymphoma include CD19, CD20, and CD22. The manufacturing and administration of CAR T cells is a complex process that entails collecting T cells from the patient (autologous) or a donor (allogeneic) via leukapheresis, T cell activation and transduction with a viral vector to express CAR, ex vivo expansion, and then infusion into the patient.8 Patients typically receive lymphodepleting chemotherapy before CAR T cell infusion to create an environment favorable for T cell proliferation. All FDA approved CAR T cell products in lymphoma are autologous CAR T cells targeting the cell surface antigen CD19 (table 1). As CD19 is expressed on normal B cells and in most B cell lymphomas and leukemias, CD19 directed therapies strike the optimal balance between tumor specificity and acceptable on-target off-tumor toxicity.14

Table 1

Summary of clinical trials of CAR T cells in diffuse large B cell lymphoma (DLBCL), mantle cell lymphoma (MCL), and follicular lymphoma (FL)

View this table:

CAR T cell associated toxicities

The potent T cell engagement and activation associated with CAR T cell therapy results in significant and potentially life threatening toxicities—namely, cytokine release syndrome (CRS) and immune effector cell associated neurotoxicity syndrome (ICANS).15 CRS is caused by the rapid expansion and activation of CAR T cells and release of pro-inflammatory cytokines from T cells and other immune cells. CRS typically occurs in the first few days after CAR T cell infusion and manifests with fever, tachycardia, hypoxia, and hypotension; hemodynamic instability and end organ dysfunction occur in severe cases. In addition to intensive monitoring and supportive care, high dose corticosteroids and the anti-interleukin 6 receptor antibody tocilizumab are the mainstay treatments for CRS.815 The pathophysiology of ICANS is less well understood, but it is also believed to be caused by a supraphysiologic immune activation state, with endothelial activation and blood-brain barrier disruption possibly playing an important role.16 ICANS commonly follows CRS and can have variable clinical presentations ranging from mild confusion and tremors to aphasia, obtundation, seizures, cerebral edema, and coma.15 Treatment is mainly corticosteroids, antiepileptics, and supportive care measures. Treatment with the currently approved CAR T cell products results in severe (grades ≥3) CRS and ICANS in up to 22% and 31% of patients, respectively.91011121718 The risk of these toxicities varies depending on many factors including the type of CAR T product used as well as patient related and lymphoma related factors.15 CAR T cell products with a 4-1bb co-stimulatory domain have lower rates and delayed onset of CRS and ICANS owing to lesser and delayed CAR T cell expansion and activation. In addition to CRS and ICANS, other toxicities associated with CAR T cells include prolonged myelosuppression, B cell aplasia, and hypogammaglobinemia. Intravenous immunoglobulin infusions are commonly used to decrease the risk of infections.15

Role of CAR T cells in DLBCL

Three CAR T cell products are approved by the FDA for patients with relapsed or refractory DLBCL who have received at least two lines of systemic therapies: axicabtagene ciloleucel (axi-cel) (October 2017), tisagenlecleucel (tisa-cel) (May 2018), and lisocabtagene maraleucel (liso-cel) (March 2021). In the phase 1/2 trials that led to their approvals, these three CAR T cell products resulted in objective response rates (ORRs) of 52% to 82%, with complete response rates of 40% to 54%, in heavily pretreated patients with relapsed or refractory DLBCL and other related aggressive B cell NHLs.91012 Long term follow-up data show durable remissions and possibly cure in 30-40% of patients treated with these CAR T cells products, particularly in those who achieve complete response.1119 Multicenter retrospective analyses show comparable safety and efficacy outcomes for patients treated with CAR T cells in routine clinical practice outside of clinical trial settings.20212223 These outcomes compare very favorably to historical controls of patients with refractory DLBCL treated with the best therapies available before CAR T cells, in which the ORR and complete response rate were 26% and 7%, respectively, with a median overall survival of only six months.24 Treatment with CAR T cells also improves the quality of life, including physical, psychosocial, and emotional functioning, of patients with relapsed DLBCL.25 The three CAR T products differ in several aspects including their co-stimulatory domains (which might affect their pharmacokinetics, efficacy, and toxicity profiles), their transfection routes, the composition of the T cell product, and the manufacturing processes (table 1).15 However, no randomized studies have directly compared the safety or efficacy of these three CAR T cell products. Across trial comparisons are cautioned given the heterogeneity of the enrolled patient populations, relatively small number of patients enrolled on each trial, and inconsistencies in defining, grading, and treating or preventing CRS and ICANS. CAR T cell product selection is also driven by practical and logistical factors such as the turnaround time between leukapheresis and the availability of the product for infusion, as well as the experience of the physician/facility with certain products.

Role of CAR T cells in MCL

The FDA approved brexucabtagene autoleucel (July 2020) for patients with relapsed or refractory MCL on the basis of the phase 2 Zuma-2 trial.17 Zuma-2 included 74 patients with high risk relapsed or refractory MCL, of whom 68 patients received brexucabtagene autoleucel. Of the 60 patients evaluable for response, the ORR was 93% (95% confidence interval 84% to 98%) including complete response in 67% (53% to 78%). With a median follow-up of 12 months, the one year PFS and overall survival were 61% and 83%, respectively. Although longer follow-up is needed, these results are very encouraging given the high response rate in these patients at high risk who otherwise have limited effective treatment options. Liso-cel is being evaluated in MCL, with preliminary results in 32 patients with high risk relapsed or refractory MCL treated on the phase 1 TRANSCEND NHL 001 trial presented in abstract form recently.18 The ORR was 84% (67% to 95%), and the complete response rate was 66% (47% to 81%); with a median follow-up of six months, the median PFS and overall survival were not reached.

Role of CAR T cells in follicular lymphoma

Axi-cel was approved by the FDA (March 2021) in patients with follicular lymphoma who have received at least two lines of systemic therapy on the basis of the phase 2 Zuma-5 trial. The full analysis of Zuma-5 has not been published yet, but results were recently presented in abstract form.13 Most patients had high risk follicular lymphoma and were heavily pretreated. In the 124 patients who received axi-cel, the ORR was 94% including complete response in 80%. With a median follow-up of 19 months, the one year PFS and overall survival were 78% and 93%, respectively. Although Zuma-5 also included a small number of patients with relapsed MZL (n=22) who achieved an ORR and complete response rate of 85% and 60%, respectively, axi-cel is not approved in relapsed MZL. Tisa-cel is also being evaluated in follicular lymphoma, with preliminary results from the phase 2 ELARA trial presented in abstract form.26 Ninety seven patients with high risk and heavily pretreated follicular lymphoma received tisa-cel, of whom 52 were evaluable for response (response assessment was conducted after three months from CAR T cell infusion). The ORR was 83% including complete response in 65%, and with a median follow-up of 10 months, the six month PFS was 73%. Whether CAR T cell therapy will result in long term remissions and possibly cure in indolent B cell NHL and MCL remains uncertain and requires longer follow-up. A recent report from the National Cancer Institute of its first-in-human clinical trial of axi-cel reported ongoing responses beyond three years in five (63%) of the eight patients with indolent B cell NHL.27

Future directions for CAR T cells in B cell NHL

The success of these CAR T cell products in patients with high risk and heavily pretreated lymphomas has encouraged clinical trials exploring their activity in earlier lines of therapy. Reviewing all ongoing and planned clinical trials of CAR T cells is beyond the scope of this article, but notable mentions include a phase 3 trial for each of tisa-cel (BELINDA, NCT03570892), axi-cel (Zuma-7, NCT03391466), and liso-cel (TRANSFORM, NCT0357535) in patients with DLBCL and related aggressive B cell NHLs at first relapse, in which patients are randomized to AHCT (the current standard of care) versus the CAR T cell product. Tisa-cel (TIGER, NCT04161118) and liso-cel (PILOT, NCT03483103) are being investigated in phase 2 studies in patients with relapsed or refractory DLBCL ineligible for AHCT owing to age, comorbidities, and/or poor performance status. Axi-cel is being studied in a phase 2 trial of patients with high risk DLBCL and other related aggressive B cell NHLs who achieve suboptimal response to first line chemoimmunotherapy (Zuma-12, NCT03761056).

Several strategies are being explored to improve the efficacy of CAR T cells, including “third generation” CAR T cells that have dual co-stimulatory domains with enhanced proliferation and persistence capabilities,2829 combination therapies to reduce immune evasion and CAR T cell exhaustion (for example, in combination with checkpoint inhibitors (NCT02706405) or BTK inhibitors (NCT04257578)), dual antigen targeting to reduce CD19 antigen loss (for example, CD19/CD22 (NCT03233854, NCT03287817) and CD19/CD20 (NCT04215016, NCT04007029)), and designing “armored” CAR T cells that secrete pro-inflammatory cytokines or express ligands that stimulate and recruit other immune cells.30313233 Allogeneic CAR T cells are also being evaluated (NCT03666000, NCT04416984) and offer the advantage of being readily available “off the shelf.” Furthermore, allogeneic CAR T cells might have superior biologic activity, as the T cells are collected from healthy donors not previously exposed to chemotherapy. The risk of graft-versus-host disease associated with allogeneic CAR T cells can be reduced by using gene editing technologies to eliminate T cell receptor expression from the allogeneic CAR T cell product.34 Other immune cells, particularly natural killer cells, are emerging as alternative vehicles for CAR engineering.35 A small phase 1/2 clinical trial showed encouraging activity of anti-CD19 CAR natural killer cells derived from cord blood.36 In addition to the CAR, the natural killer cells were engineered to express interleukin 15 to enhance in vivo expansion and inducible caspase 9 as a safety switch (to trigger apoptosis of the CAR natural killer cells in cases of unacceptable toxicities). Of the 11 patients with relapsed B cell NHLs or chronic lymphocytic leukemia (CLL) who were treated with the anti-CD19 CAR natural killer cells, eight patients achieved a response (ORR=73%) including complete response in seven (64%) patients. No cases of CRS, ICANS, or graft-versus-host disease occurred. Many clinical trials of natural killer cells and other immune cells for CAR based therapies are planned or ongoing (reviewed in depth by Basar et al35).

Bispecific antibodies

Bispecific antibodies are composed of small linker peptides connecting two different single chain variable fragments with one fragment designed to bind to CD3 on T cells and the other to a tumor associated antigen. The simultaneous binding of CD3 on T cells and the tumor associated antigen triggers T cell mediated cytotoxicity of the malignant cell. Blinatumomab, which binds to CD3 and CD19, is the first bispecific antibody to receive FDA approval in relapsed or refractory B cell precursor acute lymphoblastic leukemia. However, it has limited activity in B cell NHL.37 Several bispecific antibodies targeting CD3 and CD20 have shown very encouraging early results in relapsed B cell NHLs, with odronextamab, mosunetuzumab, epcoritamab, and glofitamab having the most data reported to date (table 2).38394041 An ongoing phase 1 trial of glofitamab recently reported on preliminary efficacy and safety in 171 patients with relapsed DLBCL, other aggressive B cell NHLs, or follicular lymphoma.41 Patients were heavily pretreated with a median number of three previous therapies, and 91% were refractory to the most recent therapy. The ORR was 54% (46% to 61%) including complete response in 37% (30% to 45%) among all treated patients and 66% (48% to 81%) including complete response in 57% (39% to 74%) in the 35 patients treated at the recommended phase 2 dose. With a median follow-up of 14 months, the median PFS was three months in aggressive NHLs including DLBCL and 12 months in follicular lymphoma. Although longer follow-up is needed, responses seemed to be durable in patients with complete response, as 73% of patients with aggressive NHLs remained in complete response at 12 months and 91% of patients with follicular lymphoma remained in complete response up to 23 months. Data from phase 1/2 trials on odronextamab, mosunetuzumab, and epcoritamab have been presented in abstract forms, with ORRs ranging from 33% to 68% including complete response in 19-55% in DLBCL and from 63% to 90% including complete response in 43-70% in follicular lymphoma and other indolent NHLs. Importantly, these responses were seen in high risk and heavily pretreated patients including patients progressing after CAR T cells, which is a particularly difficult population to treat.

Table 2

Summary of early clinical trials of bispecific antibodies in relapsed or refractory aggressive and indolent B cell non-Hodgkin’s lymphomas (NHLs)

View this table:

As expected, given T cell activation and release of pro-inflammatory cytokines, the bispecific antibodies can cause CRS and ICANS, although the risks of severe CRS and ICANS seem to be lower than with CAR T cells. Several mitigation strategies have successfully reduced the rate and severity of CRS and ICANS associated with the bispecific antibodies, including split and step-up dosing, pre-medications with steroids and/or anti-CD20 monoclonal antibodies, and subcutaneous formulations (which result in comparable bioavailability but lower peak cytokine concentrations).42 Compared with the currently available CAR T cell products, the bispecific antibodies have the advantage of being readily available “off the shelf” without the need for a production process individualized for each patient. Moreover, unlike with CAR T cells, lymphodepleting chemotherapy is not needed before treatment with the bispecific antibodies. However, the optimal duration of treatment with the bispecific antibodies is still unclear (versus one time treatment with CAR T cells), and data on durability of their response are limited.

Tafasitamab plus lenalidomide

As discussed for CAR T cells, CD19 is an attractive target in B cell malignancies given its high expression on B cells (normal and malignant) and limited expression in other tissues. Tafasitamab is a humanized anti-CD19 monoclonal antibody designed to enhance binding to Fcγ receptors on immune cells and consequently increase its antibody dependent, cell mediated cytotoxicity and antibody dependent cellular phagocytosis.43 Tafasitamab’s clinical activity as monotherapy in relapsed B cell NHLs was encouraging, but it was more remarkable when tafasitamab was combined with lenalidomide.44 Lenalidomide is an immunomodulatory agent approved in relapsed MCL (June 2013) and in combination with rituximab in relapsed follicular lymphoma and MZL (May 2019). Lenalidomide has several mechanisms of action including direct antineoplastic and antiangiogenic activities as well as immunomodulatory effects.45 Notably, lenalidomide has synergistic activity when combined with anti-CD20 monoclonal antibodies as it enhances their natural killer cell and antibody dependent, cell mediated cytotoxicity.4647 The combination of tafasitamab and lenalidomide was evaluated in the phase 2 L-MIND study, which included patients with relapsed or refractory DLBCL who were ineligible for AHCT.48 The ORR in the 80 evaluable patients was 58% (46% to 69%) including complete response in 40%. With a median follow-up of 32 months, the median PFS and overall survival were 12 and 32 months, respectively. The median duration of response was 35 months. Importantly, patients who achieved a complete response (n=32) had durable remissions, with 86% maintaining their responses at two years.49 The main toxicities were myelosuppression, fatigue, rash, and diarrhea, but treatment was generally well tolerated. On the basis of the L-MIND study results, the FDA approved the combination of tafasitamab and lenalidomide for patients with relapsed or refractory DLBCL (July 2020). Tafasitamab is being evaluated in combination with other active agents in several trials including a phase 1 trial in combination with R-CHOP and R-CHOP plus lenalidomide (NCT04134936) and a randomized phase 2/3 trial in combination with bendamustine compared with bendamustine plus rituximab in relapsed or refractory DLBCL (B-MIND, NCT02763319).

Antibody-drug conjugates

Antibody-drug conjugates (ADCs) are monoclonal antibodies covalently attached via a protease cleavable linker to cytotoxic payloads. Once the ADC attaches to its tumor associated antigen on the cell surface, the complex is internalized and directed to lysosomes where the linker is cleaved and the cytotoxic payload is released to the cytosol. Commonly used cytotoxic payloads include microtubule disrupting antimitotic agents (for example, monomethyl auristatin E (MMAE)) and DNA binding agents (for example, pyrrolobenzodiazepine dimers). ADCs combine the high selectivity and favorable pharmacokinetics of monoclonal antibodies with the antitumor potency of the cytotoxic payload and can be directed toward various tumor associated antigens. ADCs have been shown to be effective in B cell hematologic malignancies with brentuximab vedotin (anti-CD30 conjugated to MMAE) approved in Hodgkin’s and T cell lymphomas, inotuzumab ozogamicin (anti-CD22 conjugated to calicheamicin) in B cell precursor acute lymphoblastic leukemia, and moxetumomab pasudotox (anti-CD22 conjugated to Pseudomonas exotoxin (PE38)) in hairy cell leukemia.

Polatuzumab vedotin is an ADC comprised of a humanized anti-CD79b monoclonal antibody attached to MMAE. CD79b is an ideal target because of its high expression on most types of B cell NHL and limited expression on normal B cells.50 Polatuzumab resulted in modest clinical activity as monotherapy in patients with relapsed or refractory DLBCL or follicular lymphoma in small phase 1/2 clinical trials but showed more benefit when combined with chemoimmunotherapy.5152 In the two part GO29365 phase 1/2 clinical trial, polatuzumab was combined with bendamustine plus an anti-CD20 monoclonal antibody (rituximab or obinutuzumab) in patients with DLBCL ineligible for AHCT.53 Treatment was generally well tolerated, with the major adverse events being hematological (myelosuppression) and peripheral neuropathy. In the second part of GO29365, 80 patients were randomized to bendamustine plus rituximab (BR) alone or with polatuzumab. The addition of polatuzumab to BR resulted in statistically significant improvement in ORR (45% v 18%), complete response rate (40% v 18%), PFS (median 10 v 4 months), and overall survival (median 12 v 5 months). Updated data showed that 25% of patients treated with BR plus polatuzumab had ongoing responses lasting more than two years, indicating that a small subset of patients can achieve durable remissions.54 On the basis of these results, the FDA approved polatuzumab in combination with BR (June 2019) for patients with relapsed or refractory DLBCL after at least two previous therapies.

GO29365 also included a cohort of 80 patients with relapsed or refractory follicular lymphoma randomly assigned to BR alone or with polatuzumab. Interim data (presented in abstract form) did not show any improvement in response rates, PFS, or overall survival with the addition of polatuzumab, although longer follow-up is needed.55 Emerging data show encouraging results with the combination of polatuzumab and lenalidomide plus obinutuzumab as a potential non-chemotherapy option for patients with relapsed follicular lymphoma.56 We eagerly await the results of the recently concluded large phase 3 POLARIX trial that randomized 875 patients with previously untreated DLBCL to standard of care chemoimmunotherapy alone (R-CHOP) or R-CHP (without vincristine) plus polatuzumab. A similar ongoing phase 3 trial is randomizing older patients with previously untreated DLBCL to reduced dose R-CHOP (R-mini-CHOP) versus R-mini-CHP plus polatuzumab (NCT04332822). In addition, polatuzumab is being investigated in combination with R-CHOP with or without mosunetuzumab (NCT02729896) and with other chemotherapy regimens including R-GemOx (rituximab, gemcitabine, oxaliplatin) in the phase 3 POLARGO trial (versus R-GemOx alone in relapsed or refractory DLBCL).

Loncastuximab tesirine is an anti-CD19 monoclonal antibody conjugated to a pyrrolobenzodiazepine dimer that was recently approved by the FDA (April 2021) for patients with relapsed or refractory DLBCL after two or more lines of systemic therapy. In the LOTIS-2 phase 2 trial, loncastuximab resulted in an ORR of 48% (40% to 57%) including complete response in 24% among 145 patients with relapsed or refractory DLBCL and other related aggressive B cell NHLs.57 The median duration of response, PFS, and overall survival were 10, five, and 10 months, respectively. Loncastuximab resulted in responses in patients with refractory disease and high risk features, as well as in a small number of patients who had progression after CAR T cells. The most common grade 3 or 4 toxicities were myelosuppression and elevated γ-glutamyl transferase.57 Loncastuximab is being evaluated in combination with ibrutinib (NCT03684694) and in a phase 3 trial in combination with rituximab versus R-GemOx in patients with relapsed or refractory DLBCL (NCT04384484). Loncastuximab, tafasitamab plus lenalidomide, and BR plus polatuzumab are preferred treatment options for patients with relapsed or refractory DLBCL ineligible for AHCT or CAR T cells. Their roles in patients who progressed or relapsed after AHCT or CAR T cells, or as a bridge to these therapies, are undefined as only a minority of patients treated on the LOTIS-2, GO29365, and L-MIND trials had received AHCT and/or CAR T cells. The optimal sequencing of these treatments is also unclear given the lack of direct comparisons or reliable predictive biomarkers. Acknowledging the caveats of cross trial comparisons, tafasitamab plus lenalidomide resulted in higher response rates and has more data so far on durability of response compared with loncastuximab and BR plus polatuzumab.58 However, patients treated with loncastuximab on the LOTIS-2 trial and BR plus polatuzumab on the GO29365 trial had overall higher adverse risk disease features (such as higher percentage of patients with three or more previous therapies and disease refractory to most recent treatment) compared with those treated with tafasitamab plus lenalidomide on the L-MIND trial. Because both tafasitamab and loncastuximab, as well as currently approved CAR T cell products, target CD19, whereas polatuzumab targets CD79b, treatment with BR plus polatuzumab might be preferred in patients who have progressed on previous treatments with CD19 directed therapies. Treatment with BR plus polatuzumab has the advantage of being time limited (six cycles of three weeks) compared with treatment until progression with tafasitamab plus lenalidomide (although treatment with lenalidomide is limited to the first year) and for at least one year with loncastuximab. Pre-existing peripheral neuropathy, a relatively common problem in patients with DLBCL previously treated with vinca alkaloids and/or platinum agents, might limit treatment with BR plus polatuzumab.

Other ADCs are being developed or evaluated in clinical trials for patients with relapsed or refractory B cell NHLs, with early promising results for TRPH-222 (an anti-CD22 monoclonal antibody conjugated to a maytansinoid payload) and VLS-101 (an anti-ROR1 monoclonal antibody conjugated to MMAE), as summarized in table 3.6061

Table 3

Summary of early clinical trials of antibody-drug conjugates in relapsed or refractory B cell non-Hodgkin’s lymphomas (NHLs)

View this table:

CD47 blockade

CD47, a cell surface receptor overexpressed on lymphoma cells, plays an important role in inhibiting tumor cell phagocytosis by macrophages and dendritic cells.62 CD47 binds to its ligand, signal regulatory protein α (SIRP α), on phagocytic cells and provides what is known as the “do not eat me” signal, leading to reduced tumor phagocytosis and immune evasion. Blocking the CD47/SIRPα has been shown to be clinically beneficial in lymphoma and other malignancies. Magrolimab (formerly Hu5F9-G4) is a humanized anti-CD47 monoclonal antibody that has shown encouraging results in a small phase 1 trial of 22 patients with relapsed or refractory DLBCL or follicular lymphoma, most of whom were refractory to rituximab.63 In combination with rituximab, magrolimab resulted in an ORR of 50% including complete response in 36%. Treatment was well tolerated, with the major toxicities being infusion related reactions and anemia (expected on-target effect as CD47 blockade accelerates the elimination of aging red blood cells). A phase 2 study of magrolimab in combination with rituximab or rituximab plus chemotherapy is ongoing (NCT02953509). Trials of other agents targeting the CD47/SIRPα are also ongoing, including SRF231, an anti-CD47 monoclonal antibody (NCT03512340), and TTI-621, a CD47 decoy receptor (NCT02663518).

Small molecule inhibitors

Bruton tyrosine kinase inhibitors

Bruton tyrosine kinase (BTK) is a cytoplasmic kinase that plays a key role in amplifying signals from the B cell receptor and is essential for the survival, maturation, and proliferation of B cells.64 The effect of BTK loss-of-function mutations in the X linked agammaglobulinemia, leading to the near absence of B cells and severe humoral immune deficiency, highlights the crucial role of BTK in B cell maturation and survival.65 The three FDA approved BTK inhibitors are ibrutinib, acalabrutinib, and zanubrutinib, which inactivate BTK by irreversibly binding to the ATP binding pocket in the kinase domain. Common or concerning side effects of these BTK inhibitors include headache, muscle spasms, diarrhea, bleeding, neutropenia, hypertension, and atrial fibrillation. Compared with the first generation BTK inhibitor ibrutinib, the second generation BTK inhibitors, acalabrutinib and zanubrutinib, have fewer off-target effects and enhanced safety profiles, particularly lower rates of cardiovascular toxicities including atrial fibrillation.66 Ibrutinib, acalabrutinib, and zanubrutinib are FDA approved in relapsed MCL (November 2013, October 2017, November 2019, respectively), whereas only ibrutinib is currently approved in treatment naïve and relapsed lymphoplasmacytic lymphoma, as monotherapy (January 2015) and in combination with rituximab (August 2018), and as monotherapy in relapsed MZL (January 2017). Table 4 summarizes the results of the studies of BTK inhibitors that led to their approvals.

Table 4

Summary of studies of BTK inhibitors leading to their approvals in mantle cell lymphoma (MCL), marginal zone lymphoma (MZL), and lymphoplasmacytic lymphoma (LPL)

View this table:

In MCL, ibrutinib, acalabrutinib, and zanubrutinib are now generally considered the preferred treatment option for patients with relapsed or refractory disease given their high response rates and favorable safety profiles.4 Cross trial comparisons show similar efficacy for ibrutinib, acalabrutinib, and zanubrutinib in relapsed or refractory MCL, with ORRs ranging from 68% to 87% including complete response in 21-77%. Acalabrutinib or zanubrutinib may be preferred over ibrutinib in patients at significant risk for cardiovascular events 676869707172737475 However, despite the high response rates, remissions with these BTK inhibitors in relapsed MCL are short lived, with median PFS of less than two years. Further improvement is clearly needed, especially as the outcomes of MCL after progression on BTK inhibitors are poor.80 Many trials are ongoing combining BTK inhibitors with chemotherapy and/or other novel agents in MCL in the frontline and relapsed settings, including combinations offering “non-chemotherapy” options such as with venetoclax (a BCL inhibitor; NCT03112174, NCT03946878), palbociclib (a CDK 4/6 inhibitor; NCT03478514), CAR T cells (NCT04234061, NCT04484012), loncastuximab (NCT03684694), and others. In lymphoplasmacytic lymphoma, ibrutinib provides a non-chemotherapy option in the frontline and relapsed settings with a high response rate and durable remissions.7677 The ASPEN trial, the first phase 3 clinical trial to compare two BTK inhibitors directly, failed to show superior clinical efficacy for zanubrutinib over ibrutinib in lymphoplasmacytic lymphoma (complete response or very good partial response rates of 29% (20% to 40%) versus 20% (12% to 30%); P=0.12).81 Zanubrutinib had a better overall safety profile with lower incidence of diarrhea, bleeding, muscle spasms, peripheral edema, pneumonia, and atrial fibrillation but higher incidence of neutropenia. In MZL, ibrutinib resulted in an ORR of 58% including complete response in 10% and a median PFS of 16 months.7879 Zanubrutinib is being evaluated in the phase 2 MAGNOLIA trial, with initial results in 66 evaluable patients with relapsed or refractory MZL showing an ORR of 74% (62% to 84%) including complete response in 24%.82

Several other BTK inhibitors are in development in B cell malignancies. Orelabrutinib (ICP-022) and tirabrutinib (ONO-4059/GS-4059) are alternative selective and irreversible BTK inhibitors that have shown promising activity in B cell malignancies, whereas pirtobrutinib (LOXO-305), ARQ-531, fenebrutinib (GDC-0853), and vecabrutinib (SNS-062) are non-covalent reversible inhibitors that bind to different sites within the BTK.83848586 Pirtobrutinib was recently shown to have remarkable clinical activity in patients with relapsed MCL (ORR=52% (38% to 65%), complete response in 25%), lymphoplasmacytic lymphoma (ORR=68% (44% to 87%), complete response in 0%), and other B cell NHLs including patients who progressed on or were intolerant of previous irreversible BTK inhibitors (ORR=52% (38% to 66%), complete response in 25% for MCL; ORR=69% (39% to 91%), complete response in 0% for lymphoplasmacytic lymphoma).87

Phosphatidylinositol 3-kinase inhibitors

The phosphatidylinositol 3-kinases (PI3Ks) are enzymes important for various cellular functions including cellular proliferation, survival, and differentiation; intracellular trafficking; and immunity. Of the three classes of the PI3K enzymes, class I is the most important in malignancies, with the δ and γ isoforms playing prominent roles in hematopoietic cells.88 In particular, the PI3K-δ pathway is essential for normal B cell development, and its activation is implicated in the pathogenesis of lymphoma. The PI3K inhibitors exert their therapeutic effect in lymphoma through inhibiting pro-survival signaling from the B cell receptor, inhibiting cytokine signaling from the tumor microenvironment, and enhancing anti-tumor immunity through regulatory T cell inhibition and other mechanisms.88 The FDA approved idelalisib (targets PI3K-δ) (July 2014), copanlisib (pan-PI3K inhibitor with a slight selectivity for the α and δ isoforms) (September 2017), and duvelisib (targets PI3K-δ and PI3K-γ) (September 2018) in patients with relapsed follicular lymphoma who received at least two previous systemic therapies. Despite the efficacy of these three PI3K inhibitors in heavily pretreated patients with follicular lymphoma, their safety profiles and particularly the risk of immune mediated toxicities (colitis, pneumonitis, hepatitis) and opportunistic infections have limited their use.899091 More recently, umbralisib received FDA approval (February 2021) in relapsed follicular lymphoma after at least three previous systemic therapies and in MZL after at least one previous anti-CD20 based regimen. Compared with the other PI3K inhibitors, umbralisib has improved selectivity to the PI3K-δ isoform and inhibits casein kinase CK1-ε, a protein that inhibits regulatory T cell function.92 The relative preservation of regulatory T cell function and lower risk of immune mediated toxicities have distinguished umbralisib from the other PI3K inhibitors. In the UNITY-NHL phase 2 trial that included patients with relapsed follicular lymphoma (n=117), MZL (n=69), and small lymphocytic lymphoma (n=22), umbralisib resulted in an ORR of 45% (36% to 55%) including complete response in 5% (2% to 11%) in follicular lymphoma and an ORR of 49% (37% to 62%) including complete response in 16% (8% to 27%) in MZL.93 With a median follow-up of 28 months, the median PFS in follicular lymphoma and MZL was 11 months and not reached, respectively. Notable adverse events with umbralisib included diarrhea (59%, grade 3 in 10%), neutropenia (12%), transaminitis (19-20%, grade ≥3 in 7%), opportunistic infections (3%), noninfectious colitis (2%), and pneumonitis (1%). The risk of these toxicities with umbralisib compares favorably to the other approved PI3K inhibitors, although no direct comparisons in randomized clinical trials have been made. Idelalisib, duvelisib, and umbralisib are given orally, whereas copanlisib is administered intravenously. Other PI3K inhibitors with potentially improved efficacy and/or enhanced safety profiles are being developed, including PI3K-δ inhibitors (ME-401, parsaclisib, dezapelisib, acalisib, ACP-319), PI3K-δγ inhibitors (tenalisib), and pan-PI3K inhibitors (pilaralisib, pictilisib, AZD8835, voxtalisib) (reviewed in depth by Phillips et al94).


Selinexor is a first-in-class oral inhibitor of XPO1, a shuttling protein involved in the export of proteins from the nucleus to the cytoplasm, which is overexpressed in DLBCL and associated with inferior prognosis.95 XPO1 inhibition in DLBCL activates tumor suppressor proteins such as p53 and reduces the concentration of oncoproteins such as c-MYC and BCL2.96 In the phase 2 SADAL trial of patients with relapsed or refractory DLBCL ineligible for AHCT, selinexor resulted in an ORR of 28% (21% to 37%) including complete response in 12% (7% to 19%).96 With a median follow-up of 15 months, the median PFS and overall survival were three and nine months, respectively. Notable adverse events of selinexor include nausea, vomiting, anorexia, and myelosuppression. The FDA approved selinexor (June 2020) in patients with DLBCL after at least two previous systemic therapies. The modest clinical activity of selinexor favors BR plus polatuzumab, lenalidomide plus tafasitamab, and loncastuximab as preferred treatment options for patients with relapsed or refractory DLBCL who are ineligible for AHCT or CAR T cells. Selinexor is being evaluated in several clinical trials in DLBCL in combination with chemotherapy in the frontline and relapsed settings (NCT031478850, NCT02471911, NCT02741388) and in combination with novel agents such as ibrutinib (NCT02303392) and the BCL2 inhibitor venetoclax (NCT03955783).


EZH2 is a histone methyltransferase that leads to transcriptional silencing of target genes. In B cells, it enables germinal center formation by promoting cell proliferation and inhibiting plasma cell differentiation.97 Recurrent gain-of-function mutations in EZH2 occur in 14-22% of germinal center B cell DLBCL and 7-28% of follicular lymphoma.9899100101 Tazemetostat is an oral selective EZH2 inhibitor approved by the FDA (June 2020) for the treatment of patients with relapsed follicular lymphoma with an EZH2 mutation who have received at least two previous systemic therapies and for those with wild type EZH2 who have no satisfactory alternative treatment options. This approval was based on a phase 2 trial of tazemetostat in 99 patients with relapsed follicular lymphoma, of whom 45 patients had an EZH2 mutation.102 In patients with EZH2 mutations, tazemetostat results in an ORR of 69% (53% to 82%) including complete response in 13%; with a median follow-up of 22 months, the median PFS and overall survival were 14 months and not reached, respectively. Responses with tazemetostat were also seen in patients without EZH2 mutations, although with lower ORR (35%, 23% to 49%; complete response in 4%) and slightly shorter PFS (median 11 months). The responses seen in EZH2 wild type cases support results of preclinical studies showing that both mutant and wild type germinal center B cells are susceptible to EZH2 inhibition but mutant cells are more sensitive.103 Treatment with tazemetostat was exceptionally well tolerated with minimal serious toxicities. Tazemetostat’s excellent safety profile and high clinical efficacy favors its use over a PI3K inhibitor as a third line treatment in follicular lymphoma with EZH2 mutations. Tazemetostat is being investigated in combination with rituximab in a phase 2 trial (NCT04590820) and with lenalidomide plus rituximab in a phase 3 clinical trial (NCT04224493) in patients with relapsed follicular lymphoma.


BCL2 is an important anti-apoptotic molecule that has a key role in regulating the intrinsic apoptosis pathway.104 Overexpression of BCL2 due to t(14;18) is the hallmark of follicular lymphoma, being present in more than 90% of cases. BCL2 is also overexpressed in DLBCL, MCL, lymphoplasmacytic lymphoma, and CLL.104 Venetoclax is an oral selective BCL-2 inhibitor approved for the treatment of CLL and acute myeloid leukemia. Important toxicities of venetoclax include myelosuppression, gastrointestinal toxicity, and risk of tumor lysis syndrome. In a phase 1 clinical trial of patients with relapsed or refractory B cell NHL, venetoclax showed clinical activity in DLBCL (ORR=18%, n=34), follicular lymphoma (ORR=38%, n=29), MCL (ORR=75%, n=28), lymphoplasmacytic lymphoma (ORR=100%, n=4), and MZL (ORR=67%, n=3).105 Venetoclax showed encouraging results when combined with ibrutinib with or without obinutuzumab in two small clinical trials of patients with relapsed or previously untreated MCL, with an ORR of 71-75% including complete response in 67% among patients with relapsed disease and an ORR of 93% including complete response in 86% among previously untreated patients.106107 The one year PFS for patients with relapsed MCL was 75% in both studies. A small phase 1 clinical trial showed excellent clinical activity for the combination of venetoclax, lenalidomide, and rituximab in 28 patients with previously untreated MCL, with an ORR of 96% including complete response in 89%.108 However, the outcomes with venetoclax in patients with MCL treated with venetoclax outside of clinical trials (off-label use in heavily pretreated patients, most of whom received previous treatment with BTK inhibitors) were poor, with ORRs of 42-53% including complete response in 16-21% and median PFS of three to eight months.109110111 In a phase 2 trial of 31 patients with relapsed or refractory lymphoplasmacytic lymphoma, venetoclax resulted in an ORR of 87% and two year PFS of 76%.112 The phase 2 CONTRALTO clinical trial showed modest activity for venetoclax in combination with rituximab (ORR=35%) and no benefit from adding venetoclax to BR in patients with relapsed or refractory follicular lymphoma.113 Ongoing clinical trials are investigating the use of venetoclax in combination with chemotherapy in patients with newly diagnosed DLBCL (rituximab, etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin (R-EPOCH); NCT03984448) and in those with relapsed or refractory disease (rituximab, ifosfamide, carboplatin, etoposide (R-ICE); NCT03064867), as well as in combination with other novel/targeted agents (obinutuzumab, lenalidomide, ibrutinib, prednisone (NCT03223610); rituximab and ibrutinib (NCT03136497); obinutuzumab and lenalidomide (NCT02992522)).

Emerging treatments

As reviewed throughout the text, many new treatments have shown encouraging results and are anticipated to play important roles in the treatment paradigm of patients with B cell NHLs. The bispecific antibodies in particular have shown very promising results with acceptable safety profiles, and data on the long term durability of their response are eagerly awaited. Cellular therapies have already established roles in the treatment paradigms of various types of B cell NHLs, and further improvement in their efficacy and safety profiles is expected with newer generations of autologous CAR T cells as well as with other cellular products such as allogeneic CAR T and natural killer cells. ADCs have been shown to be a very effective class of therapeutic agents in B cell NHLs, with the potential for further improvement in their design and an expanding list of potential cellular targets. In addition to further refinement in the specificity of small molecule inhibitors and consequently their efficacy, safety, or both, advances in our knowledge of lymphoma biology continue to provide other potential targets for therapeutic intervention. Finally, a large number of trials are combining already approved and investigational agents to explore possible additive or synergistic activity of these agents.


The National Comprehensive Cancer Network (NCCN) guidelines are a comprehensive set of guidelines detailing the sequential management decisions and interventions that apply to 97% of cancers affecting patients in the USA ( The NCCN has a large active presence in Europe, with more than 140 000 registered users and nearly 720 000 NCCN clinical practice guidelines in oncology downloaded in 2018. The treatment approaches discussed throughout this review are generally consistent with the NCCN guidelines. As NCCN guidelines are updated periodically, very recent FDA approvals might not have been included in the guidelines yet.


The treatment landscape of B cell NHLs has evolved with the development of many novel therapies including immune based cellular and antibody therapies such as CAR T cells, bispecific antibodies, monoclonal antibodies, and ADCs and small molecule inhibitors such as BTK inhibitors, PI3K inhibitors, EZH2 inhibitors, and others. As our knowledge of lymphoma biology and cancer immunology expands, these therapies will improve further and newer exciting treatments will continue to emerge. Exploring potential combination therapies based on rational approaches and determining the optimal sequencing of these different treatments will be key to achieving higher therapeutic efficacy and acceptable toxicity. Furthermore, developing biomarkers to guide treatment selection will be crucial.

Glossary of abbreviations

  • ADC—antibody-drug conjugate

  • AHCT—autologous hematopoietic stem cell transplantation

  • BR—bendamustine plus rituximab

  • BTK—Bruton tyrosine kinase

  • CAR—chimeric antigen receptor

  • CLL—chronic lymphocytic leukemia

  • CRS—cytokine release syndrome

  • DLBCL—diffuse large B cell lymphoma

  • FDA—Food and Drug Administration

  • ICANS—immune effector cell associated neurotoxicity syndrome

  • MCL—mantle cell lymphoma

  • MMAE—monomethyl auristatin E

  • MZL—marginal zone lymphoma

  • NCCN—National Comprehensive Cancer Network

  • NHL—non-Hodgkin’s lymphoma

  • ORR—objective response rate

  • PFS—progression-free survival

  • PI3K—phosphatidylinositol 3-kinase

  • R-CHOP—rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone

  • R-CHP—rituximab, cyclophosphamide, doxorubicin, prednisone

  • R-EPOCH—rituximab, etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin

  • R-GemOx—rituximab, gemcitabine, oxaliplatin

  • R-ICE—rituximab, ifosfamide, carboplatin, etoposide

  • R-mini-CHOP—reduced dose R-CHOP

  • SIRP —signal regulatory protein α

Research questions

  • What is the optimal sequencing of currently approved treatments in patients with relapsed or refractory diffuse large B cell lymphoma (DLBCL) who are ineligible for autologous hematopoietic stem cell transplantation (AHCT)?

  • Will chimeric antigen receptor (CAR) T cells replace AHCT as standard of care treatment for patients with relapsed DLBCL?

  • Can CAR T cells lead to cure in indolent non-Hodgkin’s lymphoma (NHL) and mantle cell lymphoma (MCL)?

  • What will be the future therapeutic role of the bispecific antibodies in DLBCL, MCL, and follicular lymphoma?

    • Will the bispecific antibodies result in responses as durable as those seen with CAR T cells?

    • What is the optimal treatment duration with the bispecific antibodies?

  • A large number of clinically active small molecule inhibitors and immunotherapies are currently available in B-cell NHLs

    • How do we effectively investigate new drug combinations?

    • How should we balance efficacy, toxicity, quality of life, and cost?


  • Series explanation: State of the Art Reviews are commissioned on the basis of their relevance to academics and specialists in the US and internationally. For this reason they are written predominantly by US authors

  • Contributors: Both authors contributed to the planning, conduct, and reporting of the work in the article, and both are guarantors.

  • Competing interests: We have read and understood the BMJ policy on declaration of interests and declare the following interests: YS has received research funding from BMS, Celgene, TG Therapeutics, and Beigene and has consulted for TG Therapeutics and Epizyme outside of the submitted work; KM has received research funding from Pharmacyclics, Novartis, BMS, Celgene, and Merck and has consulted for Pharmacyclics, Janssen, Morphosys, Celgene, Beigene, Epizyme Gilead, Karyopharm, ADC Therapeutics, and Seattle Genetics outside of the submitted work.

  • Patient involvement: No patients were directly involved in the creation of this article.

  • Provenance and peer review: Commissioned; externally peer reviewed.


View Abstract