“Immunologic dream team”: BiTEs periallotransplant to redirect donor lymphocytes toward GVL Authors: Lorenzo Iovino, Mazyar Shadman, Filippo Milano, Ajay K. Gopal Published: 2026 ABSTRACT Visual Abstract Why do BiTEs fail? Why does allogeneic hematopoietic cell transplant (allo-HCT) fail? Despite allowing for significant response rates in heavily pretreated patients with B-cell non-Hodgkin lymphomas (B-NHL) and multiple myeloma (MM), bispecific T-cell engagers (BiTEs) eventually fail. There are several reasons to explain the lack of durable responses:1 BiTEs interact with the patients’ immune repertoire, and its composition enormously affects clinical outcomes.2,3 Heavily pretreated patients may have more exhausted T cells that cannot respond to further stimulation or become terminally exhausted and senescent after mediating a short response.4,5 Patients with high tumor burden and previously exposed to chemotherapy agents that are notoriously toxic to T cells, such as bendamustine, also tend to have a lower response rate and a shorter disease-free interval before relapsing.6 Allo-HCT can be curative in some patients, but posttransplant relapse remains the major cause of failure especially in patients who undergo transplant not in a complete remission status.7, 8, 9, 10 Some of the classic immunosuppressive approaches to prevent graft-versus-host disease (GVHD) have been shown to limit the capacity of donor-derived T cells to fully unleash their potential, although more recent approaches, such as posttransplant cyclophosphamide (PTCy), seem to prevent GVHD without affecting the graft-versus-tumor capacity.11 However, the composition of the immune system of the donor and other immunologic factors, including the type of donor and degree of HLA mismatch,12 play a role in preventing tumor relapse, as well as factors such as disease burden at the time of transplant and type of conditioning regimen.13 The graft-versus-lymphoma (GVL) and graft-versus-myeloma effects rely on donor-derived T-cell clones that are potentially able to recognize cancer antigens and mediate killing, but their presence in the graft is still hard to predict and, more notably, not actionable with any current available strategies. It is intriguing to think that a T-cell–redirecting strategy that binds more donor cells to residual cancer cells and mediates a more potent antitumor effect can be superior to BiTEs and allo-HCT as separate approaches. In this review, we investigate the potential role of this combination as an ideal approach for patients with aggressive B-NHL and MM otherwise untreatable. Clinical case A 56-year-old woman presented to the immunotherapy clinic of the Fred Hutchinson Cancer Center (Fred Hutch, Seattle, WA) in 2024 for relapsed/refractory mantle cell lymphoma (MCL), diagnosed in March 2023 after a 3-year history of autoimmune hemolytic anemia. At diagnosis, she had stage IVA disease (right temporal, epidural mass, extending outward from the right temporal bone and infiltrating the temporalis muscle; right parotid gland; spleen; perisplenic and retroperitoneal lymph nodes. Bone marrow biopsy positive for patchy involvement by lymphoma. MIPIb score was 7.5-7.7. TP53 mutation status at diagnosis was not checked. She had obtained only short responses after being treated with multiple lines of treatment, including R-CHOP (rituximab, cyclophosphamide, hydroxydaunorubicin, oncovin [vincristine], and prednisone), DHAOx (dexamethasone, high-dose ara-C, and oxaliplatin), R-BAC (rituximab, bendamustine, and ara-C), pirtobrutinib, and cytarabine and oxaliplatin. In February 2024, she arrived at Fred Hutchinson and underwent leukapheresis for CD19-targeted chimeric antigen receptor (CAR) T-cell manufacturing and then received pirtobrutinib and cytarabine as bridging therapy to control her rapidly growing disease. She received commercial brexucabtagene autoleucel CAR T cells in March 2024. Her post-CAR T-cell course was complicated by grade 1 cytokine release syndrome (CRS) and prolonged pancytopenia. Her day +28 positron emission tomography–computed tomography (PET-CT) scan still showed a metabolic partial response with multifocal sites of disease uptake, although a significant reduction was noted compared with the pre-CAR scans. Given the evidence of a partial response, she resumed pirtobrutinib. A day +100 PET-CT in June 2024 showed progression of disease, multifocal sites of fluorodeoxyglucose uptake involving pulmonary nodules, pleural effusion, osseous lesions, and right hilar adenopathy. She was then initiated on venetoclax. Despite that, she developed rapid disease progression and was admitted for inpatient rituximab, ifosfamide, carboplatin, and etoposide chemotherapy. She completed 3 cycles of rituximab, ifosfamide, carboplatin, and etoposide with excellent response (PET consistent with complete molecular response [CMR]). A lumbar puncture at the end of the 3 cycles in July 2024 was negative for evidence of an abnormal B-cell population. The patient underwent a matched-unrelated donor allo-HCT with fludarabine, melphalan, and 3 Gy total body irradiation conditioning regimen in October 2024, while on a complete response (CR). GVHD prophylaxis was with cyclosporine, sirolimus, and MMF.14 Her posttransplant course was relatively unremarkable, with no evidence of acute GVHD or other significant complications. Her day +28 bone marrow biopsy was negative for the involvement of MCL and showed no presence of TP53 mutation in single gene testing by NGS. Lymphoid and myeloid chimerisms were 100% donor. A PET-CT scan performed at day + 30 after transplant was consistent with CMR. Her day +84 PET-CT restaging showed 3 new hypermetabolic soft tissue nodules concerning for disease relapse. A bone marrow biopsy showed no evidence of MCL. A magnetic resonance imaging (MRI) of the cervical/thoracic/ lumbar spine showed multiple spinal enhancing lesions that were consistent with minute metastatic disease. All lesions were within the vertebral bodies, with no extranodal extension. An MRI of the brain showed multiple enhancing lesions throughout the calvarium and cervical spine. Her cerebrospinal fluid (CSF) was positive for MCL recurrence on flow cytometry. She proceeded with salvage chemotherapy with high-dose cytarabine for 4 doses and intrathecal methotrexate. A repeated brain MRI showed progressive disease with an expansile lesion in Meckel cave and a right foramen ovale up to 2.2 cm. She received brainstem radiation and had a remarkable localized improvement. In April 2025, she developed systemic progression with paraspinal masses at T7 to T8 and T11 to T12 vertebrae on PET-CT, but other sites had resolved. She completed local radiation therapy for her paraspinal masses and had significant improvement in her back pain. Our patient received her first dose of glofitamab at the end of May 2025. She has since received IV glofitamab every 3 weeks, starting from an initial ramp-up (2.5 mg on day 1 and 10 mg on day 8) up to 30 mg starting from cycle 2. She has had no evidence of CRS or immune effector cell-associated neurotoxicity syndrome thus far. A PET-CT scan obtained in August 2025 showed CMR, with a resolution of the left paraspinal mass uptakes at T7 to T8 and T11 to T12 (Figure 1). No other lymphadenopathies or suspicious lesions were detected. She remains in CMR, with no evidence of bone marrow involvement as of April 2026. The patient tolerated glofitamab without significant adverse events; she developed a diffuse skin rash after the first cycle, not felt to be GVHD on biopsy, and was more likely to be an immune activation. She is currently on a complete remission, without having experienced recurrence of central nervous system (CNS) symptoms for 7 months. To date, the patient is in a CR. The patient remains lymphopenic (absolute lymphocyte count of 0.47 per μL 2 years after allo-HCT and 10 months after initiating glofitamab), but with a slow trend toward recovery over time. Her CD4+ and CD8+ absolute lymphocyte counts remained unchanged between March 2025 and March 2026 (0.180 per μL to 0.137 per μL, and 0.136 per μL to 0.125 per μL, respectively). The clinical course of treatment is summarized in Figure 2.Figure 1.Pre- and post-glofitamab PET response. Comparison of PET-CT scans at the time of posttransplant relapse (A) and after 3 cycles of glofitamab (B) showing complete resolution of the paraspinal masses.Figure 2.Summary of the clinical course of treatment for the referenced patient. Acala + R, acalabrutinib+rituximab; AIHA, autoimmune hemolytic anemia; Brexu-cel, brexucabtagene autoleucel; Cy/VCR, cyclophosphamide/vincristine; Flu, fludarabine; HiDAC, high dose aracytine; IT MTX RT, intrathecal methotrexate radiation therapy; IVIG, IV immunoglobulins; mCR, metabolic complete remission; Mel, melphalan; MUD, matched-unrelated donor; Pirto, pirtobrutinib; PR, partial remission; TBI, total body irradiation. Pre- and post-glofitamab PET response. Comparison of PET-CT scans at the time of posttransplant relapse (A) and after 3 cycles of glofitamab (B) showing complete resolution of the paraspinal masses. Summary of the clinical course of treatment for the referenced patient. Acala + R, acalabrutinib+rituximab; AIHA, autoimmune hemolytic anemia; Brexu-cel, brexucabtagene autoleucel; Cy/VCR, cyclophosphamide/vincristine; Flu, fludarabine; HiDAC, high dose aracytine; IT MTX RT, intrathecal methotrexate radiation therapy; IVIG, IV immunoglobulins; mCR, metabolic complete remission; Mel, melphalan; MUD, matched-unrelated donor; Pirto, pirtobrutinib; PR, partial remission; TBI, total body irradiation. To the best of our knowledge, this is the first report of succesful use of glofitamab, a CD20/CD3 T-cell engager or bispecific antibody, in a patient with CNS/CSF relapse of MCL after allo-HCT. Approved BiTEs in B-NHL Currently, there are 3 Food and Drug Administration (FDA)-approved BiTEs for B-NHL. All of them engage CD20 on one side, CD3 on the other (CD20/CD3). Epcoritamab is approved for diffuse large B-cell lymphoma (LBCL) and follicular lymphoma (FL) after 2+ lines of therapy. Mosunetuzumab is approved for relapsed or refractory FL after 2+ lines of therapy. Glofitamab is approved for diffuse LBCL or LBCL arising from FL, after 2+ lines of therapy. Many other similar molecules are in the process of being tested and developed, but for the scope of this study, we will focus only on the 3 that are already approved by the FDA. The structural differences among the 3 available BiTEs in B-NHL are the following: epcoritamab and mosunetuzumab bind to 1 molecule of CD20 and 1 of CD3, whereas glofitamab binds to 2 CD20 molecules and 1 of CD3. Pharmacokinetics are different among the 3 drugs.15 BiTEs convert polyclonal T-cell responses into target-directed T-cell responses: antibody-dependent cellular cytotoxicity, the main mechanism of action of monoclonal antibodies in cancer, relies on the ability of Fc receptors present on the surface of nucleated cells. Fc receptors bind to the constant domain of immunoglobulins anchored to their targets and trigger the activation of natural killer cells and the release of perforins and granzymes. Through BiTEs, cytotoxic T cells are put in proximity to cancer cells and are induced to activate. The same mechanism promotes the activation of T helper cells, and through this engagement, T helper cells orchestrate the immune reaction. In this sense, BiTEs are superior to monoclonal antibodies.16 BiTEs have been demonstrated to be able to pass the blood-brain barrier, potentially contributing to the clearance of CNS and CSF disease.17,18 The most common side effect of BiTEs is CRS, a systemic complication that might require hospitalization and treatment with steroids and other anti-inflammatory agents such as tocilizumab, which happens in 40% to 80% of patients. Response rates with BiTEs in heavily pretreated patients are remarkable, especially compared with other available options: in their respective registration trials, epcoritamab had an overall response rate (ORR) of 63% and a CR of 38.0%,19 mosunetuzumab had an ORR of 80% and a CR of 60%,20 and glofitamab had an ORR of 66.7% and a CR of 31%.21,22 Highly promising results have been obtained with BiTEs approved for MM, including teclistamab,23 elranatamab,24 and talquetamab.25 ORR for BiTEs in MM vary from 60% to 70%, with CR/stringent complete response rates ∼30% to 39% and progression-free survival (PFS) of 11 to 17 months.26 Sequencing and combination of BiTEs with CD19-directed CAR T-cell treatments are a hot topic in the space of B-NHL: BiTEs are sometimes used as a bridge between leukapheresis and lymphodepletion in cases of very aggressive progression, especially in patients with chemorefractory disease, with the intention of leaving CD19 untouched before the exposure to commercial CAR T cells. For the same reason, considering that many cases of relapse after CAR is associated with CD19 antigen loss27, the use of CD20/CD3 BiTEs is appealing in this dramatic scenario, which is otherwise associated with extremely dismal outcomes.27,28 Some evidence suggests that BiTEs can even engage CD3 on the surface of residual CAR T cells and redirect them against CD19 and CD20 together.29,30 In post-CD19 CAR relapses, glofitamab has an ORR of 66.7% and a CR rate of 31%. The estimated PFS at 1 and 2 years were 36.6% and 24.5%, respectively. The median duration of response (DOR) was 19.7 months.21 Epcoritamab has an ORR of 68.8% and a CR of 41.7%,31 with a median DOR of 12 months. Mosunetuzumab has ORR and CR rates of 23% and 12%, respectively. Median PFS is 1.4 months (95% confidence interval, 1.3-5.3).32 Despite the good ORR and CR rates, median PFS in B-NHL varies from 3.8 to 31.1 months, and the DOR ranges from 9.7% to 25% at 2 years.33 In MM, data from the registration trials show that 50% to 65% of patients progressed or died by 2 years.34 This means that most patients with aggressive B-NHL either still do not respond to BiTEs or relapse within months, whereas longer DOR have been reported, thus far, only in trials focused on patients with FL. From these studies, it was also demonstrated that achieving a minimal residual disease (MRD) status was associated with longer PFS.35, 36, 37 Allogeneic transplant in B-NHL Allo-HCT remains a relevant but increasingly narrow option in B-NHL. The limitations to a broader application come from transplant candidacy (mostly due to comorbidities, performance status, and availability of an allogeneic donor), elevated risk of nonrelapse mortality (NRM) compared with alternative approaches, and baseline control of the underlying cancer. In B-NHL, a large multicenter study reported 5-year PFS and OS of 41% and 49%, with one-third of patients in continuous CR at up to 22 years (a durability unmatched by nontransplant salvage), although NRM reached 26% at 12 months.9 In other B-cell–derived malignancies, such as MM, the largest long-term analysis from a nonmyeloablative tandem auto-allo program (200 cGy total body irradiation ± fludarabine) shows 5-year overall survival (OS) of 54% and PFS of 31%. The 10-year OS was 41% and PFS was 19%, with low NRM (2% at day 100 and 14% at 5 years). Although patients with standard- and high-risk myeloma still had an OS >40% at 5 years, the ultrahigh-risk patients derived minimal benefit, mostly due to early relapses.7 Although more intensive than other treatment options and not suitable for everyone in light of the high risk, allo-HCT can still be curative in relapses after CD19 CAR T-cell treatment. The published 1-year OS ranges from 50% to 100%.38,39 Several separate reports clearly show that undergoing allo-HCT on a deeper disease control, possibly in a CR without evidence of MRD, is associated with improved posttransplant outcomes.10,40 The abovementioned ORR data about BiTEs in lymphoma suggest that their pretransplant use might increase the proportion of patients undergoing allo-HCT with better disease control. Although there has been progress in preventing severe and extensive GVHD, especially the chronic GVHD (cGVHD) form, by selecting the best donors and using PTCy, relapse after allo-HCT remains a major problem, and little progress has been made in increasing the GVL effect.24 We hypothesize that (a) allo-HCT, if used after BiTEs, may overcome failure of BiTEs alone, if this is due to T-cell dysfunction or depletion. The use of pretransplant BiTEs might also increase the number of patients in a CR at the time of allo-HCT. (b) If given after transplant, BiTEs might result in targeted GVL/graft-versus-myeloma. Although the risk of CRS from BiTEs in an allogeneic setting is not known, experiences from the use of BiTEs in other hematological malignancies point toward a very good safety profile. Blinatumomab, a CD19/CD3 BiTE, is indicated for B-cell acute lymphoblastic leukemia (B-ALL), including relapsed/refractory or MRD-positive disease.41 Given that B-ALL has a more frequent indication for allo-HCT, most of the available literature about the use of BiTEs around allo-HCT is on blinatumomab. The use of blinatumomab peri-HCT is well documented: patients with MRD receive blinatumomab until the start of conditioning, noting that its short ∼2-hour half-life means that the BiTE is not expected to be present at the time of allograft infusion. The advantage in terms of OS, PFS, GVHD-free and relapse-free survival (GRFS), and GVHD rates of treating B-ALL with blinatumomab pre-HCT are well documented in the reports from several clinical trials and real-world studies.42,43 It is important to note that blinatumomab, differently from CAR T-cell treatments available in B-NHL, targets the same antigen as the approved CAR T cells approved for B-ALL (tisagenlecleucel and brexucabtagene autoleucel). These 2 approaches are often observed as mutually exclusive and not very useful in sequence. This sometimes limits the application of blinatumomab in cases of post–CAR T-cell relapses with documented CD19 loss, and the same happens when offering CAR in patients who were previously treated with blinatumomab, although CD19-reduced expression is sometimes transient and there is some evidence of the effectiveness of CD19-targeted immunotherapies even in the absence of flow cytometry–documented CD19 negativity.44 Blinatumomab has been proven to be active also in post–allo-HCT relapses. Studies about planned maintenance showed promising results in terms of relapse risk, without increasing the risk of NRM and GVHD, and with an incidence of CRS in the range of 5% to 10%.45, 46, 47 The interesting finding is that antibody-dependent cellular cytotoxicity and tumor killing capacity seem to be preserved, whereas no increased risk of GVHD, a major concern when offering T-cell–directed agents in the context of allo-HCT, was reported.48 Blinatumomab has been safely given even before or after the administration of donor lymphocyte infusion, a procedure in which T cells from the donor are infused after allo-HCT to reestablish lymphoid chimerism or treat relapse and that is commonly associated with flares of GVHD.48,49 Considering that the polyclonal T-cell responses are converted into anti–B-cell responses by blinatumomab, this could reduce the risk that donor T cells with alloreactive potential could potentially target normal antigen on host tissues. Furthermore, B cells are also involved in the pathogenesis of cGVHD, to the point that CD20-directed therapies have been used to treat steroid-refractory cGVHD. Of course, toxicities are possible and have been reported, especially in a population of heavily pretreated patients. Given the much more recent experience with BiTEs in B-NHL, not much data exist for the use of CD20/CD3 engagers in periallotransplant. In the 2-year follow-up analysis of the epcoritamab registration trial, 9 of the initial 157 patients subsequently received an allo-HCT.31 The largest series on this topic has just been published by the GETH-TC Spanish group and compared 47 patients who underwent allo-HCT after BiTEs to a historical, propensity-matched scored cohort of patients. Patients received epcoritamab, mosunetuzumab, or glofitamab as monotherapy or in combination with other agents. A total of 27.7% of patients received odronextamab,50 another CD20/CD3 BiTE that is still not FDA-approved in the United States. The 2-year NRM did not differ significantly between BiTE-exposed and BiTE-naive groups. OS, incidence of GVHD, and GRFS were also comparable. In a subanalysis limited to LBCL, the cumulative incidence of relapse was inferior in the BiTE-exposed cohort.10 An interesting preclinical study shows how mosunetuzumab can interact with a population of unconventional immune cells, the γ-δ T cells, involved in anticancer responses but not mediators of GVHD.51 The abundance of this population in the graft interestingly correlates with improved posttransplant outcomes.52 Given the drastic reduction in the number of allo-HCT recently offered worldwide for MM, there are no published experiences of the use of B‑cell maturation antigen and G protein–coupled receptor class C group 5 member D BiTEs around transplant. Besides the anecdotal value of our report, all data here summarized suggest that it is time to integrate all the immunotherapy approaches available in B-NHL into clinical trials. CAR T cells, BiTEs, and allo-HCT can lead to improved long-term outcomes if sequenced in a rational way. Conclusion The reasons to think of BiTEs as a valuable addition to the peri–allo-HCT toolbox in future prospective clinical trials are multiple. First, post-CAR T-cell relapses are usually extremely aggressive, and although the ORR for BiTEs are promising, the DOR is short and the long-term PFS is still dismal. No BiTEs have shown curative potential, thus far. The use of BiTEs as a bridge could be ideal if used in a timely manner to obtain disease control and take patients to an allo-HCT while in a remission state; in addition, they are shown to be able to “revive” CD19 CAR T cells that are still circulating for patients who have residual disease or relapse after CAR T-cell treatment. Second, data from the use of blinatumomab and early data from CD20/CD3 engagers show that BiTEs around allo-HCT are safe and without an increased risk of NRM, whereas there is a signal for reduced risk of posttransplant relapse (noting that the 1- to 3-week half-life of these BiTEs may result in their persistence at the time of allograft infusion). However, patients with active B-NHL and an indication to allo-HCT may be quite debilitated, and the true risk of increased NRM in such an experimental context cannot be determined in advance. Therefore, the eligibility of candidates to such a trial should be carefully assessed according to the performance status and the predicted risk of NRM at baseline. Finally, the posttransplant blinatumomab experience has shown no signs of increased risk of GVHD and other immune complications, especially in patients previously exposed to the same agent. The mechanism of action of BiTEs can increase the GVL effect and convert/divert polyclonal T-cell responses into antitumor responses, potentially reducing the risk of GVHD. In the era of PTCy, planned posttransplant maintenance/treatment after engraftment could be translated into better GRFS. In prospective trials, a planned, posttransplant intervention with BiTEs should be ideally started as soon as the acute complications of conditioning resolve, once hematopoietic engraftment has occurred, without waiting for signs of relapse or progression. The effect of posttransplant BiTEs in B-NHL (and, potentially, in MM) should also be weighed against the use of different GVHD prophylaxis strategies, including synergistic and cumulative toxicity effects with PTCy. Of course, the actual rate of adverse events and the effects on lymphoid chimerism should also carefully be monitored, as well as the rate of infections and the effects on immune reconstitution. B-cell aplasia will for sure be affected by such a treatment, and the skewing of polyclonal T-cell responses might open the gates for more opportunistic infections. However, by keeping in mind that disease relapse or progression still represent the number one cause of deaths in these patients, we believe that the incorporation of BiTEs in the periallotransplant space may truly represent an interesting space that can help rescue a significant proportion of patients with aggressive B-cell lymphomas and myelomas. We look forward to this hypothesis being tested in prospective trials that may allow more patients with B-cell malignancies to benefit from the potentially ideal combination of BiTE and allogeneic transplantation. Conflict-of-interest disclosure: L.I. reports stock options in MustangBio; and research funding from PharmaNutra S.p.A. M.S. reports consulting or serving on advisory boards, steering committees, or data safety monitoring committees for AbbVie, Genentech, AstraZeneca, Genmab, Janssen, BeiGene/BeOne Medicines, Bristol Myers Squibb (BMS), MorphoSys/Incyte, Kite Pharma, Eli Lilly, Fate Therapeutics, Nurix, Merck, Pfizer, and Pierre Fabre; research funding from MustangBio, Genentech, AbbVie, BeiGene/BeOne Medicines, AstraZeneca, Genmab, MorphoSys/Incyte, and Vincerx; stock options in Koi Biotherapeutics; and employment (spouse) with BMS. A.K.G. reports research funding from Merck, Takeda, AstraZeneca, Agios, Janssen, BMS, Seagen, Teva, Genmab, BeiGene, Pfizer, and Umoja; consultancy/honoraria from Incyte, Kite, MorphoSys/Incyte, ADC Therapeutics, Acrotech, Merck, Karyopharm, Servier, BeiGene, Cellectar, Janssen, Compliment, Seagen, Epizyme, I-Mab bio, Gilead, Genentech, Lilly, Caribou, Fresenius Kabi, SciTech, and Sana; and stock options in Compliment Corporation and SciTech. F.M. declares no competing financial interests.