Post-Transplant Bendamustine as a Platform for Immune Modulation After Allogeneic Hematopoietic Cell Transplantation.

Methods/Scope

This is a narrative review synthesizing preclinical and clinical evidence on post‐transplant bendamustine (PT‐BEN) as an immunomodulatory strategy following allogeneic HCT. A PubMed literature search was conducted using the terms “bendamustine,” “post‐transplant bendamustine,” “haploidentical transplantation,” “GvHD prophylaxis,” “immune modulation,” and related combinations, without date restriction. Given that PT‐BEN is an emerging strategy with a nascent literature, no eligibility criteria or formal weighting framework was applied; the clinical studies summarized represent, to the best of the authors' knowledge, the totality of published and publicly reported PT‐BEN clinical experience. The preclinical literature reflects work predominantly from the originating group, as independent murine mechanistic studies of PT‐BEN specifically remain limited; this selection is acknowledged as a limitation inherent to the current state of the field rather than a bias in evidence synthesis. Evidence is presented as a hypothesis‐driven narrative synthesis and does not constitute a systematic review.

Bendamustine Pharmacology

Bendamustine (BEN) is a structurally distinctive alkylating agent comprising a nitrogen mustard alkylating group, a benzimidazole ring conferring purine analog properties, and a butyric acid side chain [ 1 ]. Like classical alkylating agents, BEN induces inter‐ and intra‐strand DNA breaks with comparatively slow repair kinetics, activates p53‐dependent apoptosis, and can trigger mitotic catastrophe in cells that evade conventional apoptotic pathways [ 1 , 2 , 3 ]. Its benzimidazole‐derived purine analog activity is less fully characterized but is thought to contribute to activity in cells resistant to conventional alkylators, potentially through alternative mechanisms of DNA damage recognition and repair inhibition. Critically for its immunomodulatory context, BEN demonstrates differential cytotoxicity across immune cell populations: it potently depletes lymphoid populations while demonstrating relative sparing of myeloid cells [ 4 , 5 ]. This selectivity, distinct from the broader lympho‐myeloid depletion characteristic of cyclophosphamide (CY) [ 6 ], provides the pharmacologic basis for BEN's capacity to reshape the post‐transplant immune milieu without ablating myeloid recovery, a property central to the mechanistic rationale for its investigation in the post‐transplant setting.

Foundational Preclinical Observations

Uncoupling GvL from GvHD in Haploidentical Models

The conceptual foundation for PT‐BEN emerged from a deliberate effort to interrogate graft‐versus‐host disease (GvHD) biology in a preclinical system that more closely approximated the clinical haploidentical transplantation setting than traditional parent‐to‐F1 models. In the original studies, a fully haploidentical F1 → F1 murine bone marrow transplantation platform was employed, incorporating T‐cell‐replete grafts and myeloablative or reduced intensity conditioning [ 7 ]. Within this stringent framework, a single dose of BEN administered post‐transplant demonstrated a striking capacity to reduce GvHD severity while preserving donor engraftment and graft‐versus‐leukemia (GvL) activity. Several features of this initial report are worth emphasizing, as they presaged many of the later clinical and mechanistic observations.

First, PT‐BEN and post‐transplant cyclophosphamide (PT‐CY) produced comparable control of early acute GvHD, indicating that BEN was not inferior to CY in suppressing early alloreactive responses (Figure 1 ). However, divergence emerged over time: animals receiving PT‐BEN showed progressive improvement in clinical GvHD scores, reduced late GvHD mortality, and improved overall survival compared with PT‐CY‐treated controls [ 7 ]. This temporal dissociation suggested that PT‐BEN was influencing not only early T‐cell proliferation but also the evolving immune milieu that governs chronic GvHD trajectories. Second, PT‐BEN preserved, and in some experimental contexts enhanced, GvL effects. In leukemia‐bearing models, PT‐BEN‐treated animals demonstrated superior leukemia control compared with PT‐CY, despite reduced GvHD. This finding directly challenged the long‐standing assumption that GvHD suppression and GvL preservation are necessarily coupled and provided an early signal that BEN might modulate alloimmunity through pathways distinct from simple cytotoxic elimination of proliferating donor T cells. Third, mechanistic interrogation in this model revealed that GvHD protection conferred by PT‐BEN was not dependent on donor regulatory T cells (Tregs). Depletion and phenotypic analyses showed that PT‐BEN did not rely on expansion or preferential survival of Foxp3 + Tregs to achieve GvHD control. Instead, PT‐BEN was associated with relative sparing of hematopoietic progenitors, earlier recovery of innate immune populations, and alterations in myeloid compartments that were consistent with an immunomodulatory, rather than purely lymphodepletive, effect [ 7 ].

Collectively, these observations established four principles that would guide subsequent work: (i) BEN given post‐transplantation preserves donor cell engraftment (ii) BEN can substitute for CY in the post‐transplant setting without loss of early GvHD control; (iii) PT‐BEN can uncouple GvHD from GvL in a stringent haploidentical system; and (iv) the dominant biology of PT‐BEN is unlikely to be explained solely by canonical Treg‐centric models developed for PT‐CY. These principles motivated a broader investigation of BEN's immunologic effects in both conditioning and post‐transplant contexts.

Immunomodulatory Insights from Conditioning Studies

The following studies constitute mechanistic antecedent evidence, biologically informative for understanding PT‐BEN biology, but indirect with respect to the post‐transplant immune modulation question that is the primary focus of this review.

Subsequent to the initial haploidentical PT‐BEN studies, preclinical work examined whether BEN exerted broader immunomodulatory effects when incorporated into the conditioning regimen itself. In a series of murine transplantation experiments directly comparing BEN with CY as a conditioning agent in combination with TBI, BEN‐TBI conditioning reproducibly limited GvHD severity while preserving donor engraftment and antileukemic activity, effects observed despite equivalent myeloablation and donor chimerism, arguing against inadequate conditioning as a confounder [ 8 ]. Mechanistic interrogation identified myeloid‐derived suppressor cells (MDSCs) as central mediators: BEN‐TBI conditioning expanded CD11b + Gr‐1hi myeloid populations with potent suppressive activity, and their depletion abrogated GvHD protection (Figure 1 ) [ 8 ]. This was particularly notable given that prevailing models of GvHD control had emphasized regulatory T cells rather than innate myeloid circuits as the dominant suppressive population. Complementary studies demonstrated that donor T cells recovered from BEN‐conditioned recipients exhibited durable functional tolerance, reduced alloreactivity with preserved leukemia clearance, without overt T‐cell deletion, supporting a model of immune reprogramming rather than cytotoxic depletion [ 9 ]. Parallel work revealed that BEN conditioning also altered dendritic cell (DC) composition, skewing host DC populations toward pre‐cDC1‐like subsets and generating regulatory DC programs characterized by altered Flt3 signaling and impaired allostimulatory capacity, independently of classical Batf3‐dependent cross‐presenting pathways [ 10 , 11 ].

Collectively, these findings suggest that BEN acts at multiple levels of the immune hierarchy, MDSCs, DCs, and downstream T‐cell differentiation, to establish a tolerogenic post‐transplant environment, and were synthesized in a dedicated review framing BEN as an agent capable of coordinated immunomodulatory effects across myeloid, dendritic, and lymphoid compartments [ 12 ]. Whether these cellular changes represent coordinated components of a unified immunologic pathway or additive independent effects of BEN has not been formally established, and resolving their causal relationships will ultimately require interrogation in human post‐transplant immune samples. While this body of conditioning evidence does not directly demonstrate the effects of BEN administered after graft infusion, it provides a coherent and testable biological rationale for PT‐BEN's clinical translation, informing the mechanistic hypotheses and dosing considerations that guided subsequent clinical work.

Beyond its post‐transplant application, BEN has been investigated as a conditioning agent across multiple transplant platforms, providing independent clinical evidence of its immunomodulatory properties and favorable hematopoietic recovery profile. In the autologous setting, BEN‐containing conditioning, most notably the BeEAM regimen (bendamustine, etoposide, cytarabine, melphalan), demonstrated an 81% complete response rate at 18 months with limited toxicity in patients with relapsed or refractory Hodgkin and non‐Hodgkin lymphoma, and offered more favorable outcomes than CBV (cyclophosphamide, carmustine, and etoposide) conditioning in mantle cell lymphoma [ 13 , 14 ]. In the allogeneic setting, BEN combined with fludarabine and rituximab as conditioning for chronic lymphocytic leukemia produced favorable neutrophil and platelet recovery with a low incidence of both acute and chronic GvHD and notably reduced myelosuppression relative to historical comparators [ 15 ]. BEN combined with fludarabine and busulfan prior to haploidentical allogeneic HCT achieved 98% engraftment with acceptable GvHD rates, though complications including cytokine release syndrome (CRS) were observed [ 16 ]. Taken together, these independent series consistently demonstrate that BEN‐containing conditioning regimens are associated with preserved hematopoietic recovery and manageable GvHD, findings that are mechanistically consonant with the myeloid‐sparing and immunomodulatory properties identified in preclinical PT‐BEN studies, though the confounding contribution of concurrently administered agents, must be acknowledged in their interpretation.

BEN has also gained recognition as a lymphodepletion agent prior to chimeric antigen receptor T‐cell (CAR‐T) therapy, an application that gained clinical momentum during a fludarabine shortage in the United States around 2020 [ 17 ]. Across multiple retrospective analyses, BEN‐based lymphodepletion demonstrated favorable CAR‐T persistence, reduced cytokine release syndrome, improved hematologic recovery, including higher nadir neutrophil and platelet counts and faster red blood cell recovery, and reduced immune effector cell‐associated neurotoxicity syndrome compared with standard cyclophosphamide/fludarabine lymphodepletion [ 17 , 18 , 19 ]. Similar findings were observed in multiple myeloma patients receiving BCMA‐directed CAR‐T therapy following BEN lymphodepletion [ 20 ]. These clinical observations in a distinct therapeutic context provide independent corroboration that BEN reshapes the post‐treatment immune environment with preferential myeloid sparing, a property central to the mechanistic rationale for PT‐BEN and consistent with the preclinical conditioning and post‐transplant data described in this review.

Validation in Humanized Xenogeneic Models

While murine transplantation models have been indispensable to elucidating GvHD pathogenesis and provided the foundational platform for PT‐BEN investigation, their translational limitations warrant explicit acknowledgment. These models are constructed around murine H‐2 histocompatibility antigens using inbred strains homozygous at each allele, enabling controlled gradation of alloreactive pressure from complete MHC mismatch, which produces the most severe GvHD, through haploidentical partial mismatch, to MHC‐matched minor antigen mismatch, each offering distinct experimental advantages [ 21 , 22 ]. The strengths of these systems include the availability of genetically defined mutant strains, intact immune organs, long‐lasting engraftment, and a complete cytokine environment amenable to mechanistic interrogation.

However, several limitations are relevant to the translational interpretation of PT‐BEN findings. Murine models depend on high‐dose irradiation for conditioning, employ splenocyte‐derived T cells rather than a clinical graft product, and are conducted under specific pathogen‐free housing conditions that restrict antigen exposure and substantially alter gut microbiome composition relative to the human clinical setting, a critical variable given the microbiome's established influence on GvHD severity [ 23 ]. Considerable inter‐laboratory variability further complicates cross‐study comparison, arising from differences in mouse vendor, environmental pathogen exposure, irradiation type and dose, and animal age [ 22 ]. Complete MHC‐mismatched models, while experimentally tractable, do not recapitulate clinical GvHD, and even haploidentical systems tend to exaggerate GvHD severity and mortality relative to what is observed in patients. The historical record, in which GvHD interventions successful in murine systems have not consistently translated to clinical benefit, underscores the importance of interpreting these findings with appropriate caution [ 23 ].

Humanized mouse models employing severely immunodeficient hosts, such as the NSG (NOD‐scid IL2Rγ null ) strain, address several of these limitations by enabling the study of GvHD induced by human immune cells in a xenogeneic system [ 24 ]. Unlike conventional mouse‐to‐mouse transplants, humanized models incorporate donor human peripheral blood mononuclear cells that reflect the genetic diversity, age, and antigen experience of real clinical graft products, features absent from inbred murine systems. These models are relatively resource‐intensive, but they provide a biologically and clinically relevant intermediate platform for evaluating the response of human immune populations to interventions such as PT‐BEN before clinical translation. An important consideration in their use is donor variability: because human peripheral blood mononuclear cells (PBMCs) grafts differ meaningfully across individuals, therapeutic effects must be evaluated across multiple donors or donor‐specific sources of differential response must be identified and accounted for [ 24 ]. Taken together, the progressive incorporation of humanized xenogeneic models as an intermediate translational step in PT‐BEN development was motivated directly by the limitations inherent in murine systems, and represents a deliberate strategy to evaluate PT‐BEN's immunomodulatory properties in a setting that more closely approximates human post‐transplant immune biology [ 25 ]. This platform recapitulates key features of human GvHD, including progressive weight loss, multiorgan infiltration by human T cells, and alloimmune‐driven lethality. In this model, post‐PBMC CY alone, BEN alone, or combined CY/BEN regimens were administered on days +3 and + 4 and each reduced xenogeneic GvHD (xGvHD) severity compared with untreated controls, with broadly comparable survival benefits [ 25 ]. Collectively, these results support the intrinsic capacity of BEN to attenuate aggressive human T‐cell–mediated pathology when delivered early after graft infusion.

The most informative and translationally relevant finding, however, emerged when calcineurin inhibition was introduced. Cyclosporine A (CSA) monotherapy improved survival compared with no treatment, consistent with its established role in GvHD prophylaxis. Strikingly, the combination of CY/CY with CSA failed to confer additive benefit and, in some analyses, performed worse than CSA alone, with evidence of worsened clinical scores and intestinal injury [ 25 ]. In contrast, CY/BEN combined with CSA produced the most favorable outcomes, significantly improving survival and reducing moderate‐to‐severe xGvHD compared with both no treatment and CY/CY + CSA (Figure 1 ). Mechanistic correlates supported these clinical observations. CY/CY + CSA was associated with markers of early gut injury, including reduced serum albumin and increased intestinal permeability, consistent with severe GvHD biology rather than nonspecific toxicity. CY/BEN + CSA, by contrast, was associated with reduced human leukocyte infiltration of target organs and lower histopathologic scores, suggesting that BEN altered the interaction between calcineurin inhibition and post‐graft immune activation [ 25 ]. These data carry two important implications. First, they demonstrate that BEN's immunomodulatory effects are preserved in a humanized immune context, strengthening the translational relevance of earlier murine findings. Second, and perhaps more importantly, they establish that the identity of the post‐transplant alkylator matters when layered onto standard GvHD prophylaxis. CY and BEN are not interchangeable placeholders within a generic “post‐transplant alkylation” paradigm; rather, they interact differently with calcineurin inhibition to shape downstream immune injury.

Reconciling Divergent Preclinical Data

Not all preclinical data have supported BEN‐based strategies. Hadjis and colleagues reported that PT‐BEN failed to prevent lethal GvHD in their murine haploidentical model, a finding that warrants serious consideration as a genuine negative result in a defined experimental system [ 26 ]. In the same study, PT‐CY uniquely restrained alloreactive CD4 + T‐cell proliferation and differentiation, leading the authors to conclude that CY's mechanism of action is not replicated by BEN under those conditions.

Several methodological features of the Hadjis model differ from systems in which PT‐BEN has demonstrated efficacy and likely account for the discordant outcome [ 7 ]. These include the use of single‐fraction higher‐dose total body irradiation, a substantially larger splenocyte inoculum generating extreme alloreactive pressure, fluoroquinolone prophylaxis altering the gut microbiome, and, critically, intraperitoneal rather than intravenous BEN administration. As a known vesicant, intraperitoneally administered BEN introduces local toxicity, altered absorption kinetics, and peritoneal inflammation not encountered with clinical intravenous delivery; the ascites and hepatic pathology observed in BEN‐treated animals in that study are consistent with route‐related confounding rather than failure of immunomodulation per se. Notably, BEN did suppress early proliferation of activated CD25 + alloreactive CD4 + T cells at day +7, confirming on‐target immunologic activity, though this was insufficient to control GvHD under the study's stringent conditions [ 26 ].

The Hadjis findings are best understood as delineating the conditions under which PT‐BEN is and is not effective, rather than invalidating the concept. Discordant preclinical results of this kind most plausibly reflect differences in conditioning intensity, alloreactive burden, microbiome context, and drug delivery route rather than irreconcilable biology [ 27 ].

Clinical Translation

From Concept to Comparative Analysis

The landscape of GvHD prophylaxis has evolved substantially, and PT‐BEN must be contextualized within this broader field. PT‐CY has been transformative, with the BMT CTN 1703 phase III trial establishing PT‐CY/tacrolimus/MMF as a benchmark in matched‐donor reduced‐intensity transplantation, and the 2025 CAST trial extending these findings to matched related donors with myeloablative conditioning [ 28 , 29 ]. Abatacept became an FDA‐approved agent for acute GvHD prophylaxis in 2021, with the ABA2 trial demonstrating a dramatic reduction in grade 3–4 acute GvHD in mismatched unrelated donor transplants [ 30 , 31 ], and a novel PT‐CY/abatacept combination showing particularly promising early results [ 32 ]. Ruxolitinib is FDA‐approved for steroid‐refractory acute and chronic GvHD and is under active investigation as prophylaxis, with early‐phase data suggesting low rates of moderate–severe chronic GvHD [ 33 , 34 ]. Additional emerging strategies include next‐generation selective JAK inhibitors, regulatory T‐cell adoptive therapy, and newer approved agents for chronic GvHD including belumosudil and axatilimab [ 35 ].

In haploidentical hematopoietic cell transplantation (haplo‐HCT), post‐transplantation PT‐CY remains the standard of care for GvHD prophylaxis [ 36 , 37 , 38 , 39 ]. However, persistent questions remain including concerns about the strength of GvL in this platform [ 40 , 41 ]. Infections are a significant concern as PT‐CY causes delays in CD4 + T‐cell reconstitution resulting in higher overall infection rates including CMV reactivation (42% by day 180 in a CIBMTR analysis), BK virus, HHV‐6, and other non‐CMV herpesviruses, particularly in the haploidentical setting, where approximately 50% of all deaths are infection‐related [ 42 , 43 , 44 , 45 ]. Cardiotoxicity is a notable adverse effect; a large retrospective study found PT‐CY independently associated with early cardiac events, predominantly heart failure, especially in patients with pre‐existing cardiovascular risk factors [ 46 , 47 , 48 ]. Hemorrhagic cystitis remains a well‐recognized complication requiring aggressive hydration and mesna prophylaxis, with no consensus on the optimal prevention strategy [ 49 , 50 ]. While dose‐reduction strategies are being explored to mitigate these toxicities [ 51 , 52 , 53 , 54 , 55 , 56 ], an alternative approach has been investigated: partial replacement of PT‐CY with PT‐BEN within an established myeloablative T‐cell‐replete haploidentical platform, guided by preclinical evidence suggesting comparable GvHD control and distinct immunomodulatory mechanisms [ 57 , 58 , 59 ].

This first‐in‐human phase I study ( NCT02996773 ) at the University of Arizona evaluated the progressive substitution of PT‐CY starting on day +4, with escalation of PT‐BEN. The regimen proved safe, with no primary graft failures and no dose‐limiting toxicities after the first 3 cohorts with replacement of day +4 PT‐CY (50 mg/kg) with PT‐BEN (90 mg/m 2 ) [ 57 , 58 ]. Completion of the dose‐escalation phase enabled formal comparison between the PT‐CY/BEN cohorts and contemporaneous PT‐CY–only controls [ 59 ]. However, these comparisons are non‐randomized, subject to selection bias and institutional practice differences, and findings should be interpreted as preliminary and hypothesis‐generating. With this important caveat in mind, three signals emerged. First, the PT‐BEN group demonstrated significantly accelerated hematopoietic recovery, including earlier myeloid and platelet engraftment, accompanied by reduced red blood cell and platelet transfusion requirements (Table 1 ). Second, while acute GvHD rates were comparable between groups, PT‐CY/BEN recipients demonstrated a reduction in chronic GvHD, resulting in significant improvement in GvHD‐free relapse‐free survival (GRFS) without apparent compromise of antileukemic efficacy, observations that require prospective randomized validation before conclusions can be drawn.

A confound inherent to the PT‐BEN partial‐substitution platform must be acknowledged when evaluating the clinical data. In the partial‐substitution platform, standard two‐dose PT‐CY (50 mg/kg on days +3 and + 4) is replaced by a single dose of PT‐CY (50 mg/kg, day +3) followed by PT‐BEN (90 mg/m 2 , day +4), effectively halving total cyclophosphamide exposure. Recent literature demonstrates that reduced‐dose PT‐CY alone accelerates engraftment and reduces toxicity including hemorrhagic cystitis and cardiotoxicity without compromising GvHD control [ 51 , 52 , 53 , 54 , 55 , 56 ]. This raises the possibility that some apparent benefits attributed to PT‐BEN reflect CY dose reduction rather than BEN‐specific immunomodulation. This confound cannot be resolved by non‐randomized single‐center comparison and applies retroactively to all currently available comparative PT‐BEN data. It represents a primary scientific rationale for the three‐arm randomized trial design proposed in the future directions section, which includes a reduced‐dose PT‐CY alone arm specifically to isolate BEN's independent contribution.

Further reduction of PT‐CY on day +3 in the Arizona phase I study (Day +3: CY 40 mg/kg and BEN 20 mg/m 2 ; Day +4: BEN 90 mg/m 2 alone) identified dose‐limiting phenomena: cytokine release syndrome (CRS) in a patient and secondary graft failure associated with CMV reactivation in another patient [ 59 ]. These events, while manageable, defined the upper therapeutic window and supported the rationale for partial (Day +4) rather than complete PT‐CY replacement.

Immune Reconstitution and TCR Repertoire Diversity Following PT‐CY/BEN

Clinical signals of reduced chronic GvHD and improved GRFS following PT‐CY/BEN substitution prompted evaluation of immune reconstitution as a potential mechanistic driver. This was addressed through integrated longitudinal immunophenotyping and high‐throughput T‐cell receptor (TCR) repertoire sequencing in patients enrolled on the phase Ia/Ib PT‐CY/BEN trial, providing a detailed human mechanistic assessment [ 63 ]. PT‐CY/BEN recipients demonstrated earlier, clinically meaningful CD4 + T‐cell recovery, with a significantly greater proportion achieving CD4 + > 50 cells/mL by days +60 and +100 compared with PT‐CY controls [ 63 ]. This reflected selective normalization of helper T‐cell reconstitution rather than nonspecific lymphocyte expansion. In parallel, PT‐CY/BEN was associated with restraint of CD8 + effector and effector‐memory expansion, resulting in sustained elevation of the CD4:CD8 ratio during the first post‐transplant year, a profile linked to reduced chronic GvHD risk.

High‐resolution TCRβ sequencing further distinguished PT‐CY/BEN from PT‐CY alone. PT‐CY/BEN recipients maintained greater TCR repertoire diversity with lower clonality, whereas PT‐CY recipients showed marked repertoire contraction driven in part by CMV‐associated expansions [ 63 ]. Preservation of repertoire diversity supports immune competence and has been implicated in protection from chronic GvHD.

Consistent with preclinical findings, trends toward enhanced suppressive myeloid populations were observed following PT‐CY/BEN, without preferential expansion of regulatory T cells, reinforcing a non‐Treg‐centric mechanism of immune modulation [ 63 ]. CMV reactivation rates were comparable between platforms after adjustment for risk, while reduced BK virus–associated hemorrhagic cystitis was observed in PT‐CY/BEN recipients [ 58 , 59 , 64 ]. Collectively, PT‐CY/BEN establishes a distinct immune trajectory characterized by earlier CD4 + recovery, restrained CD8 + effector skewing, and preserved TCR diversity, closely aligning with reduced chronic GvHD and improved GRFS and providing strong mechanistic support for further clinical evaluation.

PT‐CY immune reconstitution profile has been well‐characterized with profoundly delayed CD4 + T‐cell recovery, with absolute counts significantly lower than conventional prophylaxis throughout the first year post‐HCT, driven by depletion of rapidly proliferating alloreactive T cells and skewing toward effector memory phenotypes [ 44 ]. Beyond numerical delay, PT‐CY induces a functionally exhausted T‐cell phenotype characterized by upregulation of multiple inhibitory receptors, TIGIT, PD‐1, TIM‐3, CD38, and CD39, on both CD4 + and CD8 + T cells and reduced cytokine production upon stimulation [ 65 ]. NK‐cell recovery is similarly impaired, with predominance of functionally immature CD56 bright CD16 − subsets associated with CMV reactivation and chronic GvHD risk [ 66 ]. Importantly, emerging data demonstrate that intermediate‐dose PT‐CY alone accelerates T‐cell reconstitution and engraftment compared with standard dosing [ 53 ], and this confound must be acknowledged when attributing PT‐BEN's immune reconstitution advantages to BEN‐specific biology rather than reduced cyclophosphamide exposure.

Clinical Replication

Reproducibility beyond the Center of Origin

An important marker of maturity for any transplantation platform is reproducibility beyond the center of origin. Building upon the original trial of PT‐BEN, multiple international groups have now independently explored this approach. These subsequent studies provide critical validation but also offer new cautionary insights that refine the strategy for broader application.

An ambispective study from India evaluated the partial substitution of PT‐CY with BEN, adopting the maximum tolerated dose (MTD) established in the Arizona Phase Ia/Ib trial (CY 50 mg/kg on Day +3 followed by BEN 90 mg/m 2 on Day +4) [ 60 ]. They compared 21 patients receiving this hybrid regimen against 33 patients receiving standard PT‐CY (50 mg/kg on Days +3 and + 4). Despite a heterogeneous cohort involving haploidentical, matched sibling, and matched unrelated donors, the results were consistent with the Arizona experience; the PT‐CY/BEN arm demonstrated significantly faster neutrophil and platelet engraftment (Table 1 ). Furthermore, within the haploidentical subset, the regimen was associated with lower rates of acute GvHD and milder chronic GvHD. Importantly, BK virus–associated hemorrhagic cystitis was effectively eliminated in the PT‐BEN cohort, a crucial advantage for resource‐limited settings. The authors concluded that this approach offers a feasible, safe, and cost‐effective alternative, with the added benefits of easier administration (no mesna required) and a safer profile for patients with cardiac or renal comorbidities [ 60 ].

A prospective single‐arm study from Russia examined a combined platform of PT‐BEN and reduced‐dose PT‐CY (BEN 50 mg/m 2 + CY 25 mg/kg on Days +3 and + 4) in 50 patients with primary refractory (52%) or refractory relapsed (48%) myeloid neoplasms, a population at exceptionally high risk for both relapse and transplant‐related mortality [ 61 ]. This experience confirmed the regimen's feasibility and demonstrated potent antileukemic activity: 76% of patients achieved MRD‐negative status, with delayed relapse kinetics suggesting a preserved GvL effect, resulting in a 2‐year event‐free survival (EFS) of 22%. However, the study also identified CRS as a frequent and clinically significant toxicity (30% overall incidence, higher in haploidentical transplants), including a distinct late‐onset CRS phenotype emerging post‐engraftment (Table 1 ). Immune profiling revealed an expansion of effector‐memory CD8 + T cells and PD‐L1‐expressing myeloid populations, underscoring that BEN‐containing platforms generate immune landscapes distinct from PT‐CY alone depending on dose, timing, and disease context. The authors concluded that while this strategy shows promise for one of the most difficult‐to‐treat populations in HCT, further optimization is required, particularly for haploidentical donors. Notably, the extended time to relapse observed may provide a critical window for maintenance therapies to further improve outcomes [ 61 ].

Finally, a Phase I/II trial from MD Anderson Cancer Center explored escalating doses of PT‐BEN within reduced‐intensity conditioning platforms [ 62 ]. The study identified Dose Level 2 (DL2) as the optimal combination: CY 30 mg/kg plus BEN 30 mg/m 2 on Day +3, followed by BEN 100 mg/m 2 on Day +4. Consistent with the Russian experience, a distinct therapeutic window was evident: while this combinational approach produced encouraging results, higher pure BEN exposure (BEN 100 mg/m 2 monotherapy on Days +3 and + 4) proved toxic, resulting in grade 3–4 CRS in two of three patients and unacceptable rates of GvHD (Table 1 ). The authors concluded that PT‐BEN is a promising prophylaxis strategy for both haploidentical and matched unrelated donor transplants. Key findings at the optimal dose (DL2) included a favorable safety profile with uniform engraftment, notably lower GvHD rates with bone marrow grafts compared to peripheral blood, and excellent survival outcomes, highlighted by a 100% 1‐year overall survival in patients with acute leukemia [ 62 ].

Taken together, these studies reinforce two central themes: first, that the biological activity of PT‐BEN is reproducible across institutions; and second, that its toxicity profile is predictable and strictly dose‐ and platform‐dependent rather than idiosyncratic.

Critical Appraisal and Future Directions

Despite growing supportive data, broad adoption of PT‐BEN requires careful consideration. PT‐CY remains the dominant platform for GvHD prophylaxis because of its predictable safety profile across diverse transplant settings, anchored by large multicenter cohort data and, most recently, the BMT CTN 1703 randomized trial establishing PT‐CY/tacrolimus/MMF as a benchmark in reduced‐intensity PBSC transplantation [ 28 ]. Any strategy seeking to augment or replace this standard must demonstrate not only biological plausibility but clear clinical advantages in outcomes such as chronic GvHD, immune reconstitution, organ toxicity, and quality of life.

Interpretation of the existing PT‐BEN literature is complicated by several factors. The clinical studies summarized in this review are small and highly heterogeneous, differing in donor type, graft source, conditioning intensity, patient age, disease risk, and BEN/CY schedule, and should be interpreted as hypothesis‐generating signals rather than definitive evidence of platform‐level benefit. The Arizona experience is pediatric and young adults, haploidentical, bone marrow, and myeloablative; the India cohort mixes donor types and PBSC grafts; the Russia series focuses on refractory myeloid neoplasms with mixed donor types. The MD Anderson data remain at abstract level only, but prospective evaluation appears to be underway in a phase I/II trial ( NCT04022239 ) though no results have yet been posted.

These differences limit cross‐study comparison and raise the possibility that reported benefits reflect center‐specific practices as much as intrinsic PT‐BEN effects. A further and underappreciated interpretive challenge is that the partial‐substitution platform reduces total cyclophosphamide exposure by half, and recent literature demonstrating that reduced‐dose PT‐CY alone may improve engraftment kinetics and reduce toxicity makes it plausible that some observed benefits reflect CY sparing rather than BEN‐specific immunomodulation [ 51 , 53 , 54 , 55 ]. This confound cannot be resolved by non‐randomized single‐center comparison.

The accumulating preclinical consistency and early multi‐institutional clinical signals now provide sufficient biological plausibility to justify formal randomized evaluation of PT‐BEN against the PT‐CY standard. Continued reliance on single‐center non‐randomized series is insufficient to define the clinical value of this approach. The most rational trial design is a three‐arm comparison of standard PT‐CY, partial PT‐CY/PT‐BEN substitution, and reduced‐dose PT‐CY alone, a design that would simultaneously establish PT‐BEN's independent contribution and resolve the cyclophosphamide dose‐reduction confound that limits interpretation of all currently available data.

Stratification by graft source will be essential given the possible interaction between PT‐BEN and CRS risk. However, CRS is not unique to PT‐BEN and must be interpreted within the broader context of haploidentical HCT biology. With standard PT‐CY, CRS occurs in 70%–97% of patients, with PBSC grafts consistently associated with higher severity than bone marrow grafts [ 67 , 68 ]. Severe CRS (grade ≥ 3) carries significantly worse non‐relapse mortality and overall survival, while mild CRS may paradoxically reflect beneficial early immune activation, associated with lower relapse rates and enhanced T‐cell reconstitution in several series, suggesting a potential graft‐versus‐leukemia augmenting effect [ 69 , 70 ]. In the PT‐BEN setting, CRS appears dose‐ and graft source‐dependent, mirroring the platform‐dependent pattern established in the PT‐CY literature. Rather than representing a BEN‐specific toxicity, PT‐BEN‐associated CRS is most plausibly understood as a manifestation of shared post‐transplant innate inflammatory biology, amplified by higher BEN doses and the greater alloreactive T‐cell burden of PBSC grafts. Strategies developed in the PT‐CY setting to mitigate CRS including earlier initiation of calcineurin inhibitors and modified PT‐CY scheduling may inform analogous risk‐mitigation approaches in future PT‐BEN platforms [ 71 , 72 ].

PT‐BEN has progressed from a hypothesis‐driven preclinical observation to a reproducible early clinical signal across multiple independent centers. Whether these biological advantages translate into meaningful clinical benefit remains to be established through randomized evaluation against the PT‐CY standard.

Author Contributions

Conceptualization: E.K., R.J.S. Writing – original draft: E.K. Writing with extensive revisions: M.J.C., E.K. Review and edit of original and revised: M.J.C., E.K., R.J.S. All authors approved the final version of the manuscript.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgments

This work was supported in part by NIH/NCI R01 CA277493, R.J.S. & E.K., Courtney's Courage, and PANDA.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

1. Lalic H., Aurer I., Batinic D., Visnjic D., Smoljo T., and Babic A., “Bendamustine: A Review of Pharmacology, Clinical Use and Immunological Effects (Review),” Oncology Reports 47, no. 6 (2022): 114.

2. Montillo M., Ricci F., Tedeschi A., Vismara E., and Morra E., “Bendamustine: New Perspective for an Old Drug in Lymphoproliferative Disorders,” Expert Review of Hematology 3, no. 2 (2010): 131–148.

3. Gandhi V. and Burger J. A., “Bendamustine in B‐Cell Malignancies: The New 46‐Year‐Old Kid on the Block,” Clinical Cancer Research 15, no. 24 (2009): 7456–7461.

4. Konstantinov S. M., Kostovski A., Topashka‐Ancheva M., Genova M., and Berger M. R., “Cytotoxic Efficacy of Bendamustine in Human Leukemia and Breast Cancer Cell Lines,” Journal of Cancer Research and Clinical Oncology 128, no. 5 (2002): 271–278.

5. Leoni L. M. and Hartley J. A., “Mechanism of Action: The Unique Pattern of Bendamustine‐Induced Cytotoxicity,” Seminars in Hematology 48, no. Suppl 1 (2011): S12–S23.

6. Leoni L. M., Bailey B., Reifert J., et al., “Bendamustine (Treanda) Displays a Distinct Pattern of Cytotoxicity and Unique Mechanistic Features Compared With Other Alkylating Agents,” Clinical Cancer Research 14, no. 1 (2008): 309–317.

7. Stokes J., Hoffman E. A., Zeng Y., Larmonier N., and Katsanis E., “Post‐Transplant Bendamustine Reduces GvHD While Preserving GvL in Experimental Haploidentical Bone Marrow Transplantation,” British Journal of Haematology 174, no. 1 (2016): 102–116.

8. Stokes J., Hoffman E. A., Molina M. S., et al., “Bendamustine With Total Body Irradiation Limits Murine Graft‐Versus‐Host Disease in Part Through Effects on Myeloid‐Derived Suppressor Cells,” Biology of Blood and Marrow Transplantation 25, no. 3 (2019): 405–416.

9. Stokes J., Hoffman E. A., Molina M. S., et al., “Bendamustine With Total Body Irradiation Conditioning Yields Tolerant T‐Cells While Preserving T‐Cell‐Dependent Graft‐Versus‐Leukemia,” Oncoimmunology 9, no. 1 (2020): 1758011.

10. Molina M. S., Hoffman E. A., Stokes J., et al., “Regulatory Dendritic Cells Induced by Bendamustine Are Associated With Enhanced Flt3 Expression and Alloreactive T‐Cell Death,” Frontiers in Immunology 12 (2021): 699128.

11. Molina M. S., Stokes J., Hoffman E. A., et al., “Bendamustine Conditioning Skews Murine Host DCs Toward Pre‐cDC1s and Reduces GvHD Independently of Batf3,” Frontiers in Immunology 11 (2020): 1410.

12. Stokes J., Molina M. S., Hoffman E. A., Simpson R. J., and Katsanis E., “Immunomodulatory Effects of Bendamustine in Hematopoietic Cell Transplantation,” Cancers (Basel) 13, no. 7 (2021): 1702.

13. Chromik K., Wieczorkiewicz‐Kabut A., Zielinska P., et al., ““B” Symptoms and CBV Conditioning Negatively Impact Survival in Mantle Cell Lymphoma Patients Undergoing Autologous Stem Cell Transplantation,” Annals of Hematology 104, no. 9 (2025): 4771–4778.

14. Visani G., Malerba L., Stefani P. M., et al., “BeEAM (Bendamustine, Etoposide, Cytarabine, Melphalan) Before Autologous Stem Cell Transplantation Is Safe and Effective for Resistant/Relapsed Lymphoma Patients,” Blood 118, no. 12 (2011): 3419–3425.

15. Khouri I. F., Wei W., Korbling M., et al., “BFR (Bendamustine, Fludarabine, and Rituximab) Allogeneic Conditioning for Chronic Lymphocytic Leukemia/Lymphoma: Reduced Myelosuppression and GVHD,” Blood 124, no. 14 (2014): 2306–2312.

16. Moiseev I. S., Cherkashina A. N., Rudakova T. A., et al., “Low Incidence of Primary Graft Failure With Bendamustine, Fludarabine, and Busulfan Conditioning Prior to Haploidentical Allogeneic Hematopoietic Cell Transplantation,” Hematology/Oncology and Stem Cell Therapy 17, no. 4 (2024): 219–226.

17. Filioglou D., Husnain M., Khurana S., Simpson R. J., and Katsanis E., “Has the Shortage of Fludarabine Altered the Current Paradigm of Lymphodepletion in Favor of Bendamustine?,” Frontiers in Immunology 14 (2023): 1329850.

18. Ong S. Y., Pak S., Mei M., et al., “Bendamustine Lymphodepletion Is a Well‐Tolerated Alternative to Fludarabine and Cyclophosphamide Lymphodepletion for Axicabtagene Ciloleucel Therapy for Aggressive B‐Cell Lymphoma,” American Journal of Hematology 98, no. 11 (2023): 1751–1761.

19. Ghilardi G., Chong E. A., Svoboda J., et al., “Bendamustine Is Safe and Effective for Lymphodepletion Before Tisagenlecleucel in Patients With Refractory or Relapsed Large B‐Cell Lymphomas,” Annals of Oncology 33, no. 9 (2022): 916–928.

20. Sidana S., Hosoya H., Jensen A., et al., “Bendamustine vs. Fludarabine/Cyclophosphamide Lymphodepletion Prior to BCMA CAR‐T Cell Therapy in Multiple Myeloma,” Blood Cancer Journal 13, no. 1 (2023): 158.

21. Steven P., Perez V. L., and Sharma A., “Murine Models of Graft Versus Host Disease (GVHD): Focus on Ocular GVHD,” Ocular Surface 30 (2023): 179–186.

22. Schroeder M. A. and DiPersio J. F., “Mouse Models of Graft‐Versus‐Host Disease: Advances and Limitations,” Disease Models & Mechanisms 4, no. 3 (2011): 318–333.

23. Zeiser R., Socie G., and Blazar B. R., “Pathogenesis of Acute Graft‐Versus‐Host Disease: From Intestinal Microbiota Alterations to Donor T Cell Activation,” British Journal of Haematology 175, no. 2 (2016): 191–207.

24. Ehx G., Ritacco C., and Baron F., “Pathophysiology and Preclinical Relevance of Experimental Graft‐Versus‐Host Disease in Humanized Mice,” Biomarker Research 12, no. 1 (2024): 139.

25. Gilman K. E., Cracchiolo M. J., Matiatos A. P., Davini D. W., Simpson R. J., and Katsanis E., “Partially Replacing Cyclophosphamide With Bendamustine in Combination With Cyclosporine A Improves Survival and Reduces Xenogeneic Graft‐Versus‐Host‐Disease,” Frontiers in Immunology 13 (2022): 1045710.

26. Hadjis A. D., Nunes N. S., Khan S. M., et al., “Post‐Transplantation Cyclophosphamide Uniquely Restrains Alloreactive CD4(+) T‐Cell Proliferation and Differentiation After Murine MHC‐Haploidentical Hematopoietic Cell Transplantation,” Frontiers in Immunology 13 (2022): 796349.

27. Stokes J., Simpson R. J., and Katsanis E., “Commentary: Post‐Transplantation Cyclophosphamide Uniquely Restrains Alloreactive CD4+ T‐Cell Proliferation and Differentiation After Murine MHC‐Haploidentical Hematopoietic Cell Transplantation,” Frontiers in Immunology 13 (2022): 887648.

28. Bolanos‐Meade J., Hamadani M., Wu J., et al., “Post‐Transplantation Cyclophosphamide‐Based Graft‐Versus‐Host Disease Prophylaxis,” New England Journal of Medicine 388, no. 25 (2023): 2338–2348.

29. Curtis D. J., Patil S. S., Reynolds J., et al., “Graft‐Versus‐Host Disease Prophylaxis With Cyclophosphamide and Cyclosporin,” New England Journal of Medicine 393, no. 3 (2025): 243–254.

30. Watkins B., Qayed M., McCracken C., et al., “Phase II Trial of Costimulation Blockade With Abatacept for Prevention of Acute GVHD,” Journal of Clinical Oncology 39, no. 17 (2021): 1865–1877.

31. Norsworthy K. J., Rivera D. R., Wynne J., et al., “FDA Approval Summary: Abatacept for the Prophylaxis of Acute GVHD,” Clinical Cancer Research 31, no. 15 (2025): 3118–3123.

32. Koura D., Dykes K., Goodman A., et al., “A Prospective Clinical Trial of GVHD Prophylaxis With Posttransplant Cyclophosphamide and Abatacept,” Blood Advances 9, no. 16 (2025): 4336–4344.

33. Xie C., Shi L., Wei Z., Lai Y., and Liu R., “Efficacy and Safety of Ruxolitinib for Graft‐Versus‐Host Disease Prophylaxis in Allogeneic Hematopoietic Stem Cell Transplantation: A Systematic Review and Meta‐Analysis,” Frontiers in Medicine (Lausanne) 12 (2025): 1680188.

34. DeFilipp Z., Kim H. T., Knight L. W., et al., “Low Rates of Chronic Graft‐Versus‐Host Disease With Ruxolitinib Maintenance Following Allogeneic HCT,” Blood 145, no. 20 (2025): 2312–2316.

35. Gooptu M. and Koreth J., “New Age GVHD Prophylaxis Regimens: What Works for What Donor and Why,” Hematology. American Society of Hematology. Education Program 2025, no. 1 (2025): 458–464.

36. Luznik L. and Fuchs E. J., “High‐Dose, Post‐Transplantation Cyclophosphamide to Promote Graft‐Host Tolerance After Allogeneic Hematopoietic Stem Cell Transplantation,” Immunologic Research 47, no. 1–3 (2010): 65–77.

37. O'Donnell P. V. and Jones R. J., “The Development of Post‐Transplant Cyclophosphamide: Half a Century of Translational Team Science,” Blood Reviews 62 (2023): 101034.

38. Radojcic V. and Luznik L., “Mechanism of Action of Posttransplantation Cyclophosphamide: More Than Meets the Eye,” Journal of Clinical Investigation 129, no. 6 (2019): 2189–2191.

39. Wachsmuth L. P., Patterson M. T., Eckhaus M. A., Venzon D. J., Gress R. E., and Kanakry C. G., “Post‐Transplantation Cyclophosphamide Prevents Graft‐Versus‐Host Disease by Inducing Alloreactive T Cell Dysfunction and Suppression,” Journal of Clinical Investigation 129, no. 6 (2019): 2357–2373.

40. SR M. C., Kanakry C. G., Tsai H. L., et al., “Grade II Acute Graft‐Versus‐Host Disease and Higher Nucleated Cell Graft Dose Improve Progression‐Free Survival After HLA‐Haploidentical Transplant With Post‐Transplant Cyclophosphamide,” Biology of Blood and Marrow Transplantation 24, no. 2 (2018): 343–352.

41. Shimoni A., Peczynski C., Labopin M., et al., “Graft‐Versus‐Host Disease Is Not Associated With Reduced Incidence of Relapse Following Haploidentical Stem Cell Transplantation With Post‐Transplant Cyclophosphamide for Acute Lymphoblastic Leukaemia: A Study on Behalf of the Acute Leukaemia Working Party of European Society for Blood and Marrow Transplantation,” British Journal of Haematology 207, no. 5 (2025): 2041–2048.

42. Esquirol A., Pascual M. J., Kwon M., et al., “Severe Infections and Infection‐Related Mortality in a Large Series of Haploidentical Hematopoietic Stem Cell Transplantation With Post‐Transplant Cyclophosphamide,” Bone Marrow Transplantation 56, no. 10 (2021): 2432–2444.

43. Goldsmith S. R., Abid M. B., Auletta J. J., et al., “Posttransplant Cyclophosphamide Is Associated With Increased Cytomegalovirus Infection: A CIBMTR Analysis,” Blood 137, no. 23 (2021): 3291–3305.

44. Khimani F., Ranspach P., Elmariah H., et al., “Increased Infections and Delayed CD4(+) T Cell but Faster B Cell Immune Reconstitution After Post‐Transplantation Cyclophosphamide Compared to Conventional GVHD Prophylaxis in Allogeneic Transplantation,” Transplant Cell Therapy 27, no. 11 (2021): 940–948.

45. Tomaszewska A., Basak G. W., Peczynski C., et al., “Complications of Haploidentical Hematopoietic Cell Transplantation With Post‐Transplant Cyclophosphamide‐A Prospective Study on Behalf of the EBMT Transplant Complications Working Party,” Cancers (Basel) 17, no. 24 (2025): 4029.

46. Dulery R., Mohty R., Labopin M., et al., “Early Cardiac Toxicity Associated With Post‐Transplant Cyclophosphamide in Allogeneic Stem Cell Transplantation,” JACC: CardioOncology 3 (2021): 250–259.

47. Hadjali R., Chevillon F., Sibilia B., et al., “Risk of Posttransplant Cyclophosphamide‐Related Cardiotoxicity in Allogeneic Stem Cell Transplantation,” Canadian Journal of Cardiology (2026).

48. Yeh J., Whited L., Saliba R. M., et al., “Cardiac Toxicity After Matched Allogeneic Hematopoietic Cell Transplant in the Posttransplant Cyclophosphamide Era,” Blood Advances 5, no. 24 (2021): 5599–5607.

49. Mac S., Ngo D., Yang D., et al., “Use of High‐Dose Mesna and Hyperhydration Leads to Lower Incidence of Hemorrhagic Cystitis After Posttransplant Cyclophosphamide‐Based Allogeneic Transplantation,” Bone Marrow Transplantation 56, no. 10 (2021): 2464–2470.

50. Ngo D., Samuels D., Chen J., Koller P. B., and Al Malki M. M., “A Clinical Review of the Different Strategies to Minimize Hemorrhagic Cystitis Associated With the Use of Post‐Transplantation Cyclophosphamide in an Allogeneic Transplant,” Transplant Cell Therapy 28, no. 7 (2022): 349–354.

51. Dulery R., Goudet C., Mannina D., et al., “Reduced Post‐Transplant Cyclophosphamide Doses in Haploidentical Hematopoietic Cell Transplantation for Elderly Patients With Hematological Malignancies,” Bone Marrow Transplantation 58, no. 4 (2023): 386–392.

52. Fuji S., Nakamae H., Sugita J., et al., “Lower‐Dose Post‐Transplant Cyclophosphamide in Haploidentical Hematopoietic Cell Transplantation,” Bone Marrow Transplantation 59, no. 11 (2024): 1628–1630.

53. Hyder M. A., Dimitrova D., Sabina R., et al., “Intermediate‐Dose Posttransplantation Cyclophosphamide for Myeloablative HLA‐Haploidentical Bone Marrow Transplantation,” Blood Advances 9, no. 10 (2025): 2553–2569.

54. Pinto F. R., Salas M. Q., Suarez‐Lledo M., et al., “Modifying Post‐Transplant Cyclophosphamide Platform to Improve Safety in HLA‐Matched Allogeneic HCT: A Reduced Dose Plus Adjunctive Strategies,” Transplantation and Cellular Therapy 32, no. 3 (2026): 298 e291–298 e213.

55. Redondo S., Garcia‐Cadenas I., Esquirol A., et al., “Standard Versus Reduced‐Doses of Post‐Transplant Cyclophosphamide in HLA Matched (10/10) Peripheral Blood Stem Cell Transplantation,” European Journal of Haematology 115, no. 4 (2025): 413–422.

56. Sugita J., Kamimura T., Ishikawa T., et al., “Reduced Dose of Posttransplant Cyclophosphamide in HLA‐Haploidentical Peripheral Blood Stem Cell Transplantation,” Bone Marrow Transplantation 56, no. 3 (2021): 596–604.

57. Katsanis E., Sapp L. N., Varner N., Koza S., Stea B., and Zeng Y., “Haploidentical Bone Marrow Transplantation With Post‐Transplant Cyclophosphamide/Bendamustine in Pediatric and Young Adult Patients With Hematologic Malignancies,” Biology of Blood and Marrow Transplantation 24, no. 10 (2018): 2034–2039.

58. Katsanis E., Maher K., Roe D. J., and Simpson R. J., “Progressive Substitution of Posttransplant Cyclophosphamide With Bendamustine: A Phase I Study in Haploidentical Bone Marrow Transplantation,” EJHaem 1, no. 1 (2020): 286–292.

59. Katsanis E., Stea B., Kovacs K., et al., “Feasibility and Efficacy of Partially Replacing Post‐Transplantation Cyclophosphamide With Bendamustine in Pediatric and Young Adult Patients Undergoing Haploidentical Bone Marrow Transplantation,” Transplantation and Cellular Therapy 28, no. 7 (2022): 390 e391–390 e310.

60. Nair V., Kathrotiya M., Shirure V., et al., “Feasibility and Efficacy of Partial Replacement of Post Transplantation Cyclophosphamide With Bendamustine on Day +4 for Graft Versus Host Disease Prophylaxis in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation,” Bone Marrow Transplantation 60 (2025): 994–1001.

61. Moiseev I., Bondarenko S., Vlasova Y., et al., “Allogeneic Hematopoietic Cell Transplantation With a Combination of Posttransplantation Bendamustine and Cyclophosphamide in Refractory Myeloid Neoplasms,” Cancer 131, no. 10 (2025): e35893.

62. Khouri I., Wu S., Haddad F., et al., “Post‐Transplant Bendamustine (PT‐BEN) With or Without Cyclophosphamide for Gvhd Prophylaxis After Haplo‐Identical or Matched Unrelated Donor Transplants: A Phase I/II Trial,” Blood 146 (2025): 6022–6023.

63. Baker F. L., Stokes J., Cracchiolo M. J., Davini D., Simpson R. J., and Katsanis E., “Impact of Post‐Transplant Cyclophosphamide With Bendamustine on Immune Reconstitution in Young Patients Undergoing T‐Cell Replete Haploidentical Bone Marrow Transplantation: Results From a Phase Ia/Ib Clinical Trial,” Frontiers in Immunology 16 (2025): 1568862.

64. Filioglou D., Kovacs K., Lafleur B. J., White L. M., and Katsanis E., “Haploidentical Versus Matched Sibling Donor HCT in Racially Diverse Pediatric and AYA Patients With Hematologic Malignancies: A Single‐Center Comparison,” Transplantation and Cellular Therapy 31, no. 12 (2025): 1037 e1031–1037 e1014.

65. Zhao C., Bartock M., Jia B., et al., “Post‐Transplant Cyclophosphamide Alters Immune Signatures and Leads to Impaired T Cell Reconstitution in Allogeneic Hematopoietic Stem Cell Transplant,” Journal of Hematology & Oncology 15, no. 1 (2022): 64.

66. Rambaldi B., Kim H. T., Reynolds C., et al., “Impaired T‐ and NK‐Cell Reconstitution After Haploidentical HCT With Posttransplant Cyclophosphamide,” Blood Advances 5, no. 2 (2021): 352–364.

67. Abboud R., Wan F., Mariotti J., et al., “Cytokine Release Syndrome After Haploidentical Hematopoietic Cell Transplantation: An International Multicenter Analysis,” Bone Marrow Transplantation 56, no. 11 (2021): 2763–2770.

68. Otoukesh S., Elmariah H., Yang D., et al., “Cytokine Release Syndrome Following Peripheral Blood Stem Cell Haploidentical Hematopoietic Cell Transplantation With Post‐Transplantation Cyclophosphamide,” Transplant and Cellular Therapy 28, no. 2 (2022): 111 e111–111 e118.

69. Shapiro R. M., Kim H. T., Ansuinelli M., et al., “Cytokine Release Syndrome in Haploidentical Stem Cell Transplant May Impact T‐Cell Recovery and Relapse,” Blood Advances 7, no. 15 (2023): 4080–4088.

70. Benavente R., Montoro J., Balaguer‐Rosello A., et al., “Cytokine Release Syndrome: Implications for Transplant Outcomes in Haploidentical and HLA‐Matched HSCT Using PTCy,” Bone Marrow Transplantation 60, no. 7 (2025): 1020–1026.

71. Kurita N., Sakamoto T., Kato T., et al., “Early Administration of Cyclosporine May Reduce the Incidence of Cytokine Release Syndrome After HLA‐Haploidentical Hematopoietic Stem‐Cell Transplantation With Post‐Transplant Cyclophosphamide,” Annals of Hematology 100, no. 5 (2021): 1295–1301.

72. Tang J., Jensen R. R., Bryan B., Hoda D., and Hunter B. D., “Reduced Cytokine Release Syndrome and Improved Outcomes With Earlier Immunosuppressive Therapy in Haploidentical Stem Cell Transplantation,” Transplant and Cellular Therapy 30, no. 4 (2024): 438 e431–438 e411.

Associated Data

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.