Evolution of conditioning regimens prior to autologous stem cell transplantation in lymphoma patients

Highlights

• •Autologous stem cell transplantation (ASCT) remains the cornerstone therapy for young, high-risk lymphoma patients with chemosensitive relapsed or refractory disease.
• •Optimal selection of personalized conditioning regimens is critical for treatment success.
• •Regimen individualization should be based on lymphoma subtype, anticipated survival benefit, and treatment-related toxicity profiles.
• •This comprehensive multidimensional evaluation aims to optimize clinical outcomes while minimizing treatment risks, ultimately achieving truly individualized conditioning therapy.

Abstract

High-dose chemotherapy followed by autologous stem cell transplantation (ASCT) has become the standard salvage treatment for primary high-risk and relapsed/refractory lymphoma patients. The conditioning regimen has evolved from one that included total body irradiation (TBI) to one that includes high-dose chemotherapy, as well as combinations of novel drugs. We systematically reviewed existing research data and summarize the survival benefits, treatment-related adverse events, and hematopoietic reconstitution after various conditioning regimens. As a classical conditioning regimen, BEAM (carmustine, etoposide, cytarabine, and melphalan) can eliminate minimal residual disease and achieve long-term disease-free survival. However, considering the safety, tolerability, and accessibility of these drugs, it still needs to be improved. In recent years, with the continuous development of novel drugs, new combinations have emerged. An efficient and low toxicity regimen is crucial for the survival benefits of patients undergoing ASCT. Alternative regimens such as GBC (gemcitabine, busulfan, and cyclophosphamide)/GBM (gemcitabine, busulfan, and melphalan), BeEAM (bendamustine, etoposide, cytarabine, and melphalan), and SEAM (semustine, etoposide, cytarabine, and melphalan) are safe and feasible. Data from studies combining these regimens with novel drugs are even more attractive. Ultimately, future conditioning regimens will show low toxicity, be highly efficient, and be personalizable with respect to lymphoma subtype and comorbidities.

Graphical abstract

Introduction

Lymphoma is a group of highly heterogeneous hematological malignancies that originating from lymph nodes or extranodally; currently, certain cases persist as incurable [1]. Autologous stem cell transplantation (ASCT) may be employed as a first-line consolidation therapy for patients diagnosed with mantle cell lymphoma (MCL), aggressive peripheral T-cell lymphoma (PTCL), and young individuals with high-risk diffuse large B-cell lymphoma (DLBCL). In cases of relapsed or refractory lymphoma, ASCT serves as a salvage consolidation therapy. This treatment is indicated for patients aged 65 years or younger who exhibit no significant comorbidities, possess normal cardiac, hepatic, pulmonary, and renal function, and are free from active infections [[2], [3], [4]]. The remission rate after transplantation is about 30–60 %, the 5-year overall survival (OS) rate is about 20–50 %, and the 5-year progression-free survival (PFS) rate is about 50–60 % [5].

Conditioning regimen is the key to successful ASCT and involves enhancing cytotoxicity through high-dose chemotherapy, further clearing minimal residual disease, and achieving deeper remission. The ideal conditioning regimen should have high selective antitumor activity, maximize the clearance of residual tumor cells, minimize the disease recurrence rate, extend OS, and achieve the goal of long-term disease-free survival. It should also possess controllable toxicity to avoid treatment-related deaths [6]. Therefore, the efficacy, safety, tolerability, and accessibility of conditioning regimens are urgent practical problems to be solved. Based on multiple clinical research data, conditioning regimens of ASCT have transitioned from total body irradiation (TBI) combined with chemotherapy to high-dose chemotherapy (Fig. 1). In recent years, strategies involving additional new drugs have continued to evolve, develop, and improve.

Fig. 1.

Evolution of conditioning regimens for autologous hematopoietic stem cell transplantation in lymphoma patients. Since the 1970s, the combination of total body irradiation (TBI) and chemotherapy has been used as a conditioning regimen for autologous hematopoietic stem cell transplantation in lymphoma patients. Considering the short- and long-term toxicities of TBI, high-dose chemotherapy conditioning regimen using alkylating agents has gradually emerged since the 1980s. Representative conditioning regimens include BEAM, BuCy, and GBC. Due to the risk of inducing interstitial pneumonia and poor accessibility, alternative regimens containing semustine and bendamustine have gradually been accepted. In recent years, due to the emergence of new drugs, addition of new drugs to conditioning regimens has also been explored. Representative novel-agent regimens combined with high-dose chemotherapy encompass V-BEAM, R-BEAM, pola-BEAM, and Chi-BEAC. Conditioning regimens incorporating immunotherapy—notably those involving bispecific antibodies and chimeric antigen receptor T cells—are also under investigation. The efficacy and safety profiles of these strategies warrant further evaluation.

BEAM: carmustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan; BuCy: busulfan and cyclophosphamide; GBC: gemcitabine, busulfan, and cyclophosphamide; V: bortezomib (Bortezomib); R: rituximab; Pola: polatuzumab vedotin; SEAM: semustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan; BeEAM: bendamustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan; Chi: chidamide; TEAM: thiotepa, etoposide (Etoposide) (Etoposide), cytarabine (Cytarabine), and melphalan.

Evolution of conditioning regimens

Conditioning regimens containing TBI. Radiotherapy was added as a conditioning regimen by the Nobel Prize winner E.D. Thomas in the late 1950s [7]. Compared with chemotherapy alone, the combination with TBI may achieve greater tumor cell toxicity and better tissue permeability. In 1957, TBI was first applied as a conditioning regimen before allogeneic hematopoietic stem cell transplantation to treat acute lymphoblastic leukemia [8]. This report confirmed the feasibility of TBI, but also pointed out that irreversible damage of important organs may be an important issue. In 1976, E.D. Thomas performed a study of human leukocyte antigen-identical sibling bone marrow transplantation for patients with acute myeloid leukemia who were pretreated with cyclophosphamide and TBI [9]. This conditioning regimen was applied successfully, and the transplantation process was well tolerated. All patients were transplanted successfully, and there were no deaths within the first 50 days after transplantation. In total, 63 % of patients were in remission, and the median survival after transplantation was not <18 months. This study further confirmed the feasibility of TBI as a conditioning regimen. Since 1970, TBI-based clinical research into ASCT for non-Hodgkin lymphoma (NHL) has been ongoing [10]; however, due to significant recent and long-term toxicities, this protocol is also controversial. Approximately two-thirds of patients receiving daily TBI experienced pulmonary toxicity, which in turn affected OS [11]. In the short term, TBI can also induce nausea, vomiting, mucositis, dysphagia, diarrhea, xerostomia, and bone marrow suppression. Long-term toxicities include cognitive impairment, cataracts, pituitary dysfunction, gonadal failure, hypothyroidism, and secondary tumors, especially myelodysplastic syndrome (MDS) and acute myelogenous leukemia (AML) [[12], [13], [14], [15], [16]]. An analysis of retrospective data from the EBMT Lymphoma Working Party revealed that follicular lymphoma (FL) patients undergoing total body irradiation (TBI)-based autologous stem cell transplantation (ASCT) post-1995 exhibited an elevated risk of non-relapse mortality (NRM) and treatment-related MDS/AML compared to those receiving high-dose chemotherapy, although disease control outcomes were comparable [17]. It is essential to emphasize that in instances of refractory lymphoma exhibiting complete remission (CR) or partial remission (PR), no significant improvement in adverse prognostic outcomes was observed, despite the implementation of conditioning regimens based on TBI and ASCT [18].

The toxicities and adverse events associated with TBI are of great concern. In recent years, this protocol has been used less and less during the ASCT preprocessing stage, with use of high-dose combination chemotherapy regimens becoming more common. Usually, high-dose chemotherapy regimens are based on a single alkylating agent combined with several cytotoxic drugs. Commonly selected alkylating agents include carmustine (BCNU), busulfan (Bu), semustine, and bendamustine.

Conditioning regimens containing BCNU. BCNU is an alkylating agent that was approved by the Food and Drug Administration (FDA) in 1977 for treatment of brain tumors, Hodgkin lymphoma (HL), and myeloma [19]. It is often combined with etoposide (Etoposide), cytarabine (Cytarabine), melphalan, cyclophosphamide, and other drugs, and its representative schemes are BEAM (BCNU, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan), BEAC (BCNU, etoposide (Etoposide), cytarabine (Cytarabine), and cyclophosphamide), and CBV (cyclophosphamide, BCNU, and etoposide). Some clinical studies compared the efficacy and safety of BCNU-containing regimens. A study in 2015 aimed to evaluate the efficacy and safety of BEAM, CBV, BuCy (Bu and cyclophosphamide), and TBI conditioning regimens [20]. According to pathological subtype analysis, BEAM has the best survival benefits for patients with HL or DLBCL, with a 3-year OS rate of 79 % (P = 0.001) and 58 % (P = 0.002), respectively. In comparison to the BEAM regimen (hazard ratio [HR] 0.63, 95 % confidence interval [CI] 0.45–0.87, P = 0.006) and the CBV-high regimen (HR 0.50, 95 % CI 0.33–0.78, P = 0.002), the CBV-low regimen appears to be the most effective treatment for follicular lymphoma (FL), demonstrating a 3-year overall survival (OS) rate of 81 %. BEAM, CBV, and TBI demonstrate comparable advantages for the management of MCL, with a 3-year OS rate of 75 %, 66–80 % and 66 %, respectively (P = 0.051). The incidences of treatment-related mortality (TRM) at 1 year after transplantation in the groups that received BCNU-containing regimens deserve attention (4 %, 7 %, 8 %, 7 %, and 8 % in the BEAM, CBV-low, CBV-high, BuCy, and TBI groups, respectively). Occurrence of interstitial pneumonia (IPS) within 1 year after transplantation also should not be ignored (3 %, 3 %, 6 %, 4 %, and 5 % in these groups, respectively). Another retrospective study indicated that BEAM has better survival benefits than CBV for patients with relapsed and refractory HL [21]. In 2018, Cheol et al. retrospectively compared the hematopoietic reconstitution, adverse events, and survival of 97 patients with recurrent/refractory NHL pretreated with BEAM or BEAC [22]. In the BEAM and BEAC groups, the median time until neutrophil implantation was 11 and 12 days, respectively (P = 0.001), and the median time until platelet implantation was 13.5 and 16 days, respectively. It is worth mentioning that volume of red blood cell infusion was significantly lower in the BEAM group than in the BEAC group (3.7 units vs. 6.5 units, respectively; P = 0.037). Therefore, from the perspective of hematopoietic reconstruction after transplantation, BEAM is clearly superior to BEAC. In terms of treatment-related adverse events (TRAEs), the incidence of mucosal inflammation, nausea, vomiting, bleeding, and infection did not differ significantly between the two groups. The 2-year event-free survival (EFS) rate for the BEAM and BEAC groups was 62.4 % and 28.6 %, respectively (P = 0.001), and the 2-year OS rate was 62.4 % and 32.1 %, respectively (P = 0.003). This suggests that the toxicities of the two regimens are similar, but the survival benefits of BEAM appear to be superior.

BEAM has demonstrated better OS and PFS and fewer treatment-related events in numerous studies and thus is most commonly used for ASCT in lymphoma patients at present. However, BCNU often causes lung injury, which typically manifests as breathing difficulties, cough, and hypoxemia [23,24]. Potential mechanisms underlying the observed phenomena encompass DNA damage, glutathione depletion, infiltration of inflammatory cells within the pulmonary tissue, and alterations in immune response [25]. Characteristic pathological changes are manifested as alveolar edema and protein accumulation, diffuse hyperplasia, hypertrophy of alveolar epithelial cells, metaplasia of alveolar epithelium, proliferation of fibroblasts in the alveolar septa, and subsequent fibrosis of the alveolar septa [[26], [27], [28]]. These changes are also associated with the occurrence of IPS. The incidence of pulmonary toxicity with BCNU-based conditioning regimens is as high as 16–64 %, and the occurrence of IPS is dose-correlated [29,30]. The occurrence of IPS predicted a higher TRM rate and shorter PFS and OS. Moreover, due to the shortage of BCNU, many alternative regimens have emerged.

Conditioning regimens containing Bu. As a powerful cytotoxic bifunctional alkylating agent, Bu has been widely used to treat chronic myeloid leukemia and myeloproliferative neoplasms [31,32]. High-dose Bu (usually 16 mg/kg) in combination with other cytotoxic drugs (especially cyclophosphamide) is a common conditioning protocol for allogeneic or autologous bone marrow transplantation in the treatment of acute or chronic leukemia [33,34]. In 1998, Kroeger et al. performed a clinical study of Bu combined with cyclophosphamide and etoposide (Etoposide) as a conditioning protocol [35]. The median follow-up was 50 months and the median PFS and OS rates were 50 % and 55 %, respectively. All patients developed mucositis, and 90 % of cases were grade II. TRM at 100 days after transplantation was 5 % and patients died of hepatic veno-occlusive disease (VOD). A 2015 single-center study by Erber Ilhami et al. found no statistically significant differences between the BuCyE and BEAM regimens. The median time to neutrophil engraftment was 12 days (range, 8–14) compared to 13 days (range, 9–18) for BEAM, while platelet engraftment times were 14.5 days (range, 10–32) versus 15 days (range, 11–29), respectively. Although the BuCyE group exhibited higher incidences of mucositis (38.7 % vs. 27.3 %) and nausea (41.9 % vs. 36.4 %), these differences were not statistically significant. Furthermore, no significant differences were observed in overall survival or event-free survival between the two groups [36].

Adverse reactions of conditioning regimens containing Bu. Bu-based regimens mainly induce hepatotoxicity, with a high incidence of VOD (2.6–15 %), which is the second leading cause of death in patients receiving autologous grafts [37]. The pathophysiology of VOD is notably intricate. It involves the disruption of the equilibrium among procoagulant activity, inflammatory response, and fibrinolysis, leading to the obstruction of sinusoidal blood flow [38]. This obstruction results in damage to hepatic endothelial cells and ultimately culminates in necrosis. The characteristic clinical manifestations of VOD encompass jaundice, hyperbilirubinemia, hepatomegaly, ascites, weight gain, and pain in the upper right quadrant of the abdomen. In addition, Bu can cause neutropenia and seizures. A study conducted by Berber et al. in 2010 aimed to compare the hematopoietic reconstitution, adverse events, and safety of BuCyE and BEAM [36]. The hematopoietic reconstitution times were similar for both groups. The main TRAEs were mucosal inflammation, nausea, vomiting, diarrhea, infection, and TRM. There was no statistically significant difference in TRAE incidence between the two groups, nor was there a difference in EFS and OS. This study suggests that BuCyE can be used as an alternative regimen. Due to the small sample size, some bias, and limited follow-up time, studies with larger sample sizes are needed to confirm these findings.

In addition, a small sample clinical study of Bu combined with gemcitabine as a conditioning regimen appears to be safe and well-tolerated. The study aimed to evaluate hematopoietic reconstitution, survival benefit, and safety in ASCT patients treated with GBC (GCB, Bu, and cyclophosphamide)/GBM (GCB, Bu, and melphalan) [39]. The median time for neutrophil and platelet engraftment was 10 and 11 days, respectively. After a median follow-up of 21 months, the estimated 36–month PFS and OS rates were 87 % and 93 %, respectively. Graft failure did not occur, and no cases of VOD or TRM were found. Mild bone marrow suppression, nephrotoxicity, and elevated transaminase were induced. Grade III–IV TRAEs included neutropenia (100 %), thrombocytopenia (100 %), anemia (81.3 %), oral mucosal inflammation (45.1 %), liver dysfunction (23.1 %), diarrhea (19.8 %), nausea (14.3 %), vomiting (4.4 %), bleeding (4.4 %), fever (1.1 %), and rash (1.1 %). This study suggests that the GBC/GBM conditioning regimen is feasible and tolerable, and thus is an ideal alternative regimen. However, due to the small sample size and limited follow-up time, its safety and long-term survival benefits require further evaluation.

Conditioning regimens containing semustine. In recent years, the use of other alkylating agents, such as semustine and bendamustine, to replace BCNU has been investigated in a preliminary trial. The representative regimens are SEAM (semustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan) and BeEAM (bendamustine, etoposide (Etoposide), cytarabine (Cytarabine) (Cytarabine), and melphalan). Semustine has good curative effects on malignant melanoma, lymphoma, and brain metastases [[40], [41], [42], [43]]. A retrospective study conducted in 2022 aimed to evaluate the safety and efficacy of ASCT for lymphoma after conditioning with SEAM [44]. The median times of neutrophil and platelet engraftment were 9.5 and 12 days, respectively. The main TRAE was gastrointestinal reactions, with incidences of grade 3–4 nausea/vomiting, mucosal inflammation, and diarrhea of 21.6 %, 36.1 %, and 11.3 %, respectively. Two patients (2.1 %) suffered TRM at 100 days after transplantation. The median follow-up was 53.9 months, and 3-year PFS and OS rates were 62 % and 75 %, respectively. This study indicates that SEAM has similar effectiveness and safety as BEAM and may be an ideal alternative regimen. Another retrospective study conducted by Huang et al. in 2019 compared the survival benefits and side effects of mBuCy and SEAM in patients with poor prognosis or relapsed/refractory lymphoma [6]. Neutrophil engraftment seemed to be faster in the mBuCy group than in the SEAM group (9 days vs. 10 days, P = 0.015), and the median time of platelet engraftment was similar in the two groups (11 days vs. 12 days, P = 0.647). The incidences of mucosal inflammation and diarrhea were higher in the SEAM group than in the mBuCy group (91.2 % vs. 97.1 %, P = 0.027; 64.7 % vs. 82.4 %, P = 0.050), while the incidence of febrile neutropenia (FN) did not significantly differ between the two groups (55.9 % vs. 64.7 %, P = 0.457). The survival benefits were similar in the mBuCy and SEAM groups, with an estimated 2-year PFS rate of 79.0 % and 70.0 %, respectively (P = 0.378), and a 2-year OS rate of 81.0 % and 78.0 %, respectively (P = 0.789). Accordingly, the SEAM regimen is slower with respect to hematopoietic reconstruction; therefore, care should be exercised when using SEAM for patients with previous infections and severe bone marrow suppression after chemotherapy. Of note, approximately 23 % of patients who received SEAM conditioning had already received more than three lines of chemotherapy regimen prior to collection of hematopoietic stem cells, which can lead to bone marrow suppression and CD34+ cell damage [6,44]. Therefore, the impact of SEAM on hematopoietic reconstruction necessitates further investigation.

Conditioning regimens containing bendamustine. Bendamustine is a bifunctional molecule with alkylation and antimetabolic properties. It has been approved by the FDA and the European Medicines Agency for the treatment of indolent and invasive B-cell NHL, HL, and multiple myeloma [[45], [46], [47], [48], [49]]. It is often combined with etoposide (Etoposide), cytarabine (Cytarabine) (Cytarabine), and melphalan as BeEAM.

In 2011, a phase I/II study evaluated the safety and efficacy of BeEAM in patients with relapsed and refractory lymphoma [50]. The median times of granulocyte and platelet implantation were 10 days (8–12 days) and 13 days (8–39 days), respectively. At a median follow-up of 18 months, 81 % of patients were still in complete remission, with a 3-year PFS rate of 72 %.

Adverse reactions of conditioning regimens containing bendamustine. Non-hematologic toxicities included grade 3–4 mucositis (26 %) and grade 2–4 gastroenteritis (35 %). It is worth mentioning that no grade 3–4 renal, hepatic, cardiac, or pulmonary toxicity was observed, and the 100-day TRM rate was 0 %. BeEAM was confirmed to be safe and effective in this research. Another large-scale retrospective study conducted in 2018 further evaluated the safety of BeEAM [51]. The incidence of acute renal failure (ARF) was as high as 27.9 %, and the median time of occurrence was 6 days after conditioning. In total, 10 % of ARF cases had creatinine levels that returned to normal within a median of 10 days. ARF may be a problem that should be focused on during processing of BeEAM. Although the incidence of ARF was relatively high, it did not seem to increase TRM [52,53]. ARF may be associated with age (>55 years, 1 point), previous renal failure (2 points), and high-dose bendamustine (>160 mg/m²/day, 1 point) [51]. Through a simple 4-point scoring system, precise stratification can be achieved to reduce or avoid ARF.

Due to the pulmonary toxicity and poor accessibility of BCNU, recent research has focused on whether BeEAM can replace the classic BEAM regimen. The results of a study comparing the efficacy and safety of BeEAM and BEAM in MCL patients were published in 2020 [52]. The 3-year PFS rate of the BeEAM group was significantly better than that of the BEAM group (84 % vs. 63 %, P = 0.03), but OS did not significantly differ between the two groups (P = 0.20). The main TRAEs in the BeEAM group were ARF, mucosal inflammation, and intestinal bacterial infection. The incidence of ARF was 46 % and that of KDIGO grade 3 ARF was 32 %, which emerged a median of 5 days after transplantation. Most cases were reversible, and the median recovery time of renal function was 10 days after transplantation. This study indicated that BeEAM significantly improved PFS without increasing TRM, despite the occurrence of ARF. Another recent single-center retrospective study compared the efficacy, safety, and economic cost of BeEAM and BEAM [53]. The two groups had similar 3-year OS and PFS rates (71 % vs. 78.1 %, P = 0.296; 74.1 % vs. 71.3 %, P = 0.762). Compared with BEAM, BeEAM resulted in a saving of 21,200 Canadian dollars per patient. With its similar survival benefits and good safety, BeEAM seemed to have an economic advantage over BEAM.

Conditioning regimens containing thiotepa. Compared with BEAM, alternative regimens such as GBM/GBC, BeEAM, and SEAM are safe and feasible. In addition, the ethylene imine alkylating agent thiotepa was approved as a conditioning treatment prior to ASCT [54]. Due to its strong ability to penetrate the blood-brain barrier, thiotepa is often used as a component of high-dose therapy prior to ASCT in patients with central nervous system lymphoma [55]. Thiotepa is frequently administered in conjunction with etoposide (Etoposide), cytarabine (Cytarabine), and melphalan, collectively referred to as TEAM, with the option of incorporating cyclophosphamide to form the regimen known as TECAM. Recent studies show that the survival benefit and safety profile of TEAM and TECAM are similar to those of BEAM [54,56]. Data from patients with primary central lymphoma or central involvement are even more favorable [57]. Thus, high-dose chemotherapy regimens that include thiotepa seem to be another alternative option. More clinical data are needed to fully evaluate long-term survival benefits and safety.

Despite the expanding array of conditioning regimens, we continue to face a fundamental challenge of trading one toxicity for another, rather than achieving comprehensive optimization. Each regimen carries a distinct and significant spectrum of toxicities: pulmonary toxicity with BEAM, hepatic veno-occlusive disease with BuCyE, and renal toxicity with BeEAM. Currently, the evidence supporting most novel protocols is derived predominantly from retrospective, single-center, or small-sample studies. Consequently, these findings are susceptible to patient selection bias and are associated with a limited level of evidence.

The core objective must evolve into precision prevention, moving beyond passive toxicity management. A risk-stratified approach should guide clinical decisions: for instance, avoiding BCNU in patients with underlying pulmonary conditions and using bendamustine cautiously in older adults or those with renal impairment. Definitive breakthroughs, however, will rely on conducting rigorous randomized trials and identifying toxicity-specific biomarkers. These steps are critical to achieving truly individualized preconditioning, optimizing efficacy while curtailing toxicity.

Adding novel drugs to conditioning regimens.

With the advent of new drugs, some centers had tried to add rituximab, bortezomib (Bortezomib), Pola, or other drugs to high-dose chemotherapy in order to improve the survival and prognosis of patients.

Yttrium-90 ibritumomab tiuxetan or iodine-131 tositumomab plus BEAM. Due to the additional toxic side effects of TBI mentioned above, the trend in recent years has been to replace TBI in conditioning regimens with radio-immunotherapy (RIT). Radiolabeled anti-CD20 monoclonal antibodies show significant efficacy against both indolent and aggressive B-cell lymphomas [58]. The combination of RIT plus BEAM is at least not inferior to BEAM alone. Early studies show that for aggressive lymphoma with risk factors, Yttrium-90 ibritumomab tiuxetan combined with BEAM yielded results in a 2-yr PFS of 69 %, and an OS of 91 %, without any additional impact on the safety profile [59]. Another single-center study from Italy reached a similar conclusion; RIT plus BEAM versus BEAM alone resulted in a numerical improvement in 3-yr PFS (57 vs 48 %), although it was not statistically significant [60]. Subgroup analysis suggests that patients with early recurrence favor RIT plus BEAM, with significant differences in 3-yr PFS (78 vs 22 %, p = 0.016) and OS (83 vs 22 %, p = 0.001). Accordingly, the combination of Yttrium-90 ibritumomab tiuxetan plus BEAM seems to be a promising conditioning regimen for patients with early recurrence of B-cell lymphoma. Addition of another radiolabeled CD20 monoclonal antibody, iodine-131 tositumomab, combined with high-dose chemotherapy also yielded encouraging data. Among patients with high-risk and refractory B-cell lymphoma, the 5-year PFS and OS were 70 % and 72 %, respectively [61]. A phase 3 clinical study comparing the efficacy and safety of iodine-131 tositumomab plus BEAM or rituximab plus BEAM in recurrent DLBCL found no statistically significant difference in 2-year PFS (47.9 vs 48.6 %) or OS (61 vs 65.6 %) between the two groups [62]. However, the incidence of mucosal inflammation seemed slightly higher in the iodine-131 tositumomab group. Conditioning regimens that include RIT seem to have unique advantages with respect to improving the survival of high-risk and early recurrence patients. This hypothesis merits further examination in larger-scale prospective randomized studies.

V-BEAM.bortezomib (Bortezomib) is a proteasome inhibitor widely used to treat plasma cell tumors and MCL [[63], [64], [65], [66]]. In recent years, clinical research has explored adding bortezomib (Bortezomib) to the standard conditioning regimen for indolent NHL and MCL in order to obtain further survival benefits without increasing toxicities. In 2014, a single-center phase I/II study aimed to evaluate the safety and efficacy of adding bortezomib (Bortezomib) to the standard BEAM protocol (V-BEAM) for young patients with relapsed/refractory indolent NHL, transformed NHL, or MCL [67]. Among the 42 patients who received V-BEAM and ASCT, the PFS rates at 1 and 5 years were 83 % and 32 %, respectively, and the OS rates at 1 and 5 years were 91 % and 67 %, respectively. Grade 3 or higher TRAEs included FN (59 %), anorexia (21 %), peripheral neuropathy (19 %), and postural hypotension (16 %). Addition of bortezomib (Bortezomib) seemed to affect the implantation of hematopoietic stem cells. One patient had graft failure (2.38 %). It is worth paying more attention to peripheral neuropathy after treatment [68]. In this study, the incidence of grade 3 or higher bowel obstruction was 9 %, which may be related to peripheral neuropathy. This study compared the survival benefits with those of 26 MCL patients who had previously received BEAM. In the V-BEAM and BEAM groups, the 1-year OS rate was 96 % and 88 %, respectively, and the 5-year OS rate was 72 % and 50 %, respectively (P = 0.74). In comparison to the historical control group, V-BEAM appears not to have demonstrated a substantial benefit. Furthermore, the rise in TRAEs, notably neurotoxicity and graft failure, alongside the ongoing development of other novel therapeutics, seems to have diminished the enthusiasm for pursuing phase III randomized controlled trials.

R-BEAM. Rituximab, a human-mouse chimeric anti-CD20 monoclonal antibody, has been widely used as the first-line treatment for FL, DLBCL, chronic lymphocytic leukemia, and MCL [[69], [70], [71], [72]]. A large retrospective study in 2020 aimed to explore the effect of adding rituximab to BEAM (R-BEAM) on the prognosis of DLBCL [73]. In the R-BEAM and BEAM groups, the 4-year PFS rate was 48 % and 47 %, respectively (P = 0.61), and the 4-year OS rate was 58 % and 61 %, respectively (P = 0.83). The cumulative incidence of bacterial, viral, and fungal infections did not significantly differ between the two groups during the 100 days after ASCT. This study suggested that addition of rituximab to BEAM did not improve the survival benefits of DLBCL patients. According to the findings of the CIBMTR study, the routine use of rituximab in conjunction with the BEAM conditioning regimen is not recommended.

Pola-BeEAM. Pola is an ADC composed of a monoclonal anti-CD79b antibody coupled with a potent microtubule inhibitor that elicits survival benefits in refractory/relapsed DLBCL patients [74,75]. Data concerning combining Pola with standard conditioning regimens are limited. In 2022, a pilot study was undertaken to explore the integration of Pola with BeEAM (Pola-BeEAM) as a conditioning regimen for DLBCL [76]. All patients were successfully engraftment, with median times of platelet and neutrophil engraftment of 13 and 11 days, respectively, and no occurrence of TRM. The median follow-up was 15 months, and the PFS and OS rates were both 92 %. Common TRAEs included gastrointestinal reactions (grade 3 or higher, 58 %), FN (100 %), and ARF (17 %). The Pola-BeEAM group experienced a 58 % rate of grade III gastrointestinal toxicities. Given the study's small sample size and short follow-up, larger trials with extended follow-up are necessary to better understand the outcomes and toxicities of the Pola-BeEAM regimen in relapsed or primary refractory DLBCL patients. Additionally, advantageous population should be identified to determine which patients benefit most from this regimen.

Chi-BEAC. Chidamide, an oral selective histone deacetylase inhibitor, has been approved for the treatment of refractory/relapsed PTCL. Attempts have been made to incorporate chidamide into conditioning regimens in China. A phase II clinical trial was conducted to evaluate the safety and survival benefits of adding chidamide to BEAC (Chi-BEAC) in high-risk or relapsed/refractory invasive NHL patients [77]. The median times of neutrophil and platelet engraftment were 10 and 11 days, respectively, and most non-hematologic adverse events were grade 1–2. Grade 3 or higher TRAEs included FN (37.7 %), hyponatremia (24.6 %), and hypokalemia (21.7 %). The 100-day TRM rate was 0 %, and two patients died of pulmonary infections at 6.3 and 8 months, respectively. The 2-year PFS and OS rates were 81.1 % and 86.1 %, respectively. In this study, patients with angioimmunoblastic T-cell lymphoma (AITL), previously associated with a poor prognosis, demonstrated favorable outcomes. Of the 4 patients with advanced AITL, 3 underwent Chi-BEAC conditioning followed by ASCT. These patients were follow-up for a median duration of 31.3 months and remained in disease-free survival. Chi-BEAC seems to be an effective conditioning regimen for high-risk or recurrent NHL patients, with particular efficacy observed in those with T-cell lymphomas. It is worth noting that lung infections after transplantation can be fatal, especially in the context of the COVID-19 pandemic. Due to the small sample size and non-randomized design, this study has limitations and its findings must be confirmed by large sample studies.

Although initial data on the “BEAM-Plus” strategy appear promising, these findings should be interpreted with caution. The available studies share several methodological limitations, including small sample sizes, single-center designs, and a lack of randomization. As a result, the level of evidence remains limited, and the observed survival benefit may be influenced by patient selection bias. Moreover, the incorporation of novel agents almost invariably modifies the toxicity profile—exemplified by the neurotoxicity associated with V-BEAM and the severe gastrointestinal reactions observed with Pola-BeEAM. These potential additive toxicities represent a critical consideration in clinical decision-making. Moving forward, the central aim should shift from simply combining drugs to achieving precision in regimen selection. The focus of future research ought not to be on demonstrating which augmented regimen is more potent, but rather on identifying—through biomarkers such as lymphoma subtype, genetic profile, and minimal residual disease status—the subpopulation most likely to benefit from a specific intensified approach. This strategy would ultimately enable a balance between efficacy maximization and toxicity minimization. Among the multitude of emerging novel agents, well-designed prospective randomized controlled trials remain essential to conclusively establish the value of these personalized conditioning strategies.

The Era of Immunotherapy. Bispecific antibodies (BsAbs), a novel class of immunotherapy that recruit and activate a patient's endogenous T cells to eliminate tumor cells, represent an emerging modality for use as conditioning or bridging therapy [78,79]. Although the integration of BsAbs into conditioning regimens is still nascent, progress is accelerating. In the context of bridging therapy for patients awaiting stem cell harvest, BsAbs achieve rapid debulking of tumor burden, constituting a valuable bridge to subsequent cellular interventions. This effective cytoreduction ultimately increases the pool of eligible candidates and augments the success rates of ASCT [80]. In this setting, BsAbs act as a form of “functional conditioning.” Theoretically, integrating BsAbs before or during high-dose therapy (HDT) could exploit the patient's residual functional T cells to achieve synergistic immune-mediated tumor clearance, potentially resulting in deeper molecular responses. As potent immune activators, BsAbs hold significant promise in combination with HDT or as bridging therapy. Future clinical trials must be meticulously designed to define the optimal timing, sequence, and dosing of BsAbs, while carefully balancing efficacy against immune-related adverse events, such as cytokine release syndrome (CRS).

Chimeric antigen receptor T (CAR-T) cell therapy remains in the investigational stage as a potential preconditioning strategy prior to autologous stem cell transplantation for lymphoma. While this approach holds considerable promise, it is currently confined to the realm of clinical research [81]. CAR-T cells offer a theoretical advantage by targeting chemotherapy-insensitive tumor cells and altering the immunosuppressive microenvironment, thereby achieving deeper cytoreduction than traditional chemotherapy and thus fostering superior conditions for stem cell engraftment [82]. The paramount challenge is safety. The primary concern is the potential overlap between CAR-T-specific toxicities (e.g., CRS and immune effector cell-associated neurotoxicity syndrome) and the toxicity of conditioning chemotherapy. This cumulative toxicity could compromise patient tolerance, increasing the risk of treatment-related mortality and severe infections [83,84]. While this combined strategy holds the theoretical potential to break through current efficacy barriers, its significant toxicity profile necessitates that future clinical research first establishes the optimal timing and identifies suitable patient populations within a safety-focused framework.

The efficacy of conditioning regimens

We analyzed the OS and PFS data from existing clinical studies and compared the survival benefits of various regimens for different types of lymphoma (Table 1). As a conditioning regimen for HL, BEAM may yield more clinical benefits. The V-BEAM protocol seems to result in longer OS and PFS for FL patients; however, due to the small sample size and limited clinical trials, more research is needed. Pola-BeEAM and 131-iodine tositumomab plus BEAM may improve OS and PFS of DLBCL patients; however, due to the short follow-up, efficacy must be further confirmed. The survival benefits of various conditioning regimens are basically the same for MCL patients, and BeEAM and V-BEAM seem to improve OS and PFS.

Table 1.

Survival benefit of conditioning regimens for autologous stem cell transplantation.

Pathological subtype of lymphoma	Conditioning regimen	OS	PFS	Reference
HL	BEAM	79 % (3 y)	62 % (3 y)	Chen et al [20].
		95 % (5 y)	92 % (5 y)	William et al [21].
		84 % (10 y)	79 % (10 y)
	TECAM	90 % (3 y)	62 % (3 y)	Cohen et al [54].
	CBV	73 % (3 y)	60 % (3 y)	Chen et al [20].
		87 % (5 y)	73 % (5 y)	William et al [21].
		66 % (10 y)	59 % (10 y)
	BuCy	65 % (3 y)	51 % (3 y)	Chen et al [20].
	TBI	47 % (3 y)	43 % (3 y)
FL	BEAM	73 % (3 y)	52 % (3 y)	Chen et al [20].
	CBV	81 % (3 y)	67 % (3 y)
	BuCy	79 % (3 y)	60 % (3 y)
	TBI	71 % (3 y)	54 % (3 y)
	V-BEAM	100 % (3 y)	75 % (3 y)	William et al [67].
DLBCL	BEAM	61 % (4 y)	47 % (4 y)	Jagadeesh et al [73].
		58 % (3 y)	47 % (3 y)	Chen et al [20].
	CBV	55 % (3 y)	47 % (3 y)
	BuCy	52 % (3 y)	42 % (3 y)
	TBI	47 % (3 y)	42 % (3 y)
	R-BEAM	66 % (2 y)	49 % (2 y)	Vose et al [62].
		58 % (4 y)	48 % (4 y)	Jagadeesh et al [73].
	131-iodine tositumomab plus BEAM	61 % (2 y) 72 % (5 y)	48 % (2 y) 70 % (5 y)	Vose et al [62]. Vose et al [61].
	Pola-BeEAM	92 % (15 m)	92 % (15 m)	Stoffel et al [76].
	TECAM	75 % (3 y)	63 % (3 y)	Cohen et al [54].
MCL	BEAM	50 % (5 y)	43 % (5 y)	William et al [67].
		75 % (3 y)	62 % (3 y)	Chen et al [20].
	CBV	66 % (3 y)	57 % (3 y)
	BuCy	60 % (3 y)	47 % (3 y)
	TBI	66 % (3 y)	62 % (3 y)
	V-BEAM	82 % (3 y)	72 % (3 y)	William et al [67].
	BeEAM	93 % (3 y)	84 % (3 y)	Hueso et al [52].

OS, overall survival; PFS, progression-free survival; HL, Hodgkin lymphoma; FL, follicular lymphoma; DLBCL, diffuse large B-cell lymphoma; MCL, mantle cell lymphoma; BEAM, carmustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan; CBV, cyclophosphamide, carmustine, and etoposide (Etoposide); BuCy, busulfan and cyclophosphamide; TBI, total body irradiation; V, bortezomib (Bortezomib); R, rituximab; Pola, polatuzumab vedotin; BeEAM, bendamustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan; TECAM, thiotepa, etoposide (Etoposide), cyclophosphamide, cytarabine (Cytarabine), and melphalan; y, years.

Grade 3 or higher TRAEs and hematopoietic reconstitution

We summarized grade 3 or higher TRAEs and hematopoietic reconstitution with various conditioning regimens (Table 2). GBC/GBM, BeEAM, and Pola-BeEAM lead to a higher incidence of FN. BeEAM and Pola-BeEAM lead to a higher incidence of diarrhea. SEAM, GBC/GBM, Pola-BeEAM, and Chi-BEAC lead to the highest incidence of mucosal inflammation. The incidence of gastrointestinal reactions caused by SEAM is relatively high. TRM at 100 days is higher with BuCyE. TECAM and Pola-BeEAM result in more severe renal injury. SEAM results in the fastest implantation of granulocytes. GBC/GBM and Chi-BEAC result in faster implantation of platelets. There is less infusion of red blood cells with SEAM and less infusion of platelets with SEAM, BuCyE, and BeEAM.

Table 2.

Grade 3 or higher toxicity and hematopoietic reconstitution.

Conditioning regimen	Neutrophil engraftment (days)	Platelet engraftment (days)	Red blood cell infusion (units)	Platelet infusion (units)	Febrile neutropenia	Diarrhea	Mucositis	Nausea and vomiting		100 day TRM	Renal damage	Hepatotoxicity	Septicemia	Reference
BEAM	11	13.5	3.7	5.5	NA	NA	NA		NA	NA	NA	NA	NA	Jo Jae-Cheol et al [22].
BEAC	12	16	6.5	10.3	NA	NA	NA		NA	NA	NA	NA	NA	Jo Jae-Cheol et al [22].
BuCyE	12	14.5	2	3	NA	NA	NA		NA	6.5 %	NA	NA	NA	Berber et al [36].
GBC/GBM	10	11	NA	NA	100 %	19.8 %	45.1 %		18.7 %	NA	NA	23.1 %	NA	Liu Huimin et al [39].
SEAM	9.5	12	1.5	3	52.6 %	11.3 %	36.1 %		21.6 %	2.1 %	0 %	2.1 %	NA	Zhang Lihong et al [44].
BeEAM	11.7	12.6	2.3	4.3	100 %	NA	4.9 %		NA	2.4 %	2.4 %	NA	NA	Hahn et al [53].
	10	13	NA	NA	NA	35 %	26 %		NA	0 %	NA	NA	NA	Visani et al [50].
V-BEAM	11	16	NA	NA	59 %	4 %	0 %		2 %	NA	2 %	NA	23 %	William et al [67].
Pola-BeEAM	11	13	3	6	100 %	92 %	100 %		NA	NA	17 %	NA	NA	Stoffel et al [76].
Chi-BEAC	10	11	NA	NA	37.7 %	NA	45.1 %		18.7 %	0 %	NA	7.3 %	NA	Xia et al [77].
TECAM	12	17	NA	NA	NA	NA	63.7 %		NA	2.8 %	16.1 %	NA	30.3 %	Cohen et al [54].
TEAM	10	12	NA	NA	NA	NA	NA		NA	2.5 %	NA	NA	NA	Deveci et al [55].
131-iodine tositumomab plus BEAM	10	12	NA	NA	NA	5 %	5 %		5 %	0 %	3 %	3 %	20 %	Vose et al [61].

TRM, treatment-related mortality; BEAM, carmustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan; BEAC, carmustine, etoposide (Etoposide) (Etoposide), cytarabine (Cytarabine), and cyclophosphamide; BuCyE, busulfan, cyclophosphamide, and etoposide (Etoposide); GBC, gemcitabine, busulfan, and cyclophosphamide; GBM, gemcitabine, busulfan, and melphalan; SEAM, semustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan; BeEAM, bendamustine, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan; V, bortezomib (Bortezomib); Pola, polatuzumab vedotin; Chi, chidamide; TECAM, thiotepa, etoposide (Etoposide), cyclophosphamide, cytarabine (Cytarabine) (Cytarabine), and melphalan; TEAM, thiotepa, etoposide (Etoposide), cytarabine (Cytarabine), and melphalan.

Conclusions

The conditioning landscape for ASCT in lymphoma has evolved significantly, transitioning from TBI and high-dose chemotherapy toward novel drug combinations and, more recently, integrated immunotherapy. Owing to its substantial short- and long-term toxicities, particularly the elevated risk of secondary malignancies, TBI has been largely superseded by high-dose chemotherapy alone or combined with radioimmunotherapy. The BEAM regimen remains a cornerstone in this setting; however, limitations associated with BCNU have driven the development of alternative regimens—such as GBM/GBC, BeEAM, and SEAM—which have demonstrated safety and feasibility. Preliminary clinical data on augmenting conditioning with novel agents (e.g., polatuzumab vedotin, yttrium-90 ibritumomab tiuxetan, iodine-131 tositumomab) appear promising, though their evaluation remains limited by small sample sizes and necessitates further validation.

Looking ahead, conditioning strategies are poised to shift toward precisely orchestrated chemo-targeted-immunotherapy combinations. To realize truly individualized therapy, future research should prioritize the development of minimal residual disease–guided algorithms capable of identifying patients most likely to benefit from intensive immunotherapy-based conditioning. Such an approach would maximize efficacy while limiting toxicity and reducing healthcare costs for those unlikely to respond. Ultimately, the goal is to develop optimized, personalized regimens that are low in toxicity, highly effective, and tailored to lymphoma subtypes and patient-specific comorbidities.

Ethics and consent to participate declarations

Not applicable.

Funding declaration

This work was supported by grants from the Science and Technology Development Project of Jilin Province (20210401174YY, to W.Z.).

CRediT authorship contribution statement

Hao Tian: Writing – original draft. Ruiqi Wang: Writing – original draft. Dan Cong: Writing – review & editing, Data curation. Yuansong Bai: Writing – review & editing, Data curation. Wenlong Zhang: Writing – review & editing, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Wenlong Zhang reports financial support was provided by Science and Technology Development Project of Jilin Province. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.