We compared the hematopoietic reconstitution, economic expense, safety, and survival benefits of the two conditioning regimens for relapsed, refractory DLBCL. Our research has certain advantages. First, unlike previous comparisons of conditioning regimens in whole lymphoma populations, the selection of conditioning regimens for specific subtypes in this study allows for individualized and precise treatment. Additionally, due to the limitations of carmustine pulmonary toxicity and drug shortages, the low-toxicity and accessible bendamustine was explored as an alternative. Finally, the economics of the BeEAM regimen are explored in the context of the global economic downturn following the covid-19 epidemic. Such studies are rarely reported, and our study has clinical utility to some extent.
There seems to be little difference between BeEAM and BEAM in terms of hematopoietic reconstruction. Established studies suggest a median time to neutrophil implantation of 10 to 11.7 days for BeEAM as a conditioning regimen and a median time to platelet implantation of 11 to 15 days11, 15, 17, 18. The median time to hematopoietic reconstitution in our study was 10.2 days for neutrophil implantation and 13.23 days for platelet implantation in the BeEAM group, which is consistent with previous findings. The time to hematopoietic reconstitution in the BEAM group was also similar to the results of existing studies19. The similarity of hematopoietic reconstruction time between the BeEAM and BEAM groups ensures that BeEAM regimen does not increase the incidence of infections and bleeding events, making it a better safety profile.
TRAEs of the BeEAM group were mild and manageable. In our study, grade ≥ 3 TRAEs in the BeEAM group were found to be predominantly hematologic toxicities, including granulocyte deficiency as well as thrombocytopenia. Non-hematologic adverse events such as mucositis, gastrointestinal reactions, renal insufficiency, and transaminase elevations, most of which are ≤ grade 2, are relatively safe and manageable. Previous studies have reported that non-hematologic events in grades 3 to 4 of the BeEAM protocol were predominantly mucositis (26%-69%)17, 18. The absence of grade 3 to 4 non-hematologic adverse events in our study may be related to the small sample size of this study, the absence of predefined observational indicators in the retrospective study, and the bias in defining and documenting adverse events in the clinical data files. The adverse event of greatest concern with bendamustine was acute renal failure (ARF), which occurred at a rate of 46% with a median time of occurrence of day 5 post-transplantation and returned to normal by day 10 post-transplantation16. In a multicenter study conducted by Chantepie et al. in 2018, 27.9% of cases of lymphoma patients pretreated with the BeEAM regimen developed ARF on day 6 after the start of pretreatment, and creatinine levels returned to normal within a median time of 10 days in 86% of these patients15. Age over 55 years, prior renal insufficiency, bendamustine dosage higher than 160 mg/m2/day, and prior platinum-based therapy may increase the risk of ARF. In our study, two patients in the BeEAM group developed ARF on day 6 after autologous stem cell transplantation. one was associated with co-infection of the abdominal cavity, and the other had a urinary pH of 5.5 at the start of preconditioning, which may be related to failure to adequately hydrate and alkalinize before the start of preconditioning. These 2 patients recovered normal renal function 8–10 days after autologous stem cell transplantation after anti-infection, adequate rehydration, and supportive symptomatic treatment. The time of onset and end of ARF were consistent with the results of previous studies. ARF events did not significantly increase in the BeEAM group compared with the BEAM group in our study. The possible reasons for this are as follows: no underlying renal disease and better renal function in the enrolled patients; adequate rehydration for hydration and alkalinization before performing autologous stem cell transplantation; the dosage of bendamustine was 140 mg/m2/day, which is relatively safe; and fewer platinum-based chemical drugs were applied in the past. The 2 patients who developed ARF in the BeEAM group did not have prolonged hematopoietic reconstruction time and did not have an increased incidence of other safety events. As an alternative to BEAM, BeEAM is safe and feasible.
Although there were no statistical differences in the number of days of hospitalization and hospitalization costs, there was a numerical advantage in hospitalization costs for the BeEAM group. First, diarrhea occurred in a slightly higher proportion and lasted longer days in the BEAM group, and more patients required intravenous nutritional meal replacement. In addition, the duration of fever as well as the duration of antibiotic application was longer in the BEAM group. This may have contributed to the increased economic expenditure of the patients. The BeEAM regimen was found to be beneficial to patients in terms of financial savings in our study, and the results of future prospective studies with larger sample sizes are to be expected.
In our study, the CRR was higher in the BeEAM group than in the BEAM group after ASCT, which may be related to the bias of the disease remission status before transplantation. Previous studies have shown 3-year PFS of 42–47% and OS of 47–58% in patients with relapsed and refractory DLBCL who were conditioning with regimens such as BEAM20. In our study, the estimated 3-year PFS and OS in the BeEAM group were 85% and 90%, respectively, and the difference in values may be related to the small sample size and insufficient median follow-up time. There was no statistically significant difference in PFS and OS compared to the BEAM group in this study, suggesting that BeEAM is a valid alternative to BEAM.
This study has some limitations. First, the small sample size and lack of a random sampling component may cause selective bias in the results. In addition, patients with DLBCL are highly individualized and heterogeneous, and consistency of results across samples before conditioning and after ASCT cannot be guaranteed. Third, the short follow-up period and the loss of some patients to follow-up do not reflect the survival benefit objectively and accurately. Finally, subjective factors are unavoidable in retrospective studies, and the completeness of medical record information is not controlled by the trial design and is subject to confounding factors and bias. The results of this study still need to be confirmed by prospective clinical studies with large samples in the future.