Busulfan‐containing conditioning regimens in allogeneic hematopoietic stem cell transplantation for acute lymphoblastic leukemia: A Taiwan observational study

Abstract Background Allogeneic stem cell transplantation (allo‐HSCT) is the ultimate cure for acute lymphoblastic leukemia (ALL). Aim This study was performed to compare the outcomes of ALL patients receiving busulfan (Bu) with cyclophosphamide (Cy)‐based or total body irradiation (TBI)‐based regimen in a Chinese population. Methods We enrolled 224 adult patients with ALL who received allo‐HSCT at National Taiwan University Hospital between 1997 and 2016. Results The median age at transplantation was 33 years. Before allo‐HSCT, 75.9% of patients attained first or late complete remission. A total of 141 patients (62.9%) received Bu/Cy‐based conditioning, either myeloablative (MA) or reduced‐intensity stem cell transplantation (RIST), and 83 patients received a TBI‐based regimen (MA‐TBI). Patients receiving the MA‐Bu regimen had longer relapse‐free survival (RFS) than those receiving the MA‐TBI regimen (median, 24.1 vs. 6.7 months, p = .044). There was no difference in overall survival (OS, MA‐Bu vs. MA‐TBI vs. RIST‐Bu: 39.4 vs. 28.2 vs. 13.1 months, p = .276), treatment‐related mortality (TRM), or incidences of grade 3–4 acute graft‐versus‐host disease (GvHD). Among patients receiving identical GvHD prophylactic regimens, there was no difference between MA‐Bu and MA‐TBI groups regarding the incidence of grade 3–4 acute GvHD, grade 2–4, and all‐grade chronic GvHD. In subgroup analysis, patients receiving oral busulfan had comparable RFS and OS to the intravenous busulfan group (p = .436 and p = .236, respectively), but a higher TRM (25% vs. 9.8%, p = .016). In the multivariable analysis, disease status before allo‐HSCT was the only risk factor impacting RFS and OS. Conclusion In summary, patients receiving Bu/Cy‐based or TBI‐based regimens as conditioning had similar results in terms of OS, TRM, and acute GvHD, whereas the use of myeloablative Bu/Cy resulted in a better RFS. A Bu‐based regimen could be an alternative conditioning choice for patients who are ineligible to receive TBI. Prospective and randomized controlled trials are warranted to validate the long‐term outcomes.


| INTRODUCTION
Advances in the treatment of adult acute lymphoblastic leukemia (ALL) by small molecular agents and chimeric antigen receptor T-cell therapy have greatly improved patient survival in the last decade. [1][2][3][4][5][6][7] Meanwhile, recurrence is common in the post-therapy course, which poses a challenge to long-term remission. [8][9][10][11] Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is thus still considered the ultimate cure for ALL. [12][13][14] An essential component of allo-HSCT is the conditioning regimen, which eradicates the cancer cells and provides stem cell niches in the host bone marrow for the new stem cells. Total body irradiation (TBI) is effective against a variety of malignancies without sanctuary sites, such as the central nervous system, and therefore has been the gold standard conditioning regimen. 15,16 Complications of TBI include delayed growth and development in children, interstitial pneumonitis and secondary malignancies. 15,17,18 TBI has been widely used in the Western world, while reports of its treatment effectiveness in the Chinese population have been rare. Busulfan (Bu) is an alkylating agent that has a potent effect on leukemia and can also serve as a common conditioning agent along with cyclophosphamide. 19,20 Nevertheless, the absorption of Bu in the gastrointestinal tract is quite variable among patients. 21,22 Some adverse effects, such as sinusoidal obstruction syndrome (SOS; previously known as veno-occlusive disease [VOD]), restrict patients' and physicians' choices. [23][24][25] Previous studies have confirmed the safety and efficacy of targeted-dose Bu, which reduces patients' risks of SOS, treatment-related mortality (TRM), and relapse. [26][27][28][29] The comparison of the treatment efficacies of TBI/cyclophosphamide (Cy) and busulfan (Bu)/Cy as conditioning regimens, both of which are major options for conditioning regimens for ALL, has long been unclear. Patients receiving different regimens might have moderately different outcomes from the perspective of relapse-free survival (RFS), TRM, and the cumulative incidence of acute or chronic graft-versus-host disease (GvHD). However, no single regimen has clear benefits in terms of overall survival (OS). [30][31][32][33][34] This real-world observational study was performed to compare the outcomes of Chinese ALL patients receiving a Bu-based or a TBIbased regimen, and pre-treatment parameters and therapy modalities were analyzed for risk stratification and survival analyses.

| Patients
We enrolled 224 ALL patients who received their 1st allo-HSCT at

| Prophylaxis of graft-versus-host disease
We used cyclosporin and methotrexate for GvHD prophylaxis in patients receiving MA-TBI and MA-Bu conditioning; cyclosporin and mycophenolate mofetil were used in the RIST-Bu conditioning group.
Rabbit anti-thymoglobulin (ATG) was given to the patients who received Hematopoietic stem cells from the human leukocyte antigen (HLA)-mismatched donors or the unrelated donor. A total dose of 5 mg/kg of body weight was given to the patients who received stem cells from the matched unrelated donors, and a total dose of 6 mg/kg of body weight was given to the patients who received stem cells from the haploidentical donors or mismatched unrelated donors. The ATG was divided into 2-3 days and given before the infusion of Hematopoietic stem cells. follow-up or death, whichever occurred first. TRM was defined as a death resulted from any cause other than relapse.

| Statistical analysis
We used the Mann-Whitney U test to compare the medians of con-   Figure 1C) among all regimen groups. However, patients who took oral busulfan had a higher risk of TRM than those who received IV busulfan (25% vs.

| Graft-versus-host disease
The cumulative incidences of grade 3-4 acute GvHD at day +100 were not significantly different among the three groups (MA-Bu vs. MA-TBI vs. RIST-Bu: 15.8% vs. 10.8% vs. 15.2%, p = .445, Figure 1D), whereas the cumulative incidence of all-grade chronic GvHD in the RIST-Bu group was significantly higher than those in

| Extramedullary disease
In this cohort, patients with extramedullary disease at diagnosis did not have significantly shorter OS than those without (median, with extramedullary disease vs. without, 28.1 vs. 25.8 months, p = .898, Figure 2A). This might imply that allo-HSCT can overcome the  Figure 2C).

| Multivariable analysis
For multivariable analysis, we included parameters with a p value <.15 in univariate Cox regression analysis ( there were no significant differences between the Bu-based and TBIbased groups in terms of OS, and MA-Bu conditioning might improve RFS in eligible patients. These results imply that the Bu-based regimen might improve patient outcomes in adult patients with ALL by reducing treatment toxicity and mortality. In the meantime, strategies for the prevention and salvage of disease relapse, which accounted for more than 50% of the deaths, should also be further investigated and improved.
In summary, this study provided a risk stratification and survival analysis of ALL patients undergoing allo-SCT and demonstrated that a Bu-based regimen could be an alternative conditioning choice for patients who are ineligible to receive TBI. Larger-scale, prospective and randomized controlled trials are challenging but warranted to compare and validate the long-term outcome of patients receiving Bu-based and TBI-based conditioning before transplant.

ACKNOWLEDGMENTS
We would like to acknowledge the service provided by Department