Objectives: We evaluated the safety and efficacy of 2 conditioning regimens (busulfan/fludarabine vs modified busulfan/cyclophosphamide) in patients with acute myeloid leukemia undergoing haploidentical hematopoietic stem cell transplant.
Materials and Methods: Twenty patients with primary acute myeloid leukemia had been randomized into busulfan/fludarabine and modified busulfan/cyclophosphamide groups. We retrospectively compared hematopoietic engraftment, regimen-related toxicity, graft-versus-host disease, transplant-related mortality, leukemia-free survival, and overall survival between the groups.
Results: All patients achieved engraftment with 100% donor chimerism. The median times for the neutrophil and platelet engraftment in the busulfan/fludarabine and modified busulfan/cyclophosphamide groups were 14.1 versus 14.3 days and 12.7 versus 12.2 days, respectively. Significantly lower incidences of pretreatment toxicity, blood transfusion, and virus activation were observed in the busulfan/fludarabine group. Acute grade 1 graft-versus-host-disease developed in all patients, which was successfully controlled with methylprednisolone. There were no significant differences in engraftment, graft-versus-host disease, leukemia-free survival, and overall survival between groups. Both of these conditioning regimens achieved stable engraftment. Regimen-related toxicity in the busulfan/fludarabine group was well tolerated compared with that in the modified busulfan/cyclophosphamide group, without an increase in relapse rate.
Conclusions: Our results demonstrated that myeloablative busulfan/fludarabine might be a highly effective and low-toxicity alternative for patients with acute myeloid leukemia.
Key words : Busulfan, Cyclophosphamide, Fludarabine, Myeloablative conditioning regimen
Acute myeloid leukemia (AML) comprises a group of high-grade clonal neoplasms of myeloid progenitor cells. Until now, allogeneic hematologic stem cell transplant (allo-HSCT) has been the only curative method for adult AML patients, excluding those with acute promyelocytic leukemia.1 However, only half of patients who can benefit from the transplant can find a human leukocyte antigen (HLA)-matched sibling donor or matched unrelated donor. Transplant using umbilical cord blood is hampered by the low amount of hematopoietic stem cells.2 This is especially common in China because many families have only 1 child. Furthermore, it usually takes 3 or more months from initiation of unrelated donor search to the identification of an appropriate donor; thus, it is not realistic for a high-risk AML patient who urgently needs HSCT. Haploidentical HSCT (haplo-HSCT) is an alternative transplant strategy that is relatively quick, economic, and feasible for many patients with leukemia. Many AML patients undergoing haplo-HSCT have achieved leukemia-free survival (LFS).3
Over the past decades, the results of haplo-HSCT have improved significantly due to donor matching, conditioning regimen optimization, new prophylactic agents for graft-versus-host disease (GVHD) and virus reactivation, and more advanced supportive care. Compared with matched unrelated donor and umbilical cord blood, haplo-HSCT has several superiorities, including the following: (1) donors for haplo-HSCT are readily available, with much less time for matching and transplant preparation; (2) donor lymphocyte infusion (DLI) is convenient to obtain; and (3) embedding rate is similar to other types of HSCT and the development of GVHD can generally be controlled with a combination of immunosuppression agents. However, the same challenges facing patients who undergo allo-HSCT, including primary/secondary hematopoietic failure, transplant-related mortality (TRM), significant GVHD, prolonged immune deficiency, and primary disease relapse, remain unresolved.3,4 The TRM rate is approximately 20% to 30% during the first 100 days after transplant and is the main cause of early death in allo-HSCT patients.5 Nonmyeloablative and reduced-intensity conditioning regimens have been increasingly developed and have shown much lower regimen-related toxicities; however, relapse becomes more prevalent.6 To decrease the primary disease relapse rate, a more intensive conditioning regimen was considered, which resulted in even higher regimen-related toxicities and TRM. Consequently, it is urgent to develop more effective and well-tolerated conditioning regimens for haplo-HSCT.
Busulfan combined with cyclophosphamide is the classic myeloablative conditioning (MAC) regimen. It was introduced to clinical HSCT in the 1980s and remains a popular choice in allo-HSCT.7 Based on the classic busulfan plus cyclophosphamide combination, the modified busulfan/cyclophosphamide (cytarabine at 2 g/m2/4 d, cyclophosphamide at 1.8 g/m2/2 d, and busulfan at 3.2 mg/m2/3 d) conditioning regimen is a successful model that was reported by Xiao-Jun and associates and has obtained great success in patients undergoing haplo-HSCT.8 However, there are many weak points of this regimen, including delayed immune reconstitution, a high reactive rate of the virus, and the need for blood transfusion. Many researchers have further modified this conditioning regimen using various methods, such as adding fludarabine. Fludarabine was initially used in the reduced-intensity conditioning regimen as a purine analogue with strong antineoplastic and immunosuppressive activity.9 It can inhibit lymphocyte proliferation, induce lymphocyte apoptosis, and cause strong immunosuppression. However, no similar report exists concerning the use of MAC regimens that include busulfan/fludarabine in AML patients who have undergone haplo-HSCT.
In this study, we conducted a retrospective, randomized, single-center study to compare the modified busulfan/cyclophosphamide and the busulfan/fludarabine conditioning regimens for AML patients at their first complete remission who had been undergoing haplo-HSCT. The safety, effects, and toxicity of the 2 conditioning regimens were analyzed.
Materials and Methods
Patients and eligibility criteria
We analyzed 20 AML patients who were seen at our hospital from January 2013 to December 2016 (13 males and 7 females). Compatibility for HLA-A, HLA-B, HLA-C, HLA-DRB1, and HLA-DQB1 was assessed by low- and intermediate-resolution DNA typing. The median age was 29.6 years (range, 18-54 y). All patients were at first complete remission and had no available sibling or unrelated HLA-matching donors for allogeneic transplant. All patients had genetic data. The genetic subgroups according to the criteria were all intermediate, except one with FLT3/ITD and cKit mutation. The haplo-HSCTs were carried out from 3 to 6 chemotherapy cycles before transplant. None of the patients had active infections, and the renal and liver functions were normal. This study was performed in accordance with the modified Helsinki Declaration. Informed consent was obtained from the donors and recipients. The study was approved by Medical Ethical Committee of Shandong Provincial Hospital, which is affiliated with Shandong University (2013-048). There were no significant differences between groups regarding patient age, sex, inductive and consolidative chemotherapies, and genetic subgroups. Patient characteristics are shown in Table 1.
All patients were randomized and subgrouped into modified busulfan/cyclophosphamide (n = 10) or busulfan/fludarabine (n = 11) groups. Modified busulfan/cyclophosphamide conditioning comprises cytarabine (2 g/m2/d, from days -10 to -9), busulfan (3.2 mg/kg/d, from days -8 to -6), cyclophosphamide (1.8 g/m2/d, from days -5 to -4), semustine (250 mg/m2, day -3), and anti-human antithymocyte globulin (ATG; 2.5 mg/kg/d, from days -5 to -2). Busulfan/fludarabine conditioning comprises busulfan (3.2 mg/kg/d, from days -7 to -4), fludarabine (30 mg/m2/d, from days -6 to -2), and anti-human ATG (2.5 mg/kg/d, from days -4 to -2). From June 2014, idarubicin was added to the busulfan/fludarabine regimen (12 mg/m2/d, from days -8 to -7).
Source of hematologic stem cells
The single specific primer-polymerase chain reaction (PCR) technique was used to type HLA-A, HLA-B, HLA-C, HLA-DRB1, and HLA-DQB1. HLA matching was evaluated from the intermediate- and high-resolution results of patient and donors. All patients were transplanted from a haploidentical relative donor and received peripheral blood stem cell grafts.
Prophylaxis of graft-versus-host disease and antimicrobial and supportive
In both groups, cyclosporine, mycophenolate mofetil, and short courses of methotrexate (15 mg/m2 on day +1, 10 mg/m2 on days +3, +5, +11) were combined to prevent acute GVHD (aGVHD). The serum concentration of cyclosporine was maintained at 250 to 350 ng/mL during the first 3 months and at 200 to 300 ng/mL during months 3 to 6 after transplant. The dose of cyclosporine was decreased gradually if no obvious GVHD manifestation occurred from 6 months and thereafter. For patients with positive minimal residual disease detection, cyclosporine was decreased more quickly than planned, and DLI was considered if there was no obvious sign of GVHD above grade 2. The first-line therapy for aGVHD was methylprednisolone (about 1-2 mg/kg/d) with continuing cyclosporine; for chronic GVHD (cGVHD), prednisone with or without cyclosporine was administered.
All patients received antibacterial, antiviral, and antifungal prophylaxis. Sulfamethoxazole was used to prevent Pneumocystis jiroveci infection, and levofloxacin was used to prevent bacterial infections. Fluconazole was administered as prophylaxis for fungal infections. For patients who had a history of invasive fungal infection, voriconazole or posaconazole was used instead of fluconazole. Ganciclovir was used to prevent cytomegalovirus (CMV) infection or reactivation before transplant, with acyclovir added after stem cell infusion.
Leukocyte-depleted and irradiated blood products, including red blood cells and platelets, were given when hemoglobin levels were < 60 mg/L or platelet counts were <10 × 109/L (criteria upregulated to 70 mg/L and 20 × 109/L if the patient had fever, ischemia, and hypoxia symptoms or other bleeding complications).
All patients received granulocyte colony-stimulating factor (5 μg/kg/d) from day 6 posttransplant. Absolute neutrophil count engraftment was defined as an excess of neutrophilic granulocyte cells greater than 0.5 × 109/L for 3 consecutive days. Platelet engraftment was defined as the first day on which the platelet count was greater than 20 × 109/L for 7 consecutive days without platelet infusion. The severity of hemorrhagic cystitis (HC) was graded according to the following criteria: 1 = microscopic hematuria; 2 = macroscopic hematuria; 3 = maroscopic hematuria with the presence of clots and/or decrease in hemoglobin levels necessitating blood transfusions; and 4 = life-threatening bleeding, not responding to treatment, and necessitating surgical intervention. Acute GVHD and chronic GVHD were assessed and graded using the consensus conference guidelines.10,11 Other toxicities were scored using National Cancer Institute criteria.12
Laboratory monitoring posttransplant
Bone marrow aspiration was performed at 1, 3, 6, 9, and 12 months posttransplant. For high-risk patients or patients who were positive for minimal residual disease, bone marrow aspiration was also carried out at 4.5 months posttransplant. Minimal residual disease was detected by real-time PCR and flow cytometry. Donor-recipient chimerism was evaluated with fluorescein in situ hybridization for sex-mismatched transplants and with short tandem repeat analysis for sex-matched transplants.
Excluding formal laboratory tests such as
C-reactive protein, procalcitonin, β-(1,3)-D dextran (G test), and galactomannan, all patients received serum monitoring of CMV and Epstein-Barr (EBV) by real-time PCR twice per week until month 3 posttransplant. If HC was suspected, polyomavirus BK virus would also be tested. Cytomegalovirus reactivation was defined as a positive PCR of > 500 copies of DNA/mL, and EBV reactivation was defined as positive PCR of > 1000 copies of DNA/mL.
Primary and secondary endpoints
Primary study endpoints were engraftment time, aGVHD and cGVHD, viral reactivation, infection incidence, blood cell infusion, adverse effects of conditioning, regimen-related mortality, and TRM. Secondary endpoints were overall survival (OS), LFS, relapse, and nonrelapse mortality. The main outcome data were collected, and adverse events were assessed on days 30, 60, 100, and 180, and then at 12, 18, and 24 months after transplant, followed by once per year thereafter.
Comparisons of categorical variables were made using chi-square and Fisher exact tests for all numbers. Differences between numerical variables were calculated using a 2-sample t test. The incidence of time-dependent variables was estimated by the Kaplan-Meier method. Numerical variables were analyzed as categories based on their value below or above the median of the entire cohort. Patient age, sex, conditioning regimen, mononuclear cell dose, CD34-positive cell dose, aGVHD, and cGVHD were analyzed with univariate and multivariate analyses. Intervals were measured from day of transplant until last day of follow-up, transplant-related death, or relapse. SPSS software package 22.0 (SPSS Inc/IBM, Armonk, NY, USA) was used for all data analyses. All statistical tests were two-sided, and statistical significance was defined as P ≤ .05.
Results for the 20 AML patients who underwent haplo-HSCT are shown in Table 2 according to the 2 subgroups.
Engraftment and chimerism
For the 2 subgroups, the median doses of infused peripheral blood mononuclear cells and CD34-positive cells in busulfan/fludarabine versus modified busulfan/cyclophosphamide were 10.344 × 109/kg (range, 6.07-15.70 × 108/kg) versus 10.242 × 109/kg (range, 6.90-16.23 × 108/kg) and 3.552 × 106/kg (range, 1.71-7.69 × 106/kg) versus 3.397 × 106/kg (range, 1.38-7.21 × 106/kg), respectively (P > .05). All patients achieved engraftment with 100% donor chimerism by day 28 posttransplant. Days to initiation of neutrocytopenia were 9.1 days (range, -5 to none) versus 6.8 days (range, -5 to +8) in the busulfan/fludarabine versus modified busulfan/cyclophosphamide groups, respectively (P < .05). Most patients developed neutrocytopenia before stem cell infusion. The median duration of neutrocytopenia was 11.4 days (range, 0-21) and 14.6 days (range, 3-22) in the busulfan/fludarabine versus modified busulfan/cyclophosphamide groups, respectively (P > .05). The median times for the neutrophil and platelet engraftment in the busulfan/fludarabine versus modified busulfan/cyclophosphamide groups were 12.7 days (range, 10-24 d) versus 12.2 days (range, 10-22 d) and 14.1 days (range, 9-39 d) versus 14.3 days (range, 10-21 d), respectively (P > .05).
Adverse effects in the modified busulfan/cyclophosphamide group included noninfectious fever, nausea and vomiting, diarrhea, earlier emergence of leukocytopenia and thrombocytopenia, and mucositis. In the busulfan/fludarabine group, adverse effects included mainly nausea and vomiting, diarrhea, and mucositis. Fever mostly developed on the day that ATG was given, not excluding ATG-associated fever.
Blood cell transfusion
The infusion rates of red blood cells and platelets before engraftment were higher in the modified busulfan/cyclophosphamide group than in busulfan/fludarabine group. The median amount of red blood cell transfusions was 3 U (range, 0-8 U) in the busulfan/fludarabine group and 4.6 U (range, 2-8 U) in the modified busulfan/cyclophosphamide group (P > .05). The amount of platelet transfusion was 2.5 U (range, 0-8 U) in the busulfan/fludarabine group and 5.5 U (range, 2-8 U) in the modified busulfan/cyclophosphamide group (P < .05).
We collected the data of serum copies of CMV and EBV in 20 patients using PCR from day 28 posttransplant. Ten patients exhibited CMV-positive loading (> 500 copies/mL), and 6 patients showed EBV-positive loading (>500 copies/mL). Patient distribution in the busulfan/fludarabine and modified busulfan/cyclophosphamide groups was 3/2 (CMV/EBV) and 7/4, respectively. Five patients exhibited both CMV- and EBV-positive serum simultaneously (1 in the busulfan/fludarabine group and 4 in the modified busulfan/cyclophosphamide group; P < .05). The bacterial and fungal infection rates were similar in the 2 groups.
Grade 1 aGVHD (symptoms included maculopapular rash and/or alimentary symptoms, with other associated causes excluded) developed in all patients. Mean time to aGVHD development was at day 17 in the busulfan/fludarabine and at day 15 posttransplant in the modified busulfan/cyclophosphamide group, with all patients successful treated with methylprednisolone (1 mg/kg/d) before day 30. In the busulfan/fludarabine group, 3/10 patients developed grade 2 to 4 aGVHD, with 1/10 patients developing grade 2 to 4 aGVHD in the modified busulfan/cyclophosphamide group (P > .05). The patient in the modified busulfan/cyclophosphamide group who developed serious aGVHD had intestinal tract and liver (grade 4) involvement, relapsed rapidly, and died of cerebral hemorrhage. One patient in the busulfan/fludarabine group relapsed at 14 months posttransplant and received DLI. She developed serious grade 4 aGVHD involving the liver and intestine but was successfully treated with FK506 and anti-CD25 monoclonal antibody.
In total, 10 of 20 patients developed cGVHD in the 2 groups, including 4 in the busulfan/fludarabine group and 6 in the modified busulfan/cyclophosphamide group (P > .05). Among these patients, most cases involved organs, including the skin, oral mucosa, and liver. Prednisone combined with cyclosporine was used to treat cGVHD.
Leukemia relapse, nonrelapse mortality, leukemia-free survival, and overall
Of 20 patients, 5 relapsed within 2 years after transplant: 3 in the busulfan/fludarabine group and 2 in the modified busulfan/cyclophosphamide group (P > .05). The total relapse rates were 5% (1/20) and 20% (4/20) within 6 and 12 months after transplant, respectively. In the busulfan/fludarabine group, 1 patient relapsed at 14 months posttransplant and achieved sustained molecular remission after salvage DLI. The other 2 patients died of leukemia relapse at 8 and 10 months posttransplant. In the modified busulfan/cyclophosphamide group, one patient with FLT3/ITD and cKit mutation relapsed at 2 months and died of cerebral hemorrhage. The other relapsed at 11 months and died of serious infection and proteopathy. The total nonrelapse mortality rate was 5% (1/20). The LFS rates for patients administered busulfan/fludarabine versus modified busulfan/cyclophosphamide were 411.54 days (range, 211.12-611.96 d) versus 582.00 days (range, 327.98-836.03 d), respectively (P > .05). The OS rates were similar between the busulfan/fludarabine and the modified busulfan/cyclophosphamide groups at 2 years, with OS of 516.12 days (range, 238.27-793.98 d) and 610.56 days (range, 341.67-879.46 d), respectively (P > .05) (Figure 1).
In this study, we compared the safety and toxicity of 2 conditioning regimens for haplo-HSCT in 20 AML patients. Our data provide robust support for the safety and efficacy of the busulfan/fludarabine compared with modified busulfan/cyclophosphamide conditioning regimen. There was no increase in either GVHD or relapse. The OS and LFS rates were similar in the 2 groups. The busulfan/fludarabine conditioning regimen may be an alternative MAC regimen for AML patients undergoing haplo-HSCT.
It is a great milestone for haplo-HSCT to be introduced for clinical use because of its ability to conquer the immune barrier. Based on classic busulfan/cyclophosphamide, modified busulfan/cyclophosphamide has been confirmed to be a successful conditioning regimen that is often used in haplo-HSCT, in which cytarabine and semustine are added and busulfan is reduced to 3 days of administration. An important component in modified busulfan/cyclophosphamide is cyclophosphamide, which is a risk factor for mucositis, HC, and heart failure.13 To avoid the adverse effects of cyclophosphamide, some researchers have introduced total body irradiation, other agents such as VP16, and melphalan into busulfan/cyclophosphamide, which are aimed to decrease cyclophosphamide or substitute it.14,15 However, there is no obvious superiority in regimen-related toxicities or OS for these modified conditioning regimens compared with the classic busulfan/cyclophosphamide. Meanwhile, occurrences of relapse and late immune reconstitution have not yet been solved.
Fludarabine has a synergistic interaction with busulfan through the inhibition of DNA ligase and DNA primase and the prevention of DNA polymerization, impairing alkylator-induced damage repair. The busulfan/fludarabine combination may act synergistically on apoptosis of target cells.16 The low risk of busulfan/fludarabine toxicity is most likely related to the nonoverlapping organ toxicity.17 Previous studies have confirmed that fludarabine combined with busulfan as a MAC regimen was well tolerated without an increase in relapse. However, most studies were conducted in HLA-identical matched sibling donor or matched unrelated donor procedures.18-20 There are no similar reports about busulfan/fludarabine as a MAC regimen in AML patients undergoing haplo-HSCT.
In the present study, compared with modified busulfan/cyclophosphamide, the superiorities of busulfan/fludarabine are as follows. First, patients in the busulfan/fludarabine group showed shorter duration of neutrocytopenia (11.4 vs 14.6 d). Although there was no statistical significance in engraftment time between the 2 groups, median time for neutrophil and platelet engraftment in the busulfan/fludarabine group was shorter than that shown in the modified busulfan/cyclophosphamide group. Second, blood cell transfusion requirements (especially platelets) were much lower in the busulfan/fludarabine group than in the modified busulfan/cyclophosphamide group. Third, untoward effects of conditioning, such as mucositis or vomiting, were more severe in the modified busulfan/cyclophosphamide group than in the busulfan/fludarabine group. These results are consistent with previous studies.
In the modified busulfan/cyclophosphamide group, more mucositis accidents were shown than in the busulfan/fludarabine group. The development of mucositis, especially mouth ulcers, is associated with neutropenia, chemotherapeutic drug toxicity, and malnutrition.21 Severe mouth ulcers cause patients to have worsened appetites and decreased quality of life. Moreover, severe mouth ulcers are closely associated with aGVHD. Cytokine storms can partially explain these mechanisms, which are closely associated with the destruction of the mucosa barrier.22 In our study, a shorter duration of neutropenia and faster hematopoietic reconstitution were observed in the busulfan/fludarabine group versus the modified busulfan/cyclophosphamide group. Facchini and associates confirmed that severe mucositis and duration of neutropenia are major risk factors for early posttransplant febrile neutropenia and severe bacterial infection.23
Other serious complications posttransplant included HC, which is also often attributed to cyclophosphamide. Early-onset HC, occurring within the first 2 to 3 days after transplant, is thought to be a complication of thrombocytopenia and conditioning regimens containing high-dose cyclophosphamide and busulfan. The development of delayed-onset HC, occurring weeks or months after HSCT, is assumed to be primarily associated with GVHD, impaired immune constitution, and virus infections such as BK virus, adenovirus, or CMV.24 Uhm and associates confirmed that MAC, severe aGVHD (grade 3 or 4), and CMV viremia were attributed to the development of BK virus-associated HC.25 This might be interpreted that patients who develop severe aGVHD received higher and more prolonged doses of systemic glucocorticosteroids than patients with mild aGVHD. Peterson and colleagues26 performed a retrospective evaluation in patients receiving allo-HSCT. They found that higher-grade GVHD was common in patients with viruria and more common in patients with HC. They considered that viruria aggravated preexisting GVHD possibly by stimulating the reconstituting immune system. In turn, GVHD and its treatment aggravated viruria and HC.26 Mesna has been used in conditioning as a prophylactic reagent for many years. However, Seber and associates27 reported risk factors for severe HC following BMT in 977 patients and confirmed that there were no significant differences in HC using prophylactic mesna. Patients with certain characteristics such as grade 2 to 4 GVHD, busulfan use, and older age at transplant had a high risk of developing severe HC.27 Together, severe mucositis, GVHD, and viral reactivation cross interact and are associated with cyclophosphamide use.
The toxicity of cyclophosphamide on the cardiovascular system includes presentation of higher incidence of arrhythmia and heart failure within 1 to 3 weeks.13 Previous studies have confirmed that high-dose cyclophosphamide infusion induces reversible stage 3 heart failure in 10% of metastatic breast cancer patients, with a median decline in the ejection fraction of 31%.28 Nearly 40% of patients undergoing pediatric allo-HSCT treated with high-dose cyclophosphamide experience cardiac complications.29 Clinical studies have shown that high-dose cyclophosphamide therapy is often associated with congestive heart failure, which has been linked to extensive endothelial damage and frank myocyte death. These changes are frequently associated with life-threatening arrhythmias. On the other hand, fludarabine has been rarely associated with cardiac dysfunction. In our study, we found no significant cardiovascular adverse effects in both groups. We note that, because we did not conduct regular echocardiography posttransplant, we cannot make further conclusions regarding potential cardiovascular toxicity.
Previous studies have considered that the conditioning regimen containing fludarabine might cause a higher incidence of opportunistic infections for its immunosuppressive effect.30 Cytomegalovirus infection can involve many organs, resulting in CMV disease, which includes pneumonia, ophthalmia, hepatitis, HC, diarrhea, and cytopenia posttransplant.31 It has been confirmed that severe complications associated with CMV infection are fatal, and anti-CMV drugs such as ganciclovir and foscarnet might be toxic to hematopoietic and/or renal functions. Epstein-Barr virus reactivation is a frequent event after T-cell-depleted HSCT due to the lack of donor-derived EBV-specific cytotoxic T-cell posttransplant. Some patients with serious EBV infections progress to posttransplant lymphoma disease, which has poor prognosis and high mortality. In the present study, we did not draw a similar conclusion. The CMV and EBV activation rates were lower in the busulfan/fludarabine group than in the modified busulfan/cyclophosphamide group without a significant difference, as were bacterial and fungal infections. This may be interpreted by the quick recovery of neutrophils. More follow-up data should be collected to compare the infection rates in the 2 groups.
Busulfan/fludarabine as a MAC regimen may increase relapse rates in patients who receive allo-HSCT. However, we found no relapse differences between the groups. Bredeson and associates18 performed a retrospective matched pair analyses comparing outcomes of adult patients undergoing HLA-identical sibling HSCT using busulfan/fludarabine plus thymoglobulin versus standard oral busulfan/cyclophosphamide. Their results indicated that the busulfan/fludarabine plus thymoglobulin regimen was associated with a decreased incidence of TRM and aGVHD compared with oral busulfan/cyclophosphamide, but there were no differences in risk of cGVHD and LFS. They also suggested that the better tolerability of the busulfan/fludarabine plus TG regimen might allow for other strategies to be added to enhance the graft versus leukemia effect such as total body irradiation and DLI.18 Rambaldi and associates observed similar results in older patients (median age, 51 years) with AML who were randomly assigned 1:1 to receive intravenous busulfan/fludarabine or busulfan/cyclophosphamide.32 Ongoing studies are investigating posttransplant maintenance therapy methods such as donor stem cell infusion and early decreases in immunosuppressive agents without obvious GVHD. Despite the relapse of malignancy and infection, GVHD remains a challenge to obtaining a LFS. In our study, all 20 patients developed grade 1 aGVHD within 30 days posttransplant. However, grade 2 to 4 aGVHD was rare in the busulfan/fludarabine group, with no statistical significance. He and associates established a murine aGVHD model that simulated the clinical situation in which animals received busulfan/cyclophosphamide and busulfan/fludarabine conditioning regimens. Their results showed that busulfan/cyclophosphamide resulted in more severe aGVHD, whereas busulfan/fludarabine was associated with more extensive and long-standing bone marrow damage.33 Overall, more retrospective studies are needed to draw more accurate conclusions regarding differences between GVHD and immune reconstitution between busulfan/cyclophosphamide and busulfan/fludarabine.
Haplo-HSCT provides a potential curative method and is a promising alternative donor transplant procedure for malignant blood diseases. Our randomized retrospective trial confirmed that busulfan/fludarabine was well tolerated with reduced toxicity and similar antileukemic and immunosuppressive activity compared with modified busulfan/cyclophosphamide in patients with AML at first complete remission undergoing haplo-HSCT. Reduced conditioning-related toxicity was not associated with an increased incidence of relapse. However, our study has several limitations that should be considered, including its small number of patients. In addition, our follow-up time should be prolonged to obtain more objective and comprehensive data. It would be also more persuasive to collect additional immune reconstitution data for all patients. Post-HSCT interventions with either more agents or immune interventions may be considered. We suggest that the busulfan/fludarabine conditioning regimen would be an alternative choice for AML patients undergoing haplo-HSCT.
Volume : 16
Issue : 6
Pages : 736 - 744
DOI : 10.6002/ect.2017.0100
From the Department of Hematology, Shandong Provincial Hospital Affiliated to
Shandong University, Jinan, Shandong, China
Acknowledgements: This work was supported by research funding from National Natural Science Foundation (No. 81473486, No. 81270598); the Essential Research and Development Project of Shandong Province (No. 2015GGH318015); Natural Science Foundation of Shandong Province (No. 2009ZRB14176 and No. ZR2012HZ003); Technology Development Projects of Shandong Province (No. 2008GG2NS02018, No. 2010GSF10250, and No. 2014GSF118021); and Program of Shandong Medical Leading Talent and Taishan Scholar Foundation of Shandong Province. The authors have no conflicts of interest to disclose. *Yujie Jiang and Xiaosheng Fang contributed equally.
Corresponding author: Xin Wang, Department of Hematology, Provincial Hospital Affiliated to Shandong University, No. 324, Jingwu Road, Jinan, Shandong 250021, China
Phone: +86 0531 68776358, +86 13156012606
E-mail: email@example.com; firstname.lastname@example.org
Table 1. Characteristics of 20 Patients with Acute Myeloid Leukemia
Table 2. Subgroup Results for 20 Acute Myeloid Leukemia Patients Who Underwent Haploidentical Hematopoietic Stem Cell Transplant
Figure 1. Kaplan-Meier Curves for Leukemia-Free Survival (A) and Overall Survival (B)