Objectives: The most important problems that limit the effectiveness of allogeneic hematopoietic stem cell transplantation in patients with severe aplastic anemia are graft failure and graft-versus-host disease. Mesenchymal stem cells can support normal hematopoiesis and prevent graft-versus-host disease. We aimed to analyze the effects of combined transplant of human umbilical cord-derived mesenchymal stem cells and matched donor allogeneic hematopoietic stem cells in children with severe aplastic anemia.
Materials and Methods: We retrospectively examined 15 pediatric patients with severe aplastic anemia who received fludarabine-based reduced intensity conditioning regimen and intravenously infused human umbilical cord-derived mesenchymal stem cells at a dose of 1 × 106/kg recipient body weight within 12 to 18 hours before hematopoietic stem cells infusion. We evaluated the engraftment rate, the frequency and severity of graft-versus-host disease, and the overall survival rate.
Results: No patients had adverse events related to intravenously human umbilical cord-derived mesenchymal stem cells infusion. All patients achieved successful engraftment and sustained donor chimerism. The median time for neutrophil and platelet engraftment was 14 and 25 days, respectively. The frequency was 20% for grade III/IV acute graft-versus-host disease and 15.3% for chronic graft-versus-host disease. Patients were followed-up for a median of 33 months (range, 2-89 months). The 5-year overall survival rate was 80%.
Conclusions: Combined transplant of matched donor hematopoietic stem cells with human umbilical cord-derived mesenchymal stem cells is safe in pediatric patients with severe aplastic anemia. The achievement of engraftment in all of our patients and the acceptable frequency of acute and chronic graft-versus-host disease and survival rate are encouraging.
Key words : Acquired aplastic anemia, Allogeneic hematopoietic stem cell transplantation, Mesenchymal stem cells
Acquired aplastic anemia is a hematopoietic disorder caused by destruction of hematopoietic stem cells (HSCs) and progenitor cells, due to hematotoxic agents such as drugs, chemical agents, radiation, and virus. The severity of the disease is evaluated by the criteria published in 1976.1 Severe aplastic anemia (SAA) is a life-threatening disease, and there are 2 treatment options currently: immunosuppressive therapy and allogeneic hematopoietic stem cell transplantation (HSCT). In children and young adults with HLA-matched sibling donors (MSD) or matched family donors (MFD), a related HSCT is the initial treatment of choice. For patients with an insufficient hematological response with immunosuppressive therapy, matched unrelated donor (MUD) HSCT and haploidentical donor HSCT options should be considered.2 The most important problems that limit the effectiveness of allogeneic HSCT are graft failure and graft-versus-host disease (GVHD).3
Mesenchymal stem cells (MSCs) are characterized by self-renewal, low immunogenicity, and multidirectional differentiation into several mesenchymal lineages.4 Mesenchymal stem cells are also primary cells that support the proliferation of HSCs by providing scaffold and secreting cytokines and adhesion molecules.5 Furthermore, MSCs have immunomodulatory and immunosuppressive effects by inhibition of T-cell proliferation to alloantigens. It has been demonstrated that coinfusion of MSCs can support hematopoiesis, enhance donor engraftment of HSCs, and reduce the incidence of GVHD following HSCT.5,6 Bone marrow (BM) stroma is damaged by hematotoxic agents, such as drugs, chemical agents, radiation, and viruses; this could poorly affect hematopoietic engraftment in patients with SAA who are undergoing HSCT. Therefore, combined transplant of MSCs plus HSCTs in patients with SAA is quite rational.7
In this study, we retrospectively analyzed the efficacy and safety of human umbilical cord-derived (hUC)-MSC infusion in 15 pediatric patients with SAA who received matched (related and unrelated) donor HSCT. The primary endpoint of our study was to evaluate the engraftment rate and the frequency and severity of acute GVHD (aGVHD) and chronic GVHD (cGVHD). The secondary endpoint was to determine the overall survival rate.
Materials and Methods
From June 2014, we performed combined transplant with hUC-MSCs and HSCs in 15 pediatric patients with SAA who underwent allogeneic HSCT. The use of MSCs was approved by the National Ministry of Health for each patient. Informed consent was obtained from parents or legal guardians before treatment. Patients met the following inclusion criteria: (1) were in line with criteria of the International Aplastic Anemia Study Group, aplastic anemia diagnostic criteria for SAA1; (2) were age <18 years old; and (3) had received MUD HSCT and did not respond to at least 1 immunosuppressive therapy cycle. Exclusion criteria included patients who did not fulfill all of the inclusion criteria.
Of 15 SAA pediatric patients enrolled in the study, 10 were males and 5 were females with a median age of 11 years (range, 2-18 years). Twelve patients (80%) had not responded to previous immunosuppressive therapy, including 1 or more courses of steroids ± cyclosporine ± anti-human antithymocyte globulin (ATG), and all were transfusion dependent. The median time from diagnosis to HSCT was 17 months (range, 2-172 months) (Table 1).
Donors and HLA matching
Seven patients had matched related donors (6 were MSD, 1 was MFD), and 8 patients had MUD (HLA 10/10: 5 patients, HLA 9/10: 3 patients). ABO blood group mismatching was detected in 8 patients (Table 1).
All patients received pretransplant reduced intensity conditioning regimens with cyclophosphamide, ?udarabine, and ATG (rabbit ATG), with 50 mg/kg of cyclophosphamide on days -7, -6, -5, and -4; 30 mg/m2 of fludarabine on days -5, -4, -3, and -2; and 2.5 to 7.5 mg/kg ATG on days -5, -4, -3, and -2 (Table 1).
Prophylaxis and diagnosis of graft-versus-host disease
The prophylactic therapy for GVHD was adjusted based on the allogeneic HSCT donor, stem cell source, and pretransplant history of patients. Recipients of MSD HSCT procedures usually received cyclosporine and short-term methotrexate, whereas recipients of MUD HSCT usually received cyclosporine and mycophenolate mofetil (MMF). Details are shown in Table 1. Diagnosis and clinical grading of GVHD were performed according to Glucksberg-Seattle criteria.
Isolation and preparation of human umbilical cord-derived mesenchymal stem cells
Human umbilical cord tissue was donated by third-party donors with their consent. Production of hUC-MSCs was performed at the Acibadem Labcell Cell Laboratory, which meets Good Manufacturing Practice standards. Preparation of hUC-MSCs was as previously reported.8 The final product was prepared for application after being packaged in a 20-mL vial containing 2 × 106 cells/mL or in a 5-mL vial containing 5 × 106 cells/mL.
Infusion of human umbilical cord-derived mesenchymal stem cells
Ten patients were administered a single dose of hUC-MSCs on day -1, and 5 patients were administered 2 doses of hUC-MSCs on day -1 and on day +5. Patients were monitored for vital signs and symptoms of allergy during transfusion. A suspension of hUC-MSCs was intravenously infused within 10 minutes. The MSC target dose for infusion was 1 × 106/kg recipient body weight. The first dose of MSCs was infused intravenously within 12 to 18 hours before HSC infusion. Median first dose of hUC-MSCs was 1.0 × 106 cells/kg (range, 1-1.47 × 106 cells/kg), and median second dose was 1.0 × 106 cells/kg (range, 1-1.15 × 106 cells/kg) (Table 2).
Source and count of hematopoietic stem cells
Nine patients received unmanipulated BM, 5 patients received unmanipulated peripheral blood stem cells (PBSCs), and 1 patient received BM + PBSCs. The median total nucleated cell (TNC) and CD34+ cell counts were 4.29 × 108/kg and 3.8×106/kg, respectively. In patients who received BM transplant, the median TNC count was 3.91 × 108/kg and that of CD34+ cells was 3.35 × 106/kg. In patients who received PBSC transplant, the median TNC count was 6.23 × 108/kg and that of CD34+ cells was 5.02 × 106/kg (Table 2).
Definitions and assessment of engraftment and chimerism
Neutrophil engraftment was defined as the first of 3 consecutive days with an absolute neutrophil count above 0.5 × 109/L, and platelet engraftment was defined as the first day of a week with the platelet count exceeding 20 × 109/L in the absence of transfusion. Hematopoietic chimerism was assessed using peripheral blood samples of the patient and donor via short-tandem repeated sequence-polymerase chain reaction DNA fingerprinting for all pairs. A BM sample from each patient was analyzed for hematopoietic chimerism every 30 days until 90 days after HSCT.
The Statistical Package for Social Sciences statistical package program (SPSS version 16.0) was used for analysis of the data. Data were evaluated using descriptive statistical methods (median, mean, minimum, maximum). Continuous data are presented as median (range) due to the overall small sample size. Categorical data are presented as count and percentage. The cumulative incidence of aGVHD and cGVHD was also evaluated by the Kaplan-Meier estimate. Overall survival rate was evaluated with Kaplan-Meier survival analysis.
Engraftment and chimerism
All 15 patients (100%) achieved hematopoietic reconstitution after HSCT. The median time for neutrophil and platelet engraftment was 14 days (range, 9-25 days) and 25 days (range, 15-95 days), respectively. Although 14 patients achieved full donor chimerism within 30 days after HSC plus hUC-MSC transplant, 1 patient (patient 3) achieved and sustained stable mixed donor chimerism (Table 2). Another patient (patient 5) who was full chimeric on day 30 after HSCs plus hUC-MSCs transplant and had mixed chimerism on day 90 was administered donor lymphocyte infusion, with full chimerism then obtained. None of the patients developed poor graft function.
Of the 15 patients, 5 (33.3%) experienced aGVHD, including 2 with grade I/II and 3 (20%) with grade III/IV aGVHD, which occurred at a median time after HSCT of 22 days (range, 14-39 days). Acute GVHD involved the skin in 7.1% of patients (1/14), gastrointestinal system and skin in 14.3% of patients (2/14), skin and liver in 7.1% of patients (1/14), and gastrointestinal system and liver in 7.1% of patients (1/14). The frequency of cGVHD was 15.3% (2/13 patients) (Table 2).
Infectious complications in the 100-day period after hematopoietic stem cell transplant
Severe bacterial sepsis occurred in 3 patients (20%), with 1 patient recovering after antimicrobial treatment and 2 patients who died (Table 2). Invasive fungal infection was seen in 2 patients. Hepatosplenic candidiasis occurred in 1 patient (patient 14). Another patient with a history of invasive pulmonary aspergillosis (patient 11) before HSCT developed an intracerebral fungal lesion, which required surgery on day 19. Both patients were successfully treated. Cytomegalovirus (CMV) reactivation was detected by DNA testing in 8 patients, with 7 patients treated successfully with ganciclovir. However, 1 patient (patient 7), who had transplant-associated thrombotic microangiopathy and renal failure, died during month 7 from CMV pneumonia. BK viruria was detected in 5 patients. None of the patients had positivity for Epstein-Barr virus, parvovirus, adenovirus, or human herpesvirus 6.
Infusion-related adverse event
No adverse events related to intravenous hUC-MSCs infusion were detected in any patient.
Survival and causes of death
Patients were followed-up for a median of 33 months (range, 2-89 months). The 5-year overall survival rate was 80% (Figure 1). A total of 3 patients died (Table 2). Two patients (patients 13 and 15) died of severe bacterial sepsis 54 and 62 days after HSCT. The other patient (patient 7) died of CMV pneumonia. In patients who died, median time from diagnosis to HSCT was longer than in surviving patients at, respectively, 47 and 16.5 months.
Because of major improvements in the HSCT procedure in recent years, superiority of allogeneic HSCT over immunosuppressive therapy in SAA patients has been demonstrated, with allogeneic HSCT considered the primary therapy for the effective cure of this disease. Although fludarabine-based reduced intensity conditioning regimens including fludarabine + cyclophosphamide + ATG ± TBI have been shown to reduce rejection and achieve better outcomes in HLA-matched donor HSCT,9-11 major causes of poor outcomes of HSCT including death are still graft rejection and GVHD.10 Therefore, studies have focused on the prevention of graft rejection and prevention/minimization of GVHD, including the use of MSCs as a cell-based therapeutic approach.12
Mesenchymal stem cells are widely used in cellular therapy in allogeneic HSCT. They may create a more favorable BM microenvironment. In addition, MSCs could prevent and treat GVHD.13 The number of studies on the use and efficacy of MSCs in adults and children with SAA, especially in haploidentical transplants, has increased over time.14-17 However, studies in the literature have shown heterogeneity in terms of source (BM-MSCs versus hUC-MSCs) and dose of MSCs, frequency of MSC infusions (once vs twice), administration time (before vs after HSCs infusion), and study population (adults vs children). Most of the studies are not randomized controlled studies but are case series. In addition, results of HSCT procedures with different donor types have been presented in the same study. Therefore, it is difficult to compare the results of the studies with each other and to make correct inferences. There are also studies in the literature in which MSCs infusion alone was used in patients with SAA resistant to traditional treatments.18,19 In these studies, MSCs treatment was found to be safe, although not enough by itself to recover BM. There is little experience in pediatric patients.
Allogeneic HSCT provides a rapid hematopoietic reconstruction in patients with SAA. Lower engraftment rates have been reported with allogeneic HSCT alone.20,21 In a study that compared 48 children with SAA transplanted from MSD versus 38 children transplanted from MUD, engraftment was achieved in 93.75% of patients after MSD HSCT and in 86.8% of patients after MUD HSCT.22 In addition, Bacigalupo and colleagues reported 17% graft failure in a retrospective study of 100 patients with SAA who underwent HSCT alone from an alternative donor.10 However, rapid and sustained hematopoietic reconstitution has been reported in nearly all HSCTs in which MSCs combined transplant was performed. Wang and colleagues reported that engraftment was achieved in all 19 children with SAA who received MUD HSCT with UC-MSCs 1 hour before HSCs infusion.23 In our study, median time for neutrophil and platelet engraftment was 12 (range, 9-21 days) and 14 (range, 8-24 days) days, respectively. However, Wang and colleagues noted that delayed rejection occurred in 1 case at 4 months after HSCT. In our study, allogeneic HSCT combined with infusion of hUC-MSCs resulted in successful engraftment of HSCs in all 15 patients (100%). Although the number of cases in our study is small, it is encouraging that no patient had graft failure, especially in those with MUD. The longer engraftment times in our study compared with the study from Wang and colleagues may be related to our use of BM as a stem cell source in 9 of 14 patients.
Because myeloablative conditioning regimens are not used in patients with SAA, aGVHD is not a serious problem, especially with MSD HSCT. However, high aGVHD rates have been reported in MUD (43%)24 and haploidentical HSCT (39.2% and 42.1%)25,26 in patients with SAA who had HSCT alone. In the literature, different aGVHD rates have been reported in studies where MSCs coinfusion was performed with HSCT. Fu and colleagues reported that no pediatric patients who received MSCs infusion for MUD HSCT developed severe GVHD.27 Wang and colleagues reported, in 19 MUD (10/10 matched = 9 patients, 9/10 matched = 6 patients, 8/10 matched = 4 patients) who had HSCT plus hUC-MSCs procedures, grade I aGVHD in 9 patients (47.3%) and grade III aGVHD in 1 patient (5.2%).23 They infused PBSCs, and patients received cyclosporine + methotrexate+ MMF for GVHD prophylaxis. Our study showed that the frequency of grade I to IV aGVHD and grade III/IV was 33.3% and 20%, respectively. All but 1 of the patients without aGVHD were patients who received hUC-MSCs plus HSCT from a 10/10 matched donor (MSD= 6 patients, MFD= 1 patient, MUD = 2 patients). All of our patients who developed aGVHD had MUD HSCT, and we administered MMF instead of methotrexate with cyclosporine/tacrolimus in all but 1 patient. Patients who developed grade III/IV aGVHD were also patients who did not receive methotrexate. This may be related to the use of cyclosporine/tacrolimus + MMF, without use of methotrexate for GVHD prophylaxis. In 1988, Bacigalupo and colleagues reported that the combined use of cyclosporine and methotrexate is a predictor for favorable HSCT outcomes.28
In a recently published meta-analysis of patients who received haploidentical HSCT and hUC-MSCs, the rates of aGVHD, grade II-IV aGVHD, cGVHD, 2-year survival, and engraftment were similar to those who only received HSCT.29 Thus, an important question is whether combined transplant of MSCs has a beneficial effect for another subgroup of patients, especially for those who have matched donors. In our study, aGVHD was not detected in any of our patients who received MSD HSCT. Another question is whether combined transplant for patients with MSD is really necessary. In our opinion, well-designed controlled clinical trials and a meta-analysis seem necessary to find the real effects of combined transplant of hUC-MSCs and HSCs on GVHD rates in patients with SAA who had transplant with matched donors.
Chronic GVHD is a major cause of morbidity in long-term survivors. Chen and colleagues30 and Perez-Albuerne and colleagues,24 in studies of HSCT alone in patients with SAA, reported frequencies of cGVHD of 28.3% and 35%, respectively. In our study, the frequency of cGVHD was 15.3%. which is markedly lower than most pediatric studies of patients who received matched donor HSCT alone. Wang and colleagues reported a rate of cGVHD of 5.2%.23 This low rate can be explained by the low rate of severe aGVHD in that study.
Cytomegalovirus viral load and the risk of CMV disease increase with MSC infusions.31 In our study, the reactivation rate of CMV in patients was 46.6%, with CMV-related death occurring in 1 patient. Wang and colleagues reported a higher CMV reactivation rate (78.9%).23 Interestingly, Si and colleagues reported a very low CMV reactivation rate (5.4%) in children with SAA who received hUC-MSCs at 7 to 10 days after allogeneic HSCT from matched and haploidentical donors.32 The effect of time of infusion of hUC-MSCs during the procedure on CMV reactivation may be debatable.
In our study, no adverse events related to intravenous hUC-MSCs infusion was detected in any patient. However, mild and self-limited side effects such as fever, chills, headaches, hypoxemia, mild dyspnea and diarrhea have been reported after MSCs infusion.19,33
Our study reported a mortality rate of 20% and a 5-year overall survival rate of 80%. Median follow-up time was 33 months, and the longest survival was up to 7 years. Three patients died due to infection. None of our study patients had severe GVHD resulting in death.
As a result of the observations in this study, we planned to omit MMF but add methotrexate to the GVHD prophylaxis in patients with MUD. On the other hand, we planned to make an individualized decision for each patient until certain results were reported with controlled studies and meta-analyses regarding the necessity of MSCs coinfusion in patients with MSD.
This pediatric single-center study has several limitations, including a small sample size and lack of a control group. Most of the similar studies in the literature do not have a control group, although some have used a historical control group. This problem may be overcome by performing meta-analysis of suitable case series. Despite these limitations, the achievement of engraftment in all of our patients and the acceptable frequency of aGVHD and cGVHD and survival rate are encouraging.
Our study supported that combined transplant of hUC-MSCs plus HSCs can be used safely and may increase the efficiency of the transplant procedure in pediatric patients with SAA who undergo matched donor HSCT.
DOI : 10.6002/ect.2021.0027
From the 1Pediatric Hematology and Oncology/Bone Marrow Transplantation Unit, Department of Pediatrics, School of Medicine, Acibadem Mehmet Ali Aydinlar University; and the 2Pediatric Hematology and Oncology, Koç Univercity Hospital; and the 3Acibadem Labcell Laboratory, Istanbul, Turkey
Acknowledgements: The authors thank all of the Labcell laboratory team members who contributed to the production of MSCs, especially Muhammet Yilanci, PhD, and Merve Kongur, PhD. The authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no declarations of potential conflicts of interest.
Corresponding author: Arzu Akçay, Acibadem University, School of Medicine, Altunizade Acibadem Hospital, Department of Pediatric Hematology and Oncology/Bone Marrow Transplantation Unit, Istanbul, Turkey
Phone: +90 533 646 77 06
Table 1. Characteristics of Pediatric Patients With Severe Aplastic Anemia Who Received Hematopoietic Stem Cell Transplantation (N = 15)
Table 2. Clinical Outcomes of Hematopoietic Stem Cell Transplantation in 15 Pediatric Patients With Severe Aplastic Anemia
Figure 1. Overall Survival of Children With Severe Aplastic Anemia After Hematopoietic Stem Cell Transplantation (Kaplan-Meier)