Outcomes of Transplantation with Related- and Unrelated-Donor Stem Cells in Children with Severe Thalassemia

Introduction 

Allogeneic bone marrow transplantation (BMT) from a genotypically identical family donor is the only curative therapy for patients with thalassemia. However, the probability of finding a histocompatibility (HLA)-identical donor within the family is <30%. For patients with thalassemia without an HLA-identical sibling donor, there was until recently no treatment other than chronic transfusion and iron chelation. Within the past few years, there has been a steady increase in the number of unrelated-donor BMTs in a variety of disorders, mainly due to the increase in the number of volunteer donors worldwide [1]. There are only 2 case-series reports, by La Nasa et al. [2] from Italy and Hongeng et al. [3] from Thailand, that are available about unrelated BMT in thalassemia and both series have indicated favorable outcomes. The present study investigated outcomes of related and unrelated stem cell transplantations (SCTs) and whether these compare favorably in children with severe thalassemia that was treated at a single institution (Ramathibodi Hospital, Mahidol University, Bangkok, Thailand).

Methods 

We studied 49 consecutive patients who received related-donor (n = 28) or unrelated-donor (n = 21) stem cells for severe thalassemia between September 1992 and May 2005. Some of the present patients were previously reported [3, 4, 5]. Patients and donors or their parents signed an informed consent before treatment. Characteristics of the 2 groups are presented in Table 1. Some of our patients could not be classified with the Pesaro risk classification because we did not perform routine pretransplantation liver biopsy before 2000 [6]. Although some of our patients did not have pretransplantation liver biopsy, these patients had severe anemia, marked hepatosplenomegaly, multiple blood transfusions, and high serum ferritin levels. Most of them received no or inadequate iron-chelation therapy. Therefore, these patients were classified as moderate to high risk (Lucarelli class 2 or 3) [6].

Table 1. Characteristics of Patients, Frequencies of GVHD, Procedure-Related Complications, and Causes of Death in Related- and Unrelated-donor SCT⁎

		Related (n = 28)	Unrelated (n = 21)
	Median age at transplantation, y (range)	7.2(0.5-18.7)	4.0(0.7-12.0)
	Male/female	13/15	14/7
	Type of thalassemia
	β/β thalassemia	8	8
	β thalassemia/hemoglobin E	20	13
	Lucarelli class
	1	15	13
	2/3	13	8
	Source of stem cells
	Bone marrow	12	21
	Peripheral blood	12	—
	Cord blood	4	—
	HLA matching
	6/6 HLA antigen match	23	—
	5/6 HLA antigen match	3	—
	4/6 HLA antigen match	2	—
	6/6 HLA allele match	—	13
	5/6 HLA allele match	—	6
	4/6 HLA allele match	—	2
	Myeloablative
	Bu 16 Cy 200†	20	—
	Bu 16 Cy 200 ATG 40‡	4	21
	Nonmyeloablative
	Bu 8 Flu 175 ATG 20 TLI 500 cGy§	4	—
	GVHD
	Acute, grade II-IV	9(32%)	9(42%)
	Acute, grade III-IV	3(11%)	3(14%)
	Chronic	4(14%)	3(14%)
	Extensive chronic	2(7%)	1(4%)
	Procedure-related complications
	Hemorrhagic cystitis	3(11%)	4(19%)
	Sepsis	3(11%)	3(14%)
	Autoimmune hemolytic anemia	—	1(4%)
	Seizure	2(7%)	4(19%)
	Veno-occlusive disease	6(22%)	5(23%)
	Causes of death
	Graft failure	1	—
	Bleeding	—	1
	Pneumonitis	—	1
	Sepsis	1	—
	Chronic GVHD	—	1
	Median follow-up, mo (range)	51(6-157)	35(6-55)

⁎ No statistical comparison showed significant differences in incidence of acute and chronic GVHD and procedure-related complications. † Busulfan 16 mg/kg, cyclophosphamide 200 mg/kg. ‡ Busulfan 16 mg/kg, cyclophosphamide 200 mg/kg, antithymocyte gloubulin 40 mg/kg. § Busulfan 8 mg/kg, fludarabine 175 mg/m2, antithymocyte gloubulin 20 mg/kg, total lymphoid irradiation 500 cGy.

Results 

Patients in the 2 groups were slightly different in age and sex. Median ages at transplantation were 4 years in the unrelated group and 7.2 years in the related group, and more female patients were included in the related group (Table 1). Because we did not perform liver biopsy before transplantation before 2000, we classified patients who had >1 Lucarelli risk factor as Lucarelli class 2/3. Of the 28 patients who received stem cells from a related donor, 23 had a 6/6 HLA antigen match, 3 had a 5/6 HLA antigen match, and 2 had a 4/6 HLA antigen match. Of the 21 patients who received marrow from an unrelated donor, 13 had a 6/6 HLA allele match, 6 had a 5/6 HLA allele match, and 2 had a 4/6 HLA allele match by molecular typing.

In the related group, 12 of 28 patients received stem cells from bone marrow, 12 from peripheral blood, and 4 from cord blood; in the unrelated group, all 21 patients received stem cells from bone marrow.

The course of transplantation in patients who received grafts from an unrelated donor did not differ substantially from that in the related group. Mean times to myeloid cell engraftment were identical (16 days; ranges, 12-35 and 11-30, respectively), although platelet engraftment was delayed in recipients of unrelated-donor stem cells (mean, 50 days; range, 20-250; versus 25 day; range, 15-75; P = .01). Frequencies of GVHD were similar, as were frequencies for other post-transplantation complications (Table 1). Because we administered preemptive therapy for CMV and EBV infections by CMV and EBV viral load monitoring, there was no incidence of CMV disease and EBV post-transplant lymphoproliferative disease in our study. Karnofsky or Lansky performances in patients who survived >1 year were similar (related-donor group: mean, 98; range, 85-100; in unrelated-donor group: mean, 95; range, 80-100).

For the first transplantation, the 2-year thalassemia-free survival rate for the 49 patients with thalassemia was 77% (95% confidence interval, 61-86) and overall survival rate was 89% (95% confidence interval, 72-94). The 2-year thalassemia-free survival rate estimated for recipients in the related-donor group was 82% (95% confidence interval, 61-92) compared with 71% (95% confidence interval, 45-85) in the unrelated-donor group (P = .42; Figure 1). The 2-year overall survival rate estimated for recipients in the related-donor group was 92% (95% confidence interval, 70-97) compared with 82% (95% confidence interval, 53-94) in the unrelated-donor group (P = .43; Figure 2). The rejection rate after first transplantation in the related-donor (n = 4) and unrelated-donor (n = 3) groups was the same (14%). Of 4 patients in the related group whose graft failed after the first transplantation, 2 underwent a second transplantation with stem cells from the same donors and achieved successful engraftment with full (100%) donor chimerism; the second myeloablative regimen consisted of busulfan 1 mg/kg for 4 days, cyclophosphamide 50 mg/kg for 4 days, and ATG 10 mg/kg/d for 4 days; 1 patient died of graft failure; and 1 awaits a second transplantation. Three patients in the unrelated group whose graft failed had autologous recovery and are still alive. Two patients in the related group and 3 in the unrelated group died (P = .63). The causes of death are listed in Table 1. We did not analyze the survival rate after the second transplantation.

View Large Image Download to PowerPoint

Figure 1. Thalassemia-free survival analysis of related- and unrelated-donor transplantation groups.

View Large Image Download to PowerPoint

Figure 2. Overall survival analysis of related- and unrelated-donor transplantation groups.

Four patients in the related group who underwent nonmyeloablative transplantation had full donor engraftment (100% donor chimerism) at 1 month, 6 months, and 1 year after transplantation.

Of 4 patients who received cord blood stem cells, 1 had primary graft failure and 3 achieved full donor chimerism. One patient died of salmonella sepsis at 1 year after transplantation, and 2 are still alive without thalassemia.

All patients in the 2 groups who were alive and free of thalassemia had full (100%) donor chimerism at the time of this report. Median follow-up times were 51 months (range, 6-157) in the related group and 35 months (range, 6-55) in the unrelated group.

Discussion 

Most patients with thalassemia do not have an HLA-identical donor within the family. For this reason, BMT centers that have been engaged in a cure for thalassemia have been under increasing pressure from patients and families to explore the possibility of transplantation with alternative donors. Use of partly matched family donors carries a high risk of rejection, GVHD, and mortality and had been confined to particular clinical situations [13].

Recently published reports have indicated that outcome of unrelated-donor transplantation in patients with leukemia is currently comparable to that of transplantation from identical family donors. This is due in part to the better compatibility between donor and recipient afforded by DNA methods [14].

La Nasa et al [2] reported thalassemia-free survival, rejection, and mortality rates of 66%, 12.5%, and 19%, respectively, for 32 patients from 4 BMT centers for a median of 30 months after unrelated-donor BMT. In our study, 2-year thalassemia-free survival rates were 82% in the related group and 71% in the unrelated group (P = .42). Median follow-up times were 51 months in the related group and 35 months in the unrelated group. Our study is the first to support the success of unrelated-donor SCT in thalassemia as a case series that is comparable to that offered by a related-donor group in a single institution, even though we did not perform extended haplotype analysis as La Nasa et al [2] did. Apart from platelet engraftment, which was delayed by 25 days in the unrelated group, the end points of our study (engraftment, frequency of acute and chronic GVHD, rejection rate, performance status, thalassemia-free survival rate, and mortality rate) were unaffected by the donor source of stem cells.

Our previous report demonstrated a favorable outcome, with a 100% thalassemia-free survival rate for the first 11 patients who underwent unrelated-donor SCT [3]. Because there was rejection in 3 patients in the unrelated group after the first 11 patients had been enrolled, we plan to add fludarabine at a dose of 30 mg/m2/d for 5 days with the same dose of busulfan in the conditioning regimen to ensure an improved and sustained engraftment of donor cells for unrelated-donor transplantation in a future study.

The present study has shown that outcomes of related- and unrelated-donor SCT in children with severe thalassemia are favorably comparable. We believe the reason for this favorable result in unrelated transplantation patients is the younger age of this sample. Because there were multiple differences in patient characteristics and treatment for the 2 groups, this study was not meant to be comparison of the same regimen for different donors. Rather, the outcomes show that reasonably comparable results can be achieved. Due to the small numbers of cases, we cannot rule out small differences in outcomes between groups.

The present study provides an additional evidence to support the view that it is quite reasonable to consider unrelated-donor SCT as an acceptable therapeutic approach in thalassemia, at least for younger patients who are not fully compliant with conventional treatment and do not yet show irreversible severe complications of iron overload.