We retrospectively reviewed data on 22 consecutive TM or SCD patients who received allogeneic HSCT from July 2012 to December 2015 at the Department of Pediatrics, Catholic Blood Hospital, The Catholic University of Korea. This study received approval from the institutional review board (IRB). The study group included 15 patients with TM and 7 patients with SCD (11 males, 11 females) (Table 1). The median age at transplantation was 9.0 years (range, 1.6–16.9). All patients were of Middle Eastern ethnicity. The median ferritin level of TM patients was 1,803 ng/mL (range, 1,103–3,096), and 13 patients received iron chelation therapy. All SCD patients had previously experienced vaso-occlusive crisis, and received hydroxyurea therapy at the time of HSCT, except for 1 patient, intolerant to the medication.

All donor–recipient pairs were matched according to high resolution typing for HLA-A, B, C, and DRB1 alleles. Donor types in the HSCTs were as follows: 18 matched sibling donors (MSD), 3 mismatched related donors (including 2 haploidentical and 1 HLA-DR 1-antigen mismatched donor), and 1 matched unrelated donor (MUD). Although 12 donors had β-thalassemia trait and 3 donors had sickle cell trait, none of the donors showed evidence of hemoglobinopathy (Table 2).

Twenty patients received the following conditioning regimen: intravenous (IV) busulfan 130 mg/m² once daily for 4 days, cyclophosphamide 60 mg/kg once daily for 2 days and anti-thymocyte globulin (ATG) (Thymoglobulin; Genzyme, Cambridge, MA, USA) 2.5 mg/kg once daily for 3 days. One patient received the same conditioning regimen except that cyclophosphamide was administered at 50 mg/kg for 4 days. One SCD patient, who underwent haploidentical HSCT, received total body irradiation 400 cGy, busulfan 130 mg/m² once daily for 2 days, fludarabine 30 mg/m² for 5 days and ATG 2.5 mg/kg for 3 days. The other TM patient received haploidentical HSCT from a homozygous HLA donor with 4/8 match in the graft-versus-host direction, and 8/8 match in the host-versus-graft direction; this patient received the same conditioning as MSD HSCT patients.

Graft-versus-host disease (GVHD) prophylaxis consisted of IV cyclosporine (3 mg/kg/day IV initially, followed by 5 mg/kg/day oral dose, when oral intake was possible) starting from day −1 and 4 infusions of short course methotrexate (5 mg/m²) at days 1, 3, 6, 11. In the absence of GVHD, cyclosporine was maintained at therapeutic levels (100–200ng/mL) until 9 months post-transplantation, then subsequently tapered and stopped at 1 year post-transplantation. Antimicrobial prophylaxis included oral acyclovir from the beginning of conditioning until day 42, and oral trimethoprim-sulfamethoxazole from the beginning of conditioning until day −3, and from neutrophil engraftment until at least 1 year after transplantation. Antifungal prophylaxis consisted of IV micafungin (1 mg/kg/day) from the beginning of conditioning until neutrophil engraftment, at which point oral fluconazole was administered and maintained for at least 6 weeks. G-CSF was administered from day 5 until absolute neutrophil count (ANC) ≥3.0×10⁹/L.

Detection of post-HSCT cytomegalovirus (CMV) DNAemia and Epstein-Barr virus (EBV) DNAemia was performed with real-time quantitative monitoring of plasma samples after engraftment at weekly and fortnightly intervals, respectively, until 3 months after transplantation, and at longer intervals subsequently. Preemptive treatment for EBV DNAemia was not provided, and rituximab was only administered with diagnosis of post-transplantation lymphoproliferative disease (PTLD).

Immune reconstitution and chimerism analyses were performed at 1, 3, 6, 9 and 12 months post-HSCT. Chimerism was determined with short tandem repeat polymerase chain reaction (STR-PCR) analysis of peripheral blood, with full donor chimerism defined as >95% donor hematopoietic cells, and mixed donor chimerism defined as 5–95% donor-derived cells.

Date of neutrophil engraftment was defined as the first of 3 consecutive days with ANC ≥0.5×10⁹/L, and the date of platelet engraftment was defined as the first of 3 consecutive days with platelet count ≥50×10⁹/L without platelet transfusions in the preceding week. Acute GVHD was graded according to published consensus criteria, and chronic GVHD was classified according to revised 2014 NIH criteria [7,8]. Major toxicities were graded according to NCI CTCAE (version 4.0). Hepatic veno-occlusive disease (VOD) was diagnosed according to clinical criteria as the presence of two of the following: 1) hyperbilirubinemia (bilirubin >2 mg/dL), 2) painful hepatomegaly, and 3) unexplained weight gain (>5% from baseline).

Primary outcomes analyzed were the incidences of graft failure and treatment-related mortality (TRM) after HSCT. Secondary endpoints included the incidence of VOD and acute and chronic GVHD. Overall survival (OS) was calculated using the Kaplan–Meier method. Both acute and chronic GVHD were calculated using the cumulative incidence function, with TRM without respective GVHD as a competing risk [9]. Patients alive were censored at date of last follow-up, but no later than April 30, 2017. Statistical analyses were performed with R package version 2.14.2 (R Foundation for Statistical Computing, Vienna, Austria).

Graft failure was not observed, and all patients showed neutrophil and platelet engraftment at a median time of 14.5 days (range, 8–25), and 12.5 days (range, 8–65), respectively. Median donor chimerism at 1 month post-transplantation was 99% (range, 93–100) (Table 3). Twenty-one out of 22 patients showed complete donor chimerism at this time point. During the last follow-up, 2 MSD bone marrow transplantation (BMT) recipients, who initially achieved complete donor chimerism, showed a decrease in chimerism down to 94% and 81%, while 1 patient with initial partial donor chimerism converted to complete donor chimerism. All TM patients became transfusion-independent with normal complete blood counts.

As TM and SCD are at high risk for VOD, and as 1 of the 2 initial patients showed VOD, prophylaxis against VOD was started for the third HSCT patient with alprostadil (Eglandin, Welfide Korea, Hwaseong, Korea) at a dose of 0.1 µg/kg/day with continuous infusion starting 1 day prior to conditioning until 30 days after HSCT. No significant difference in VOD incidence was observed according to median ferritin levels: 36.4% [95% confidence interval (CI), 21.9–50.9] for lower ferritin levels and 45.5% (95% CI, 30.5–60.5) for higher ferritin levels (P=0.636). VOD occurred in 8 out of 15 patients with TM (53.3%) and in none of the SCD patients (Table 3). Severity of VOD was as follows: 1 mild, and 7 moderate. Median time to VOD onset was 13 days (range, 9–26), and median highest total bilirubin was 3.9 mg/dL (range, 1.7–11.2). All 7 patients with moderate VOD showed abdominal pain, and disseminated intravascular coagulopathy requiring antithrombin III treatment. Median weight gain percentage was 15% (range, 6.5–27.5). All VOD patients received either recombinant tissue plasminogen activator (TPA) or defibrotide as anticoagulant therapy, with 12.5 days median duration of treatment (range, 9–26). Supportive care also included glutathione and vitamin E treatment. All patients showed resolution of VOD.

Hemorrhagic cystitis occurred in 1 patient in each of the TM and SCD patient groups.

Overall, 8 patients were diagnosed with and treated for acute GVHD of grade 2 or above with a cumulative incidence of 36.4% (95% CI, 25.8–47.0) (Fig. 1). Four of these patients experienced grade 3 or 4 acute GVHD (1 TM and 3 SCD patients) and 1 SCD patient died of complications resulting from grade 4 gut GVHD. PB was the cell source for these 5 patients. Cumulative incidence of acute GVHD of grade 2 or above was higher in the 14 patients who underwent peripheral blood stem cell transplantation (PBSCT, 50.0%, 95% CI, 36.6–63.4), compared with the 8 patients, who received BMT (12.5%, 95% CI, 0.8–24.2), although the number of patients was too low to determine statistical significance (P=0.072). Other factors such as primary disease (TM vs. SCD) and donor type (MSD vs. others) did not significantly influence acute GVHD incidence.

Eight patients were diagnosed with chronic GVHD and the severity of chronic GVHD was as follows: 1 mild, 2 moderate, and 5 severe. One patient with de novo mild chronic GVHD showed oral GVHD and dry eye symptoms, but showed rapid resolution of disease with oral prednisolone administration. The cumulative incidence of moderate to severe chronic GVHD was 32.7% (95% CI, 22.2–43.2) (Fig. 2). The cumulative incidence of chronic GVHD of at least moderate severity was higher for PBSCT recipients compared with BMT recipients, although without statistical significance (38.9%, 95% CI, 25.1–52.7 vs. 25.0%, 95% CI, 9.7–40.3, P=0.467). Of the 7 patients with chronic GVHD of moderate severity or above, 2 were diagnosed with de novo chronic GVHD, while 5 showed preceding acute GVHD. The following organs were involved in chronic GVHD: the lungs (5 patients), the skin (3 patients), the oral cavity (3 patients), and the eyes (2 patients). Three patients had lung GVHD alone, of whom 1 PBSCT recipient died of this complication. Four patients continued to receive immunosuppressive therapy during the last follow-up, while GVHD treatment ceased for 3 patients without disease progression.

CMV DNAemia was detected in 8 patients (36%), with initial preemptive ganciclovir treatment resulting in resolution of DNAemia in all instances (Table 3). EBV DNAemia was detected in 8 patients and 1 patient with thalassemia progressed to PTLD, which resolved with rituximab therapy.

The estimated 2-year OS for the entire cohort was 90.9±6.1%, with a median follow-up of 27 months (range, 16–58) (Fig. 3). All TM patients survived, while 2 patients with SCD died; 1 from sepsis subsequent to grade 4 acute gut GVHD after MSD PBSCT, and another patient from severe chronic lung GVHD after MSD PBSCT. All surviving SCD patients remained free of SCD-related complications, such as vaso-occlusive crisis.