Allogeneic HSCT in children with relapsed/refractory acute lymphoblastic leukemia: ambulatory or in-hospital? Open Access

Acute lymphoblastic leukemia (ALL) is the most common childhood oncohematological disease worldwide¹; it represents 70% to 80% of leukemias in children.² Children respond more favorably to treatment than adults.³ Initial treatment consists of multidrug induction regimens with steroids, vincristine, asparaginase, an anthracycline, and intrathecal therapy. Later, postinduction intensification consists of chemotherapy and methotrexate.⁴ 

Most children with ALL achieve complete remission, and ∼90% reach the elective cessation of chemotherapy 2 years after beginning treatment⁵; a subset is either refractory to the initial treatment or experience a relapse. Patients with refractory or relapsed (R/R) ALL have a worse prognosis; allogeneic hematopoietic stem cell transplantation (allo-HSCT) has been shown to improve prognosis, remaining relevant in most of the world where novel therapies like monoclonal antibodies and chimeric antigen receptor T cells are inaccessible.⁶ Ideally, an HLA-matched donor is elected, but it is difficult to have an HLA-identical sibling donor.⁷ In addition to limitations imposed by donor selection, limited economy decreases the possibility of an HSCT to children living in low- and middle-income countries (LMICs).⁸ Several strategies have been implemented to overcome these barriers: one is expanding the number of donors by choosing HLA-matched unrelated donors or haploidentical donors,⁹ and another is the modification of the HSCT procedure to render it more affordable and realistic for persons living in LMICs, such as the outpatient conduction.¹⁰ 

We report the results of a combined strategy employing HLA-identical or haploidentical donors in allogeneic reduced-intensity conditioned HSCT, conducted either in the hospital or as outpatients in the treatment of children with R/R ALL.

Children with R/R ALL underwent allografting using reduced-intensity conditioning regimens based on the Mexican conditioning method^8,11; HLA-identical siblings or haploidentical donors were chosen. All patients from Clínica Ruiz de Puebla underwent grafting as outpatients, whereas those from the Hospital Universitario de Nuevo León underwent grafting as inpatients. In Clínica Ruiz de Puebla, all transplants are performed on an outpatient basis, whereas patients in the Hospital Universitario de Nuevo León underwent transplantation in-hospital; the only reason for selecting this was that this was the center where the procedure was performed. All patients were in remission at the time of HSCT, as defined by flow cytometry. The “outpatient” transplantation schedule employed fludarabine (150 mg/m²), pretransplant cyclophosphamide (Cy; 2000 mg/m²), and posttransplant Cy on days +3 and +4, either a conventional (50 mg/kg) or a reduced dose (25 mg/kg); per protocol, post transplant cyclophosphamide (PTCy) was employed in patients who received grafts from matched siblings. The “inpatient” transplant used fludarabine (75 mg/m²), pretransplant Cy (1500 mg/m²), IV busulfan (0.8 mg/kg per dose, 2-3 doses) or IV melphalan (140-200 mg/m², 1 dose), and posttransplant Cy on days +3 and +4. The selection of busulfan or melphalan was made depending on the availability of drugs. Cyclosporine-A was used for graft-versus-host disease (GVHD) prophylaxis in outpatients and tacrolimus or cyclosporine-A, mycophenolate (MMF), and Cy for inpatients. Total body irradiation was not employed because it is not available. Antibiotics, antifungals, and antivirals were administered in all patients. For inpatients, levofloxacin 10 mg/kg every 24 hours, trimethoprim/sulfamethoxazole 5 mg/kg 3 times per week, voriconazole 8 mg/kg every 12 hours, and acyclovir 30 mg/kg every 12 hours were used; whereas for outpatients, itraconazole 100 mg every 24 hours, acyclovir 400 mg every 24 hours, and trimethoprim/sulfamethoxazole 160/800 mg every 24 hours. Outpatients who develop neutropenic fever were admitted to hospital to be given meropenem 1 g every 12 hours and to perform microbiological tests; red blood cells were given to patients with low hemoglobin levels and symptoms of anemia, whereas platelet transfusions were given for those with <20 × 10⁹/L. The protocol was approved by the institutional review board of the Clínica Ruiz (CEI-30-11-24-02); all patients signed an informed consent form. Overall survival (OS) was defined as the interval from the date of transplantation to death. The probability of OS was estimated using the Kaplan-Meier method. The long-rank test was used to compare groups. The study's primary end point was OS, whereas the secondary end points were the prevalence of both acute and chronic GVHD.

Seventy-two consecutive patients were entered in the study; 31 who received grafts as outpatients and 41 in-hospital, in a 23-year period (1999-2022), Table 1 summarizes their salient data. There were no significant differences between these 2 groups in terms of age, sex, time to granulocyte count recovery, prevalence of acute GVHD (25.8% in the outpatient group vs 36.5% in the inpatient group), and follow-up duration (P = .006). In the inpatient group, 33 of 41 (80.5%) transplants were from haploidentical donors vs 10 of 31 (32.2%) in the outpatient group, the difference being significant (P = .00003). The prevalence of chronic GVHD was higher in the inpatient group (48.7% vs 9.6%; P = .0004), and time to recovery of >20 × 10⁹/L platelets was shorter in the outpatient group than in inpatients (7.5 vs 15 days; P = .03); the shorter time to platelet recovery in outpatients may be explained by the fact that most were from identical donors (21/31), whereas in inpatients, most were from haploidentical donors (33/41), as HLA disparities can affect the time to engraftment.¹² The likelihood of completing the HSCT as outpatients was higher if grafts were obtained from matched siblings rather than haploidentical donors (67.7% vs 32.25%). The OS of children who received allografts in-hospital was 48.7% at 72 months, whereas in outpatients, 38.7% at 173 months (P = .1), a nonsignificant difference (Table 1; Figure 1). The OS of patients allografted from identical donors was 39.9% at 173.2 months, and from haploidentical donors was 47.7% at 153.5 months (P = .2), a nonsignificant difference (Figure 2). Median OS was 33.7 months in outpatients vs 27.5 months in inpatients, a nonsignificant difference. The nonrelapse mortality was 16% for outpatients and 17% for inpatients; 10% of outpatients and no inpatient died before day 100 (P = .05). A total of 18 outpatients (58%) and 16 inpatients (39%) relapsed (P = .34). Bacterial infections were recorded in 22% of outpatients and 29% of inpatients (P = .625); cytomegalovirus reactivation was shown in 4% and 24% of outpatients and inpatients, respectively, and BK virus reactivation was recorded in 0 of 31 outpatients and 3 of 41 inpatients.

All-HSCT is predominantly conducted within a hospital setting, and this intervention typically entails substantial financial burdens for both patients and health care facilities¹³; as a result, this poses challenges for its application in LMICs.¹⁴ We have been conducting outpatient autologous, allogeneic, and haploidentical HSCT since 1993,^15,16 with the primary goal of decreasing the costs of the procedures and rendering them more available to more individuals.¹⁶ According to previous studies in our centers, the cost of an outpatient HSCT is between 12 000 to 18 000 USD, which is ∼50% of the in-hospital procedure, whereas in high-income countries, such as the United States, it ranges between 150 000 and 400 000 or more.¹⁷ Additional advantages of outpatient HSCT include an extended OS in patients receiving allografts in non-hospital environments, reduced infection rates, a decreased incidence of GVHD, and lower pharmaceutical disbursements.¹⁸ We have shown that the prevalence and severity of GVHD decrease if the transplant is performed on an outpatient basis.^19,20 In Sweden, Svahn et al²¹ reported that 66% of patients accepted outpatient care after allo-HSCT: outpatients had lower transplantation-related mortality rates and incurred lower costs. Campos et al²² analyzed 14 patients, who underwent an ambulatory allo-HSCT and confirmed its safety and feasibility, with significant cost reductions, avoiding the difficulty in hospitalizing every patient, and the increased associated risk of hospital-acquired infections. In our sample, the incidence of bacterial infections was not different between in-hospital patients and outpatients, whereas the incidence of viral infections was higher for those children kept in the hospital; this difference probably being related to the hospital exposure to infectious agents.

The data we have presented compare the outcome of children with relapsed or refractory ALL managed with an allo-HSCT, either as in-hospital or outpatients. The salient data of our study are that in outpatients, platelet recovery is faster and the prevalence of chronic GVHD is lower, thus confirming our previous observations in adults.²³ This observation can be biased because of the lower proportion of cases in the outpatient arm that underwent a haploidentical transplant. In addition, undergoing an allo-HSCT as outpatients was significantly improved when grafts were sourced from matched siblings as opposed to haploidentical donors. Data have indicated that the prevalence of chronic GVHD is higher in recipients of haploidentical cells²⁴ and we have shown that haploidentical transplants can be conducted entirely on an outpatient basis,¹¹ and that the prevalence of chronic GVHD does not substantially increase.²⁵ Accordingly, the outpatient conduction of the procedure may somehow counteract the increase in GVHD associated with the use of haploidentical donors.

We have found that allo-HSCT can be entirely conducted in children as outpatients, employing either HLA-identical or haploidentical donors, and that the procedure is safe enough to consider it as a therapeutic option. This observation may be critical in LMICs because the HSCT procedure is rendered more affordable by being conducted as outpatients. Additional studies are needed to either ratify or rectify this observation.

Contribution: O.L.-L. performed research, analyzed data, and wrote the manuscript; S.C.-M. performed research and wrote the manuscript; J.C.O.-G. performed research, analyzed data, and wrote the manuscript; M.R.-N. performed research and wrote the manuscript; M.M.G.-P. performed research and analyzed data; E.J.H.-F. performed research and wrote the manuscript; O.G.-L., Y.V.J.-A., D.G.-A., A.G.-D.-L., and C.H.G.-A. performed research and contributed vital analytical tools; G.E.C.-P. and M.A.V.-L. performed research and wrote the manuscript; and G.J.R.-D. and G.J.R.-A. designed research, contributed vital analytical tools, and wrote the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Guillermo J. Ruiz-Argüelles, Centro de Hematología y Medicina Interna, Clínica Ruiz de Puebla, 8B Sur 3719, 72530 Puebla, Mexico; email: [email protected].