Haploidentical Stem Cell Transplantation With TCR-αβ⁺/CD19⁺ Depletion in High-risk Pediatric Leukemias: Experience From a Referral Center in Peru

Abstract

Haploidentical hematopoietic stem cell transplantation (haplo-HSCT) using TCR αβ+/CD19+ depletion provides an alternative treatment for patients with high-risk (HR) leukemias without a matched donor, especially in developing nations with limited donor registries. We present the outcomes of 36 patients <16 years with HR leukemia who underwent haplo-HSCT with TCR αβ⁺/CD19⁺ depletion between 2018 and 2022 at a referral center in Peru. Survival probabilities and cumulative incidence functions were calculated using the Kaplan-Meier method. Patients were followed for a median of 17.38 months (range: 2.34 to 60.36 mo). The 5-year overall survival (OS), 5-year event-free survival (EFS), and non-relapse mortality rates were 72.1%, 72.2%, and 16.7%, respectively. The incidence of relapse for the entire group was 11.1%. Acute graft versus host disease (GvHD) was observed in 36.1% of the patients, with only 2.8% experiencing grade III-IV acute GvHD. No patients developed chronic GvHD. Among all patients, CMV reactivations were observed in 27.78%, HHV-6 reactivations in 33.33%, and ADV or BK virus reactivations in 16.67%. Our study suggests that haplo-HSCT with TCR αβ+/CD19+ depletion is a safe and effective treatment for HR pediatric leukemias. Adopting this approach in major transplant centers throughout the country could improve outcomes for this group of patients.

HSCT plays a significant role in treating HR leukemias.¹ Allogeneic HSCT (allo-HSCT) is an effective alternative that allows cell recovery after high-dose chemotherapy and triggers the activation of an antitumor response, known as the graft versus leukemia effect (GvL).² A matched sibling remains the preferred donor for allo-HSCT due to the lower incidence of GvHD and transplant-related mortality.³ However, only one-third of patients will have a fully matched donor at the time of transplantation.⁴ Haploidentical HSCT expands the donor pool for patients who require an urgent HSCT without a matched-related or unrelated donor. This type of transplant has become more essential for minority ethnic groups and non-Caucasians who have lower representation in donor registries.⁵

Ex vivo depletion of TCR αβ⁺ cells reduces the alloreactive T cells involved in the pathogenesis of GvHD while maintaining the GvL.^6–8 In contrast, CD19⁺ cell depletion prevents the development of EBV-related posttransplant lymphoproliferative disorder (PTLD).^6–8 This approach has demonstrated an improvement in OS and EFS up to 69% (95% CI: 48%-76%) and 64% (95% CI: 48%-76%), respectively.⁷ Moreover, TCR αβ⁺ depleted transplant decreases the incidence of relapse by up to 21% (95% CI: 11%-34%) with minimal severe forms of GvHD.⁷

This transplant approach has been successfully used worldwide to treat pediatric HR leukemias, showing promising results.^7,9–20 However, no reports for Peru and Latin America are available to date, limiting its potential applicability in other transplant referral hospitals throughout the country and continent.

This study aimed to describe the outcomes of TCR αβ⁺/CD19⁺–depleted haploidentical HSCT (TCR αβ⁺/CD19⁺–depleted haplo-HSCT) for the treatment of HR acute leukemias at the Instituto Nacional de Salud del Niño San Borja (INSN-SB), a pediatric referral center in Lima-Peru.

MATERIALS AND METHODS

Patients

This study included patients under the age of 16 who underwent TCR αβ⁺/CD19⁺–depleted haplo-HSCT at INSN-SB from February 2018 to September 2022. The patient’s demographic and clinical information was collected retrospectively from their medical records.

Patients referred for HSCT were classified as high-risk based on established clinical criteria, which included:

• Failure to achieve a sufficient response to prednisone, evidenced by a bone marrow status classified as M3 on day 14 posttreatment.

• Sustained positive minimal residual disease (MRD) following induction A and/or induction B therapy.

• Detection of positive MRD at any time after an initial period of negative MRD.

• Diagnosis of Philadelphia chromosome–positive acute leukemia.

• Identification of molecular abnormalities associated with an unfavorable prognosis.

None of the patients received posttransplant immunosuppressive therapy. This study was approved by the Institutional Review Board of the INSN-SB under the code PI-763-2022.

Donor Selection, Stem Cell Source, and Graft Manipulation

The selection process for optimal hematopoietic stem cell donors utilized an analysis of KIR haplotypes. Donors displaying the best, better, or neutral KIR haplotypes were classified as the most suitable candidates. Potential donors were restricted to immediate family members, including the father, mother, or siblings.

The source of stem cells for all patients was peripheral blood mobilized with granulocyte colony–stimulating factor (GCSF). Donors received 5 µg/kg of GCSF every 12 hours for 5 days.

Apheresis procedures were performed using the Spectra Optia device (Terumo BCT) following the manufacturer’s recommended protocols. The anticoagulant employed was acid-citrate-dextrose formula A (ACD-A), administered at a standardized input volume to the anticoagulant ratio of 12 to 24:1. The maximum infusion rate for ACD-A was set at 1.0 mL/min/L of total blood volume, with an allowable variation of ±0.2 mL/min. The target dose for CD34⁺ cells was defined at 8 ×10⁶ cells/kg of the recipient, with an acceptable range of 4 ×10⁶ to 12 ×10⁶ cells/kg.

Grafts were manipulated to selectively deplete TCR αβ⁺/CD19⁺ using the CliniMACS Plus device (Miltenye Biotec). Graft goals were ≥4 ×10⁶ CD34⁺/kg, ≤1 ×10⁵ TCR αβ⁺/kg, and ≤1 ×10⁵ CD19⁺/kg.

Conditioning Regimen

The conditioning regimen consisted of rabbit anti-thymocyte globulin (rATG), fludarabine (FLU), cyclophosphamide (CY), busulfan (BU), and thiotepa (TT) (for patients <4 y of age), and fractionated total body irradiation (TBI) 800 cGy. Chemotherapy doses and frequencies varied depending on the primary disease (Table 1).

TABLE 1.

Conditioning Regimens Used for TCR αβ⁺/CD19⁺–depleted Haplo-HSCT

Diagnosis	Conditioning regimen
Acute lymphoblastic leukemia	rATG 1.5 mg/kg/d × 2 + FLU 30 mg/m2/d × 6 + TBI 200 cGy/d × 4 + CY 50 mg/kg/d × 2
Acute myeloid leukemia	rATG 1.5 mg/kg/d × 2 + FLU 30 mg/m2/d × 6 + TBI 200 cGy/d × 3 + CY 50 mg/kg/d × 2
Acute leukemia in patients <4 y	rATG 2.5-7.5 mg/kg given over 2 d + FLU 30 mg/m2/d × 6, BU 4 mg/kg/d × 4 + TT 5 mg/kg/d × 2

BU indicates busulfan: CY, cyclophosphamide: FLU, fludarabine: rATG, rabbit anti-thymocyte globulin: TBI, total body irradiation: TT, thiotepa.

Immunoglobulin infusion was administered selectively, specifically in cases where IgG levels were below 400 mg/dL, and there was a history of recurrent infections. Due to prior experience with early viral reactivation (CMV and HHV6 primarily) causing graft failures, we decided to use a therapeutic dose of ganciclovir for 14 days (last dose day: −2) during the conditioning regimen.

Acyclovir and cotrimoxazole were utilized as prophylactic agents against herpes simplex virus and Pneumocystis jirovecii, respectively. In addition, posaconazole was employed for antifungal prophylaxis, commencing after the conditioning regimen.

Viral Infections Surveillance and Management

CMV and HHV6 reactivations were monitored biweekly through real-time PCR assays in patients undergoing hematopoietic stem cell transplantation. In contrast, weekly surveillance was implemented for ADV, EBV, and BK virus. Upon identification of viral reactivation within the first 3 months posttransplantation, patients received foscarnet at a dosage of 90 mg/kg administered every 12 hours for the treatment of CMV and HHV6. For BK virus and ADV reactivations, cidofovir was prescribed at a dosage of 5 mg/kg on a weekly basis. These proactive treatment strategies successfully resolved viral infections in the patient cohort.

Statistical Analysis

The continuous variables were described in terms of median and interquartile range (IQR) values, while categorical variables were reported as frequencies and percentages.

The probability of OS, EFS, and relapse-free survival (RFS) was estimated using the Kaplan-Meier method. OS was calculated from the time of transplantation until death from any cause. EFS was calculated from the time of transplantation to either relapse or death from any cause. RFS was calculated from the time of transplantation until the date of relapse.

The cumulative incidence of relapse (CIR) and GvHD were calculated using competing risk analysis, with death as a competing event. Non-relapse mortality (NRM) was defined as death after HSCT without evidence of relapse or disease progression.

Platelet engraftment was defined as achieving a platelet count exceeding 20,000/mm³, independent of platelet transfusion support. Neutrophil engraftment was established as having a neutrophil count >500 cells/mm³​​​​​​ without the administration of GCSF. Immune reconstitution was evaluated through the measurement of natural killer (NK) cells 1 month post-HSCT, using flow cytometry.

Statistical analysis was performed by using R studio v4.3.1. A P value of <0.05 was considered statistically significant.

RESULTS

Patients and Graft

Thirty-six children were included in the study. The median age at transplantation was 5.87 years (IQR: 6.01; range: 2.72 to 15.12 y) (Table 2). Twenty-three of the patients (64%) were males. Median follow-up time was 17.38 months (IQR: 15.57; range: 2.34 to 60.36 mo).

TABLE 2.

Patient and Graft Characteristics

Variable	N (%)
Sex
Female	13 (36.1)
Male	23 (63.9)
Median age (IQR; range)	5.9 (6.01; 2.72-15.12)
Diagnosis
Acute lymphoblastic leukemia	27 (75)
Acute myeloid leukemia	9 (25)
Cytogenetics
Normal karyotype	10 (27.8)
Structural abnormalities	4 (11.1)
t(1;19)(q23;p13.3)	1 (2.8)
t(9;22)(q34;q11.2)	1 (2.8)
t(1;11)(q21;q23)	1 (2.8)
t(8;21)(q22;q22)	1 (2.8)
Numerical abnormalities	1 (2.8)
-Y	1 (2.8)
Structural and numerical abnormalities (<3 abnormalities)	3 (8.3)
-Y, t(2;14)(q35;q11.2)	1 (2.8)
-Y, t(8;21)(q22;q22)	2 (5.5)
Complex karyotype (≥3 abnormalities)	5 (13.9)
Insufficient metaphases	5 (13.9)
Data not available	8 (22.2)
Gene fusions
ETV6-RUNX1	2 (5.6)
BCR-ABL1	2 (5.6)
TCF3-PBX1	1 (2.8)
AML1-ETO*	3 (8.3)
None of the above	22 (61.1)
Data not available	6 (16.7)
Disease status
CR1	33 (91.6)
CR2	2 (5.6)
CR3	1 (2.8)
CMV serostatus (donor/recipient)
Donor/recipient−	3 (8)
Donor/recipient+	33 (92)
Median cell dose (range)
CD34+ × 106/kg	11.58 (4.7-23.93)
TCR αβ+ × 105/kg	0.65 (0.08-2.95)
CD19+ x 105/kg	0.27 (0.03-5.52)
TBI regimen
Yes	30 (83.3)
No	6 (16.7)

^*AML1-ETO fusion was only evaluated in the 9 patients diagnosed with AML.

CR1 indicates first complete remission; CR2, second complete remission; CR3, third complete remission; IQR, interquartile range; TBI, total body irradiation.

Acute lymphoblastic leukemia (ALL) accounted for 75% (27 patients) of the transplants. Among the 9 patients with acute myeloid leukemia (AML), 3 (8.3%) exhibited the AML1-ETO fusion. Within the group of ALL patients, 2 harbored the ETV6-RUNX1 (5.6%) fusion, 2 the BCR-ABL1 fusion (5.6%), and 1 (2.8%) the TCF3-PBX1 fusion. Most of the patients displayed a normal karyotype. Among those with cytogenetic abnormalities, 13.9% presented with a complex karyotype (≥3 abnormalities), while 11.1% and 2.8% exhibited only translocations and single chromosome loss, respectively. In addition, 8.3% of patients harbored translocations accompanied by the loss of a single chromosome.

Of the 36 patients, 33 (91.7%) were in first complete remission (CR1), 2 (5.6%) in second complete remission (CR2), and 1 (2.8%) in third complete remission (CR3). All patients had undetectable MRD by flow cytometry at the time of transplantation.

The median cellularity count of CD34⁺, TCR αβ⁺, and CD19⁺ were 11.58 ×10⁶ cells/kg (IQR: 5.16; range: 4.7 to 23.93 × 10⁶ cells/kg), 0.27 ×10⁵ cells/kg (IQR: 0.73, range: 0.03 to 5.52 × 10⁵ cells/kg), and 0.65 ×10⁵ (IQR: 0.49, range: 0.08 to 2.95 × 10⁵ cells/kg), respectively.

Engraftment and Graft Failure

Thirty-one patients (86.1%) achieved engraftment. Four patients younger than 4 years old with ALL and 1 patient transplanted for acute myeloid leukemia (AML) were diagnosed with primary graft failure. Neutrophils and platelets engraftment occurred at a median of 15 days (IQR: 2.5; range: 12 to 24 d) and 11 days (IQR: 2; range: 6 to 18 d), respectively. No patients presented secondary graft failure.

The characteristics of the patients who experienced graft failure are detailed in Table 3. Only 1 patient received a second haploidentical HSCT, while the others underwent autologous recovery and achieved remission.

TABLE 3.

Characteristics of the patients with graft failure

Diagnosis	Age	Sex	CD34+ dose (cells × 106/kg)	CD19+ dose (cells × 105/kg)	TCR αβ+ dose (cells × 105/kg)	Conditioning regime	CMV serostatus (donor/recipient)	Viral reactivation	Outcome after HSCT
ALL	3.5	F	7	0.06	1.9	BU/FLU/TT	+/+	No	Dead
ALL	2.9	M	14	0.15	0.18	BU/FLU/TT/rAGT	+/+	Yes (HHV6)	Autologous recovery
ALL	3.5	F	14	0.07	0.39	BU/FLU/TT/rAGT	+/+	No	Autologous recovery
ALL	3.4	F	5.22	0.1	1	BU/FLU/TT/rAGT	+/+	No	Autologous recovery
AML	9.0	M	10.60	5.52	1.61	FLU/ rAGT/TBI/CY	+/+	No	Dead

ALL indicates acute lymphoblastic leukemia; AML, acute myeloid leukemia; BU, busulfan: CY, cyclophosphamide: F, female; FLU, fludarabine: M, male; rATG, rabbit anti-thymocyte globulin: TBI, total body irradiation; TT, thiotepa.

Survival, Relapse, and Events

Of the 36 patients transplanted for acute leukemia, 26 were alive and well until the last follow-up. The 5-year OS, EFS, and RFS were 72.1% (95% CI: 58.8%-88.4%), 72.2% (95% CI: 59%-88.4%), and 87.5% (95% CI: 76.7%-99.8%), respectively (Figs. 1A–C).

FIGURE 1.

Principal outcomes of TCR αβ⁺/CD19⁺–depleted haplo-HSCT in HR pediatric acute leukemias. A, OS; B, EFS; C, RFS; D, NRM.

During the first year post-HSCT, 4 patients relapsed. The CIR was 11.1% (95% CI: 3.5%-23.8%), while NRM was 16.7% (95% CI: 6.7%-30.6%) (Fig. 1D).

Ten patients (27.8%) died after HSCT. Among this group, 8 patients were diagnosed with systemic infections caused by either CMV or ADV virus, 1 patient had an invasive fungal infection and tuberculosis, and 1 patient experienced primary graft failure, which was further complicated by infections. The median time to death was 6.47 months (IQR: 4.97; range: 2.33 to 14.36 mo).

The 3-year OS rate for patients with ALL was 77.8% (95% CI: 63.6%-95.2%), while for patients with AML, it was 55.6% (95% CI: 31%-99.7%) (Fig. 2A). EFS was 77.8% for ALL (95% CI: 57.1%-89.3%) and 55.6% for AML (95% CI: 20.4%-80.5%) (Fig. 2B). There were no statistical differences in the OS and EFS rates between ALL and AML (P = 0.24 and P = 0.25, respectively, Figs. 2A, B). The median follow-up time for patients with ALL and AML was 17.50 months (IQR: 14.56; range: 2.34 to 43.57 mo) and 16.13 months (IQR: 11.76; range: 2.99 to 60.28), respectively. In patients with ALL, the CIR and NRM were 7.4% (95% CI: 5.1%-21.4%) and 14.8% (95% CI: 4.5%-30.8%), respectively. For patients with AML, both the CIR and NRM were 22.2%, with a 95% CI of 2.6%-53.7% for CIR and 2.8%-53.1% for NRM.

FIGURE 2.

Comparison of OS (A) and EFS (B) between ALL and AML.

Graft Versus Host Disease

The cumulative incidence of acute GvHD (aGvHD) on day 100 was 36.1% (95% CI: 20.8%-51.7%). Only 1 patient, who had severe infectious complications (HHV6, ADV, and pulmonary aspergillosis), developed severe aGvHD grade III-IV. The patient had a poor response to steroids and died from multiorgan failure and hemorrhagic cystitis caused by ADV.

aGvHD was resolved after treatment with topical steroids. No cases of chronic GvHD (cGvHD) were reported during the follow-up period.

Infections

Reactivation of CMV, HHV6, and ADV or BK virus was observed in 27.8%, 33.3%, and 16.7% of the patients, respectively. The median time to reactivation was 32.5 days for CMV, 20.5 days for HHV6, and 58.5 days for ADV or BK virus.

Immune Reconstitution

The CD4+, CD8+, CD19+, and NK cell counts are detailed in Table 4. Normal counts of NK and CD19+ cells were achieved 1 and 3 months after HSCT, respectively, with median counts of 295 and 272 cells/μL, respectively. In contrast, normal counts of CD4+ and CD8+ cells were achieved 9 months after HSCT, with median counts of 544 and 350 cells/μL, respectively.

TABLE 4.

Immune Recovery After HSCT

	Months after HSCT
	1 mo		3 mo		6 mo		9 mo		12 mo
Cell population	Median	Range	Median	Range	Median	Range	Median	Range	Median	Range
CD4+ (cell/μL)	22	3-277	128.5	38-383	301	119-1085	544	215-1190	720	355-1400
CD8+ (cell/μL)	69	4-368	112.5	17-513	211.5	35-766	350	157-756	541.5	266-683
CD19+ (cell/μL)	7	0-49	272	0-556	420	0-778	459	162-682	525	179-1987
NK (cell/μL)	295	30-1697	439	100-1549	280	97-968	183	103-1160	233	152-643

DISCUSSION

Haploidentical HSCT with TCR αβ⁺/CD19⁺ depletion is a valuable treatment option for patients with hematologic malignancies who do not have a matched sibling donor or a matched unrelated donor available. Depletion of TCR αβ⁺/CD19⁺ cells can aid in preserving crucial cell populations such as CD34 progenitors, TCRγδ⁺ cells, and NK cells.²¹ These cells do not exhibit alloreactivity and can offer protection against infections and leukemia.^22–27 Furthermore, prophylactic depletion of CD19⁺ cells can prevent EBV PTLD.^28,29

This study is the first to report the outcomes of pediatric patients with HR leukemias who underwent TCR αβ⁺/CD19⁺–depleted haplo-HSCT in Peru and Latin America. It has been documented that Peruvian patients with acute leukemia who experience disease relapse have a 5-year OS rate of <35% with the current treatment regimens.³⁰ Although >70% of the population had HR ALL, most patients included in this study were transplanted in CR1 and MRD-negative disease.

The outcomes reported in this study, including OS, EFS, and RFS, exceeded the rates reported in previous research. For instance, previous studies on pediatric patients showed OS rates of 51%, 55.1%, and 29±5%, whereas EFS and RFS rates were 50% or lower.^13,31–33 We hypothesize that the improved outcomes observed in our patients may be attributed to the early identification of high-risk cases and prompt transplantation in patients with complete response, in addition to other factors related to the disease and the transplant process (such as relapse rate, immune recovery, administration of antiviral prophylaxis, infusion dose of TCR αβ+/CD19+, among others).

Of all the patients included in the study, only 11% experienced a relapse, which is a significant improvement compared to the findings of Yanir et al.³⁴ According to their study, the 3-year incidence of relapse was 47%, which was suggested as the cause of the low EFS (35%).³⁴ The lower incidence of relapse in our cohort may be attributed to the patients being referred to transplantation in the best possible health status. Most of our patients received HSCT in CR1 and without MRD detected.

NRM after HSCT is mainly caused by infections and GvHD. TCR αβ⁺ depletion significantly reduces the incidence of severe acute and chronic GvHD, thereby improving NRM. Patients who received haploidentical TCR αβ⁺/CD19⁺ depleted grafts show lower NRM rates compared to those who received mismatched unrelated donor grafts (9% vs. 28%, P<0.001).²⁰

Our study found that the NRM rate was significantly better compared to other studies conducted by Klingebiel et al³¹ and Dadi et al,³² who reported TRM rates of 37±4% and 36% (95% CI: 18%-53%), respectively. This positive outcome could be attributed to the low rates of CMV reactivation, which was made possible by the introduction of CMV prophylaxis during the conditioning regimen and the absence of posttransplant immunosuppressive therapy. Moreover, we observed a rapid recovery of NK cells in the first month post-HSCT, which could have contributed to the improved NRM due to their GVL and antiviral activity.³⁵ Proper patient treatment, adequate complications management, transplant center experience, and timely use of antivirals are additional factors that may contribute to lower NRM rates, all of which were successfully addressed in this study population.³⁴

Of all the patients who underwent transplant, 86% achieved engraftment. Bielorai et al³⁶ highlighted the importance of TBI in achieving successful engraftment. Three of 4 patients who failed to engraft did not receive TBI, which confirms the crucial role of TBI in TCR αβ⁺/CD19⁺–depleted haplo-HSCT. According to a European Society for Blood and Marrow Transplantation report, primary and secondary graft failures were 9% and 4%, respectively.³¹ However, our study did not observe any secondary graft failure. The incidence of primary graft failure was 13.9%, likely due to the absence of TBI in younger patients.

TCR αβ⁺ lymphocytes have been identified as the main contributors to the development of GvHD.³⁷ Previous clinical experiences in pediatric cohorts who underwent TCR αβ⁺/CD19⁺–depleted haplo-HSCT have shown low rates of GvHD, with mostly cutaneous manifestations and almost no cases of visceral forms of acute and chronic GvHD.^38,39 In our cohort, we observed an incidence rate of aGvHD of 36%, characterized exclusively by cutaneous manifestations. However, these were resolved without the need for systemic immunosuppressive therapy. Only 1 patient developed grade III-IV aGvHD, and we had no incidence of cGvHD.

It has been suggested that the risk of developing grade II-IV GvHD increases with the volume of TCR αβ⁺ cells in the graft.⁴⁰ To mitigate the risk, a previously established residual TCR αβ⁺ count threshold of 1×10⁵ cells/kg is recommended.²⁰ In our cohort, we followed this recommendation to ensure that all patients received grafts containing <1×10⁵ cells/kg, thereby reducing the risk of developing GvHD.

Reactivations of viral infections were frequently observed in our cohort, with CMV and HHV6 being the most common. Im et al³³ also reported a similar incidence of CMV reactivation (34±8%) during the first 38 days post-HSCT. This similarity may be due to the use of similar antiviral prophylaxis schemes, among other factors such as matched serology between donors and recipients and early NK cell recovery.

The time for immune reconstitution is a critical factor for successful TCR αβ⁺/CD19⁺–depleted haplo-HSCT. Studies have shown that recovery of TCR γδ⁺ cells and NK cells occurs early after the transplant.^21,41 TCR γδ⁺ cells start to recover from day +7, while NK cells reach normal levels around day +30.^21,41 In contrast to TCR γδ+ cells and NK cells, B and T lymphocytes displayed a delayed recovery after HSCT. CD19+ B lymphocyte counts returned to normal levels 3 months after HSCT, while CD4+ and CD8+ T lymphocyte counts normalized almost 1 year after HSCT.²¹

In line with these findings, our study observed an early recovery of NK cells just 1 month after the HSCT procedure, which could have contributed to improved survival rates in the cohort.

Although this study has some limitations, such as a relatively small sample size and a short follow-up period, it is important to note that the pediatric group included in this study was the first in Peru to undergo an HSCT with TCR αβ⁺–depleted grafts. As the INSN-SB serves as a pediatric referral center, the knowledge gained from this study can be useful for other pediatric health care institutions performing HSCT across the country.

The cost of the depletion procedure can be compensated by eliminating the need for immunosuppressive therapy after HSCT. The economic implications associated with the cellular depletion procedure can be significantly alleviated by the subsequent discontinuation of immunosuppressive therapy following hematopoietic stem cell transplantation (HSCT). An internal cost analysis conducted at our institution indicates that the cumulative costs related to the procurement of intravenous tacrolimus and mycophenolate, including blood level monitoring over a 20-day period and an additional 6 months of oral administration alongside routine blood evaluations, are comparable to the expenses incurred from the use of cell separation kits and the cellular depletion technique. In addition, the limited availability of intravenous formulations of tacrolimus and mycophenolate in the region presents notable challenges to the implementation of optimal post-HSCT immunosuppressive therapies.

Depletion plays a crucial role in optimizing graft composition, which is essential for preventing complications such as graft rejection, disease recurrence, infections, and GvHD. All of these factors have a significant impact on the OS rate and enhance the quality of life of patients. With reduced hospital visits, patients can spend more time at home and attend follow-up appointments better, which is crucial for a successful transplant, especially in Peru, where most patients come from low-resource families.

The implementation of TCR αβ⁺/CD19⁺–depleted haplo-HSCT at our health facility presented several challenges. Notably, there was an initial shortage of essential medications necessary for the treatment of viral and fungal infections, as well as a lack of laboratory assays for the timely identification of viral infections. Furthermore, optimizing the parameters for cell depletion was critical to achieving a target of <1 ×10⁶ αβ+/kg. In this regard, it is imperative to highlight that ex vivo depletion of TCR αβ+/CD19+ cells should be conducted only in health care institutions that possess substantial expertise, as well as the requisite financial and logistical resources to ensure successful outcomes. These institutions should be capable of ensuring the timely and appropriate management of any potential complications that may arise from this type of transplant. Furthermore, such institutions should possess adequate financial capacity to acquire or lease a cell separation instrument. In our specific case, the CliniMacs instrument was not independently acquired during the implementation of TCR αβ+/CD19+–depleted haplo-HSCT; instead, it was provided by the kit manufacturer as part of the procurement of cell separation kits. Stem cell processing was conducted by an external entity, facilitating hospital personnel’s concurrent training.

On the other hand, intravenous tacrolimus and mycophenolate, which are typically employed in post-HSCT immunosuppressive therapy, are currently unavailable in Peru. A preliminary analysis conducted at our institution comparing the costs and outcomes of cell depletion versus conventional post-HSCT immunosuppressive therapy suggests that the expenses associated with these medications—including necessary blood level monitoring over a 20-day period, followed by 6 months of oral treatment and blood evaluations—are approximately comparable to the costs of procuring cell separation kits and executing the cellular depletion process (unpublished data from INSN-SB). Moreover, this approach appears to be associated with lower rates of GVHD and relapse (unpublished data from INSN-SB), potentially leading to improved DFS and OS. Therefore, the use of cell depletion as a prophylactic strategy against GVHD may not only be cost-equivalent to the combined expenses associated with post-infusion cyclophosphamide, tacrolimus, and mycophenolate over a 6-month period, but it may also yield superior clinical outcomes.

In summary, our results suggest that TCR αβ⁺/CD19⁺–depleted haplo-HSCT represents a safe and effective treatment for HR pediatric leukemias as it prevents the development of severe acute and chronic GvHD that may require systemic therapy. Peruvian health authorities should consider implementing this technique in other pediatric transplant centers across the country to expand the chances of survival of children in need of an urgent HSCT.

Further studies are needed in other pediatric populations across Latin America to support the use of this transplantation technique throughout the continent. Moreover, to increase our understanding regarding the efficacy and safety of TCR αβ+/CD19+–depleted haplo-HSCT in HR pediatric leukemias, it is recommended that additional investigations be carried out on a more diverse ethnic population with larger sample sizes, longer follow-up periods, and comprehensive data on the immune recovery of all relevant cellular populations, as well as all types of infections.