EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Therapy-related acute myeloid leukemia often exhibits adverse biologic features and treatment resistance. Allogeneic hematopoietic stem cell transplant as consolidation in remission status after initial therapy offers the greatest possibility of long-term disease control. For patients without options of human leukocyte antigen-matched family donor or unrelated human leukocyte antigen-matched donor or killer immunoglobulin-like receptor-favorable haploiden-tical donor, an alternative strategy using vaccine hematopoietic stem cell transplant such as comple-mentary haplo-cord hematopoietic stem cell transplant could have good graft-versus-leukemia effect and enable long-term remission. Here, we reported a successful case of more than 3 years of leukemia-free survival in a patient with therapy-related acute myeloid leukemia who underwent haplo-cord hematopoietic stem cell transplant with vaccine effect.

Key words : Haploidentical transplantation, Killer immunoglobulin-like receptor, Umbilical cord blood stem cell transplantation

Therapy-related acute myeloid leukemia (t-AML) often exhibits adverse biologic features and treat-ment resistance.¹ Allogeneic hemopoietic stem cell transplant (HSCT) as consolidation after remission being achieved by initial therapy offers the greatest possibility of long-term disease control.² For patients without human leukocyte antigen (HLA)-matched family donors or HLA-matched unrelated donors, the Pediatric Transplantation and Cellular Therapy Consortium ONC1401 trial has previously demons-trated superior HSCT outcomes and comparable 1-year disease-free survival from killer immunog-lobulin-like receptor (KIR)-favorable haploidentical donor and TCRαβ+/CD19+-depleted peripheral blood stem cell transplant.³ However, for patients without a KIR-favorable haploidentical donor, an alternative strategy using vaccine HSCT such as complementary haplo-cord HSCT with major histocompatibility complex-restricted peptide-specific vaccine effect^4-7 (Figure 1, Figure 2) enables long-term remission.^8-10 Here, we reported a successful case of long-term remission in a patient with t-AML who underwent haplo-cord HSCT due to unavailability of an alternative donor.

A 15-year-old girl with metastatic left shoulder osteosarcoma achieved complete remission after treat-ment according to the Hong Kong Pediatric Hematology and Oncology Study Group Osteosarco-ma 2009 protocol,¹¹ which comprised methotrexate, doxorubicin, and cisplatin. The patient experienced t-AML 18 months after completion of chemotherapy.¹² With KMT2A-MLLT³ rearrangement and karyotype 46,XX,t(9;11)(p²²;q²³)[29]/46,XX[1], intermediate prognosis was conferred according to 2017 European LeukemiaNet risk stratification.^13,14 The TP53 muta-tion had not been demonstrated. Because prior treatment was complicated with methotrexate-induced transaminitis, cisplatin-induced renal tubulopathy, and bilateral grade 2 sensori-neural hearing loss, the patient received low-intensity regimen with venetoclax and azacitidine to induce complete remission of leukemia. Measurable residual disease was 0.01% by flow cytometry before HSCT. Premorbid status prior to HSCT reflected the following details: Eastern Cooperative Oncology Group performance status of 2, Karnofsky perfor-mance scale score of 50, Charlson comorbidity index of 2, and hematopoietic cell transplant-specific comorbidity index of 3. Because HLA-matched family donors or HLA-matched unrelated donors were not available and because both parents were not KIR-favorable candi-dates (no KIR-ligand mismatch and KIR B-score of 0), haplo-cord transplant was planned. The father was medically unfit for donation, and the 7/8-matched mother (C allelic mismatch) was chosen as the stem cell donor. A 5/8-matched umbilical cord blood (UCB) unit (A, B, and C antigen mismatched with the shared maternal-child antigens; total nucleated cells 3.8 × 107 cells/kg, CD34+ cells 2.1 × 105 cells/kg) was selected (Figure 1). Conditioning included cytarabine 100 mg/m²/day for the 7-day period from day -9 to day -3, cyclophosphamide 15 mg/kg/day for the 2-day period from day -8 to day -7, intravenous busulfan 100 mg/m²/day for the 4-day period from day -6 to day -3 and fludarabine 40 mg/m²/day for the 3-day period from day -4 to day -2. Unmanipulated fresh granulocyte colony-stimulating factor-mobilized peripheral blood stem cells (total nucleated cells 10 × 108 cells/kg, CD34+ cells 8.8 × 106 cells/kg) were infused on day 0. The UCB was infused on day +6 to augment the graft-versus-leukemia (GvL) effect and prevent relapse. Cyclophosphamide 50 mg/kg/day and fludarabine 40 mg/m²/day were given for the 2-day period from day +3 to day +4 to induce tolerance of the maternal graft with regard to the recipient and the cord. Prophylaxis for graft-versus-host disease (GvHD) with tacrolimus and mycop-henolate mofetil was started on day +7. Neutrophil engraftment was achieved on day +15, and 100% UCB chimerism had been achieved and maintained since day +13. Serial bone marrow examinations at post-HSCT months 1, 3, 6, and 12 demonstrated morphological and molecular remission with undetectable minimal residual disease. At the time of this writing, the patient remained alive and leukemia-free for more than 3 years after HSCT.

Management approach of therapy-related myeloid neoplasms
All patients who have previously been treated with cytotoxic chemotherapy or radiotherapy are at risk to develop therapy-related myeloid neoplasms (t-MNs) including AML, myelodysplastic syndromes, or myeloproliferative neoplasms due DNA damage. It has been established that t-MNs often exhibit adverse biologic features and treatment resistance and usually confer worse outcomes than de novo myeloid neoplasms, especially for those with mutated TP53 as recognized by the International Classification Consensus as a new diagnostic category. Risk of t-MNs associated with topoisomerase II inhibitors, such as doxorubicin used in the osteosar-coma protocol in our patient, appears to be constant across age spectrum with shorter latency period (1-3 years after treatment as AML without antecedent myelodysplastic syndromes or myeloproliferative neoplasms), in contrast to alkylation agents and radiotherapy, which usually increases with age (usually developing 5-7 years after treatment). After 10 years of treatment, no increased risk of t-MNs has been shown for patients treated for nonhematolo-gical malignancies versus 3-fold to 6-fold increased risk for patients with Hodgkin lymphoma, non-Hodgkin lymphoma, and multiple myeloma. Abnormal karyotype has been demonstrated in more than 90% of t-MN cases, as in our case, reflecting possible underlying pathophysiology of mutational events induced by cytotoxic therapy and clonal hematopoiesis of chemotherapy-resistant aberrant clones. Translocations involving KMT2A at chromosome 11q²³.3 (as in our patient), RUNX¹ at 22q²².1, or RARA at 17q²¹.2 are common aberrations associated with t-MNs induced by topoisomerase II inhibitors. For medically fit patients, defined as Eastern Cooperative Oncology Group performance status ≤2, Karnofsky performance scale score >50, Charlson comorbidity index ≤2, and/or hemato-poietic cell transplant-specific comorbidity index ≤3, with favorable or intermediate prognosis, induction of remission with intensive therapy followed by allogeneic HSCT offers the best chance of long-term survival, as shown in our patient.

Bridging chemotherapy for allogeneic hemato-poietic stem cell transplant
Intensive therapy, such as the use of anthracy-cline-containing regimens (eg, conventional 7+3 cytarabine/anthracycline regimen or the CPX-351 liposomal formulation), is commonly used to induce remission. However, some patients are ineligible because of prior anthracycline therapy at a high cumulative dose and/or substantial toxicity; therefore, lower intensity treatment with hypo-methylating agent azacitidine and B-cell lymphoma 2 inhibitor venetoclax can be used to achieve disease remission prior to HSCT, as in our case.

Donor selection and human leukocyte antigen vaccine effect
Unrelated UCB transplant (UCBT) represents a crucial option in the treatment of high-risk pediatric leukemia, as demonstrated by the first clinical trial published by Joanne Kurtzberg and colleagues in 1996.15 Unrelated UCBT offers a viable source for grafts in allogeneic HSCT, particularly when a matched donor is not available, and a primary advantage is the ready availability of cord blood units. Moreover, UCBT is associated with a lower incidence of GvHD versus unrelated bone marrow HSCT, a critical consideration in the pediatric population for whom long-term complications can significantly affect the quality of life. The patient’s disease status, degree of HLA match, and the cell dose provided by the cord blood unit were the important known factors affecting the outcome. Double-unit UCBT was introduced to overcome the cell dose limitation inherent in single-unit UCBT. Early studies have demonstrated the safety and feasibility of double-unit UCBT, further fueling the interest in this approach. However, a truly intriguing observation is that double-unit UCBT may confer an additional benefit beyond merely increasing the cell dose; that is, double-unit UCBT may confer a lower rate of leukemia relapse. This finding has prompted speculation that the GvL effect is potentiated by the graft-versus-graft (GvG) effect. It has been hypothesized that the immunological interaction during the double-unit series that induces the in vivo dominancy GvG effect may serve to enhance the GvL effect. Subsequent investigations have provided compelling evidence for this notion, revealing the presence of alloreactive effector CD8+ T cells, and subsequently also CD4+ T cells from the engrafted unit, capable of mounting a robust immune response against the common HLA antigens shared between the non-engrafted unit and the recipient cells. That is, the non-engrafted cord blood unit may act as a “vaccine” to potentiate the GvL effect of the engrafted cord blood unit when non-engrafted cord blood unit shares the same HLA antigen(s) with the host, mismatched with the engrafted cord blood unit. The vaccine effect was defined as the non-engrafted cord blood unit sharing the same HLA antigen(s) with the host but that HLA was not found in the engrafted cord blood unit in a double-unit UCBT.

Rationale for complementary (haplo-cord) transplant
Allogeneic HSCT as consolidation after remission for t-AML offers the greatest possibility of long-term disease control. For patients without HLA-matched family donors or HLA-matched unrelated donors, a KIR-favorable haploidentical donor with a TCRαβ+/CD19+-depleted graft appears to be a reasonably good alternative.⁵ However, for patients without a KIR-favorable haploidentical donor, a different strategy has to be considered. Unrelated UCBT may be a good option because of potentially higher anti-leukemia potential, especially when 2 units are used. Previous studies on double-unit UCBT have suggested that GvG immunological interaction between the 2 grafts may enhance the GvL effect associated with the rejected graft-specific CD4+ and CD8+ T cells. Compared with double-unit UCBT, complementary haplo-cord transplant shortens cytopenia phase and possibly confers enhanced GvL effect, low GvHD risk, and reduced risk of disease relapse (Figure 2). In contrast to selection of cord blood unit and haploidentical donor based on mutual haploidentical donor mismatched antigen, defined by the Beijing group as a major HLA locus mutually harbored by the UCB and recipient but mismatched with the haploidentical donor in the haplo-versus-UCB or conventional GvH direction in the haplo-HSCT system¹⁶, selection of cord in our patient was based on HLA mismatches between the cord and the shared major HLA loci of the mother donor and the patient, a process that facilitates the major histocompatibility complex-restricted peptide-specific vaccine effect (Figure 1) and presumably augments the GvL effect and prevents disease relapse. Our transplant regimen followed the description from Li and colleagues,⁹ who have reported many patients with UCB engraft-ment and 93% 1-year leukemia-free survival.

Haplo-cord transplant is a promising approach to improve outcomes for hematological malignancies including t-AML, presumably attributed to the GvG effect, which enhances the GvL effect and hence reduces relapse risk. This relatively new transplant strategy has the potential for wider clinical use especially in very high-risk patients with leukemia who lack a suitable conventional donor, and further investigations are warranted.

Volume: 24 Issue: 6 June 2026

FULL TEXT