Venetoclax-Azacitidine Bridging PTCy-haplo-PBSCT for Refractory Acute Myeloid Leukemia with IDH2 Mutations

Venetoclax and azacitidine combination therapy (VEN+AZA) is a promising novel therapy for elderly or unfit patients with acute myeloid leukemia (AML), owing to its excellent response and tolerability [1, 2]. Recently, VEN+AZA and subsequent allo-hematopoietic stem cell transplantation (allo-HSCT) have emerged as a superior treatment, rather than VEN+AZA maintenance alone, owing to a superior outcome in patients and good tolerability for elderly and fragile patients [3, 4]. However, the duration of response achieved by VEN+AZA is relatively short, and it is crucial to promptly find donor sources for patients who receive VEN+AZA as a bridging treatment to allo-HSCT. Owing to ease of accessibility to donor sources and good tolerability of transplantation, human leukocyte antigen (HLA)-haploidentical peripheral blood stem cell transplantation using posttransplant cyclophosphamide (PTCy-haplo-PBSCT) from related donors was chosen as a suitable treatment in this instance. Here, we report two elderly patients with refractory AML harboring an IDH2 mutation, who were successfully treated with VEN+AZA bridging PTCy-haplo-PBSCT.

A 69-year-old female, who had received adjuvant chemotherapy for myxofibrosarcoma of the right femur 7 years ago, was diagnosed with therapy-related AML. Genetic analysis of the bone marrow (BM), using next-generation sequencing (NGS), revealed that the tumor clone harbored IDH2^R140Q, NRAS, and KRAS mutations (Table 1). Daunorubicin and cytarabine combination therapy (DNR+Ara-C) was initiated as induction therapy. Unfortunately, the patient developed severe pneumonia on day 2, and induction therapy was suspended on the same day. Antibiotic therapy with methylprednisolone pulse followed by steroid therapy was initiated. Although she recovered from severe pneumonia, normal hematopoiesis did not recover, and bone marrow aspiration revealed that 44% of the BM cells were myeloblasts. Therefore, we decided to start VEN+AZA as a salvage chemotherapy. Although grade 4 leukopenia and neutropenia were prolonged for about 3 weeks, normal hematopoiesis gradually recovered, and non-hematological toxicity was tolerable by VEN+AZA. After one cycle of VEN+AZA treatment, hematological complete remission (CR) was achieved. Another cycle of VEN+AZA and one cycle of azacitidine monotherapy were added, and subsequently, she received PTCy-haplo-PBSCT from a related donor who was HLA 4/8 matched. Fludarabine 30 mg/m² and busulfan 3.2 mg/kg were administered for 5 days and 2 days, respectively, and 4 Gy of total body irradiation was administered as reduced-intensity conditioning. Posttransplantation cyclophosphamide levels were 40 mg/kg on day 3 and day 4, and tacrolimus and mycophenolate mofetil (TAC+MMF) was started on day 5 as graft-versus-host disease (GVHD) prevention. Engraftment was performed smoothly on day 17, and no acute GVHD appeared after engraftment. The patient’s clinical course after transplantation was uneventful, and she was discharged on day 47. The clinical course from the start of VEN+AZA treatment to transplantation is shown in Figure 1. It also shows the levels of Wilms tumor 1 (WT1) mRNA in peripheral blood (PB) and variant allele frequency (VAF) of IDH2^R140Q mutation in the BM. In this case, WT1 mRNA of PB on day 30 after transplantation was <50 copies/μg RNA, and hematological CR was maintained on day 180 after transplant. Unfortunately, the patient died on day 193 due to thrombotic microangiopathy.

A 62-year-old male was diagnosed with AML with intermediate risk, according to the ELN 2017 classification [5], indicating that the tumor clones harbored IDH2^R172K, DNMT3A, BCOR, and BCORL1 mutations (Table 2). DNR+Ara-C was initiated as induction therapy, during which he developed a skin infection of Stenotrophomonas maltophilia and recovered after administration of sulfamethoxazole/trimethoprim. Unfortunately, normal hematopoiesis did not recover, and bone marrow aspiration on day 28 revealed that 29% of BM cells were myeloblasts. Therefore, we decided to start VEN+AZA as a salvage chemotherapy. Although grade 4 leukopenia and neutropenia were prolonged for about 4 weeks, normal hematopoiesis gradually recovered, and non-hematological toxicity was tolerable by VEN+AZA. After one cycle of VEN+AZA treatment, hematological CR was achieved. An additional cycle of VEN+AZA was added, and the patient subsequently received PTCy-haplo-PBSCT from a related donor who was HLA 4/8 matched. Fludarabine 30 mg/m² and busulfan 3.2 mg/kg were administered for 5 days and 2 days, respectively, and 4 Gy total body irradiation was administered as reduced-intensity conditioning. Posttransplantation cyclophosphamide was administered at 40 mg/kg on days 3 and 4, and TAC+MMF was initiated on day 5 for GVHD prevention. Engraftment was performed smoothly on day 14, and no acute GVHD appeared after engraftment. The patient’s clinical course after transplantation was uneventful, and he was discharged on day 48. The clinical course from the start of VEN+AZA treatment to transplantation is presented in Figure 2. It also shows levels of WT1 mRNA of PB and VAF of IDH2^R172K mutation of BM. In this case, WT1 mRNA of PB on day 30 after transplantation was <50 copies/μg RNA, and hematological CR was maintained on day 360 after transplantation. One and a half years have passed since the transplant, and the patient is still alive without relapse.

The duration of response to VEN+AZA was relatively short if exclusively VEN+AZA was continued as maintenance therapy, and subsequent allo-HSCT using cord blood and a matched sibling donor for newly diagnosed AML patients older than 60 years was reported to be promising [4]. In contrast, PTCy-haplo-PBSCT has advantages, including donor source availability, high tolerability associated with early engraftment, and low incidence of severe acute GVHD among alternative donor sources. These properties seem suitable for subsequent allo-HSCT after VEN+AZA treatment, particularly in elderly AML patients. Notably, PTCy-haplo-PBSCT was promptly performed after bridging VEN+AZA therapy and showed good tolerability in elderly patients with AML. As mentioned above, cord blood is a reasonable alternative graft for post-VEN+AZA transplantation [4]; however, which graft is better remains undetermined.

Generally speaking, hematological toxicity of VEN+AZA is severe, and grade 4 leukopenia and neutropenia are often prolonged and immediate transplant after achievement of hematological CR seems to be a reasonable option to avoid additional risk. However, we needed time to prepare the donor for PB stem cell collection and we considered additional chemotherapy as consolidation therapy to prevent relapse. Then, we decided to administer one more cycle of VEN+AZA and one cycle of AZA in the case 1 and one more cycle of VEN+AZA in the case 2. On the other hand, maintenance therapy of IDH2 inhibitor after transplantation has been studied as effective approach to prevent relapse [6], but IDH2 inhibitor was not approved in Japan and we did not perform maintenance therapy by IDH2 inhibitor.

However, both AML patients harboring IDH2^R140Q and IDH2^R172K mutations were refractory to conventional chemotherapy, although they showed good response with VEN+AZA. Recently, it was reported that IDH1/2 mutations are favorable response markers for AML patients treated with VEN+AZA, with both IDH2^R140 and IDH2^R172 mutations presenting similar outcomes [7]. This study included patients with untreated AML patients with IDH2 mutations, and VEN+AZA for the first-line chemotherapy was reasonable option, but genetics analysis using NGS took time to see results in our institute, and we selected conventional chemotherapy instead of waiting for the results of genetic analysis in both cases. On the other hand, genetic analysis revealed the IDH2 mutations on the day when we selected salvage chemotherapy in both cases, and we considered possibility that AML patient with IDH2 and RAS mutations in the case 1 might be refractory to VEN+AZA [8], but she was old and fragile and there was no effective and tolerable salvage regimen except VEN+AZA.

AML patients with IDH1/2 mutations exhibited a longer duration of response to VEN+AZA than those without mutations [7]. However, the median duration of composite CR was reported as 29.5 months [7], and the benefit of allo-HSCT for AML patients with IDH1/2 mutations has been confirmed [9], which seems to be necessary for patients who are eligible for allo-HSCT.

In our cases, WT1 mRNA of PB and VAF of both IDH2 mutations in the BM were correlated with the marrow blast ratio. Due to its low cost and availability, WT1 mRNA of PB might be a more useful marker for treatment response to VEN+AZA than NGS of BM. Certainly, multiparametric flow cytometry was recommended to evaluate measurable residual disease [10], but multiparametric flow cytometry was not approved in Japan and we could not perform it to evaluate disease status. In addition, WT1 mRNA levels of less than 50 copies/μg RNA on day 30 after transplant were reported as a possible positive prognostic factor for PTCy-haplo-PBSCT [11] and may also be useful for the assessment of disease status after transplantation in VEN+AZA bridging PTCy-haplo-PBSCT.

In conclusion, this report suggests the efficacy and safety of VEN+AZA as a bridging treatment for PTCy-haplo-PBSCT in elderly patients with refractory AML harboring IDH2 mutations; however, this case report only included 2 cases with limited data; thus, further studies and evidence are warranted. The CARE Checklist has been completed by the authors for this case report, attached as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000533749).

We would like to thank all the staff at Chugoku Central Hospital of the Japan Mutual Aid Association of Public School Teachers for their contributions to this report. We would like to thank all staff at the Department of Pathology and Tumor Biology, Kyoto University, for providing data on genetic analysis using NGS. We would like to thank Editage (www.editage.com) for English language editing.

The authors certify that they have obtained all appropriate patient consent forms. Written informed consent for case 1 was obtained from the patient for publication of the details of their medical case and any accompanying images prior to their passing away. Written informed consent for case 2 was obtained from the patient for publication of the details of their medical case and any accompanying images. The completed consent form is available to the editor if requested and will be treated confidentially. Ethical approval is not required for this study in accordance with local or national guidelines.

The authors declare no potential conflicts of interest regarding the publication of this study.

No funding was received for this study.

Y.D. and H.S. managed clinical practice and authored this report. T.I., T.K., S.O., N.N., T.M., and N.S. provided advice on this paper. Y.N. and S.O. provided data on genetic analysis using NGS and advised on the paper. M.T. supervised the clinical practice.