Hematopoietic Cell Transplantation in Adult With Acute Myeloid Leukemia

Allogeneic Transplant or Chemotherapy Only

Identifying an individual patient in whom transplant will achieve livable cure requires knowing the best strategy for transplant with least possible morbidity and mortality; it is important to realize that all that is discussed and done is known. Despite an agreement that allo-HCT is the most effective anti-AML therapy in the majority of patients, expert opinion differ about its deployment in CR1. Such differences are based on genuine concerns mainly attributed to limitation in data and absence of reliable benefit and risk prediction methods. In real world practice, the decision for or against either type of consolidation approaches is influenced by large number of factors, including apprehension of unknown variables. With chemotherapy-only approach, heightened relapse risk in postponing the allo-HCT, suboptimal efficacy of salvage regimen at the time of relapse, and losing the opportunity of transplant because of morbidity or mortality are major concerns. There are few retrospective data that indirectly address some of these concerns. For instance, few studies reported high CR2 numbers and long LFS in patients who had specific favorable prognostic variables available at the time of CR1 such as young age (vs. older patient) and favorable cytogenetics (not favorable NPM1) [inv(16) or t(8;21)] that may assist in consolidation selection decision. Also, in an analysis by Burnett et al. of 8909 patients who entered the Medical Research Council (MRC) AML10, AML12, and AML15 trials, 1271 of 3919 patients aged 16 to 49 years who did not receive a transplant in CR1 relapsed. Of these, only 55% of patients who relapsed entered CR2. This percentage of CR2 varied by risk group as follows: favorable (82%), intermediate (54%), adverse (27%), and unknown (53%), which resulted in 5-year survivals of 32%, 17%, 7%, and 23%, respectively. About 67% of CR2 received an allogeneic transplant that delivered superior survival compared with patients who did not receive a transplant (42% vs. 16%). A more stringent assessment of a transplant by using delayed-entry (Mantel-Byar) analysis confirmed the benefit of transplant overall and within intermediate- and high-risk groups but not the favorable subgroup.

Another analysis focusing on AML patient who underwent allo-HCT in CR2 reported 3-year overall survival (OS) rates of 19%. The study highlighted five adverse pretransplantation variables, which significantly influenced survival: CR1 < 6 months, circulating blasts, donor other than HLA-identical sibling, Karnofsky or Lansky score less than 90, and high-risk cytogenetics. Patients who had a predictive score of 0 had 42% OS at 3 years, whereas OS was 6% for a score > or = 3.

In summary, despite all the concerns against allo-HCT in CR1, few data indicate worse survival after allo-HCT, so the key question in our opinion is not whether to pursue allogeneic transplantation or not in intermediate- and high-risk AML but to determine the best timing of allo-HCT (CR1 vs. CR2+) and the optimal strategy of transplant to derive the best possible outcomes.

Data in Young Adults

Several metaanalyses and prospective multicenter trials compared results of allo-HCT to chemotherapy alone or auto-HCT as postremission therapy in AML. These trials compared outcomes based on donor availability, or “biologic randomization” (not true randomization) and reported lower relapse rates and better survival in allogeneic transplant cohort. In most studies, the benefit appears limited to patients with intermediate-risk cytogenetics. Patients with favorable- and high-risk features did not benefit from a transplant in all studies. The major limitations are selection biases operating in the absence of true randomization and dropout of subjects who have an identified donor but who never received a transplant. The results from metaanalyses favor allo-HCT for young adults with intermediate- and high-risk AML in CR1 ( Table 13.2 ).

Data in Older Adults

AML primarily affects older adults; allo-HCT is underused in these patients because of perceived higher risk of TRM. Recent advances in AML and transplant arena have allowed greater proportion of older adults to be considered for allo-HCT. A prospective phase II study, Cancer and Leukemia Group B (CALGB) 100103 (Alliance for Clinical Trials in Oncology)/Blood and Marrow Transplant Clinical Trial Network (BMT CTN) 0502, assessed outcomes of allo-HCT in CR1 for patients aged 60 to 74 years who underwent fludarabine/busulfan (fludarabine at 30 mg/m ² per day intravenously for 5 days and busulfan 0.8 mg/kg intravenously every 6 hours for 8 doses) reduced intensity regimen (RIC). The 2-year LFS was 42%, nonrelapse mortality (NRM) of 14%, and OS of 48%. A second prospective phase III OSHO 2004 study on behalf of the East German Study Group compared RIC based allo-HCT versus chemotherapy in patients aged 60 to 75 years. In the Cox regression analysis, improved LFS was seen across cytogenetic risk groups with allo-HCT, but OS was uninfluenced by allo-HCT.

Research and education about utility of allo-HCT in this patient population and prevention of relapse are important unmet needs.

Molecular Aberrations and Controversies

Accurate risk stratification of AML patients has a critical role in facilitating patient selection with respect to the optimal postremission therapy. In 2022, the latest European LeukemiaNet (ELN) recommendations on diagnosis and management of AML in adults were published, identifying three risk groups, largely based on pretreatment cytogenetic abnormalities and molecular aberrations ( Table 13.3 ). Most recently, it could be shown that in the context of allo-HCT, the ELN risk categorization maintained its prognostic utility, with patients with favorable risk having the best prognosis, enabling a risk-adapted approach. Furthermore, presence of monosomal karyotype or TP53 mutation were associated with increased relapse rates, within the adverse-risk group.

Additional prognostic characteristics in cytogenetically normal AML with and without NPM1 have been reported in retrospective and prospective nonrandomized studies. For instance, patients with concurrent mutations of NPM1 and IDH have longer LFS compared with mutated NPM1 without mutated IDH , and in similar context of mutated NPM1 but , with high levels of ERG expression, predicts for adverse outcome compared with low expression of ERG . The consensus of best postremission therapy in these types of scenarios and others with co-mutations are lacking highlighting the fact that one size does not fit all.

Relevance With Germline Mutation on Type of Consolidation

Germline mutations in CEBPA, DDX41, RUNX1, GATA2, ETV6 , and ANKRD2 6 have been recognized as myeloid malignancy predisposition syndromes. Of note, there are some variants (e.g., frameshift DDX41 variants such as p.D140fs) that are virtually always germline, but also several that can occur as either somatic or germline variants. Decision of whether a patient requires additional testing for germline variants on the basis of a cancer-based next-generation sequencing (NGS) panel requires some understanding of what features make a variant more likely germline versus somatic mutation.

Upon conformation of germline mutation in otherwise healthy individual, consideration for preemptive allo-HCT is reasonable; however, the optimal timing of allogeneic transplant remains an individualized and a difficult decision. In addition, the presence of germline mutation in patient has additional bearing on selection of a related donor (discussed in chapter).

FLT3 Mutation and Allogeneic Hematopoietic Transplantation

There is a controversy around the role of allo-HCT for normal karyotype patients with NPM1 mutation plus low-allele ratio (<0.5) FLT3-ITD . Some retrospective studies have shown the benefit of allo-HCT in CR1 for these patients, while others showed no benefit. Interestingly, there is absence of randomized data showing survival advantage with allo-HCT over alternative approach in CR1 for patients with FLT3-ITD AML. However, it is well known that FLT3- ITD AML patients have an inferior survival compared to AML patients without the mutation, primarily because of short remission rates. Single-center retrospective studies have repetitively shown improvement of LFS and OS in this group of patients with deployment of allo-HCT in CR1. It remain to be seen if newer FLT3-ITD inhibitors will spare patients from allo-HCT in CR1.

Therapy-Related or Secondary Acute Myeloid Lymphoma

The CIBMTR analysis (1990–2004) of allo-HCT for t-AML and therapy-related myelodysplastic syndrome (t-MDS) showed 5-year OS and disease-free survival (DFS) of 22% and 21%, respectively. Contemporary supportive care measures, GVHD prophylaxis, and RIC regimens have led to improved tolerability of allo-HCT. Therefore CIBMTR revisited the same type of analysis in 1531 (t-MDS [n = 759] and t-AML [n = 772]) patients transplanted between the period of 2000 and 2014. In patient with t-AML, the median age was 52 years (range, 18–77 years). A myeloablative conditioning (MAC) regimen was used in 61%. The NRM and relapse rate at 5 years was 34% (95% CI, 30%–37%) and 43% (95% CI, 40%–47%), respectively. This study found an improved NRM of 24% at 1 year and 34% at 5 years. This is encouraging, especially when considering that the median age was >15 years older than in the previous analysis (56 years vs. 40 years). Nonetheless, high rates of relapse persist. The 5-year OS and DFS were 25% (95% CI, 22%–28%) and 23% (95% CI, 20%–26%), respectively. In multivariate analysis, OS and DFS was worse for those with previous auto-HCT, higher-risk cytogenetics, and a partially HLA-matched unrelated donor. Intensified conditioning regimen should be considered if tolerated based on age, performance status, and comorbidities in CR1 for t-AML.

Measurable Residual Disease

Numerous studies have investigated the value of MRD in AML and have consistently shown that MRD negativity, as defined by specified cutoff values, is highly prognostic for outcome. A recent metaanalysis reported a 5-year estimated OS of 68% for the MRD-negative group compared to only 34% for the MRD-positive group. Of importance, selection of the optimal technique and target is highly individualized and each MRD detection platform has its advantages and limitations mainly because of: (1) complex genetic heterogeneity of AML and (2) institutional expertise. Two methods are currently widely applied (i.e., multiparameter flow cytometry [MFC] and real-time quantitative polymerase chain reaction [qPCR]), and newer technologies, including digital PCR and NGS, are emerging and evolving.

Measurable Residual Disease–Based Consolidation Assignment

In the GIMEMA AML1310 study, young patients with National Comprehensive Cancer Network (NCCN) 2009 intermediate-risk AML were allocated to allo-HCT versus auto-HCT based on the presence or absence of MRD by MFC assay after 1 cycle of consolidation. Using this “risk-adapted” approach, no difference in DFS or OS was observed between MRD positive and negative groups. These findings suggest that an MRD-directed selection of allo-HCT consolidation may overcome the negative effect of positive MRD.

European LeukemiaNet Genetic-Risk and Measurable Residual Disease Classification

A recent study showed that ELN genetic-risk classification could be combined with monitoring of the MRD after induction therapy. The 2017 ELN classification included the new response category “CR without MRD.” The ELN 2022 response criteria expanded MRD classification to include patients achieving CR with partial hematologic recovery (CRh) or CR with incomplete count recovery (CRi) without MRD (CRh _MRD2 or CRi _MRD2 ) The application of MRD, detected by either MFC or qPCR for specific molecular markers, may further refine conventional morphologic assessment of response. MRD may be detected at time points early after induction treatment to assess the remission status of the AML but also after postremission/pre–allo-HCT and post–allo-HCT.

Measurable Residual Disease and Decision Making in Acute Myeloid Leukemia

Assessment of MRD in AML is challenging. When used for risk stratification, it is important to note that the specific impact of MRD status on relapse risk is contingent on type of therapy, methodology, reporting methods, turnaround time, cost, kinetics and depth of cytogenetic and molecular aberration, age-related clonal hematopoiesis indeterminate potential (CHIP), and timing(s) of assessment among other factors. Recently published consensus guidelines by the ELN-MRD working party provided a framework for future studies. As previously stated, that not all mutations are suitable as MRD markers. For instance, presence of CHIP-related somatic mutations such as D NMT3A, T ET2, and A SXL1 (DTA) are commonly found in healthy individuals and most often represent preleukemic clonal hematopoiesis and not relapse-initiating cells.

In summary, the MRD methods in AML await standardization for routine clinical practice and the optimal timing of allogeneic transplant in MRD-positive patients remains an individualized and a difficult decision.

High-Risk Subgroup in Core Binding Factor-Acute Myeloid Leukemia—a New Entity or Not?

Numerous studies have tried to identify a high-risk feature in otherwise favorable core binding factor (CBF)-AML subgroup to attain superior outcomes. Analysis of features such as high white blood cell (WBC) at diagnosis, therapy-related disease, loss of the Y chromosome, fms-like tyrosine kinase 3 internal tandem duplication (FLT3-ITD) or KIT or NRAS mutation, and/or MRD positivity has been evaluated.

The negative impact of KIT mutation appears to be heterogenous among CBF-AML. KIT mutations were associated with reduced OS among patients with t(8;21) but not among patients with the inv(16)/t(16;16) alteration. Rucker et al. showed that both reduction of RUNX1 - RUNX1T1 transcript levels and achievement of MRD negativity at defined time points are of significant prognostic importance. Notably, 19 of the 93 (20%) patients achieving MRD negative relapsed compared with 14 of the 19 (74%) patients not achieving MRD-negative state ( P < .0001). A collaborative study by the German AML Intergroup and CIBMTR showed that lower relapse rates using MSD in t(8;21) AML were offset by higher TRM when compared to chemotherapy alone; a high-risk feature by itself was not an absolute indication for early allo-HCT. However, this cooperative group study did not provide enough evidence to analyze the impact of co-occurring genetic mutations like KIT mutation. A prospective study of allo-HCT in patients with t(8;21) demonstrated lower relapse and better DFS in MRD-positive but lower OS in MRD-negative patients, both compared to chemotherapy.

According to most recent ELN guidelines, in RUNX1-RUNX1T1 and CBFB-MYH11 mutated AML MRD should be assessed preferentially in PB after two cycles of chemotherapy, in the marrow at end of consolidation treatment, and in PB every 4 to 6 weeks for 24 months after the end of consolidation.