Detection, prevention and treatment of acute myeloid leukemia (AML) relapse.

National Heart, Lung, And Blood Institute

Investigators

Linked publications, trials & patents

Abstract

The fundamental interest of the Laboratory of Myeloid Malignancies is the detection, prevention, and treatment of acute myeloid leukemia (AML) relapse. Our work has focused on clinical trials of novel biomarker and immunotherapy approaches, and the development of molecular and genomic laboratory methods (Measurable Residual Disease, MRD) to predict development or recurrence of myeloid malignancy. Foundational to our objective has been the development of high sensitivity biomarkers for residual AML in those patients who have been treated to apparent remission but remain at risk of clinical relapse. Previously we have demonstrated the ability to risk stratify AML patients into groups with either high or low leukemic relapse rates, based on a pre-transplant peripheral blood sample, prior to allogeneic hematopoietic cell transplantation (alloHCT). We have extended our previous work on the BMT-CTN 0901 Phase 3 randomized controlled trial. Firstly, we published the impact of detection of FLT3, IDH1, IDH2, JAK2, KIT, NPM1, NRAS, RUNX1, SF3B1 and/or TP53 mutations in the pre-conditioning blood of patients with myelodysplastic syndrome (MDS) undergoing alloHCT (JCO Precision Oncology, Ref 1). We have shown that testing positive was associated with increased rates of relapse and decreased overall survival. In those testing positive, relapse rates were higher and relapse-free survival was lower in reduced intensity versus myeloablative conditioning arms. Testing additional genes, including those associated with MDS, did not improve prognostication. This helps identify a high-risk group of patients who may particularly benefit from additional interventions before and/or after reduced intensity conditioning alloHCT. Secondly, we compared the survival of all MDS and AML patients testing MRD positive prior to reduced-intensity conditioning alloHCT in our analysis of BMT-CTN 0901 with that seen in another newly reported randomized phase 3 trial (Journal of Clinical Oncology, Ref 2). Thirdly, we reported long-term clinical follow-up of patients on BMT-CTN 0901 which confirmed a benefit for myeloablative conditioning (TCT, Ref 3). It is currently unknown if these genomic methods measure the same residual AML cells as those detected by flow cytometric methods, with some reporting examples of discrepancy between these techniques. We have used single-cell sequencing with antibody-oligonucleotide conjugate staining to allow reconciliation of results from these two AML MRD detection methods, including the first description of whole-genome sequencing (WGS) informed patient-personalized scDNA-seq (Blood Cancer Discovery, Ref 4). This personalized single cell proteogenomics approach allowed us to clearly distinguish between AML and nonmalignant clonal hematopoiesis by confident description of the complex clonal architecture, highlighting situations where either sequencing or flow cytometry approaches would incompletely describe the total leukemic burden. We have continued collaborative research with extramural colleagues specifically focusing on AML MRD (JAMA Oncology, Ref 5 and Leukemia, Ref 6) but also AML more generally (Refs 7, 8, 9 and 10). We have also continued to collaborate with intramural colleagues at the NIH Clinical Center who treat patients with non-malignant hematological conditions of whom a subset may develop myeloid malignancy. In collaboration with the group of Dr. Neal Young, patients with severe aplastic anemia (SAA) treated at NHLBI with anti-thymocyte globulin or alemtuzumab immunosuppressive therapy (IST) between 1989 and 2019 were reviewed (n=666). A total of 56 had high-risk clonal evolution (defined as diagnosis of myeloid malignancy, or acquisition of chromosome 7 abnormality or complex karyotype) of whom 15 occurred 5 years or later after IST. Of these 15 late events, 7 had normal karyotype of whom samples from time of myeloid malignancy diagnosis were available from 4. We were able to detect somatic variants present in bone marrow samples from at least 1-2 years prior to clonal evolution but not at time of initial SAA diagnosis (Haematologica, Ref 11). We have also helped design and test an assay for somatic mutation detection in a newly described disease with hematological manifestations (NEJM, Ref 12). In summary, the primary interest of the Myeloid Malignancies Section remains the detection, prevention and treatment of AML relapse, in particular the development of molecular and genomic laboratory methods to predict development or recurrence of myeloid malignancy. Going forward we will continue to integrate genomic and immunophenotypic approaches on carefully annotated clinical samples to better understand myeloid malignancy and will also expand the evidence for clinical utility of AML MRD testing using molecular testing of large informative patient cohorts.