Acute myelogenous leukemia (AML) affects more than 20,000 adult patients in the US per year causing greater than 11,000 untimely deaths. Standard AML therapy comprises chemotherapy induction to achieve leukemia debulking, followed by a few cycles of consolidation chemotherapy alone and/or followed by allogeneic bone marrow transplantation. This general therapeutic approach to AML is risk-adapted relying principally on well-established prognostic factors, including the type of AML, age, gene mutations and cytogenetic results. Despite best efforts, AML remains incurable in the majority of afflicted patients. A principle barrier to AML cure is disease relapse despite achieving a clinical complete remission (CR) following standard chemotherapy regimens. Of substantial interest therefore, is the identification and accurate measurement of residual AML that persists during remission in AML patients, as such residual disease is likely the source of relapse. Several important questions about AML relapse are only in early stages of satisfactory answers, including i) how to best measure minimal residual disease (MRD) in AML, ii) whether all MRD is indeed caused by residual disease or rather as suggested by recent data comprises a mixture of leukemia, pre- leukemia and age-related clonal hematopoiesis, iii) what constitutes the cellular source of relapse in AML, iv) if such relapse-causing cells can be better targeted using novel therapeutic approaches, and v) under what circumstances is MRD prognostic and useful for MRD risk-adapted AML clinical management. Given lack of well-suited techniques to reliably answer the relevant questions detailed above, we have optimized droplet digital PCR (ddPCR), a novel ultra-high sensitivity assay for detecting genomic MRD in AML. In a detailed recently published study (Parkin et al, JCI 2017) we found that AML frequently relapses from rare cells residing in remission marrows that based on mutation analysis resemble AML blast cells detected at diagnosis. We demonstrated the feasibility for detecting AML-associated gene mutations at allele frequencies as low as 0.002% and have also provided important novel prognostic insights. In this application, we are proposing complete characterization of mutational MRD and aberrant cellular clusters in AML remission bone marrows using a combination of sophisticated cell sorting and single cell transcriptome analyses complemented with mouse xenografting and ex vivo colony forming assays. Anticipated findings will improve the functional characterization of cells that carry AML-associated gene mutations in an attempt to identify and better characterize the source(s) of AML relapse. Using AML samples from two clinical validation cohorts, we will define the prognostic utility of ddPCR-based MRD assessments. Overall, data will lay the foundation for future real-time genomic MRD-guided clinical AML trials aiming at monitoring and improving consolidation and maintenance therapy and ultimately survival outcome in AML.

Public Health Relevance

Acute myelogenous leukemia (AML) affects more than 20,000 adult patients in the US per year causing greater than 11,000 untimely deaths. In this application, we are proposing the in-depth biological characterization of minimal residual disease in AML in complete remission bone marrows and association of mutant allele levels in these remission samples with remission duration and relapse rates. Resulting findings will likely guide the development of novel treatment approaches for this lethal leukemia.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Research Project (R01)
Project #
Application #
Study Section
Cancer Biomarkers Study Section (CBSS)
Program Officer
Agrawal, Lokesh
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Michigan Ann Arbor
Internal Medicine/Medicine
Schools of Medicine
Ann Arbor
United States
Zip Code