Acute myeloid leukemia (AML) is a genetically heterogeneous disease, which occurs de novo or evolves from pre-existing myeloid disorders. The incidence of AML increases with age, and with increasing age, prognosis and overall survival decline. Treatment failure in many patients is attributed to relapsed disease and remains the fundamental clinical challenge in these patients. The exact biological basis of relapse remains unclear. One prevailing theory is that inherently chemotherapy resistant leukemia stem cells (LSCs) survive and repopulate the disease. However genetic background must also have an influence on LSC functions since specific mutations are associated with more frequently relapsing disease. It has also been observed that AMLs undergo clonal evolution with relapse, suggesting that pre-existing subclones or chemotherapy induced natural selection contribute to this phenomenon. Our preliminary data strongly suggest that epigenetic mechanisms play a fundamental role in this process. Specifically: i) we observe extensive epigenetic deregulation in all AML patients, ii) specific epigenetic signatures and classifiers are associated with poor outcomes, iii) there is a relative paucity of genetic mutations in AML as compared to solid tumors (based on TCGA data), iv) many of the common genetic lesions in AML directly alter epigenetic gene regulation. Moreover in preliminary studies we observe that AMLs at relapse display large scale epigenetic changes, whereas TCGA finds relatively subtle genetic changes at relapse. Of note, many of the epigenetic changes we observe are in common between patients, suggesting that AMLs may respond and resist chemotherapy using shared or stereotyped mechanisms. Based on these and other considerations we hypothesize that AML relapse and adaptation to chemotherapy is encoded by a combination of epigenetic and genetic lesions that are either present in a subset of AML cells at diagnosis or occur de novo in response to exposure to cytotoxic drugs. We predict that these lesions preferentially disrupt specific biological pathways that directly mediate chemotherapy resistance, and that detection of these lesions at diagnosis will be predictive of unfavorable outcome. In order to test these hypotheses we will perform a genome wide integrative genetic and epigenetic comparison of seventy matched diagnosis/relapse AML specimens including the use of novel methods developed by the applicant. Results will be validated, and the clinical significance determined in an independent cohort of 750 patients all enrolled in a single phase II clinical trial. The biologial function of relapse-associated lesions will be confirmed in functional assays, with the ultimate goal of developing targeted therapies to prevent leukemia relapse.

Public Health Relevance

Acute myeloid leukemia remains a mostly fatal disease, with most patients relapsing even after high dose chemotherapy. The goal of this proposal is to identify the epigenetic and genetic factors that determine the ability of AML cells to relapse and determine their clinical and biological significance. The ultimate goal is to target these lesions o prevent AML relapse and improve patients'clinical outcomes.

Agency
National Institute of Health (NIH)
Institute
National Cancer Institute (NCI)
Type
Clinical Investigator Award (CIA) (K08)
Project #
5K08CA169055-02
Application #
8535705
Study Section
Subcommittee G - Education (NCI)
Program Officer
Ojeifo, John O
Project Start
2012-09-01
Project End
2017-08-31
Budget Start
2013-09-01
Budget End
2014-08-31
Support Year
2
Fiscal Year
2013
Total Cost
$148,320
Indirect Cost
$10,987
Name
Weill Medical College of Cornell University
Department
Internal Medicine/Medicine
Type
Schools of Medicine
DUNS #
060217502
City
New York
State
NY
Country
United States
Zip Code
10065
Will, Britta; Vogler, Thomas O; Bartholdy, Boris et al. (2013) Satb1 regulates the self-renewal of hematopoietic stem cells by promoting quiescence and repressing differentiation commitment. Nat Immunol 14:437-45
Li, Sheng; Garrett-Bakelman, Francine E; Akalin, Altuna et al. (2013) An optimized algorithm for detecting and annotating regional differential methylation. BMC Bioinformatics 14 Suppl 5:S10