Acute myeloid leukemia (AML) is a clinically and molecularly heterogeneous disease.1 This concept has been supported by more than 4 decades of studies showing distinct outcomes of subsets of patients that differ in age, disease type (primary vs secondary vs therapy-related), and cytogenetic and molecular aberrations. Indeed, both clinical and genetic features have replaced blast morphology as the basis for newer AML classifications, such as the 2008 World Health Organization2 or the European LeukemiaNet classifications.3
Of all patient characteristics at diagnosis, genetic aberrations (ie, cytogenetics and gene mutations) are among the strongest predictors for clinical outcome, are used for treatment guidance, and have allowed for a deeper understanding of the molecular mechanisms of myeloid leukemogenesis.4 However, while there has been a remarkable agreement among the results of large studies investigating the clinical impact of the recurrent cytogenetic aberrations and of the more common genetic mutations, it is striking that the outcome of patients with seemingly identical genetic “makeup” is not uniform. In fact, not all patients with favorable cytogenetic and/or genetic molecular risk achieve long-term survival, whereas a small but clinically meaningful percentage of patients remain alive despite harboring adverse cytogenetic and/or molecular aberrations.
Although this may be explained by individual variations in drug metabolism and/or mechanisms of chemoresistance, it also may be a direct consequence of molecular aberrations yet to be discovered that will further segregate patients currently in the same genetic groups into different smaller molecular subsets. Should this be proven true, virtually all AML patients may need to be approached with specific treatments tailored to their unique molecular profiles. While this is an exciting perspective, its feasibility remains to be tested, given the relatively few molecular targeting drugs available and the lack of ready information on toxicity and clinical activity of combinations of such drugs.
Cancer Genome Atlas Research Network
The recent introduction of next-generation sequencing technology has offered a tantalizing opportunity to address these questions by scanning the entire genome in search of previously unreported genetic and epigenetic variations that offer new insights into the clinical relevance of molecular heterogeneity in AML, explain the conundrum of the different outcomes of patients classified within the same cytogenetic/genetic subgroups, and ultimately may provide ground for “individualized” therapeutic approaches. Dr. Ley, leader of the recently reported Cancer Genome Atlas Research Network (CGARN) study,5 and his colleagues have pioneered the field of next-generation sequencing in AML and provided seminal observations.6-8
These investigators have shown how to deconvolute the complexity of huge amounts of bioinformatic data into biologically and clinically intelligible information,5 have discovered previously unknown mutations (eg, IDH1, IDH2 and DNMT3A),5,6 and have shown the temporal evolution of minor clones during the dynamics of disease relapse or treatment refractoriness.8 These results have created high expectations regarding what next-generation sequencing can do and how it could change our approach to leukemia patients.
The CGARN study has now gone beyond its initial results and provided information that not only defines the landscape of gene and microRNA mutations but also correlates them with specific gene and microRNA expression and DNA methylation profiles in a relatively large cohort of AML patients.5 The enormous amount of data that has been generated in this project is summarized by the following key points.
First, the AML genomes have been shown to harbor fewer mutations relative to most other adult cancers, with only 23 genes found to be “significantly” mutated. Second, nearly all patients had at least one nonsynonymous mutation in categories of genes that likely contribute to leukemogenesis. Third, mutations were associated with specific gene and microRNA expression and DNA methylation profiles. Fourth, patterns of cooperation and mutual exclusivity among mutations and gene categories allowed for identification of novel molecular subsets of patients.
So, could one say that we now have a new and deeper understanding of AML and that we are ready to devise “molecular-driven individualized” approaches for the vast majority of patients in the near future? To answer this question, we should consider the following points.
‘Ready for Prime Time’?
First, the relatively small number of mutations identified in the analyzed patients is somewhat surprising. Even more surprising is that the majority of the 23 genes that were indicated as “significantly” mutated were already known, and very few of them were discovered through next-generation sequencing.4,5,9
Second, several of the mutations were found to be mutually exclusive and to be directly linked to specific gene, microRNA, and methylation profiles. But doesn’t this indeed confirm results that were already partly reported in previous studies? Almost a decade ago, German and Dutch groups reclassified patients into novel, clinically relevant “clusters” using a combination of cytogenetics, gene mutations, and microarray-derived gene-expression profiles.10,11 Similarly, Figueroa et al have previously reported DNA methylation profiles associated with specific mutations.12,13 We and others have demonstrated microRNA signatures that are highly associated with distinct cytogenetic aberrations and gene mutations.14-19
Could it therefore be said that our biologic and clinical knowledge of AML, albeit incomplete, was perhaps not that far from the picture that the CGARN study has now depicted? Or instead, is it true that the CGARN study has carved out previously unidentified subsets of patients, thereby leading the way to new AML classifications, risk-stratification strategies, and treatment guidelines? Fortunately, although these questions have not been addressed in the CGARN paper, the data are now available for the scientific community to find the answers.
Undoubtedly, the CGARN study is an important step forward in providing an integrated “snap-shot” of the biologic puzzle that is AML. However, it also requires additional steps to provide the missing pieces. In addition to in silico validation (performed using a computer simulation) of the presented data in independent cohorts of patients, it will be necessary to functionally validate these results and understand how mutations and their associated gene and microRNA epigenetic and expression profiles impact on protein expression, posttranslational modifications, and oncogenic or tumor-suppressor functions.
Furthermore, while the integrated CGARN data were derived from “bulk” blasts, it may be necessary to assess which of the genetic and epigenetic changes also occur in those minute cell populations that are enriched for leukemia stem cells and may be functionally relevant for treatment resistance or disease relapse. Moreover, as it is possible that like other types of cancer, AML may be partly caused by “field defects,” the contribution of the microenvironment to the acquisition of the reported epigenetic/genetic alterations or—vice versa—how these changes affect the microenvironment needs to be dissected and the results considered for future treatment design.
Translating to the Clinic
Finally, as the sequencing techniques become more affordable and broadly available and are brought into clinical diagnostic laboratories, the wealth of knowledge from the CGARN and similar studies must be translated rapidly into the clinic. But this may not be sufficient to transform our approach to AML without eliminating other barriers. In fact, progressively smaller molecular groups of patients will be identified by next-generation sequencing assays as eligible for “individualized” therapies.
Thus, the main condition for success in treating AML will be that all stakeholders involved in translational and clinical leukemia research work collaboratively to guarantee access to such therapies. Moreover, the traditional approaches to clinical trial design and drug evaluation will need to rapidly be adapted to this new biologic and clinical reality. ■
Dr. Marcucci is Professor of Medicine, Charles Austin Doan Chair of Medicine, and Associate Director of Translational Research at The Ohio State University Comprehensive Cancer Center, and Dr. Bloomfield is Distinguished University Professor, William G. Pace III Professor of Cancer Research, Cancer Scholar and Senior Advisor at The Ohio State University Comprehensive Cancer Center, Columbus.
Disclosure: Drs. Marcucci and Bloomfield reported no conflicts of interest.
References
1. Dombret H: Gene mutation and AML pathogenesis. Blood 118:5366-5367, 2011.
2. Vardiman JW, Thiele J, Arber DA, et al: The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: Rationale and important changes. Blood 114:937-951, 2009.
3. Mrózek K, Marcucci G, Nicolet D, et al: Prognostic significance of the European LeukemiaNet standardized system for reporting cytogenetic and molecular alterations in adults with acute myeloid leukemia. J Clin Oncol 30:4515-4523, 2012.
4. Marcucci G, Haferlach T, Döhner H: Molecular genetics of adult acute myeloid leukemia: Prognostic and therapeutic implications. J Clin Oncol 29:475-486, 2011.
5. Cancer Genome Atlas Research Network: Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl J Med 368:2059-2074, 2013.
6. Mardis ER, Ding L, Dooling DJ, et al: Recurring mutations found by sequencing an acute myeloid leukemia genome. N Engl J Med 361:1058-1066, 2009.
7. Ley TJ, Ding L, Walter MJ, et al: DNMT3A mutations in acute myeloid leukemia. N Engl J Med 363:2424-2433, 2010.
8. Ding L, Ley TJ, Larson DE, et al: Clonal evolution in relapsed acute myeloid leukaemia revealed by whole-genome sequencing. Nature 481:506-510, 2012.
9. Patel JP, Gönen M, Figueroa ME, et al: Prognostic relevance of integrated genetic profiling in acute myeloid leukemia. N Engl J Med 366:1079-1089, 2012.
10. Bullinger L, Döhner K, Bair E, et al: Use of gene-expression profiling to identify prognostic subclasses in adult acute myeloid leukemia. N Engl J Med 350:1605-1616, 2004.
11. Valk PJ, Verhaak RG, Beijen MA, et al: Prognostically useful gene-expression profiles in acute myeloid leukemia. N Engl J Med 350:1617-1628, 2004.
12. Figueroa ME, Lugthart S, Li Y, et al: DNA methylation signatures identify biologically distinct subtypes in acute myeloid leukemia. Cancer Cell 19:13-27, 2010.
13. Figueroa ME, Abdel-Wahab O, Lu C, et al: Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell 18:553-567, 2010.
14. Jongen-Lavrencic M, Sun SM, Dijkstra MK, et al: MicroRNA expression profiling in relation to the genetic heterogeneity of acute myeloid leukemia. Blood 111:5078-5085, 2008.
15. Li Z, Lu J, Sun M, et al: Distinct microRNA expression profiles in acute myeloid leukemia with common translocations. Proc Natl Acad Sci USA 105:15535-15540, 2008.
16. Becker H, Marcucci G, Maharry K, et al: Favorable prognostic impact of NPM1 mutations in older patients with cytogenetically normal de novo acute myeloid leukemia and associated gene- and microRNA-expression signatures: A Cancer and Leukemia Group B study. J Clin Oncol 28:596-604, 2010.
17. Marcucci G, Maharry K, Wu YZ, et al: IDH1 and IDH2 gene mutations identify novel molecular subsets within de novo cytogenetically normal acute myeloid leukemia: A Cancer and Leukemia Group B study. J Clin Oncol 28:2348-2355, 2010.
18. Mi S, Li Z, Chen P, et al: Aberrant overexpression and function of the miR-17-92 cluster in MLL-rearranged acute leukemia. Proc Natl Acad Sci USA 107:3710-3715, 2010.
19. Mendler JH, Maharry K, Radmacher MD, et al: RUNX1 mutations are associated with poor outcome in younger and older patients with cytogenetically normal acute myeloid leukemia and with distinct gene and MicroRNA expression signatures. J Clin Oncol 30:3109-3118, 2012.
