A unifying genetic basis has been sought to explain the complex and heterogeneous nature of myeloid neoplasms since before Janet Rowley’s quinacrine banding discovered the Philadelphia chromosome (Nature. 1973;243:290-3). In the decades following that discovery, groundbreaking work has uncovered new chromosomal abnormalities, new gene fusions, new recurrent mutations – often with prognostic implications, but rarely with therapeutic ones.
The recent work by Elli Papaemmanuil, PhD, of Memorial Sloan Kettering Cancer Center, New York, and her colleagues reaffirms the genetic heterogeneity of AML based on molecular profiling of patients from three large European trials. Yet the most insightful aspect of this reclassification is not just the detail of the genetic resolution but the realization that, even within a gene such as NRAS, the genetic background for acquisition of a codon 12/13 mutation is mutually exclusive with clones where NRAS codon 61 occurs.
When speaking with relapsed patients, I often say that, while we are very good at cutting down trees in AML, we still have not done very well with getting rid of the roots. Admittedly, this metaphor grossly oversimplifies cancer stem cell biology, but it gets at the real importance of the work by Dr. Papaemmanuil and her colleagues. The interactions of gene mutations such as NPM1 and DNMT3A are not uncommon and their co-mutation in isolation has an intermediate prognosis. The clonal acquisition of a codon 12/13 mutation in NRAS seems to result in a more favorable prognosis – lending to the likelihood that the tumor is simply more chemosensitive. In contrast, the acquisition of FLT3-ITD by the NPM1/DNMT3A co-mutant clone results in a very poor prognosis likely due to chemoresistance.
The real power of this study’s findings is the potential for building a toolbox of agents to push against the innate clonal selection and force the “tree” to grow in a direction that is detrimental to its survival. One could consider using FLT3 inhibitors in the wild-type setting of a genetic background primed towards FLT3-ITD evolution to prevent this resistant outgrowth. Of course, such an approach needs to be studied first in a laboratory setting, but similar therapeutic strategies have been applied to BRAF in melanoma. Peter Nowell urged “controlling the evolutionary process in tumors before it reaches the late stage,” and this new ordinal understanding of AML may help to do just that.