Therapeutic advances in AML have come through a better understanding of the molecular genetics of the disease. New recommendations are now in place that use disease-associated cytogenetic and molecular changes that impact outcome, both for general practice purposes as well as standards for clinical trials. However, the precise mechanisms whereby these molecular changes impact on remission induction, relapse and long-term survival are unknown for the most part, and within each cytogenetic risk category, there are large individual variations in outcome that are poorly understood. Overall outcome in human AML is still dismal, with most patients succumbing to chemoresistant disease within a relatively short time. One goal in leukemia research is to understand the underlying molecular mechanisms driving these variations, to allow better stratification of patients and to give additional insight into those signaling pathways that could be amenable to targeted therapy. MLL-AF9 (MA9) is a frequent chromosomal abnormality in AML, a result of translocations affecting chromosome 11q23, and is associated with an intermediate outcome in response to standard chemotherapy. We have demonstrated that expression of this protein in primary human cord blood (CB) CD34+ cells induces AML with a median latency of 10 weeks. The genetic aberrations FLT3-ITD (associated with a poor outcome) and activated N-ras (no clear effect on outcome in patients) are both found associated with primary 11q23(+) samples in humans. We have data showing that both FLT3-ITD and N-Ras cooperate efficiently with MA9 and accelerate leukemogenesis in human cells. We have established an in vivo induction chemotherapy regimen using the two mainstays of AML treatment, cytarabine and doxorubicin, and show that we effectively target leukemia cells in the xenograft model. Interestingly, the MA9+FLT3-ITD leukemia is refractory to chemotherapy while the MA9+N-ras leukemia is chemosensitive, mimicking in some respects the response seen in AML patients in the clinic. Recent data has demonstrated that the surviving cell in leukemia xenograft experiments is frequently the clone that presides in human patient relapse, indicating that a xenograft approach has significant potential to identify those clones that pose the greatest risk in human disease. The gene expression profiles (GEP) of these clones could be highly informative with regard to the specific signaling cascades that correlate with chemoresistance. We hypothesize that a signature can be identified that will predict chemotherapeutic response and the potential for relapse. We will use both primary patient AML samples and our own inducible human AML xenograft models to study the varying chemosensitivity associated with therapy response and failure in AML, to identify the genotypic contribution of defined cooperating mutations to chemoresistance. This high-risk/high-yield project is based on solid experimental approaches developed in my lab and has great potential for high impact findings with rapid translation to the clinic.
Our studies support the concept that chemoresistance in AML is a discernable problem that can be addressed using defined experimental model systems based on human cell studies in the xenograft setting. Chemoresistance associated with particular genetic aberrations in human AML has very strong clinical correlations, but the nature of the signals associated with this chemoresistance has yet to be understood at the molecular level. This proposal will address this outstanding question, and attempt to identify specific signatures that can be used for prognostic stratification of patients at initial presentation in th clinic.
|Wunderlich, Mark; Brooks, Ryan A; Panchal, Rushi et al. (2014) OKT3 prevents xenogeneic GVHD and allows reliable xenograft initiation from unfractionated human hematopoietic tissues. Blood 123:e134-44|