Skip to main content

Understanding the mechanisms of human acute myeloid leukaemia (AML) evolution

Periodic Reporting for period 3 - MAMLE (Understanding the mechanisms of human acute myeloid leukaemia (AML) evolution)

Reporting period: 2020-02-01 to 2021-07-31

Acute myeloid leukemia (AML) is a heterogeneous cancer characterized by excessive clonal expansion of hematopoietic progenitor cells in the bone marrow. It is not only the most common but also remains to be the deadliest of all types of leukemia among adults. The 5-year overall survival rate is roughly 30% in adult AML patients. These poor outcomes highlight the unmet need for a better understanding of leukemogensis and novel therapies to replace chemotherapy, which has not been changed for nearly five decades. In recent years it has become apparent that prior to the diagnosis of AML a subclinical pre-leukemic phase can be identified. In this latent phase, pre-leukemic stem and progenitor cells (preL-HSPC), carrying pre-leukemic mutations (pLM), exhibit clonal expansion and gradually accumulate further mutations over time. In the majority of cases, preL-HSPC will benignly expand in elderly individuals through clonal hematopoiesis, but will not lead to leukemia. This phenomenon has been termed Age related clonal hematopoiesis (ARCH). By definition ARCH occurs among the elderly with no blood malignancies, however, it is becoming evident that ARCH is associated with a large number of pathological states and increased risk for blood cancers. It is not clear, whether the process is deterministic and little is known with regard to why certain individuals go on to develop malignancies while others do not. Even if tools to identify those individuals with ARCH predestined to develop AML existed, it is unclear how this condition should be treated and whether the benefits of the treatment would outweigh potential risks.
In this study we attempted to answer some of these questions. We sequenced genes related to AML from the peripheral blood (PB) of 124 individuals with AML from the EPIC project. Blood was available, on average, 7 years prior to AML diagnosis. Results were compared to 681 controls. In brief, we found that compared to healthy controls, in pre-AML cases more mutations were found, with larger clones based on variant allele frequency (VAF), in specific positions of the genome which are recurrently mutated in patients with hematologic malignancies. Most strikingly, spliceosome machinery mutations (SMM) were almost exclusively found in pre-AML cases and tended to arise at an earlier age when compared to individuals who carry them but do not proceed to develop the disease. Collectively, our findings provide new insights into the pre-clinical evolution of AML and support the hypothesis that individuals at high risk of AML development can be identified years before they develop overt disease.

Another main problem is that many AML patients will relapse after initial successful therapy and succumb to their disease. In our studies, we showed that the origins of AML relapse are heterogeneous but converge to stem cell properties of the relapsing clones In order to prevent relapse, leukemic stem cells should be eradicated in early stages of treatment (induction/remission) before they expand and acquire additional resistance mechanisms. It has been suggested that combination therapy which targets several phenotypes of the leukemic cells could be useful in overcoming AML heterogeneity. AML, which is mainly a disease of the elderly, the older individuals are less tolerable to the additive side effects of combination therapy. Therefore applying the right drug with minimal toxicity, the right set of patients, might be a valid approach. In this study, our global aim was to identify therapies and biomarkers, that can predict which drugs might be added to induction therapy to prevent relapse in specific subtypes of AML. Here, we studied patterns of in vitro drug response, which correlates with molecular characteristics of primary AML patients who achieved Complete Remission. In the current study we discovered that samples sensitive in vitro to dasatinib (a multi kinase inhibitor) had a specific in vitro drug sensitivity pattern, gene expression signature, and were enriched in FLT3/ITD and PTPN11 mutations. Dasatinib demonstrated anti LSC effect in vivo and the response to dasatinib among carriers of FLT3/ITD could be predicted based on gene expression of dasatinib targets.
1. We have tracked the complex evolutionary history of AML within individual patients from the early stages of pre-leukaemic development to diagnosis and through progression to relapse. Here, through combined genetic and functional analysis of purified subpopulations and xenografts from paired diagnosis/relapse samples, we identify therapy-resistant cells already present at diagnosis and two major patterns of relapse. In some cases, relapse originated from rare leukaemia stem cells with a haematopoietic stem/progenitor cell phenotype, while in other instances relapse developed from larger subclones of immunophenotypically committed leukaemia cells that retained strong stemness transcriptional signatures (Shlush et al.2017 Nature 547, pages104–108).
2. To identify ARCH mutations with the high probability of AML transformation, we compared the mutational landscape of ARCH in two different groups: 95 individuals who later developed AML with a median of 6.3 years before AML diagnosis (pre-AML group) and 414 age/gender-matched individuals who did not develop AML with an average of 11.6 years follow-up (control group) (Fig 1). Error-corrected deep sequencing of recurrently mutated genes in AML showed that the pre-AML group had significantly higher prevalence of ARCH, higher variant allele frequency (VAF) of detected ARCH mutations, and a higher number of ARCH mutations per individual compared to the control group. We then calculated the relative risk of individual ARCH mutation for AML development and validated the results in a validation cohort. This analysis revealed a remarkable heterogeneity of AML risk among ARCH mutations and identified TP53 (hazard ratio [HR] = 12.5 95% CI: 5.0–160.5) and U2AF1 (HR = 7.9 95% CI: 4.1–192.2) mutations as having the highest risk for AML development, while the risk was relatively small in DNMT3A mutations (Fig 2). Additionally, the presence of two or more ARCH mutations per individual and ARCH mutations with VAF ≥ 9% were also associated with a high risk of AML transformation. (Abelson S,…..Shlush L,. 2018. Nature 559, pages400–404)
3. We studied patterns of in vitro drug response among primary AML patients who achieved complete remission (CR). Results from the current study were compared to the Beat AML study. We performed unsupervised hierarchical clustering analysis for 46 drug ex vivo sensitivity on primary AML samples from diagnosis and relapse and found a "responder" group to many tyrosine kinase inhibitors. We examined the response to the multi tyrosine kinase inhibitor, dasatinib and the correlation with exome sequencing and transcriptome expression from our dataset and the Beat AML dataset. Mutations in FLT3-ITD and PTPN11 were enriched in the dasatinib sensitive samples as oppose to mutations in TP53 that were significantly more common in the resistance group to dasatinib (Fig 3). AML patients carrying PTPN11 mutations express higher level of LYN (a target of dasatinib) as compared to wild type. We found that LYN, HCK, CSK and EPHB2 (targets of dasatinib) genes were significantly overexpressed in the dasatinib responders carrying FLT3/ITD (Fig 4). Expression levels of these genes could predict which carriers of FLT3/ITD will respond to dasatinib . Finally, we demonstrated that dasatinib treatment targeted in vivo leukemic stem cells (LSCs) (Fig 5).
1.We are planning to finalise the study on the inhibition of FLT3/ITD and PTPN11 mutated acute myeloid leukemia cells over-expressing SRC tyrosine kinases by dsatinib as described above. The data from this study suggest that in FLT3/ITD mutated AML with overexpression of SFK adding dasatinib to chemotherapy can prevent development of AML and maybe prevent the relapse. Treatment of multi tyrosine kinase inhibitor, like dasatinib will be effective especially in AML with mutations in growth signaling pathways like FLT3-ITD and PTPN11 with a proliferative growth advantage.
2. We are currently studying the mechanism which might lead to preleukemic deletions in myeloid malignancies. Results of this study will have an high impact on potentially aid in preventing somatic deletions that characterize clonal expansion in high risk individuals.