Periodic Reporting for period 1 - ONCODESTROYER (Targeting cancer vulnerabilities in acute leukemia)
Okres sprawozdawczy: 2020-10-01 do 2022-03-31
The overall objective of the action (ONCODESTROYER) is to evaluate the therapeutic potential of a new class of anti-cancer drugs in acute myeloid leukemia (AML), using clinical study samples and mouse models. Acute myeloid leukemia (AML) is a relatively infrequent cancer, yet accounts for 2 percent of all cancer death. Despite some recent remarkable developments in treating certain AML types, the standard therapy of the disease hasn’t changed much in the past 50 years. AML, like many cancer types, harbors distinctive vulnerabilities, prominent among which are cell death resistance control and transcriptional addiction. Dysregulation of gene expression programs due to genetic alterations can create dependencies on transcriptional regulators, making tumor cells more sensitive to inhibition of these regulators than normal cells, which are inherently more robust to perturbations. Novel small molecule kinase inhibitors developed in our lab selectively kill leukemia cells and show capability to cure leukemia in several mouse models. We showed that therapeutic efficacy of the inhibitors denoted “oncodestroyers” (ODs), is largely due to blocking a unique array of protein kinases that are essential for building or maintaining de novo generated leukemia driving elements. Thus the main idea behind the project is to further develop and use ODs to simultaneously abolish the expression of many leukemia drivers and vulnerability protectors.
Further, we wish to learn which molecular properties are essential for making an oncodestroyer and what is the mechanistic basis by which ODs disrupt oncogenic elements. We also plan to develop means of assessing OD vulnerabilities and resistance in clinical samples of leukemia patients. We plan to further diversify our set of ODs to cover a wide disease spectrum and improve the therapeutic window. Whereas the major effort of anti-cancer drug development today is directed at tackling individual cancer drivers, guided by precision therapy, ODs represent a different approach, aiming to affect cancer hubs coordinating multiple oncogenes and cancer cell vulnerability protectors at once. Whereas our project addresses AML, the OD properties are also relevant to other types of cancer and indeed, proved efficacious in preliminary studies of a melanoma mouse model.
Further, our lead compound - A51, proved to be highly effective in p53 non-mutated samples, we are interested in understanding how critical is p53 activation for leukemia therapy. Loss of function (LOF) of p53 occurs through mutation in the TP53 gene, and they are present in only ∼8% of de novo (primary) AML cases, but occurs in ∼30% of secondary or therapy related AML and ∼70% in complex karyotype AML. We aim to understand the effects of LOF of wildtype (WT) p53 and the gain of function (GOF) of mutant p53 in leukemia progression and their response to A51 treatment. Importantly, A51 is currently under an open-label, multi-center, first-in-human Phase 1 study, to evaluate its safety in patients with R/R AML or high-risk MDS (ClinicalTrials.gov Identifier: NCT04243785). The physicians have identified the range of recommended Phase 2 dose in patients and the trials are very likely to be approved for Phase 2 clinical trials.
Further, we wish to learn which molecular properties are essential for making an oncodestroyer and what is the mechanistic basis by which ODs disrupt oncogenic elements. We also plan to develop means of assessing OD vulnerabilities and resistance in clinical samples of leukemia patients. We plan to further diversify our set of ODs to cover a wide disease spectrum and improve the therapeutic window. Whereas the major effort of anti-cancer drug development today is directed at tackling individual cancer drivers, guided by precision therapy, ODs represent a different approach, aiming to affect cancer hubs coordinating multiple oncogenes and cancer cell vulnerability protectors at once. Whereas our project addresses AML, the OD properties are also relevant to other types of cancer and indeed, proved efficacious in preliminary studies of a melanoma mouse model.
Further, our lead compound - A51, proved to be highly effective in p53 non-mutated samples, we are interested in understanding how critical is p53 activation for leukemia therapy. Loss of function (LOF) of p53 occurs through mutation in the TP53 gene, and they are present in only ∼8% of de novo (primary) AML cases, but occurs in ∼30% of secondary or therapy related AML and ∼70% in complex karyotype AML. We aim to understand the effects of LOF of wildtype (WT) p53 and the gain of function (GOF) of mutant p53 in leukemia progression and their response to A51 treatment. Importantly, A51 is currently under an open-label, multi-center, first-in-human Phase 1 study, to evaluate its safety in patients with R/R AML or high-risk MDS (ClinicalTrials.gov Identifier: NCT04243785). The physicians have identified the range of recommended Phase 2 dose in patients and the trials are very likely to be approved for Phase 2 clinical trials.
The following results of the proposed action contributes to the overall objective:
Aim 1 - Diversify our OD library and develop OD-based Proteolysis Targeting Chimera (PROTAC): We expanded the OD library by synthesizing new PROTACs around our lead compound, A51. These ODs were tested for anti-leukemic activity in human leukemia cell lines (MV4-11 and THP1). Based on this screening, we have two potent ODs that were selected for the in vivo experiments in MLL-AF9 AML model in “humanized cereblon” mice. These mice have germ-line mutation CrbnI391V – that enables them to bind thalidomide analogues. BM cells isolated from these mice, were infected with MLL-AF9 retrovirus and transplanted to WT mice to create MLL-AF9 driven leukemia mouse model. These ODs are currently being tested for PK and therapeutic responses – in short term and long term in vivo experiments.
Aim 2 - Elucidating oncodestroyer mechanisms of action and resistance; what distinguishes an oncodestroyer?
To evaluate how critical is p53 activation for leukemia therapy with A51, we
(i) performed in vitro studies, using congenic MOLM-13 human leukemia cell lines harboring either WT p53 or CRISPR-engineered null and hotspot p53 mutations - R175H & R273H obtained from Dr. Benjamin Ebert. Here we see that activation of wild type p53 is indeed critical to achieve full potential of A51 response in short and long term therapeutic response.
(ii) developed in vivo mouse models of MLL-AF9 AML with the p53 mutant R172H (homolog of human R175H) or R273H (homolog of human R275H). We generated four different MLL-AF9 driven AML mouse models from mice harboring WT p53 (p53WT/WT) or with germ-line mutation resulting in p53 knockout (p53-/-); or harboring two different hotspot point mutations in the TP53 gene, R172H and R270H (p53R172H/R172H and p53R270H/R270H). Preliminary results show that A51 treatment can induce mutant p53 leukemia cell killing but much less efficient than that in leukemia cells with WT p53 suggesting that p53 is essential to obtain complete cure upon long-term treatment.
Further, to trace and understand A51 resistance and relapse mechanisms, we developed resistant leukemic cell lines (in THP1 cells) using incremental concentrations of A51. After 8 cycles of treatment, we obtained cells with significantly higher LD50 compared to A51 Naïve cells. We also noticed these A51-resistant THP1 cells have many phenotypic and similarities with A51-relapsed MLL-AF9 mice. In both cases, the resistance seems to be induced by epigenetic drivers – rather than genetic drivers and also involve drug induced “persistence” state.
Aim 3 - Exploring translational aspects of OD treatment - A first-in-human study of A51 to study the safety and efficacy of in patients with relapsed or refractory AML demonstrated an acceptable safety profile and promising monotherapy antileukemic activity in patients with heavily pre-treated R/R AML and HR-MDS. Bayesian optimal interval design was used to define the maximum tolerated dose (MTD). Interestingly, MIC-1, a marker of p53 activation, was increased in patients treated with the optimal dose. We also obtained the Dnmt3afl-R878H/+ Mx1-cre+ NPM1frt-cA/+ R26FlpoER Cre+ mice from Dr. Jennifer Trowbridge and established a model to mimic human clonal evolution from clonal hematopoiesis to MDS to AML in our lab.
Aim 1 - Diversify our OD library and develop OD-based Proteolysis Targeting Chimera (PROTAC): We expanded the OD library by synthesizing new PROTACs around our lead compound, A51. These ODs were tested for anti-leukemic activity in human leukemia cell lines (MV4-11 and THP1). Based on this screening, we have two potent ODs that were selected for the in vivo experiments in MLL-AF9 AML model in “humanized cereblon” mice. These mice have germ-line mutation CrbnI391V – that enables them to bind thalidomide analogues. BM cells isolated from these mice, were infected with MLL-AF9 retrovirus and transplanted to WT mice to create MLL-AF9 driven leukemia mouse model. These ODs are currently being tested for PK and therapeutic responses – in short term and long term in vivo experiments.
Aim 2 - Elucidating oncodestroyer mechanisms of action and resistance; what distinguishes an oncodestroyer?
To evaluate how critical is p53 activation for leukemia therapy with A51, we
(i) performed in vitro studies, using congenic MOLM-13 human leukemia cell lines harboring either WT p53 or CRISPR-engineered null and hotspot p53 mutations - R175H & R273H obtained from Dr. Benjamin Ebert. Here we see that activation of wild type p53 is indeed critical to achieve full potential of A51 response in short and long term therapeutic response.
(ii) developed in vivo mouse models of MLL-AF9 AML with the p53 mutant R172H (homolog of human R175H) or R273H (homolog of human R275H). We generated four different MLL-AF9 driven AML mouse models from mice harboring WT p53 (p53WT/WT) or with germ-line mutation resulting in p53 knockout (p53-/-); or harboring two different hotspot point mutations in the TP53 gene, R172H and R270H (p53R172H/R172H and p53R270H/R270H). Preliminary results show that A51 treatment can induce mutant p53 leukemia cell killing but much less efficient than that in leukemia cells with WT p53 suggesting that p53 is essential to obtain complete cure upon long-term treatment.
Further, to trace and understand A51 resistance and relapse mechanisms, we developed resistant leukemic cell lines (in THP1 cells) using incremental concentrations of A51. After 8 cycles of treatment, we obtained cells with significantly higher LD50 compared to A51 Naïve cells. We also noticed these A51-resistant THP1 cells have many phenotypic and similarities with A51-relapsed MLL-AF9 mice. In both cases, the resistance seems to be induced by epigenetic drivers – rather than genetic drivers and also involve drug induced “persistence” state.
Aim 3 - Exploring translational aspects of OD treatment - A first-in-human study of A51 to study the safety and efficacy of in patients with relapsed or refractory AML demonstrated an acceptable safety profile and promising monotherapy antileukemic activity in patients with heavily pre-treated R/R AML and HR-MDS. Bayesian optimal interval design was used to define the maximum tolerated dose (MTD). Interestingly, MIC-1, a marker of p53 activation, was increased in patients treated with the optimal dose. We also obtained the Dnmt3afl-R878H/+ Mx1-cre+ NPM1frt-cA/+ R26FlpoER Cre+ mice from Dr. Jennifer Trowbridge and established a model to mimic human clonal evolution from clonal hematopoiesis to MDS to AML in our lab.
The above results will be followed up with the following experiments to achieve the rest of the aims in the proposal:
Aim 1 - Diversify our OD library and develop OD-based Proteolysis Targeting Chimera (PROTAC):
The pharmacokinetics properties of the new ODs will be evaluated and the therapeutic efficiency in short-term and long-term response will be studied.
Aim 2 - Elucidating oncodestroyer mechanisms of action and resistance; what distinguishes an oncodestroyer?
To understand the mechanistic basis by which ODs disrupt super enhancers and phase separation properties of BRD4 condensates, we plan to first understand (i) the proteomic changes in the cell upon A51 treatment and (ii) modifications on BRD4 and co-interacting proteins. We plan to do mass spectrometric analysis of whole cell lysate of cells with and without A51 treatment and analyze BRD4 centric changes by immunoprecipitation with anti-BRD4 antibody. This will be followed by BRD4 ChIP-Seq to correlate SE functional activity before and after OD treatment.
As A51 treatment proved to be most efficient in MLL-AF9 AML mouse model with wild-type p53, but not in MLL-AF9 AML mouse model with mutant p53, we wish to investigate ways to improve the treatment efficiency by finding a suitable drug for combination treatment. We plan to use the congenic MOLM-13 in vitro system to screen for a suitable drug for combination treatment along with A51 and translate the most synergistic combinations for in vivo treatment.
To understand the immuno-cooperation for the inhibitor responses in AML therapy we wish to (i) study TCR repertoire diversity and clonality to identify anti-leukemic T-cells by TCR sequencing and (ii) to study the fictional state of memory CD8 T cells, we will use immune profiling technologies from 10x Genomics.
Aim 3 - Exploring translational aspects of OD treatment. Even though we established the mouse model of normal hematopoiesis to MDS, and followed the mice for one year, these mice have not developed AML. Our hypothesis is that there might be a leukemia suppressive microenvironment, preventing this AML transformation and we will investigate the mechanism of suppression. Further experiments will be performed to investigate the reciprocal interactions between cancer cells and immune microenvironment to form an inflammatory tumor microenvironment. In parallel, we will also propose a treatment regime to eliminate mutated ARCH clones in this mouse model, which may pave the way for OD treatment as means of ARCH control in future clinical trials.
Aim 1 - Diversify our OD library and develop OD-based Proteolysis Targeting Chimera (PROTAC):
The pharmacokinetics properties of the new ODs will be evaluated and the therapeutic efficiency in short-term and long-term response will be studied.
Aim 2 - Elucidating oncodestroyer mechanisms of action and resistance; what distinguishes an oncodestroyer?
To understand the mechanistic basis by which ODs disrupt super enhancers and phase separation properties of BRD4 condensates, we plan to first understand (i) the proteomic changes in the cell upon A51 treatment and (ii) modifications on BRD4 and co-interacting proteins. We plan to do mass spectrometric analysis of whole cell lysate of cells with and without A51 treatment and analyze BRD4 centric changes by immunoprecipitation with anti-BRD4 antibody. This will be followed by BRD4 ChIP-Seq to correlate SE functional activity before and after OD treatment.
As A51 treatment proved to be most efficient in MLL-AF9 AML mouse model with wild-type p53, but not in MLL-AF9 AML mouse model with mutant p53, we wish to investigate ways to improve the treatment efficiency by finding a suitable drug for combination treatment. We plan to use the congenic MOLM-13 in vitro system to screen for a suitable drug for combination treatment along with A51 and translate the most synergistic combinations for in vivo treatment.
To understand the immuno-cooperation for the inhibitor responses in AML therapy we wish to (i) study TCR repertoire diversity and clonality to identify anti-leukemic T-cells by TCR sequencing and (ii) to study the fictional state of memory CD8 T cells, we will use immune profiling technologies from 10x Genomics.
Aim 3 - Exploring translational aspects of OD treatment. Even though we established the mouse model of normal hematopoiesis to MDS, and followed the mice for one year, these mice have not developed AML. Our hypothesis is that there might be a leukemia suppressive microenvironment, preventing this AML transformation and we will investigate the mechanism of suppression. Further experiments will be performed to investigate the reciprocal interactions between cancer cells and immune microenvironment to form an inflammatory tumor microenvironment. In parallel, we will also propose a treatment regime to eliminate mutated ARCH clones in this mouse model, which may pave the way for OD treatment as means of ARCH control in future clinical trials.