Long non-coding RNA and n6-methyladenosine in tyrosine kinase inhibitor resistance

Abstract

A method of obtaining a prognosis for a subject who has or has had leukemia who has been or is being treated with a tyrosine kinase inhibitor (TKI) is described. The method includes determining the levels of one or more of 1) total N6-methyladenosine (m6A), 2) long non-coding RNA (lncRNA) bearing m6A sites, 3) lncRNA-specific m6A, 4) the fat mass and obesity-associated gene (FTO) in a biological sample from the subject; and assigning a poor prognosis or poor TKI responder status to the subject if the level of m6A, lncRNA being m6A sites, lncRNA-specific m6A, and / or FTO is higher than the one or more corresponding control values obtained from a healthy subject. Methods of treating a subject having TKI-resistant leukemia and methods of predicting TKI resistance in a subject having leukemia are also described.

Description

LONG NON-CODING RNA AND N6-METHYLADENOSINE IN TYROSINE KINASE INHIBITOR RESISTANCESTATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

[0001] This invention was made with government support under Grant Numbers R01CA248019 and R01CA266256, awarded by the National Institutes of Health. The Government has certain rights in this invention.CROSS REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Application No. 63 / 663,269, filed on June 24, 2024, which is incorporated herein by reference.BACKGROUND

[0003] Leukemia is an aggressive malignancy frequently driven by the constitutively active tyrosine kinases (TKs), for example, the growth and progression of chronic myeloid leukemia (CML) is attributed to a hybrid protein BCR / ABLl,! resulting from the t(9;22)(q34;ql l) chromosomal translocation. Multiple TK inhibitors (TKIs), such as imatinib, nilotinib, dasatinib, bosutinib and ponatinib, have emerged as leading compounds to treat CML. Initial clinical response of CML patients to TKIs is often excellent. However, relapse and disease progression characterized by drug resistance occurs even with continuous drug administration, and patients with TKI resistant CML may still succumb to their disease. Studies regarding TKI resistance in leukemia have mostly centered on cancer genetic alterations (Khorashad et al., Blood 121, 489-498 (2013)), for example, acquired mutations (which happen in about 1 / 3 of progressive cases) in the kinase domains of BCR / ABL that reduce or even fully impair TKIs’ binding to ABL kinase. However, it is less known whether and how dynamic and reversible RNA epitranscriptomic mechanisms play a role in TKI resistance, particularly for resistant patients without acquired mutations.

[0004] N6-methyladenosine (m6A) is the most common internal modification on RNAs, including long non-coding RNAs (IncRNAs). The m6A is installed by a methyltransferase complex, erased by the demethylases (e.g., FTO), and recognized by m6A-binding proteins. Cumulative evidence suggests that m6A methylation occurs frequently in a dynamic andreversible manner, and significantly regulates RNA half-life, RNA splicing and protein translation (Wang et al., Nature 505, 117-120 (2014)), thus determining target expression levels. Defects in m6A methylation affect diverse biological processes, including cancer initiation, development and progression. Li et al., Cancer cell 31, 127-141 (2017). We recently demonstrated that a dynamic FT0 / m6A axis emerges as a key epigenetic driver of reversible TKLtolerance state making contribution to acquired TKI resistance through activating cell proliferation / anti-apoptotic genes. Yan et al., Cell research 28, 1062-1076 (2018). Yet, it is not well understood whether and how the FT0 / m6A axis empowers leukemia cells to survive prolonged drug exposure through non-protein coding genes.

[0005] LncRNAs are transcripts with lengths of > 200 nucleotides and have no apparent protein-coding potential. As they crucially activate or repress transcription of protein-coding genes, IncRNA dysregulations have been linked to cancer progression and drug resistance. Fang etal., Cancer. Genomics, proteomics & bioinformatics 14, 42-54 (2016). Previous studies suggest that functions of IncRNAs are subjected to m6A regulation through, for example, potentiating IncRNAs-mediated transcriptional repression. Patil et al., Nature 537, 369-373, (2016). These findings, together with the observations that upregulation of certain IncRNAs fuels the survival and proliferation of cancer cells (Huarte, M, Nature medicine 21, 1253-1261 (2015)), raise the possibility that, upon exposure to TKIs, the dynamic m6A methylation allows a set of proliferation / anti-apoptosis-relevant IncRNAs bearing m6A motifs to be rapidly upregulated, thus helping a subpopulation of cells avoid TKI killing.SUMMARY OF THE INVENTION

[0006] Disease recurrence characterized by drug resistance is a primary cause of failure in the management of cancers, including BCR / ABL-positive chronic myeloid leukemia (CML). Long non-coding RNAs (IncRNAs) and RNA N6-methyladenosine (m6A) have been linked to cancer drug resistance. The inventors evaluated if and how IncRNAs and m6A coordinately regulate resistance and develop effective regimens to treat relapsed / refractory patients.

[0007] Resistant leukemia models were established by sequential exposure of CML cells to the physiologically achievable concentrations of tyrosine kinase inhibitors (TKIs). The inventors performed transcriptomic and epitransciptomic profiling in resistant CML cells to identify pathways required for the development and maintenance of drug resistance. Western blot,qPCR, immunoprecipitation and dotblot were performed to characterize changes in proteins, DNA and RNAs of the lead candidates and elucidate the engaged mechanisms. Phenotypic assays such as CCK-8, flow cytometry and clonogenic assays were also performed to assess the effects of ectopic expression, and genetic silencing of genes and IncRNAs. The expression of IncRNAs was examined in CML and AML patients pre- vs post- treatment, responders vs non-responders, blast crisis / advanced phases vs chronic phases. The therapeutic effects of PI3K inhibitor were tested in leukemic mouse models.

[0008] Many differentially expressed IncRNAs enrich m6A, and more IncRNAs tend to have higher m6A content in TKI resistant cells. The top-ranked IncRNAs, including PROX1-AS1, SENCR and LN892, are highly elevated in resistant cell lines, non-responding patients and leukemia patients with blast crisis when compared with their respective counterparts. Patients with higher levels of IncRNAs (PROX1-AS1, SENCR, LN892) survive shorter than those with lower expression. Knockdown of these IncRNAs impairs resistant cell growth and renders resistant cells sensitive to TKI-induced cell death. Mechanistically, IncRNA upregulation is attributed to FTO-dependent m6A hypomethylation that stabilizes IncRNA transcripts, and empowers resistant cell growth through the activated PI3K signaling. Treatment with PI3K inhibitor alpelisib sensitizes TKI resistant cells to TKI killing in vitro, and increases the survival time of leukemic mice as a monotherapy via suppressing PI3K signaling in vivo.

[0009] The inventor’s findings add a new layer to the complexity of mechanisms regulating leukemia cell fate under TKI selection and indicate that the m6A-regulated IncRNAs represents a new non-genetic factor to regulate TKI resistance; the discoveries identify a promising therapeutic target, the m6A-lncRNAs axis, for specifically the most challenging patient subpopulations who are TKI non-responders / relapsed but do not carry the acquired mutations on top of BCR / ABL. The results also uncover a strong predictor, m6A-lncRNA-PI3K axis, for poorer prognosis and failure in drug response which might be a pan-cancer mechanism.BRIEF DESCRIPTION OF THE FIGURES

[0010] The present invention may be more readily understood by reference to the following figures, wherein:

[0011] Figures 1 A- 1 F provide graphs showing the generation and characterization of leukemia cells with acquired resistance to nilotinib or imatinib (A and B) CCK-8 assays for cellproliferation and flow cytometry assays for cell apoptosis in parental and resistant cells treated with 1 pM nilotinib or imatinib for 72 hours. The data of CCK-8 represent two independent experiments with 6 repeats in total, but flow cytometry assays represent three independent experiments. (C and D) Colony-forming assays for K562, Kasumi-1 and MV4-11 parental and resistant cells in drug free medium. (E) Flow cytometry assays to measure cell size in K562 resistant and parental cells. (F) Western blotting of parental and resistant cells growing in drugcontaining medium. In Figure A-E, data are expressed as mean values ± S.E.M. of triplicate samples. In Figure F, the data are representative of 3 independent experiments. Par, parental; NIR, nilotinib resistant; IMR, imatinib resistant; NI, nilotinib; IM, imatinib; MV, MV4-11; Kas, Kasumi-1. *p < 0.05, **p < 0.01, ***p < 0.001.

[0012] Figures 2A-2E provide graphs showing the differentially expressed IncRNAs in nilotinib resistant cells bear m6A motifs (A) Plot distribution of log2 change of enrichment across m6A peaks from parental to nilotinib resistant cells. The average log2-transformed normalized signal for the duplicated m6A-seq was used to generate a histogram of read counts values. (B) The heatmap based on the sorted log2-fold change of 40 upregulated IncRNAs. Fourty differentially expressed IncRNAs (annotated by gencode vl9) are ranked by log2 (fold change) with different m6A methylation levels. (C) IncRNAs PROX1-AS1, SENCR and LN892 with m6A-seq track are highlighted. (D) qPCR of total RNA for the expression of indicated IncRNAs in parental and resistant K562 cells. (E) qPCR for IncRNA expression in poly-A and non poly-A RNA of K562 P, IR and NR cells. Data are expressed as mean values ± S.E.M. of duplicate samples from three independent experiments. P, parental; NR, nilotinib resistant; IR, imatinib resistant; IP, immunoprecipitation, IN, Input; Nilo, nilotinib.

[0013] Figures 3A-3E provide graphs showing IncRNA upregulation predicts unfavorable outcomes in leukemia patients (A) BM cell morphology of CML patients. The classical morphological images were obtained from some CML-CP and CML-BP patients. Bone marrow cells were cytospinned onto slides followed by Wright-Giemsa staining 20 minutes. The images were obtained under the light microscope. Upper, the chronic phase; lower, The blast phase (AML and ALL). (B) Changes in the number of BCR / ABL transcripts prior and post nilotinib or imatinib therapy. Green, yellow and red areas represent optimal, warning and failure results, respectively. (C) qPCR for IncRNA expression in CML (n = 40) patients who are classified as responder and inadequate responders. Median values are depicted by the horizontal lines. (D) qPCR of RNA extracted from AML patients (n = 30) receiving nilotinibtherapy. (E) The association of IncRNA expression with overall survival (OS) in leukemia patients analyzed by the Kaplan-Meier estimate. Re, Responder; Nre, Non-responder. *p < 0.05, ** / ? < 0.01.

[0014] Figures 4A-4I provide graphs showing IncRNAs are partially regulated by FTO- dependent m6A methylation in resistant cells (A) Left: Dotblotting for global m6A amounts (lower) and Western blot for FTO protein expression (upper) in K562, MV4-11 and Kasumi-1 cells resistant to nilotinib or imatinib. Data represent three independent experiments. Right: qPCR for FTO expression in resistant CML patient with / without BCL / ABL mutations. (B and C) m6A immunoprecipitation (IP) was performed in mRNA / poly-A (B) and non-poly-A RNA(C). The eluted RNA was converted to cDNA and IncRNA expression was assessed by qPCR.(D) qPCR for the m6A-IP relative expression of IncRNAs normalized by the total levels. (E) Dotblotting for global m6A methylation (lower) and Western blot for FTO protein expression (upper) in clones with FTO knockdown. Data represent three independent experiments. (F) qPCR for IncRNA expression in clones with FTO knockdown. (G) qPCR for IncRNA expression in resistant cells treated with meclofenamic acid (50 pM) and FB23-2 (10 pM) for 6 hours. (H) m6A IP was performed in mRNA and non-poly-A RNA in clones with FTO knockdown. The eluted RNA was converted to cDNA and expression of PROX1-AS1, SENCR, LN892, normalized by total RNA, was assessed by qPCR. (I) qPCR for IncRNA expression in scrambled and FTO knockdown clones treated with 5 pg / ml actinomycin-D for the indicated time points. The gene expression was normalized to GAPDH. In all qPCR, data are mean + S.E.M. Par, parental; NIR, nilotinib resistant; IMR, imatinib resistant; KD, knockdown.

[0015] Figures 5A-5E provide graphs showing FTO-regulated IncRNAs contribute to nilotinib sensitivity (A) Colony-forming assays in K562 clones with FTO knockdown. (B) CCK-8 assays in FTO knockdown or control clones treated with nilotinib for 48 hours. The data represent two independent experiments with 6 repeats in total. (C) Resistant K562 cells were infected with IncRNA virus for 48 hours, and qPCR was used to assess IncRNA expression. (D) Colony-forming assays in K562 resistant cells with knockdown of indicated IncRNAs. (E) CCK-8 assays in IncRNA knockdown or control clones treated with indicated doses of nilotinib for 48 hours. The data represent two independent experiments with 6 repeats in total. KD, knockdown. **p < 0.01, ***p < 0.001.

[0016] Figures 6A-6F provide graphs showing the PI3K signaling pathway mediates IncRNA- sustained resistant cell growth (A and B) KEGG pathway analysis identifies a network of genes related to PI3K signaling that is responsible for IncRNA-sustained cell growth. Functional pathway analysis was performed in differentially expressed genes upon SENCR knockdown using DAVID 6.8 / KEGG software. Up-regulated (A) and down-regulated (B) gene pathways were determined by comparing SENCR knockdown and scrambled control cells. Count (%) means the ratio of genes involved in the pathway. (C) The expression of indicated genes was assessed by qPCR in K562 resistant clones with knockdown of PROX1-AS1, SENCR or LN892. (D) The expression of indicated genes was assessed by qPCR in K562 parental and resistant (nilotinib, imatinib) cells. (E) The expression of ITGA2, COL6A1, cyclin DI, PKN1, PDGFRA, F2R, and HSP90AB1 was assessed by qPCR in TKI responding and resistant CML patients. (F) The association of ITGA2, COL6A1, cyclin DI, PKN1, PDGFRA, F2R, and HSP90AB 1 expression with overall survival (OS) in leukemia patients analyzed by the Kaplan-Meier estimate using online tool PROGgene or GEPIA2. In F, the curves (ITGA2, F2R, COL6A1, cyclin DI) were adjusted for age and cohort divided at median of gene expression. In all qPCR, data are mean ± SD.

[0017] Figures 7A-7N provide graphs showing the inactivation of PIK3 signaling by alpelisib in TKI resistant cells achieves leukemia remission in vitro and in vivo (A) CCK-8 assays in K562 parental and resistant cells to imatinib (IMR) or nilotinib (NIR) with various concentrations of alpelisib for 72 hours. (B) CCK-8 assays in K562 IMR or NIR resistant cells with various concentrations of imatinib or nilotinib with / without alpelisib (100 pM) for 72 hours. (C-E) IMR or NIR resistant K562 were treated with alpelisib (Alp; 100 pM), imatinib (30 pM) or nilotinib (30 pM) alone or combination for the indicated time points. The treated cells were subjected to flow cytometry (C, D) for cell apoptosis or Western blot (72 hours; E). Graphs are the quantification of apoptotic cells presented as percentage. (F and G) About 0.5 x 106nilotinib resistant K562 cells were injected through tail-vein into sub-lethally irradiated NSGS mice (n = 5 mice / group). The BM cells were isolated for (F) FACS analysis of the engrafted recipient BM cells (stained by CD44 antibody) from the representative disease mice (5 weeks after cell injection). Graphs are the quantification of CD45+ cells (folds). (G) H&E staining of lung, liver or spleen sections (original magnification x40) from healthy (norm) or leukemic mice bearing parental (Par) or nilotinib resistant (NIR) cells. (H-N) about 0.5 x 106BM cells isolated from F were injected via the tail-vein into sub-lethally irradiated NSGS mice(second recipient). Five days after cell injection, the mice were randomly grouped and treated with Alp. (H, I) Representative external views of the spleens from the leukemic mice (H) and the weight of spleen and liver (I). (J) H&E staining of lung, liver or spleen sections (original magnification x40) from healthy (normal), leukemic mice treated with vehicles (control) or Alp (n = 3) (magnification x 40). (K) Representative images of Wright-Giemsa-stained cytospins of BM cells from leukemic mice (left; magnification x 40) and graph (right) shows quantification of post-mitotic cells from (e). (L) Graphs show the changes in body weight of vehicles or alp-treated mice (n = 10). (M) Effects of alpelisib on survival of leukemia-bearing mice were determined by the Kaplan-Meier estimate (log-rank test). (N) BM cells isolated from treated mice were subjected to qPCR for the indicated IncRNA targets using primers of human genes. In A and B, the data represent two independent experiments with 8 repeats in total; in C, D and E, data represent three independent experiments. Note, the survival time is from the start of leukemia cell injection. Data are shown as mean values ± SD, *P < 0.05; **P < 0.01; ***P < 0.001.DETAILED DESCRIPTION OF THE INVENTION

[0018] The present invention provides a method of obtaining a prognosis for a subject who has or has had leukemia who has been or is being treated with a tyrosine kinase inhibitor (TKI). The method includes determining the levels of one or more of 1) total N6-methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, 3) IncRNA- specific m6A, 4) a fat mass and obesity-associated gene (FTO) in a biological sample from the subject; and assigning a poor prognosis or poor TKI responder status to the subject if the level of m6A, IncRNA being m6A sites, IncRNA-specific m6A, and / or FTO is higher than the one or more corresponding control values obtained from a healthy subject. Methods of treating a subject having TKI-resistant leukemia and methods of predicting TKI resistance in a subject having leukemia are also provided.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which these exemplary embodiments belong. The terminology used in the description herein is for describing particular exemplary embodiments only and is not intended to be limiting of the exemplary embodiments. As used in the specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearlyindicates otherwise. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety.

[0020] A “subject,” as used herein, can be any animal, and may also be referred to as the patient. Preferably the subject is a mammal, such as a research animal (e.g., a monkey, rabbit, mouse or rat) or a domesticated farm animal (e.g., cow, goat, horse, pig) or pet (e.g., dog, cat). In some embodiments, the subject is a human.

[0021] As used herein, the term "diagnosis" can encompass determining the likelihood that a subject will develop a disease, or the existence or nature of disease in a subject. The term diagnosis, as used herein also encompasses determining the severity and probable outcome of disease or episode of disease or prospect of recovery, which is generally referred to as prognosis).

[0022] The term “prognosis” refers to a forecast as to the probable outcome of the disease as well as the prospect of recovery from the disease as indicated by the nature and symptoms of the case. Accordingly, a negative or poor prognosis is defined by a lower post-treatment survival term or survival rate. Conversely, a positive or good prognosis is defined by an elevated post-treatment survival term or survival rate. Usually, prognosis is provided as the time of progression free survival (PFS) or overall survival (OS). Prognosis, as used herein, can also refer to the likelihood that a subject with response to a particular type of drug treatment (e.g., treatment with a tyrosine kinase inhibitor).

[0023] “Treat", "treating", and "treatment", etc., as used herein, refer to any action providing a benefit to a subject at risk for or afflicted with a condition or disease such as cancer, including improvement in the condition through lessening or suppression of at least one symptom, delay in progression of the disease, prevention or delay in the onset of the disease, etc.

[0024] “Preventing,” as used herein, refers to any action that decreases the risk that a subject will develop leukemia, inhibits the growth of leukemia, or decreases the incidence of leukemia recurrence. Cancer prevention can be done in subjects who have an increased risk of developing cancer. Also intended to be encompassed by this definition is the prevention of metastasis of malignant cells or to arrest or reverse the progression of malignant cells. Subjects can have an increased risk of developing leukemia as a result of smoking, exposure to ionizing radiation or petrochemicals (such as benzene), prior chemotherapy, Down syndrome, and family history.

[0025] The terms “therapeutically effective” and “pharmacologically effective” are intended to qualify the amount of each agent which will achieve the goal of decreasing disease severity while avoiding adverse side effects such as those typically associated with alternative therapies. The therapeutically effective amount may be administered in one or more doses.

[0026] The term “determining the level of’ at least one biomarker in a sample, control or reference, as described herein, shall refer to the quantification of the presence of said at least one biomarker in the tested sample. For example, the concentration of the first and second biomarker in said samples may be directly quantified via measuring the amount of m6A and / or IncRNA as present in the tested sample. Moreover, it is also possible to quantify the amount of the biomarker directly via assessing the modified RNA nucleotide, for example by mass spectrometry. However, it is also possible is to quantify the amount of the biomarker indirectly via assessing the gene expression of the encoding gene of the biomarker, for example by quantification of the expressed mRNA encoding for the respective biomarker. How to determine the level of a particular biomarker is well known to the skilled artisan. The present invention shall not be restricted to any particular method for determining the level of a given biomarker, but shall encompass all means that allow for a quantification, or estimation, of the level of said first and second biomarker, either directly or indirectly. “Level” in the context of the present invention is therefore a parameter describing the absolute amount of a biomarker in a given sample, for example as absolute weight, volume, or molar amounts; or alternatively “level” pertains to the relative amounts, and preferably to the concentration of said biomarker in the tested sample, for example in mol / 1, g / 1, g / mol etc. In preferred embodiments the “level” refers to the concentration of the tested biomarkers in g / 1.

[0027] The term “control” or "reference value," as used herein, refers to a value that statistically correlates to a particular outcome when compared to an assay result. In preferred embodiments the reference value is determined from statistical analysis of studies that compare microRNA expression with known clinical outcomes. The reference value may be a threshold score value or a cutoff score value. Typically, a reference value will be a threshold above which one outcome is more probable and below which an alternative outcome is more probable.

[0028] "Nucleic acid" or "oligonucleotide" or "polynucleotide", as used herein, may mean at least two nucleotides covalently linked together. The depiction of a single strand also defines the sequence of the complementary strand. Thus, a nucleic acid also encompasses thecomplementary strand of a depicted single strand. Many variants of a nucleic acid may be used for the same purpose as a given nucleic acid. Thus, a nucleic acid also encompasses substantially identical nucleic acids and complements thereof. A single strand provides a probe that may hybridize to a target sequence under stringent hybridization conditions. Thus, a nucleic acid also encompasses a probe that hybridizes under stringent hybridization conditions.

[0029] Nucleic acids may be single- stranded or double-stranded or may contain portions of both double-stranded and single-stranded sequence. The nucleic acid may be DNA, both genomic and cDNA, RNA, or a hybrid, where the nucleic acid may contain combinations of deoxyribo- and ribo-nucleotides, and combinations of bases including uracil, adenine, thymine, cytosine, guanine, inosine, xanthine hypoxanthine, isocytosine and isoguanine. Nucleic acids may be obtained by chemical synthesis methods or by recombinant methods.

[0030] Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit unless the context clearly dictates otherwise, between the upper and lower limit of that range and any other stated or intervening value in that stated range, is encompassed within the invention. The upper and lower limits of these smaller ranges may independently be included in the smaller ranges, and are also encompassed within the invention, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the invention.

[0031] As used herein, the term "about" refers to + / - 10% deviation from the basic value.

[0032] All scientific and technical terms used in the present application have meanings commonly used in the art unless otherwise specified. The definitions provided herein are to facilitate understanding of certain terms used frequently herein and are not meant to limit the scope of the present application.Leukemia Prognosis

[0033] In one aspect, the present invention provides a method of obtaining a prognosis for a subject who has or has had leukemia who has been or is being treated with a tyrosine kinase inhibitor (TKI), comprising: determining the levels of one or more of 1) total N6- methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, 3) IncRNA-specific m6A, 4) the fat mass and obesity-associated gene (FTO) in a biological sample from the subject; and assigning a poor prognosis or poor TKI responder status to the subject if the level of m6A, IncRNA being m6A sites, IncRNA-specific m6A, and / or FTO is higher than the one or more corresponding control values obtained from a healthy subject.

[0034] Leukemia is a group of blood cancers that usually begin in the bone marrow and produce high numbers of abnormal blood cells. These blood cells are not fully developed and are called blasts or leukemia cells. There are four main types of leukemia: acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL) and chronic myeloid leukemia (CML), as well as a number of less common types. Leukemia is generally diagnosed using a blood test or bone marrow biopsy.

[0035] In some embodiments, the leukemia is acute myeloid leukemia (AML) or chronic myeloid leukemia (CML). In myeloid or myelogenous leukemias, the cancerous change takes place in a type of marrow cell that normally goes on to form red blood cells, some other types of white cells, and platelets.

[0036] Acute leukemia is characterized by a rapid increase in the number of immature blood cells. The crowding that results from such cells makes the bone marrow unable to produce healthy blood cells resulting in low hemoglobin and low platelets. Immediate treatment is required in acute leukemia because of the rapid progression and accumulation of the malignant cells, which then spill over into the bloodstream and spread to other organs of the body. Acute forms of leukemia are the most common forms of leukemia in children.

[0037] Chronic leukemia is characterized by the excessive buildup of relatively mature, but still abnormal, white blood cells (or, more rarely, red blood cells). Typically taking months or years to progress, the cells are produced at a much higher rate than normal, resulting in many abnormal white blood cells. Whereas acute leukemia must be treated immediately, chronic forms are sometimes monitored for some time before treatment to ensure maximum effectiveness of therapy. Chronic leukemia mostly occurs in older people but can occur in any age group.

[0038] In some embodiments, the prognosis includes characterizing the susceptibility of the leukemia to treatment with a tyrosine kinase inhibitor. Such staging involves determining the expected level of effectiveness of a TKI in treating the leukemia. In other embodiments, themethod determines or provides information that helps to determine the stage of the leukemia. Chronic myeloid leukemia can be staged as chronic phase, accelerated phase, and blast phase (a.k.a. blast crisis). Acute myeloid leukemia has 8 subtypes (M0-M7) based on FAB classification, but can also be staged as untreated, in remission, or relapsed / refractory.

[0039] The method includes determining the levels of one or more of 1) total N6- methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, 3) IncRNA- specific m6A, and 4) the fat mass and obesity-associated gene (FTO). Collectively, these various chemical entities can be referred to herein as biomarkers.

[0040] With a modification rate of up to 0.5% of all adenosines, the modified mRNA nucleotide N6-methyladenosine (m6A) is the most abundant modification in mRNA. m6A is co-transcriptionally deposited on nascent RNAs during their transcription by RNA polymerase II by the m6A methyltransferase complex (m6A-MTC), which consists of METTL3, METTL14, WTAP, VIRMA, RBM15 ZC3H13. Thereby, m6A imprints mRNAs with splicing information that is read by tRNAs.

[0041] The level of m6A is typically determined using immunoprecipitation, as described in detail in the Example provided herein. However, other methods of determining m6A levels are known to those skilled in the art. See for Example U.S. Patent Publication 20220364173, which described additional methods for detecting levels of m6A in RNA molecules by incubating the RNA molecule with a methyltransferase enzyme and a S-adenosyl-1 -methionine (SAM) analog comprising a functional group under conditions sufficient to attach the functional group to the m6A, thereby generating a modified m6A, attaching a functional group to the RNA, and generating a complementary nucleic acid molecule, which includes mutations based on the presence of m6A.

[0042] In some embodiments, the total m6A level is determined. The total m6A level is the total amount of m6A modified polynucleotide (e.g., RNA) found in the biological sample being evaluated. Alternately, or in addition, the level of IncRNA-specific m6A is measured. IncRNA- specific m6A is m6A that is found only on IncRNA, rather than an overall level of m6A in the polynucleotides (e.g., RNA) being examined.

[0043] In some embodiments, the prognosis is based on the level of long non-coding RNA (IncRNA). Long-non-coding RNAs are a type of RNA molecule that are over 200 nucleotideslong and that do not code for proteins. In some embodiments, the IncRNAs have a size of at least 200 nucleotides, while in other embodiments the IncRNAs have a size of at least 500 nucleotides. While not intending to be bound by theory, it is thought that IncRNAs act as molecular scaffolds, bringing together different proteins and RNA to facilitate their interactions and regulate gene expression, to guide proteins and other molecules to specific DNA or RNA targets, as decoys to bind to proteins or other molecules to prevent their normal actions, and in epigenetic modification (e.g., DNA methylation). See Mattick et al., Nature Reviews Molecular Cell Biology, 24, 430-447 (2023). In some embodiments, the IncRNA are long noncoding RNA (IncRNA) bearing m6A sites; i.e., in which one or more of the adenosine nucleotides have been modified to m6A.

[0044] The inventors have identified 3215 upregulated and 97 downregulated IncRNAs in TKI resistant samples, as compared with healthy samples, as described in the Example provided herein. In some embodiments, the method involves determining the level of IncRNA have m6A binding sites that are changed during TKI treatment or are present in responders vs. nonresponders or in blast crisis / advanced versus chronic phases of leukemia. The inventors have also identified 40 upregulated IncRNAs bearing m6A sites in nilotinib resistant samples, ranging in size from 985 to 9545 nucleotides in length. These upregulated IncRNAs were identified by the genenames PROX1-AS1, LINC00989, SENCR, LINC00892, HAGROS, PCAT7, MSC-AS1, FOXN3-AS 1, NOXA11-AS, LINC00654, TSPEAR-AS1, FER1L6-AS 1, TRPM2-AS, LINC01353, LINC01410, LINC01436, LINC01305, PSMG3-AS1, CCNYL2, TSPEAR- AS2, LINC01366, FGF14-AS2, CD27-AS1, CASC9, LINC00528, LINC01547, DGCR9, ARHGAP5-AS, GLYCTK-AS1, USP46-AS1, LINC00491, LOH12CR2, LIPE-AS1, ERVK13-1, TAPT1-AS 1, and MMP25-AS1. In some embodiments, the IncRNA is selected from the group of top-ranked IncRNAs, while in further embodiments the top-ranked IncRNAs consist of PROX1-AS1, SENCR, and LN892.

[0045] The Fat Mass and Obesity-Associated gene, or FTO, also known as alpha-ketoglutarate- dependent dioxygenase FTO, is a gene strongly linked to obesity and body mass index (BMI). It is located on chromosome 16 and is associated with increased fat mass and obesity risk. The amino acid sequence of the transcribed FTO protein is known and shows high similarity with the enzyme AlkB which oxidatively demethylates DNA. FTO has been demonstrated to efficiently demethylate the related modified ribonucleotide, N6,2'-O-dimethyladenosine, and to an equal or lesser extent, m6A, in vitro. Methods of detecting genes such as FTO are knownto those skilled in the art, and include PCR, DNA sequencing, or gene expression profiling, as well as methods of detecting the FTO protein. In some embodiments, the FTO is measured and compared between responders and non-responders, or between subjects having leukemia in blast crisis / advanced phase versus chronic phase.

[0046] The method includes the step of assigning a poor prognosis or poor TKI responder status to the subject if the level of m6A, IncRNA bearing m6A sites, IncRNA-specific m6A, and / or FTO is higher than the one or more corresponding control values (i.e., test values) obtained from a healthy subject. Poor TKI responder status corresponds to characterizing the susceptibility of the leukemia to treatment with a tyrosine kinase inhibitor as low or below average.

[0047] A test value, expression level or other calculated test level of a biomarker (i.e., the level of m6A, IncRNA bearing m6A sites, IncRNA-specific m6A, and / or FTO) in a subject's undiagnosed biological sample. The test level may be compared to that of a control sample, or may be analyzed based on a reference standard that has been previously established to provide information to establish a prognosis. A test sample or test amount can be either in absolute amount (e.g., number of copies / ml, nanogram / ml or microgram / ml) or a relative amount (e.g., relative intensity of signals).

[0048] A control value, expression level or other calculated level of a biomarker may be any amount or a range of amounts to be compared against a test amount of a biomarker. A control level may be the amount of a marker in a healthy or non-diseased state. For example, a control amount of a marker can be the amount of a marker in a population of patients with a specified condition or disease or a control population of individuals without said condition or disease. A control amount can be either in absolute amount (e.g., number of copies / ml, nanogram / ml or microgram / ml) or a relative amount (e.g., relative intensity of signals). A corresponding control value represents the control value for a particular biomarker; e.g., total m6A levels and the total m6A control value.

[0049] The level of the biomarker used in the methods for diagnosis, prognosis or monitoring leukemia described may be measured, quantified and / or detected by any suitable RNA detection, quantification or sequencing methods known in the art, including, but not limited to, reverse transcriptase-polymerase chain reaction (RT-PCR) methods, microarray, serial analysisof gene expression (SAGE), next-generation RNA sequencing (e.g., deep sequencing, whole transcriptome sequencing, exome sequencing), gene expression analysis by massively parallel signature sequencing (MPSS), immune-derived colorimetric assays, in-situ hybridization (ISH) formulations (colorimetric / radiometric) that allow histopathology analysis, mass spectrometry (MS) methods, RNA pull-down and chromatin isolation by RNA purification (ChiRP), and proteomics -based identification (e.g., protein array, immunoprecipitation) of IncRNA. In one embodiment, the method of measuring an IncRNA transcript or IncRNA bearing m6A sites level includes performing quantitative / gel-based electrophoresis PCR or non-PCR-based molecular amplification methods for detection.

[0050] The term “biological sample” as used herein refers to a sample that was obtained and may be assayed for any one of the biomarkers as disclosed with the present invention, or their gene expression. The biological sample can include a biological fluid (e.g., blood, cerebrospinal fluid, urine, plasma, serum), tissue biopsy, and the like. In some embodiments, the sample is a tissue sample, for example, tumor tissue, and may be fresh, frozen, or archival paraffin embedded tissue. Preferred samples for the purposes of the present invention are bodily fluids, in particular plasma samples.

[0051] In a preferred embodiment, the biological sample is a sample of a subject comprising a tissue sample or a body liquid sample, such as a sample of a group of cells from a tumor, a tumor tissue, an RNA sample, a DNA sample, a blood sample, a serum sample, a plasma sample, a urine sample, a lymph fluid sample, a pleural fluid sample, or a brain liquor sample, preferably a sample of a group of cells from a tumor or a tumor tissue.

[0052] Sample collection procedures and devices known in the art are suitable for use with various embodiments of the present invention. Examples of sample collection procedures and devices include but are not limited to: phlebotomy tubes (e.g., a vacutainer blood / specimen collection device for collection and / or storage of the blood / specimen), dried blood spots, Microvette CB300 Capillary Collection Device (Sarstedt), HemaXis blood collection devices (microf uidic technology, Hemaxis), Volumetric Absorptive Microsampling (such as CE-IVD Mitra microsampling device for accurate dried blood sampling (Neoteryx), HemaSpot™-HF Blood Collection Device, a tissue sample collection device; standard collection / storage device (e.g., a collection / storage device for collection and / or storage of a sample (e.g., blood, plasma, serum, urine, etc.); a dried blood spot sampling device. In some embodiments, the VolumetricAbsorptive Microsampling (VAMS™) samples can be stored and mailed, and an assay can be performed remotely.

[0053] The method includes evaluation of a subject who has or has had leukemia who has been or is being treated with a tyrosine kinase inhibitor. In some embodiments, the subject is one who has had leukemia but whose leukemia is in remission, while in other embodiments the subject is one who currently has leukemia. In some embodiments, the subject is currently being treated with a tyrosine kinase inhibitor, while in other embodiments the subject has previously been treated with a tyrosine kinase inhibitor. In further embodiments, wherein the subjects are AML and / or CML patients who have achieved complete remission but are still being treated with a tyrosine kinase inhibitor.

[0054] Tyrosine kinase can play a role in cancer as a result of autocrine or paracrine stimulation, mutations, or overexpression. In some embodiments, the subjects comprise AML and / or CML patients with acquired mutations in a tyrosine kinase domain. For examples of such mutations, see Du Z, Lovly CM, Mol Cancer, 17( 1) :58 (2018).

[0055] The presence or absence of a mutation may, for example, be determined by isolating nucleic acids obtained from cells, amplifying the nucleic acids and determining the presence or absence of the mutation in the amplified nucleic acids. In some of the embodiments, the presence or absence of the mutation is determined by allele-specific polymerase chain reaction (AS-PCR). As a standard value a corresponding property, in particular the base sequence, of a corresponding section of genomic DNA is suitable, which does not have the mutation, hereinafter also referred to as wild type, wild-type state or reference DNA. For example, genomic DNA from normal tissue, in particular healthy tissue, of the same individual is suitable as wild-type. This can be, for example, the preferred wild type with respect to the determination of a somatic mutation. As a wild-type, a reference genome can also be used. A suitable reference genome includes the human genome version of the Genome Reference Consortium Genome Reference Consortium Human Build 38 patch release 2 (GRCh38.p2) as of Jul. 16, 2015. For example, a reference genome may be the preferred wild type for determining a germline mutation and / or somatic mutation.Predicting Tyrosine Kinase Inhibitor Resistance

[0056] Another aspect of the invention provides a method of predicting tyrosine kinase inhibitor (TKI) resistance in a subject having leukemia or having an increased risk of developing leukemia. The method includes determining the levels of one or more of 1) total N6-methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, 3) IncRNA-specific m6A, 4) a fat mass and obesity-associated gene (FTO) in a biological sample from the subject; and characterizing the subject as more likely to exhibit resistance to a TKI if the level of m6A, IncRNA being m6A sites, IncRNA-specific m6A, and / or FTO is higher than the one or more corresponding control values obtained from a healthy subject. The biomarkers can be any of the biomarkers described herein. In some embodiments, the biomarkers comprise IncRNA bearing m6A sites, and the IncRNA is selected from the group consisting of PR 0X1- AS1, SENCR, and LN892.

[0057] Tyrosine kinase inhibitors (TKIs), also known as tyrphostins, are a class of drugs that block the activity of tyrosine kinases. This can occur through four different mechanisms, including blocking the tyrosine kinase active site. More than 50 tyrosine kinase inhibitors have been approved by the U.S. Food and Drug administration for the treatment of cancer. A number of tyrosine kinase inhibitors are known to those skilled in the art. Examples of TKIs include imatinib, asciminub, adavosertib, bevacizumab, bosutinib, crizotinib, nilotinib, lapatinib, dasatinib, erlotinib, pexidartinib, ripretinib, sunitinib, sorafenib, trametinib, trastuzumab, and pazopanib. See Ebrahimi et al., Cell Mol Life Sci., 80(4): 104 (2023). Preferred TKIs for treatment of leukemia include imatinib, dasatinib, nilotinib, bosutinib, and asciminub. See Kennedy, IA. and Hobbs, G. Curr Hematol Malig Rep., 13(3) :202-211 (2018). In some embodiments, the TIK being evaluated for resistance is imatinib or nilotinib.

[0058] TKI resistance refers to the ability of cancer cells (e.g., leukemia cells) to proliferate despite treatment of the patient with a TKI. This can occur through various mechanisms, including mutations in the target protein, activation of alternative signaling pathways, or changes in drug uptake or efflux. For example, cancer cells can acquire mutations that prevent the TKI from binding to its target protein or interfering with its function. A specific example of this are mutations in the BCR-ABL kinase domain, which prevent the TKI from binding effectively. For a review of TKI resistance, see Yang et al., Signal Transduct Target Then, 7(1):329 (2022).

[0059] In some embodiments, the subject having leukemia, or having an increased risk of developing leukemia, is being treated with a TKI, while in other embodiments the subject having leukemia, or having an increased risk of developing leukemia, is not currently being treated with a TKI. In further embodiments, the subject has been treated for leukemia, but is currently in remission. In some embodiments the leukemia is acute myeloid leukemia (AML) or chronic myeloid leukemia (CML).Treatment of TKI-resistant Leukemia

[0060] Another aspect of the invention provides a method of treating a subject having TKI- resistant leukemia, comprising administering to the subject a therapeutically effective amount of a TKI inhibitor and an effective amount of a compound that decreases the total level of m6A, the level of IncRNA bearing m6A sites and / or IncRNA-specific m6A in the subject. In some embodiments, the IncRNA that are decreased are PROX1-AS1, SENCR, and / or LN892. In further embodiments, the leukemia is acute myeloid leukemia (AML) or chronic myeloid leukemia (CML).

[0061] The inventors performed a pathway analysis for proteins that were differentially expressed in TKI-resistant leukemia, which showed that the phosphatidylinositol 3-kinase (PI3K) pathway played an important role in mediating IncRNA-sustained TKLresistance. Based on this, the inventors demonstrated the ability of a PI3K inhibitor to overcome TKI resistance. Accordingly, in some embodiments, the compound used to suppress TKLresistance by decreasing the level of IncRNA is a PI3K inhibitor.

[0062] A number of PI3K inhibitors are known to those skilled in the art. PI3K inhibitors include pan-PI3K inhibitors, isoform-selected inhibitors, and deal PI3K / mT0R inhibitrs. Examples of PI3K inhibitors include alpelisib, copanlisib, duvelisib, idelalisib, umbralisib, and leniolisib. In some embodiments, the PI3K inhibitor is alpelisib. See Sirico et al., Cancers (Basel), 15(3):703 (2023).

[0063] Other compounds that decrease the level of IncR A bearing m6A sites and / or IncRNA- specific m6A are antisense oligonucleotides. Antisense oligonucleotides are tools for use in inhibiting the expression of target genes in a sequence- specific manner, and can be targeted to inhibit IncRNA. Different types of anti-mRNA strategies, for example, the use of single stranded antisense-oligonucleotides, the triggering of RNA cleavage through catalyticallyactive oligonucleotides referred to as ribozymes, and RNA interference induced by small interfering RNA molecules, have been attempted.

[0064] Antisense oligonucleotides may, for example, consist of from about 15 to about 20 nucleotides, which are complementary to their target mRNA. Phosphorothioate (PS) oligodeoxynucleotides are one type of oligonucleotide used for inhibiting gene expression. In this class of oligonucleotide, one of the non-bridging oxygen atoms in the phophodiester bond is replaced by sulfur. The introduction of phosphorothioate linkages into oligonucleotides was primarily intended to enhance their nuclease resistance.

[0065] In some embodiments, the method further comprises treating the subject with an additional therapeutic method such as surgical resection, chemotherapy, cryotherapy, radiation therapy, or immunotherapy. The use of an additional method of cancer treatment can be used before, during, or after treatment with the TKI inhibitor and PI3K inhibitor.

[0066] Chemotherapy involves treatment of cancer with an anticancer agent. Examples of chemotherapeutic agents include alkylating agents (e.g., doxorubicin), antimetabolites (e.g., methotrexate), topoisomerase inhibitors (e.g., etoposide), mitotic inhibitors (e.g., vincristine), platinum-based compounds (e.g., carboplatin), and anthracycline antibiotics (e.g., epirubicin).

[0067] Surgical resection refers to the surgical removal of all or part of the cancer. Cryotherapy includes, but is not limited to, therapies involving decreasing the temperature, for example, hypothermic therapy.

[0068] Radiation therapy includes, but is not limited to, exposure to radiation, e.g., ionizing radiation, UV radiation, as known in the art. Exemplary dosages include, but are not limited to, a dose of ionizing radiation at a range from at least about 2 Gy to not more than about 10 Gy or a dose of ultraviolet radiation at a range from at least about 5 J / m2to not more than about 50 J / m2, usually about 10 J / m2.

[0069] Immunotherapy either modulates the immune system or in some embodiments regulates immune checkpoints. In further embodiments, the immunotherapy comprises, or consists essentially of, or yet further consists of an immune checkpoint inhibitor, such as a Cytotoxic T-Lymphocyte Associated Protein 4 (CTLA4) inhibitor, or a Programmed Cell Death 1 (PD-1) inhibitor, or a Programmed Death Ligand 1 (PD-L1) inhibitor. In yet furtherembodiments, the immune checkpoint inhibitor comprises, or consists essentially of, or yet further consists of an antibody or an equivalent thereof recognizing and binding to an immune checkpoint protein, such as an antibody or an equivalent thereof recognizing and binding to CTLA4 (for example, Yervoy (ipilimumab), CP-675,206 (tremelimumab), AK104 (cadonilimab), or AGEN1884 (zalifrelimab)), or an antibody or an equivalent thereof recognizing and binding to PD-1.

[0070] The effectiveness of cancer treatment may be measured by evaluating a reduction in tumor load or decrease in tumor growth in a subject in response to treatment. The reduction in tumor load may represent a direct decrease in mass, or it may be measured in terms of tumor growth delay, which is calculated by subtracting the average time for control tumors to grow over to a certain volume from the time required for treated tumors to grow to the same volume.

[0071] In certain embodiments, the methods described herein may include a step of monitoring or assessing the progression of leukemia in a subject; monitoring or assessing a response to treatment in a subject having leukemia; monitoring or assessing a metastatic spread of leukemia in a subject; monitoring or assessing a remission state or a recurrence of melanoma in a subject or a combination thereof. Such monitoring or assessing may include an individual's response to a therapy, such as, for example, predicting whether an individual is likely to respond favorably to a tyrosine kinase inhibitor (TKI), is unlikely to respond to a TKI, or will likely experience toxic or other undesirable side effects as a result of being administered a therapeutic agent; selecting a therapeutic agent for administration to an individual, or monitoring or determining an individual's response to a therapy that has been administered to the individual.Drug Identification

[0072] Another aspect of the invention provides a method of identifying a drug for suppressing TKI resistance in leukemia treatment. The method includes administering an effective amount of a test compound to a subject having TKI-resistant leukemia; evaluating whether the test compound decreases the level of 1) total N6-methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, and / or 3) IncRNA-specific m6A, in the subject; and determining that the test compound is a candidate for treatment of TKI-resistant leukemia if the compound decreases the level of 1) total N6-methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, and / or 3) IncRNA-specific m6A in the subject.

[0073] As described herein, the inventors have determined that the FT0 / m6A axis plays an important role in resistance to TKI in leukemia treatment, and more specifically, that suppressing long non-coding RNA (IncRNA) bearing m6A sites helps to overcome TKI resistance. This discovery provides the basis for a method for identifying drugs for suppressing TKI resistance in leukemia treatment by seeing if a test compound can decrease the levels of total N6-methyladenosine (m6A), long non-coding RNA (IncRNA) bearing m6A sites, and / or IncRNA-specific m6A. A test compound is any pharmaceutically acceptable compound that may, or is suspected of having, an effect on m6A and / or IncRNA.

[0074] In some embodiments, the subject is an animal model for leukemia. Animal models play a crucial role in leukemia research, particularly in understanding disease mechanisms and developing new treatments. Mice are the most common animal model for leukemia, due to their genetic and hematopoietic similarities with humans. Other models include rats, zebrafish, and Fruit flies (Drosophila melanogaster). Leukemia models can be chemically induced, transgenic or knock-in models, or xenograft models in which leukemia cells are transplanted into immunodeficient mice.

[0075] The IncRNA that is used in the method can be any of the IncRNA described herein. In some embodiments, the IncRNA have m6A binding sites and it has been observed that their levels change during TKI treatment or in TKI responders vs. non-responders, or in blast cris / advanced leukemia compared with chronic phase leukemia. In some embodiments, the IncRNA is selected from the group consisting of PROX1-AS1, SENCR, and LN892.Formulations

[0076] The pharmaceutical compositions used in the present invention comprise an active agent such as a TKI inhibitor or PI3K inhibitor, or pharmaceutically acceptable salts thereof, as the active ingredient. The pharmaceutical compositions may also contain a pharmaceutically acceptable carrier and optionally other therapeutic ingredients.

[0077] The term "composition", as in pharmaceutical composition, is intended to encompass a product comprising the active ingredient(s), and the inert ingredient(s) that make up the carrier, as well as any product which results, directly or indirectly, from combination, complexation or aggregation of any two or more of the ingredients, or from dissociation of one or more of the ingredients, or from other types of reactions or interactions of one or more of the ingredients.Accordingly, the pharmaceutical compositions of the present invention encompass any composition made by admixing a compound of the present invention and a pharmaceutically acceptable carrier.

[0078] The compositions include compositions suitable for oral, rectal, topical, parenteral (including subcutaneous, intramuscular, and intravenous), ocular (ophthalmic), pulmonary (nasal or buccal inhalation), or nasal administration, although the most suitable route in any given case will depend on the nature and severity of the conditions being treated and on the nature of the active ingredient. They may be conveniently presented in unit dosage form and prepared by any of the methods well-known in the art of pharmacy. Because of its convenience, in some embodiments oral administration is used.

[0079] The active agents(s) (e.g., TKI inhibitor) can be combined as the active ingredient in intimate admixture with a pharmaceutical carrier according to conventional pharmaceutical compounding techniques. The carrier may take a wide variety of forms depending on the form of preparation desired for administration, e.g., oral or parenteral (including intravenous). In preparing the compositions for oral dosage form, any of the usual pharmaceutical media may be employed, such as, for example, water, glycols, oils, alcohols, flavoring agents, preservatives, coloring agents and the like in the case of oral liquid preparations, such as, for example, suspensions, elixirs and solutions; or carriers such as starches, sugars, microcrystalline cellulose, diluents, granulating agents, lubricants, binders, disintegrating agents and the like in the case of oral solid preparations such as, for example, powders, hard and soft capsules and tablets, with the solid oral preparations being preferred over the liquid preparations.

[0080] Because of their ease of administration, tablets and capsules represent the most advantageous oral dosage unit form, in which case solid pharmaceutical carriers are obviously employed. If desired, tablets may be coated by standard aqueous or non-aqueous techniques. Such compositions and preparations should contain at least 0. 1 percent of active compound. The percentage of active compound in these compositions may, of course, be varied and may conveniently be between about 2 percent to about 60 percent of the weight of the unit. The amount of active compound in such therapeutically useful compositions is such that an effective dosage will be obtained. The active compounds can also be administered intranasally as, for example, liquid drops or spray.

[0081] The tablets, pills, capsules, and the like may also contain a binder such as gum tragacanth, acacia, corn starch or gelatin; excipients such as dicalcium phosphate; a disintegrating agent such as corn starch, potato starch, alginic acid; a lubricant such as magnesium stearate; and a sweetening agent such as sucrose, lactose or saccharin. When a dosage unit form is a capsule, it may contain, in addition to materials of the above type, a liquid carrier such as a fatty oil.

[0082] The active agent may also be administered parenterally. Solutions or suspensions of these active compounds can be prepared in water suitably mixed with a surfactant such as hydroxy-propylcellulose. Dispersions can also be prepared in glycerol, liquid polyethylene glycols and mixtures thereof in oils. Under ordinary conditions of storage and use, these preparations contain a preservative to prevent the growth of microorganisms.

[0083] The pharmaceutical forms suitable for injectable use include sterile aqueous solutions or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersions. In all cases, the form must be sterile and must be fluid to the extent that easy syringability exists. It must be stable under the conditions of manufacture and storage and must be preserved against the contaminating action of microorganisms such as bacteria and fungi. The carrier can be a solvent or dispersion medium containing, for example, water, ethanol, polyol (e.g. glycerol, propylene glycol and liquid polyethylene glycol), suitable mixtures thereof, and vegetable oils.

[0084] The term "pharmaceutically acceptable salts" refers to salts prepared from pharmaceutically acceptable non-toxic bases or acids including inorganic or organic bases and inorganic or organic acids. Salts derived from inorganic bases include aluminum, ammonium, calcium, copper, ferric, ferrous, lithium, magnesium, manganic salts, manganous, potassium, sodium, zinc, and the like. Particularly preferred are the ammonium, calcium, magnesium, potassium, and sodium salts. Salts in the solid form may exist in more than one crystal structure, and may also be in the form of hydrates. Salts derived from pharmaceutically acceptable organic non-toxic bases include salts of primary, secondary, and tertiary amines, substituted amines including naturally occurring substituted amines, cyclic amines, and basic ion exchange resins, such as arginine, betaine, caffeine, choline, N,N’-dibenzylethylenediamine, diethylamine, 2-diethylaminoethanol, 2-dimethylaminoethanol, ethanolamine, ethylenediamine, N-ethyl-morpholine, N-ethypiperideine, glucamine, glucosamine, histidine,hydrabamine, isopropylamine, lysine, methylglucamine, morpholine, piperazine, piperidine, polyamine resins, procaine, purines, theobromine, triethylamine, trimethylamine, tripropylamine, tromethamine, and the like.

[0085] An Example has been included to more clearly describe a particular embodiment of the invention and its associated cost and operational advantages. However, there are a wide variety of other embodiments within the scope of the present invention, which should not be limited to the particular example provided herein.EXAMPLE

[0086] The inventors tested the hypothesis that epitranscriptomic diversity in heterogeneous tumor cell populations may generate divergence in the expression of cell fate determination genes / pathways that can swiftly avoid drug-induced cell death. More specifically, we examined the potential interactions between m6A methylome and IncRNAs in resistant leukemia cells and found that this interaction shapes the balance between TKI resistance and sensitivity. We dissected its pathological relevance in leukemia patients and found that m6A-upregulated IncRNAs are novel biomarkers associated with TKI non-responders and predict worse outcomes.ResultsThe activities of the targeted kinase signaling are not required for the survival and proliferation of TKI-resistant cells

[0087] To dissect the molecular mechanisms of acquired resistance to inhibitors of TK pathway, we generated drug-resistant derivatives of leukemia cells K562 (BCR / ABL), MV4- 11 (FLT3) and Kasumi-1 (c-KIT) by long-term culture in the presence of TKIs nilotinib (second generation) or imatinib (first generation) (10, 30, 100, 300, 1,000 nM). Cells cultured in parallel without drugs serve as parental / sensitive controls. Cells are considered resistant when they can routinely grow in medium containing 1 p M nilotinib or imatinib, respectively. We then characterized the resistant phenotypes by measuring the survival rate of parental and resistant cells upon transient exposure to TKIs. All parental controls displayed a significant decrease in cell viability (Figure 1A) and an increase in apoptotic cells (Figure IB), but the resistant cells could proliferate in the drugs with no obvious changes of cell apoptosis. Resultsfrom colony-forming assays were heterogeneous (Figure 1C and D), in which nilotinib- resistant K562, Kasumi-1 and MV4-11 cells have significantly higher colony-forming potential with more colonies and larger colony sizes. However, imatinib-resistant K562 cells did not show significant changes in colony number and size compared with parental cells. Flow cytometry detected different cell sizes between parental and resistant K562 cells, in which imatinib resistant cells are the largest (Figure IE). Finally, we examined the activities of the targeted TKs, and found that, compared to parental cells, the phosphorylation of KIT, FLT3 and BCR / ABL is decreased without obvious changes of total protein expression when growing in drug-containing medium, leading to dephosphorylation of STAT5 (Figure IF), a shared downstream signaling mediator. These results suggest that survival and proliferation of TKI resistant cells are independent of the targeted TK signaling.Transcriptome-wide m6A sequencing unveils distinct IncRNAs in nilotinib resistant versus sensitive leukemia cells

[0088] Our previous studies (Yan et al., Cell research 28, 1062-1076 (2018)) showed that, compared with parental cells, most changes of m6A peaks in resistant cells come from IncRNAs as well as CDs and 3’UTR with less profound changes at 5’UTR. These findings imply that m6A-associted IncRNAs may partially affect the development of TKI resistance. To test this, first we analyzed the m6A-peak fold enrichment from normalized read counts mapped to peak region focusing on the changes of IncRNAs. We found that the distribution of log2 transformed fold change show three peaks, the highest peak in the middle part of the distribution indicating that most IncRNAs have comparable m6A levels with log2 fold change close to 0, while the peaks of the positive log2 fold change located on the right side of the distribution are much higher than those of the negative log2 fold change on the left side, indicating that relative more IncRNAs tend to have higher m6A in resistant cells (Figure 2A). These observations suggest that, compared with parental cells, the overall m6A peak fold enrichment is decreased, but IncRNAs, in general, have a trend of higher m6A content in resistant cells. Second, we analyzed m6A RNA sequencing data, and identified 315 upregulated and 97 downregulated IncRNAs in resistant versus parental cells. By intersection analyses of differentially expressed IncRNAs with changed m6A peaks, we identified 40 upregulated IncRNAs bearing m6A sites in Nil (nilotinib resistant) samples. The heatmap in Figure 2B was plotted based on the sorted log2- fold change of top 40 overlapped IncRNAs. For all immunoprecipitation (IP) and input (IN) samples, the sorted bam files along with m6A site coordinate information stored in bed fileswere loaded into Integrative Genomics Viewer (IGV) to compare expression levels of interested intergenic IncRNAs (e.g., LN892) across different conditions (Figure 2C).

[0089] We conducted qPCR for total RNA from K562 cells, and verified the alterations of topranked IncRNAs (Figure 2C), for example, upregulation of LN989, PROX1-AS1, SENCR, LN892, LN504, and KIF25-AS 1; downregulation of LN659, VPS9D1-AS1, LNC200, UCA1 and WASIR2; but no change in LN1270 and MAP3K14-AS1. Given that our RNA-seq is done in the fraction of poly-A-associated RNAs, and as IncRNAs have both poly- A and non-poly-A tailed RNAs (Yang et al., Genome biology 12, R16 (2011)), we separated poly-A tailed RNAs from non-poly-A tailed RNAs, and measured the IncRNA levels. We observed that the expression of PROX1-AS1, LN892 and SENCR is upregulated as a similar pattern in both poly-A and non-poly-A tailed fraction (Figure 2D). Pathway analysis revealed that PROX1- AS1 regulated pathways are mostly negatively enriched, but LN892 and SENCR regulated pathways are mostly positively enriched; pathways in red boxes are positively enriched pathways regulated by LN892 and SENCR; MYC pathway in blue box is significantly negatively enriched and regulated by all three IncRNAs; PI3K-AKT / mT0R signaling is negatively enriched and regulated by PROX1-AS1 but has trends toward being enriched when regulated by SENCR and LN892. Taken together, these data suggest that m6A- associated IncRNAs are involved in a non-genetic mechanism that underlies the emergence and maintenance of acquired TKI resistance.Expression of the m6A-associated IncRNAs is particularly elevated in leukemia patients at diagnosis or not-responding to TKIs, and could serve as an independent predictor for poor prognosis

[0090] To explore the clinical implications of the dysregulated IncRNAs, first, we analyzed the gene expression profiling of AML collected at diagnosis, and compared the expression of FTO, LIN892, LIN989, PROX1-AS1 and SENCR among different response groups. In these AML patients, a trend towards higher levels of expression of LIN989 and SENCR was observed in the ELN risk groups intermediated and adverse. In line, expression of SENCR at diagnosis was higher in AML patients that did not achieve complete remission (CR) after induction chemotherapy. Furthermore, we performed an exploratory analysis of matched samples from CML patients at diagnosis in chronic phase, during hematologic remission and at the time of progression to blast phase CML. Expression of FTO was highest at blast phaseand lowest at hematologic remission. Also, differential expression of IncRNAs was overserved in this cohort. Such heterogeneous correlation between the expression of given genes and drug responses could be attributed to the variations of blast percentage, genetic mutations, age and sex, which will be validated in a larger cohorts of patients. To further clarify the role of FTO, LIN892, LIN989, PR0X1-AS1 and SENCR in drug resistance, we obtained CML PB cells from patients after nilotinib or imatinib therapy. The definition of responding and inadequately responding was based on the number of BCR / ABL transcripts post therapy, but the chronic phase (Figure 3A, upper panel; n = 3) versus blast crisis (Figure 3A, lower panel; n = 3) was determined by the cell morphology and relative clinical information. When the BCR / ABL international score (IS) meets response milestones at 3, 6, and 12 months (< 10% BCR / ABL IS at 3 and, < 1% BCR / ABL IS at 6 months and < 0.1% BCR / ABL IS at >12 months) post therapy, these patients are classified as TKI responders, and others as inadequate responders (Figure 3B) (O’Brien et al., J Natl Compr Cane Netw 7, 984-1023 (2009)). We then performed qPCR in normal and CML patient cells for LN989, PROXI -AS 1 , SENCR, LN892 and KIF25- AS1, which are consistently elevated in resistant K562 cells. All aforementioned IncRNAs are significantly upregulated in CML patients when compared with normal blood cells. We also employed online tools GEPIA (Tang etal., Nucleic acids research 45, W98-W102 (2017)) and BloodSpot (Bagger et al., Nucleic acids research 44, D917-924 (2016)) to further validate the expression patterns of the aforementioned IncRNAs in leukemia patients and normal donors. Consistently, expression of LN892 and SENCR in GEPIA as well as PROX1-AS1 and KIF25- AS 1 in Bloodspot (Leukemia MILE Study) is much higher in leukemia patients than those in normal donors with barely or undetectable expression of LN892 in both datasets. When compared with TKI responders, CML patients with inadequate response had significantly higher expression of LN989, PROX1-AS1, SENCR, LN892, and KIF25-AS1 (Figure 3C). When compared with chronic disease, patients with blast crisis had a trend toward higher IncRNA expression. Notably, to strengthen the conclusion, we employed both 18S and ABL as internal controls to normalize IncRNA expression in patients. The same conclusions were made when using ABL, the most cited internal control, although more obvious difference was observed when 18S was used.

[0091] To further substantiate the clinical significance of the m6A-assicated IncRNAs in leukemia, we also examined the expression of LN989, PROX1-AS1, SENCR, LN892 and KIF25-AS1 in AML patients, who received nilotinib twice daily after induction and consolidation chemotherapy. We observed that nilotinib (in combination with chemotherapy)upregulates LN989, PR0X1-AS1, SENCR, LN892 and KIF25-AS1 (Figure 3D). Finally, we used online tool GEPIA to explore the association between the expression of these IncRNAs and patient survival. As expected, overexpression of LN989, SENCR, LN892 and KIF25-AS1 significantly or has a trend to predict shorter survival time (Figure 3E). There were no survival plots for PR0X1-AS1 in GEPIA. Moreover, higher levels of LN989, SENCR, LN892 and KIF25-AS1 tend to predict poorer survival in leukemia patients from the TCGA study using two different comparisons. The Kaplan-Meier test was not performed on PROX1-AS1 due to its extremely low expression in TCGA leukemia patients (median of FPKM < 0.003). Together, these results indicate that upregulation of m6A-associated IncRNAs could be a common vulnerability and prognostic factors in CML and AML patients post TKI therapy.LncRNA upregulation in resistant cells takes place through the enhanced RNA stability by FTO-mediated m6A demethylation

[0092] Given that m6A binding motifs are enriched within IncRNAs, and as m6A methylation has been found to regulate the stability of RNA transcripts (Geula et al., Science, 347(6225): 1002-6 (2015), IncRNA upregulation in resistant cells may result from prolonged half-life of RNA transcripts, due to m6A demethylation. To this end, we first measured the levels of global m6A amounts and FTO expression, and observed a decrease of m6A abundance (Figure 4A, left) and an increase of FTO expression (Figure 4A, left) in K562, MV4-11 and Kasumi- 1 cells resistant to nilotinib or imatinib, in line with our previous reports. Further, FTO levels were significantly elevated in nilotinib-treated AML patients and in CML patients nonresponding to imatinib or nilotinib therapy. Most notably, TKI non-responders without acquired kinase domain mutations have significantly higher FTO expression than those carrying mutations (Figure 4A, right). As IncRNAs include poly-A and non-poly A tailed, we then performed RNA IP using anti-m6A antibody in both poly-A and non-poly A fraction, converted the eluted RNA to cDNA and carried out qPCR with primers covering m6A binding sites. As shown in Figure 4b and 4c, the m6A amounts for PROX1-AS1 , SENCR and LN892 were increased without obvious changes in SENCR expression in nilotinib-resistant K562 cells when using GAPDH as internal control. For imatinib-resistant K562 cells, the amounts for PROX1-AS 1 with poly-A tailed and LN892 without poly-A tailed were decreased. Because the total levels of these IncRNAs are dramatically increased in resistant versus parental cells (ref. Figure 2D, left), to accurately reflect the changes of m6A amount specific for these IncRNAs, we normalized the m6A-IP related alterations to the changes of total IncRNA levels,and observed that the m6A levels of PR0X1-AS1, SENCR and LN892 are decreased in resistant versus parental controls, in parallel with the changes in global m6A content (Figure 4D). These findings suggest that m6A hypomethylation accounts for IncRNA upregulation in resistant cells.

[0093] To study whether FTO, a key m6A demethylase, plays a role in IncRNA dysregulation, we infected K562 cells with scramble or FTO shRNAs (shRNA-TRCN0000183897, TRCN0000179651, TRCN0000180978) viruses, and selected TRCN0000180978 for further investigation, because TRCN0000180978 had the highest efficacy to knock down FTO. Colony-forming assays revealed that FTO knockdown impairs colony potential as evidence by a decrease of colony number and size. Then single clones were selected and expanded for further investigations. Firstly, knockdown of FTO and increase of m6A methylation were confirmed in each clone (Figure 4E). Secondly, qPCR analysis disclosed that expression of PROX1-AS1, SENCR and LN892 is significantly inhibited in clones with FTO knockdown compared with scrambled controls (Figure 4F). Thirdly, in line with FTO knockdown, pharmacological inhibition of FTO by meclofenamic acid (MA) (Huang et al., Nucleic acids research 43, 373-384 (2015)) and FB23-2 (Huang et al., Cancer cell 35, 677-691 e610 (2019)) downregulated these IncRNAs in K562 cells resistant to imatinib or nilotinib (Figure 4G). Notably, no obvious changes in FTO protein expression were observed suggesting the enzymatic inhibition by MA and FB23-2; the increased downregulation of IncRNAs by FB23- 2 was consistent with its higher and more selective FTO enzymatic inhibition when compared with MA. To assess changes of m6A on individual IncRNAs, we performed m6A IP in total RNAs, poly-A and non-poly-A tailed RNAs in clones with scrambled and FTO knockdown. qPCR analysis revealed that m6A content of PROX1-AS 1, SENCR, and LN892 is significantly increased when normalized to the expression of IncRNAs from total RNAs (Figure 4H). These results suggest that FTO upregulates IncRNAs through decreasing m6A amount of specific IncRNAs.

[0094] It has been shown that m6A methylation predominantly and directly decreases transcript stability. To address why the m6A-containing IncRNAs are downregulated in FTO knockdown cells, we examined the RNA stability in FTO knockdown and control cells. Namely, we treated FTO knockdown or control clones with antinomycin-D, a transcription inhibitor. The total RNA was extracted and the IncRNA levels were assessed by qPCR, which was normalized to GAPDH. As expected, the decay of IncRNAs was shortened in FTO knockdown cellscompared to scrambled cells (Figure 41). Collectively, these findings suggest that IncRNA upregulation in resistant cells is attributed to longer RNA half-life, which is mediated by FTO- dependent m6A demethylation.Genetic suppression of the FTO-lncRNA axis promotes nilotinib sensitivity, and decreases proliferation and growth of resistant cells

[0095] Having demonstrated the regulatory roles of FTO and m6A in IncRNA expression, we proceeded to address the impacts of IncRNAs on TKI sensitivity using K562 clones with stable FTO knockdown. We first confirmed that FTO knockdown impairs colony potential as evidence by a decrease of colony number and size compared to scrambled controls (Figure 5A); expression of IncRNAs, including LN989, LN892, LN504, PROX1-AS1, SENCR, and KIF25-AS1, was significantly decreased in cells with FTO knockdown. We then examined the survival of FTO knockdown or scrambled clones upon transient exposure to nilotinib. As shown in Figure 5B, the scrambled clones had IC50 values to nilotinib several orders of magnitude larger than those exhibited by FTO knockdown. Although all FTO knockdown clones displayed significant and dose-dependent decreases of cell viability, the scrambled clones could proliferate at drug concentrations much larger than the IC50 value, in agreement with our previous findings that FTO positively regulates TKI sensitivity in leukemia cells.

[0096] We have shown that IncRNA upregulation in TKI non-responding CML patients predicts worse outcomes. If IncRNAs confer increase of TKI protection to leukemia cells, then ablating IncRNAs in these cells should confer increased sensitivity to TKIs. To test this hypothesis, we knocked down LN892, PROX1-AS1 or SENCR in K562 resistant cells carrying IncRNA upregulation (Figure 5C), because these three IncRNAs carry m6A motifs and are highly expressed with a marked decrease of their m6A methylation in resistant cells. Colony assays showed that knockdown of PROX1-AS 1, SENCR or LN892 reduces colony number and size (Figure 5D). To test whether IncRNA abundance affects nilotinib sensitivity, we treated IncRNA knockdown clones with various concentrations of nilotinib for 72 hours. We found that knockdown of IncRNAs renders the resistant clone sensitive to nilotinib-inhibited cell proliferation, as supported by showing that the IC50 values in clones with PROX1-AS1 (AS1521), SENCR (S1238) orLN892 (ln-619) depletion are several orders of magnitude lower than those in scrambled (Figure 5E). Collectively, these results suggest that IncRNAs enable culture leukemia cells to better withstand nilotinib-induced cell death.LncRNA-fueled growth of resistant cells is linked to the active PI3K-AKT signaling pathway

[0097] The function of PROX1-AS1, SENCR or LN892 in regulating TKI resistance may be through a change in a specific gene program. To gain further insights of IncRNA-conferred resistant cell growth, we first analyzed a public database (GSE51878) using LncRNA2Target v2.0, in which SENCR was knocked down in cancer cells, to identify IncRNA downstream targets. We focused on SENCR, but not PROX1-AS 1 and LN892, because the pathogenic roles of SENCR have been documented in cancer, but limited knowledge is available for PROX1- AS1 and LN892. Importantly, our findings revealed that SENCR is also subjected to regulation by FTO-dependent m6A hypomethylation, and affects response to TKIs in vitro and in patients and impacts patient survival. The signature of SENCR consisted of 663 differentially expressed genes. Using DAVID 6.8 for functional annotation, KEGG analysis provided insights into biological processes enriched in SENCR knockdown cells, and revealed many functional pathways, which involve 291 SENCR-regulated genes.

[0098] To select the most important pathways for further investigations, we performed functional pathway analyses for all differentially expressed genes together (n = 663), or dividing them into up- (n = 337) or downregulated (n = 326) groups. All these strategies nominated the phosphatidylinositol 3-kinase (PI3K) pathway as mediating IncRNAs-sustained resistant cell growth (Figure 6A and B). This was especially interesting given that the PI3K signaling plays an important role in leukemia pathogenesis with limited knowledge in drug resistance. Wohrle et al., Leukemia 27, 118-129 (2013). Although 13 up- and 12 down- regulated genes were enriched in the PI3K signaling, we focused on 7 top-ranked downregulated genes (e.g., ITGA2, COL6A1, cyclin DI, PKN1, PDGFRA, F2R, HSP90AB1), because SENCR behaved as an oncogenic IncRNA in TKI resistance. Further, these genes were either downregulated to the highest levels in cells with SENCR knockdown, or well-established resistant genes and / or annotated to have a role in sustaining cancer cell survival and proliferation. Shen et al., Clin Cancer Res., 23(20):6254-6266 (2017). qPCR on RNAs from independent samples confirmed that, compared with scrambled controls, ITGA2, COL6A1, cyclin DI, PKN1, F2R, PDGFRA, and HSP90AB1 are downregulated in imatinib- and nilotinib-resistant K562 clones with knockdown of LN892, PROX1-AS1 or SENCR (Figure 6C). These findings place PI3K signaling as downstream of all three InsRNAs PROX1-AS 1, SENCR and LN892 in resistant cells.

[0099] Finally, the merits of the aforementioned genes within PI3K signaling warrant further insights before we could abrogate this signaling and assess the consequences to TKI resistance. We examined the expression of ITGA2, COL6A1, cyclin DI, PKN1, F2R, PDGFRA, and HSP90AB 1, and found that these genes are significantly upregulated in imatinib or nilotinib- resistant versus sensitive K562 cells (Figure 6D) and in TKI resistant CML patients compared to responding counterparts (Figure 6E). To address the clinical relevance of these PI3K signaling downstream effectors that we unearthed in vitro, we first compared the expression of ITGA2, COL6A1, cyclin DI, PKN1, F2R, and PDGFRA among patients at a diagnosis with follow-up information available. In AML patients, a trend towards higher levels of expression of most genes was observed in the ELN risk groups intermediated and adverse. In line, a trend towards higher expression of COL6A1, F2R and ITGA2 at diagnosis in AML patients that did not achieve CR after induction chemotherapy was observed. In line, a trend towards higher expression of COL6A1, F2R and ITGA2 was seen in blast phase CML compared to samples at initial diagnosis. Notably, the predictable capacity is heterogeneous and not robust, which warranty further verification by co-founding with multiple factors, including blast percentage, genetic mutations, age and sex, in a larger cohorts of patients. We then analyzed public datasets (GSE12417 or TCGA-AML) to examine whether expression of these targets is associated with patient survival. We found that upregulation of ITGA2, COL6A1, cyclin D1 / CCND1, PKN1, PDGFRA and F2R genes predicts or at least has trends toward worse outcomes (Figure 6F). This result was further verified by the findings from GEPIA-mediated analysis. Collectively, these results provide compelling evidence that key components of the PI3K-AKT signaling are responsible for IncRNA-sustained TKI resistant cell growth.TKI resistant cells are sensitive to PIK3 inhibitor alpelisib in vitro and in vivo

[0100] Having shown that PI3K signaling is essential for leukemia TKI resistance, we reasoned that pharmacological targeting of PI3K signaling could override resistant cells. To test this, we selected alpelisib, an FDA approved first PI3K inhibitor for breast cancer, which has not been tested in leukemia. When parental or resistant K562 cells were treated with alpelisib in vitro for 72 hours, the growth of drug-resistant cells was significantly inhibited by alpelisib in a dose-dependent manner with ICso values in parental cells that are 2.7-fold or 4.3-fold higher than that in IMR or NIR cells (Figure 7A). The lower ICso values in resistant cells indicated the higher activation of PI3K signaling and thus more sensitive to alpelisib treatment. Further, the addition of alpelisib lowered the ICso of imatinib by 3.5 folds and the ICso of nilotinib by625 fold in imatinib- or nilotinib-resistant cells (Figure 7B). We then treated resistant cells with alpelisib alone or in combination with imatinib or nilotinib for 24, 48, 72 or 96 hours. Although alpelisib, imatinib or nilotinib as single drugs marginally slowed down resistant cell growth, their combination resulted in more pronounced impairment of cell proliferation, as demonstrated by the highest rate of cell apoptosis (Figure 7C and D). Mechanistically, combination of alpelisib with nilotinib led to more reduction of PIK3 signaling mediators, such as PKNI, F2R and Cyclin DI (Figure 7E). These results imply that PIK3 signaling is a therapeutic target for overcoming TKI resistance.

[0101] To further determine the clinical implications of PI3K signaling in TKI resistance, we established a CML mouse model by injecting 0.5 x 106nilotinib resistant K562 cells via the tail-vein into sublethally irradiated (2.5 Gy) triple transgenic NSG-SGM3 (NSGS) mice (male). The successful engraftment of CML cells was verified by the identification of human CD45+ cells in mouse bone marrow (BM) cells (Figure 7F) and the infiltration of CML K562 cells to mouse organs (e.g., liver, lung, spleen) (Figure 7G). Then the BM cells (0.5 x 106) were isolated from mice bearing K562 resistant cells and injected into the second recipient NSGS mice 4 hours after sublethal irradiation. These leukemic mice were randomly grouped and sequentially given 30 or 50 mg / kg of alpelisib (Delestre et al., Sci Transl Med 13, eabg0809 (2021)) in PEG400 and saline (ratio 15:38:47) intraperitoneally twice a week for a total of eight doses. The leukemic mice injected with only vehicle served as controls. Although the size and weight of spleen and liver did not show significant difference (Figure 7H and I), H&E staining revealed that, compared to the alpelisib-treated mice, the vehicle group displayed an increased infiltration of leukemic cells into the spleens, lungs and livers of recipients, leading to considerable damage to these organs (Figure 7 J). The BM histopathology from alpelisib-treated mice identified more differentiated cells containing metamyelocytes, bands, and segmented neutrophils compared to vehicle-treated mice (Figure 7K). Alpelisib-treatment significantly slowed the decrease of body weight (Figure 7L), importantly, increased the survival time of leukemic mice (Figure 7M). No toxicity was observed for the tested drug dosage and schedule because we did not see any evident change in capability of moving and getting food and water when comparing alpelisib-treated mice to vehicle groups. Mechanistically, alpelisib administration in mice remarkably downregulated PI3K signaling mediators, including ITGA2, COL6A1, cyclin D1 / CCND1, PKNI, PDGFRA or F2R (Figure 7N). Of note, the leukemic mice were bearing human CML K562 cells, thus we used primers for human, but not mouse, genes to detect the aforementioned genes. This approach could verify the limited toxicity ofalpelisib to leukemic mice and demonstrate the specificity of CML cell killing by alpelisib in vivo. Together, these findings provide critical means of targeting PIK3-AKT signaling to treat patients with refractory leukemia post TKI therapy.Discussion

[0102] Treatment with TKIs often leads to development of resistance that has been a major hurdle to successful cancer treatment. However, mechanisms that generate and sustain drugresistant cell populations as well as molecular predictors for prognosis and drug response are still unclear, particularly for resistant patient subpopulation without acquired TK kinase domain mutations. By dissecting the impact of a dynamic m6A methylome on TKI sensitivity, we demonstrate that the m6A-regulated IncRNAs are deregulated molecules whose overexpression in TKI resistant cell lines and in patients with inadequate response to TKI therapy is important not only for resistant cell proliferation, but also for insensitivity to TKI treatment and patient survival. We also show that the FTO-dependent m6A demethylation can regulate IncRNAs by increasing their RNA stability, and IncRNA-activated PI3K signaling renders insensitivity to TKIs and sustain resistant cell growth. Further, we observe that, among refractory or relapsed population, FTO levels are significantly higher in patients without kinase mutations compared to those with mutations; IncRNA upregulation and the activated PI3K signaling detected in patients at a diagnosis is associated with the follow-up bad drug response or unfavorable outcomes. Our findings add a new layer to the complexity of mechanisms regulating leukemia cell fate under TKI selection, and raise the possibility that the m6A- regulated IncRNAs represents a new non-genetic factor to affect the development and maintenance of TKI resistance; our discoveries identify a promising therapeutic target, FTO, for specifically, the most challenging patient subpopulations who are TKI non- responders / relapsed but do not carry the acquired BCR / ABL mutations; our results uncover a strong predictor, m6A-regulated lncRNA-PI3K axis, for poorer prognosis and failure in drug response in cancers.

[0103] Acquired resistance to TKIs is commonly attributed to genetic mechanisms. However, evidence is emerging that the development of drug resistance is accomplished without genetic alterations. Easwaran et al., Molecular cell 54, 716-727 (2014). Targeting epigenetic machinery has become a major thrust in the development of new therapeutic strategies; unfortunately, these strategies have yielded mixed results in clinical trials. One barrier toprogress in harnessing epigenetic therapies to overcome drug resistance is the limited understanding of how cancer cells can rapidly escape TKI killing. We recently demonstrated that a dynamic and reversible m6A methylome represents a non-genetic driver of a TKI- tolerance state in heterogeneous leukemia cells, yet the molecular mechanisms by which the TKI-altered m6A methylome affects drug sensitivity are incompletely understood. Because transcriptome-wide m6A sequencing disclosed that most changes of m6A peaks in resistant cells come from IncRNAs, this study was designed to test the hypothesis that m6A- associated IncRNAs could be additional key non-genetic factor to fuel and sustain resistant cell growth. As expected, we identified and validated many annotated IncRNAs, which are differentially expressed with an obvious change of m6A content in TKI resistant cell lines and nonresponding leukemia patients.

[0104] LncRNA defects have been found to affect drug resistance (Bester et al., Cell 173, 649- 664 e620 (2018)), but it is unclear whether an interaction between m6A and IncRNAs exists in leukemia resistance against TKIs. To test whether m6A-assicated IncRNAs play a role in TKI response, we initially examine their role in leukemia patients. We observed that these IncRNAs, including PROX1-AS1, SENCR or LN892, are elevated in leukemia patients vs normal donors, and in CML patients not responding to TKIs or with blast crisis disease compared with TKI responders or with chronic disease. We further observed that patients with higher IncRNA expression at least have a trend toward worse prognosis than those with lower levels. These findings support PROX1-AS1, SENCR or LN892 as prognostic biomarkers in leukemia therapy. We next knocked down the upregulated PROX1-AS1, SENCR or LN892 in resistant cells, and found that IncRNA knockdown inhibits resistant cell growth, and sensitizes nilotinib resistant cells to nilotinib killing. These findings suggest that IncRNA upregulation is required, at least partially, to support resistant cell growth and to preserve TKI insensitivity.

[0105] It is increasingly appreciated that IncRNAs are key regulators in cancer pathogenesis and drug resistance. However, they are not the direct contributors, but determine cancer cell fate through up- or down-regulating cancer cell determinant genes. Further, very few studies have been found to address the role of PROX1-AS1, SENCR or LN892 in drug resistance. By analyzing a gene expression profile resulting from SENCR knockdown, we identified a downstream mediator, PI3K signaling, which represents a central regulatory node controlling cell proliferation and apoptosis, and plays a critical role in leukemia drug resistance. Airiau et al., Cell Death Dis., 4(10):e827 (2013). The key genes (i.e., ITGA2, COL6A1, cyclin DI,PKN1, PDGFRA and F2R), which are changed the most and likely involved in the PI3K pathway, could be downstream targets of PROX1-AS1, SENCR or LN892, because knockdown of these IncRNAs impairs the aforementioned target expression. These IncRNA- regulated genes are overexpressed in TKI resistant cells and in imatinib or nilotinib nonresponding patients compared to good responders. Leukemia patients with overexpression of ITGA2, C0L6A1, cyclin DI, PKN1, PDGFRA or F2R predicts or has trends toward worse prognosis. However, it will be interesting to see in the future whether these PI3K targets are prognostic biomarkers in larger cohort of patients, which should also be justified by age, mutation status, sex, disease stages and induction pre-treatment.

[0106] The molecular basis of IncRNA dysregulation in resistant leukemia is still obscure. We focused on m6A methylation in regulating IncRNAs, because 1) a dynamic m6A methylome represents a bona fide defense mechanism in developing TKI resistance; 2) half of the differentially expressed IncRNAs have m6A motifs; 3) m6A reduction and IncRNA upregulation co-exist in resistant cells; 4) the abundance of m6A plays a key role in RNA stability; and 5) the transcriptional repression by IncRNAs is enhanced by m6A methylation. Patil et al., Nature 537, 369-373 (2016). Our studies revealed that, in parallel with global m6A hypomethylation, m6A content specific for upregulated IncRNAs, like PROX1-AS1, SENCR or LN892, is significantly decreased in resistant cells. Given that upregulation of these IncRNAs affects TKI sensitivity and patient survival, these results support a crosstalk between m6A methylome and IncRNA pathway in TKI resistance. Mechanistically, FTO serves as a new IncRNA regulator through FTO-dependent m6A hypomethylation in leukemia resistant cells. In support of this possibility was the downregulation of PROX1-AS1, SENCR or LN892 and the reduction of their m6A levels when FTO is inactivated by either gene knockdown or inhibitor treatment. This is followed by an increased RNA half-life of IncRNA transcripts. These findings provide an alternative explanation for why many IncRNAs are differentially expressed in TKI resistant cells. It would also be important to know whether all identified IncRNAs are direct targets of FTO and whether FTO dependent m6A affects RNA stability of all these targets.

[0107] While TKIs (e.g., imatinib, nilotinib) have revolutionized leukemia treatment, TKI don’t cure all leukemia patients, due to the development of TKI resistance. Further, a subpopulation of patients (30-50% in CML) who achieve complete molecular remission, must take TKIs for the rest of their lives, which can lead to severe side effects and substantialfinancial burden. Thus, the development of novel therapies to overcome resistance are major goals in the field. After demonstrating the key role of lncRNA-PI3K axis in regulating TKI resistance, we explored the therapeutic potential of PI3K inhibitor alpelisib in treating patients with resistant CML. Notably, PI3K signaling is an important downstream mediator in BCR- ABL-driven leukemia, but alpelisib has not been tested, particularly, for refractory CML patients. We showed that alpelisib sensitizes imatinib or nilotinib resistant cells to imatinib or nilotinib-induced cell death, and its combination with imatinib or nilotinib results in more pronounced cells apoptosis in vitro. Most importantly, alpelisib-therapy in NSGS mice bearing nilotinib-resistant K562 cells significantly reduces leukemia burden and increases the survival time of leukemic mice, mechanistically through suppression of PI3K signaling mediators like F2R, CyclinDl and PDGFR. These findings support alpelisib as a novel therapeutic reagent for TKI resistant leukemia, opening promising new avenues to treat patients with recurrent disease.

[0108] In summary, this study discovers a critical role that m6A-regulated IncRNAs play in sustaining resistant growth, maintaining resistant phenotypes and in promoting TKI insensitivity in vitro and in CML patients receiving TKI therapy. It reveals a novel mechanism, the FTO-m6A-lncRNAs axis, to explain why IncRNAs are elevated in TKI resistant cells, and also a pathway, IncRNA-regulated PI3K signaling, to answer the question of how IncRNA dysregulation affects TKI resistance. Given that m6A-regulated IncRNAs affect TKI response and survival of leukemia patients, it will be interesting to see in the future whether these hitherto underappreciated roles for m6A-lncRNAs-PI3K cascade in anti-cancer resistance could serve as a foundation to design therapeutics that might overcome TKI resistance for this devastating cancer.MethodsExperimental model and subject detailsMice

[0109] All animal experiments were approved by the Institutional Animal Care and Use Committees of the University of Minnesota and were in accordance with the US National Institutes of Health Guide for Care and Use of Laboratory Animals. For survival studies, micewere sacrificed when they showed any signs of distress (i.e., breathing disorders, weight loss or immobility).

[0110] The NOD / SCID / ycnull immunodeficient NOG mice (female, male; 4-6 weeks old) were purchased from Charles River and sublethally irradiated. About 0.1 x lO6nilotinib resistant K562 cells were injected into the irradiated mice through the tail vein 4 hours after irradiation. Recipient mice were monitored weekly for signs of leukemia beginning on day 7 after transplantation. Mononuclear cells collected from bone marrow (BM) of the euthanatized mice were stained with mouse anti-human CD45 antibody (Bioligned) for flow cytometry sorting. The presence of a CD45+ population was considered as human leukemia engraftment. The development of leukemic disease was verified by H&E staining. Then BM cells were isolated for further investigations.

[0111] For preclinical testing of PI3K inhibitor: Alpelisib was prepared just prior to administration by dissolving in DMSO to provide a clear solution and then diluting with 2- Hydroxypropyl-0-cyclodextrin (HPCD) and PBS (ratio 2.5:75:22.5). About 0.5 x 106BM cells isolated from the aforementioned leukemic mice will be injected into the sublethally irradiated NSGS mice (second recipient). The leukemia-bearing mice were randomized and administrated alpelisib intraperitoneally every three days at 100 mg / kg for 5 weeks. The administration of vehicle (DMSO, HPCD, PBS; ratio 2.5:75:22.5) was used as a negative control. The survival time was analyzed by Kaplan-Meier estimate, and comparison between groups was analyzed by log-rank tests. The survival time was from the start of leukemia cell injection. Cytospin preparations of BM cells were processed for Wright-Giemsa staining. The lungs, spleens, and livers were immediately fixed in 10% neutral-buffered formalin and stained with H&E.AML and CML patient samples

[0112] The current study was approved by the Institutional Review Board of Mayo Clinic and the second hospital of Shanxi Medical University and conducted in accordance with the Declaration of Helsinki. The diagnoses of acute myeloid leukemia (AML) and chronic myeloid leukemia (CML) were made according to the criteria of World Health Organization. Mononuclear cells from BM or PB samples of AML and CML patients were prepared by Ficoll-Hypaque (GE Healthcare #71-7167-00) gradient centrifugation. The patient cells were frozen in 10% DMSO plus 90% FBS and directly used for molecular biological assays withoutfurther cell culture. All patients signed an informed consent document approved by the Institutional Review Board before entering any study.Cell lines and cell culture

[0113] Leukemia cell lines, K562, MV4-11 and Kasumi-1, were newly purchased from American Type Culture Collection with no further authentication or testing for mycoplasma. Cell lines were grown in RPMI-1640 (GE Healthcare #SH30027.01) supplemented with 20% (Kasumi-1) or 10% (K562, MV4-11) fetal bovine serum (FBS, Gibco by Life Technologies™ #16140-071) and Antibiotic-Antimycotic (Gibco by Life Technologies™ #15240062) at 37°C under 5% CO2. No cell line used in this Example is listed in the database of commonly misidentified cell lines maintained by ICLAC (International Cell Line Authentication Committee).In vitro adaption of TKI resistant cells

[0114] Cell lines, K562, Kasumi-1 and MV4-11, were passaged with low concentration of imatinib or nilotinib (0.1 pM) and sequentially cultured in increasing concentrations of these TKIs (0.3, 1 LIM) for 8-10 weeks. Cells cultured in parallel in drug-free medium were used as parental / sensitive controls. Cells were considered resistant when they could routinely grow in medium containing 1 pM imatinib, or nilotinib, respectively.Method detailsPlasmid design and construction

[0115] Three shRNAs against FTO (TRCN0000183897, TRCN0000179651,TRCN0000180978) and the negative control vectors (pLKO.l) were obtained from BMGC RNAi (University of Minnesota), but the shRNA against IncRNAs PROX1-AS1 (NR_037850.2; AS7, AS1521, AS3299), SENCR (NR_038908.1; S16, S785, S1238) and LN892 (NR_038461.1; LN370, LN619, LN576) were designed using an online tool, synthesis by Integrated DNA Technologies Inc, and cloned into pLKO.l vector.Lentivirus vector, virus production and virus infection

[0116] For virus production, HEK-293 (3.8 x 106) cells were planted in a 10 cm cell culture dish for 24 hours, and transfected with 6 pg of target or scrambled control plasmids using calcium phosphate transfection reagent (CalPhos™ Mammalian Transfection Kit), following the manufacture’s instruction. The lentiviruses were harvested at 48 and 72 hours after transfection and concentrated using the protocol of the Lenti-X™ Concentrator (Clotech #631232). For virus infection, leukemia cells (1 x 106) were infected by the lentiviruses using Polybrene (final concentrate 4 pg / ml) in 1 ml medium and Puromycin (final concentration 2 pg / ml) was added to select the stable transformants 24 hours post-infection.Colony-forming assays

[0117] Colony -forming assays were performed using MethoCult® medium (Stem Cell Technologies #03434) as previously reported. Yan et al., Cell research 28, 1062-1076 (2018). Briefly, the cells were suspended in 0.3mL of IMDM medium (Stem Cell Technologies #36150), mixed with 3 ml MethoCult® medium and then dispensed into 35 mm dishes. The colony count and size were recorded after 7-10 days. Generally, 10 single clones were picked up into 96 well plate per each cell line in order to get the stable clone after virus infection and puromycin selection. m6A Dotblotting

[0118] The mRNA was extracted using GenElute™ direct mRNA Miniprep Kit (Sigma #DMN7O-1KT). The m6A RNA dotblotting was performed with a Bio-Dot Apparatus (BioRad #170-6545). About 500 ng mRNA was diluted in 50 pl RNase-free water and mixed with 150 pl RNA incubation solution (1 ml mix: 657 pl formamide, 210 pl 37% formaldehyde solution and 133 pl 10 x MOPS). Then the RNA was denatured at 65°C for 10 min and mixed with 200 pl ice-cold 20 x SSC. The RNA loading membrane was baked at 80°C for 5 min, UV crosslinked, blocked with 5% non-fat milk at room temperature for 45 min, and incubated with m6A antibody for overnight. After being washed 3 times with lx PBST, the membranes were incubated with a HRP-conjugated secondary antibody anti -rabbit in 5% non-fat dry milk. The signal was detected by enhanced chemiluminescence. RNA spotted membrane was stained with 0.02% methylene blue (Sigma #1808) in 0.5 M sodium acetate (pH 5.0) for loading control.Western Blotting

[0119] The whole cellular lysates were prepared by harvesting the cells in 1 x cell lysis buffer [20 mM HEPES (pH 7.0), 150 mM NaCl and 0.1% NP40] supplemented with 1 mM phenylmethane sulfonyl fluoride (PMSF, Sigma #10837091001), 1 x Phosphatase Inhibitor Cocktail 2 and 3 (Sigma #P5726, P0044), and 1 x protease inhibitors (protease inhibitor cocktail set III, Calbiochem-Novabiochem #539134). The proteins were resolved by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, transferred onto PVDF membranes (GE Healthcare #10600023), blocked by 5% non-fat milk followed by probing with first and HRP-conjugated secondary antibodies.RNA isolation, cDNA preparation and quantitative PCR (qPCR)

[0120] According to manufacturers’ instructions, the total RNA was isolated using miRNeasy Kit (Qiaqen #217004), and complementary DNA (cDNA) synthesis was performed using SuperScript® III First-Strand Synthesis System (Invitrogen #18080-051). The expression of target genes was assessed by SYBR Green qPCR (Applied Biosystems #4309155). The expression of the targets was analyzed using the ACT approach. The levels of GAPDH or 18s were used as normalization in cell lines, but ABL was used as internal control in patient sample according to BCR / ABL IS. m6A immunoprecipitation (IP)

[0121] The RNAs were diluted in 200 pl IPP buffer (150 mM NaCl, 0.1% NP-40, 10 mM Tris- HC1, pH 7.4) and fragmented into 100-nucleotide-long fragments using sonication. The fragmented RNAs were incubated for 12 hours at 4 °C with 5 pl anti-m6A in IPP buffer. The mixture was then immunoprecipitated by incubation with Dynabeads™ Protein G (ThermoFisher #10004D) at 4 °C for additional 3 hours. After extensive washing by IPP buffer, 75 pl 42°C pre-heated Elution Buffer (0.02 M DTT, 0.15 M NaCl, 0.05 M Tris-HCl, pH 7.4, 0.001 M EDTA, 0.1% SDS) were added into m6A-positive RNA solution for 5 min at 42°C, and this step was repeated 2 times. Finally, the enriched m6A RNA was eluted from the beads into 225 pl solution and precipitated by adding 2.5 times volume of 100% ethanol.Cell proliferation and apoptosis assays

[0122] Cell proliferation assays were performed using Cell Counting Kit-8 (CCK-8, DojindoMolecular Technologies #CK04-ll) following manufactures’ instruction. Briefly, the parentaland resistant cells with various treatment (1.5 x 104) in RPMI-1640 medium (100 l) were dispensed into 96-well flat-bottomed microplates and incubated for 24 hours. The cells were cultured for another 24 or 48 hours, and CCK-8 reagent (10 pl) was added to each well. The microplates were incubated at 37°C for additional 2~4 hours. Absorbance was read at 450 nm using a microplate reader and the results were expressed as a ratio of the treated over untreated cells (as 100%). Five wells were sampled per experimental group in a given experiment. Averages are reported +SD. Cell apoptosis assays were performed using Annexin V-PI Apoptosis Detection Kit I (BD Pharmingen™ #556547) according to the manufacturer’s instruction, and followed by flow cytometry analysis.Identification of differentially expressed annotated IncRNAs

[0123] For all IP and IN samples, the RNA-seq raw reads were aligned to the UCSC human reference genome (GRCh38 / hg38) by Hisat2 (Kim et al., Nature methods 12, 357-360 (2015)) with parameter -U and then sorted and indexed by SAMtools (Li et al., Bioinformatics 25, 2078-2079 (2009)), The mapped reads were counted using featureCounts vl.6.1 (Liao et al., Bioinformatics 30, 923-930 (2014)) and a differential expression analysis comparing resistance and parental samples was performed using R packages, edgeR (Robinson et al., Bioinformatics 26, 139-140 (2010)) and limma. (Ritchie et al., Nucleic acids research 43, e47 (2015). The differentially expressed IncRNAs (annotated by GENCODE vl9) with an absolute value of log2FC >0 and FDR < 0.05 were considered for future analysis. Top 40 most differential expressed LncRNAs having m6A sites in nilotinib samples were selected from the upregulated IncRNAs and a heatmap of these IncRNAs ranked by log2FC in a decreasing order was plotted by R package pheatmap.Peak reads count analysis for m6A sequencing

[0124] Peaks were called for each of the three groups by using MeTPeak R package as described previously and were further filtered by choosing only those assigned to annotated IncRNAs. For IncRNAs that have more than one peaks, the mean of fold enrichment of all its peaks was used as its fold enrichment. To compare m6A enrichment of IncRNAs between resistant and parental samples, the IncRNAs occurred only in one sample were assigned 0 fold enrichment in the other sample, and then log2 (resistant fold enrichment) - lo 2 (parental fold enrichment) was calculated to obtain log2 Fold Change (if one of these two fold enrichmentvalues is 0, add 1 to both before log? transformation), finally, R package ggplot2 was used to plot a histogram for the log2 Fold Change calculated above.Hematoxylin & eosin (H&E) staining

[0125] Mouse tissues (lung, liver, spleen) were fixed in 10% neutral-buffered formalin, deparaffinized, hydrated, and stained with H&E (Thermo-Scientific) staining was performed at the image center, the MetroHealth System, Case Western Reserve University.Cytospin / Wright-Giemsa staining

[0126] The mouse BM cells (0.1 x 106) were isolated and placed in the Shandon EZ Single Cytofunnel (Thermo Electron Corporation). Samples were centrifuged at 1,000 rpm for 8 min. The slides were air-dried and stained with Hema-3 Kit (22-122-911, Fisher Scientific, Hampton, NH). Stained slides were viewed and photographed using a Leica microscope mounted with a high-resolution spot camera with Image-Pro Plus software. Morphologic differentiation was determined by calculating the percentage of post-mitotic cells containing metamyelocytes, bands and segmented neutrophils within six visual fields per slide (original magnification x200).Analysis of gene expression omnibus (GEO) data and functional pathways

[0127] LncRNA and gene expression profile data were downloaded and analyzed for the expression of IncRNAs. These samples were normalized, managed and analyzed by GraphPad Prism 5 Software. Further, pathway and gene-enrichment analyses that are associated with SENCR expression were conducted using DAVID 6.8 (KEGG_pathway) software. The significance of the association between target genes and canonical pathway was computed using two parameters: 1) a ratio of the number of target genes from the dataset that map to the pathway divided by the total number of genes that constitute the canonical pathway; and 2) a - logic (p-value) determining the probability that the association between the DEGs in the dataset and the canonical pathway is due to chance alone. In the present analysis, the computed -logw (p-value) of 3.0 (p < 0.05) and above was considered as statistically significant.Analysis of MLL database

[0128] The cohort comprised a total of 985 patients: 64 healthy individuals, 774 AML patients, and 121 CML patients at initial diagnosis. Furthermore, sequential samples of 12 CML patients with progression from chronic to blast phase CML were analyzed. Bone marrow or peripheral blood (PB) samples from these patients had been sent to MLL Leukemia Laboratory between 2006 and 2023 for diagnostic work-up. The respective diagnosis was established based on cytomorphology, immunophenotyping, cytogenetics, and molecular genetics following WHO guidelines. AML patients were grouped according to European Leukemia Net (ELN) 2022 risk group.— All patients gave their written informed consent for scientific evaluations. The study was approved by the Internal Review Board and adhered to the tenets of the Declaration of Helsinki. RNA was extracted using the MagNA Pure 96 Cellular RNA LV Kit (Roche LifeScience, Mannheim, Germany). For transcriptome analysis the TruSeq Total Stranded RNA kit was used, starting with 250 ng of total RNA, to generate RNA libraries following the manufacturer’ s recommendations (Illumina, San Diego, CA, USA). 2x100 bp paired-end reads were sequenced on the NovaSeq 6000 (Illumina, San Diego, CA, USA) with a median of 50 million reads per sample. Using BaseSpace’s RNA-seq Alignment app (v2.0.1) with default parameters reads were mapped with STAR aligner (v2.5.0a) to the human reference genome hgl9 (RefSeq annotation). Estimated read counts for each gene were normalized applying Trimmed mean of M- values (TMM) normalization method and the resulting log2 counts per million (CPMs) were used.Statistical analysis

[0129] The statistical analysis was performed using the Student’s t test. All analyses were performed using the GraphPad Prism 5 Software, p < 0.05 was considered statistically significant. All p values were two-tailed. No blinding or randomization was used. No samples or animals were excluded from analysis. All criteria were pre-established. No statistical method was used to predetermine sample size and the sample size for all experiments was not chosen with consideration of adequate power to detect a pre-specified effect size. Variations were compatible between groups. In vitro experiments, such as qPCR, Western blotting, cell proliferation assays, dotblotting etc. were routinely repeated three times unless indicated otherwise in Figure legends or main text. For every Figure, the statistical tests were justified as appropriate.

[0130] The complete disclosure of all patents, patent applications, and publications, and electronically available material cited herein are incorporated by reference. The foregoing detailed description and examples have been given for clarity of understanding only. No unnecessary limitations are to be understood there from. The invention is not limited to the exact details shown and described, for variations obvious to one skilled in the art will be included within the invention defined by the claims.

Claims

CLAIMSWhat is claimed is:

1. A method of obtaining a prognosis for a subject who has or has had leukemia who has been or is being treated with a tyrosine kinase inhibitor (TKI), comprising: determining the levels of one or more of 1) total N6-methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, 3) IncRNA- specific m6A, 4) the fat mass and obesity-associated gene (FTO) in a biological sample from the subject; and assigning a poor prognosis or poor TKI responder status to the subject if the level of m6A, IncRNA being m6A sites, IncRNA-specific m6A, and / or FTO is higher than the one or more corresponding control values obtained from a healthy subject.

2. The method of claim 1, wherein the leukemia is acute myeloid leukemia (AML) or chronic myeloid leukemia (CML).

3. The method of claim 2, wherein the subjects comprise AML and / or CML patients with acquired mutations in a tyrosine kinase domain.

4. The method of claim 2, wherein the AML and / or CML patients have achieved complete remission but are still being treated with a tyrosine kinase inhibitor.

5. The method of claim 1, wherein the prognosis includes determining the stage of the leukemia.

6. The method of claim 1, wherein the IncRNA is selected from the group consisting of PR0X1-AS 1, SENCR, and LN892.

7. A method of predicting tyrosine kinase inhibitor (TKI) resistance in a subject having leukemia, or having an increased risk of developing leukemia, comprising determining the levels of one or more of 1) total N6-methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, 3) IncRNA-specific m6A, 4) the fat mass and obesity-associated gene (FTO) in a biological sample from the subject; and characterizing the subject as more likely to exhibit resistance to a TKI if the level ofm6A, IncRNA being m6A sites, IncRNA-specific m6A, and / or FTO is higher than the one or more corresponding control values obtained from a healthy subject.

8. The method of claim 7, wherein the leukemia is acute myeloid leukemia (AML) or chronic myeloid leukemia (CML).

9. The method of claim 7, wherein the IncRNA is selected from the group consisting of PROX1-AS1, SENCR, and LN892.

10. The method of claim 7, wherein the subject is being treated with a TKI.

11. The method of claim 7, wherein the TKI is imatinib or nilotinib.

12. A method of treating a subject having TKLresistant leukemia, comprising administering to the subject a therapeutically effective amount of a TKI inhibitor and an effective amount of a compound that decreases the level of IncRNA bearing m6A sites and / or IncRNA-specific m6A in the subject.

13. The method of claim 12, wherein the leukemia is acute myeloid leukemia (AML) or chronic myeloid leukemia (CML).

14. The method of claim 12, wherein the compound is a PI3K inhibitor.

15. The method of claim 14, wherein the PI3K inhibitor is alpelisib.

16. A method of identifying a drug for treating suppressing TKI resistance in leukemia treatment, comprising: administering an effective amount of a test compound to a subject having TKI- resistant leukemia; evaluating whether the test compound decreases the level of 1) total N6- methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, and / or 3) IncRNA-specific m6A, in the subject; and determining that the test compound is a candidate for treatment of TKI-resistantleukemia if the compound decreases the level of 1) total N6-methyladenosine (m6A), 2) long non-coding RNA (IncRNA) bearing m6A sites, and / or 3) IncRNA-specific m6A in the subject.

17. The method of claim 16, wherein the subject is an animal model for leukemia.

18. The method of claim 16, wherein the IncRNA is selected from the group consisting of PR0X1-AS1, SENCR, and LN892.

Images