Prognostic and treatment response predictive method

Abstract

Methods of predicting whether a subject with chronic myelogenous leukaemia (CML) is at risk of relapse following cessation of a treatment for CML, whether a subject with CML will achieve TFR following cessation of a treatment for CML, and selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML are disclosed herein. The methods comprise determining the expression of one or more genes regulated by PRC1. Also provided are kits and antigen-binding molecules suitable for use in the disclosed methods.

Description

[0001] PROGNOSTIC AND TREATMENT RESPONSE PREDICTIVE METHOD

[0002] This application claims priority from SG 10202404017S filed 20 December 2024, the contents and elements of which are herein incorporated by reference for all purposes.

[0003] Technical Field

[0004] The present disclosure relates to the field of cancer therapy, more specifically, methods of diagnosis and prognosis and methods of medical treatment and prophylaxis.

[0005] Background

[0006] The achievement of treatment-free remission (TFR) is a new and important goal in the management of patients with chronic myeloid leukaemia (CML). TFRs occur when patients with stable (> 2 years) deep molecular responses (DMR) are able to stop their treatment (e.g. BCR-ABL1 -targeted therapies (BTT)) without suffering a molecular relapse. TFRs are highly desirable for both patients and healthcare systems due to improvements in quality of life, avoidance of long-term BTT toxicities, and significantly decreased drug costs, but are only achievable in a minority of patients. Tyrosine kinase inhibitor (TKI) cessation trials have shown that only ~50% of CML patients can maintain stable remission after TKI cessation (Figure 1).

[0007] The major obstacle to improving TFR rates is an almost complete lack of understanding surrounding the biologic factors which determine relapse following BTT cessation. In addition, it remains unclear if these biologic factors are already present at diagnosis (deterministic) or emerge during the course of therapy (acquired). These knowledge gaps preclude the rational development of therapeutic strategies to enhance TFR rates, as well as the identification of biomarkers predicting relapse following BTT cessation. Furthermore, while CML leukaemic stem cells (LSC) are the likely reservoir for relapse, studies of LSCs annotated for TFR are hampered by technical limitations due to their rarity, sample availability, and the lack of appropriate technologies to interrogate small numbers of cells. Multiple aspects of the bone marrow microenvironment, immune system, as well as intrinsic features of the LSC contribute to their ability to persist (Figure 2) (Ng et al. 2022 and Patterson et al. 2023).

[0008] Summary

[0009] In a first aspect, the invention relates to a method for predicting whether a subject with chronic myelogenous leukaemia (CML) is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by protein regulator of cytokinesis 1 (PRC1) in CD34+ CD38- cells in a sample obtained from the subject, wherein an increased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML.

[0010] In a second aspect, the invention relates to a method for predicting whether a subject with chronic myelogenous leukaemia (CML) is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by protein regulator of cytokinesis 1 (PRC1) in CD34+ CD38- cells in a sample obtained from the subject, wherein a decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML.

[0011] In a third aspect, the invention relates to a method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in CD34+ CD38- cells in a sample obtained from the subject, wherein a decreased expression of the one or more genes regulated by PRC1 relative to a predetermined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

[0012] In a fourth aspect, the invention relates to a method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in CD34+ CD38- cells in a sample obtained from the subject, wherein an increased expression of the one or more genes regulated by PRC1 relative to a predetermined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

[0013] In some embodiments, the method further comprises: determining the level of immune response activation in CD34+CD38- cells in a sample obtained from the subject.

[0014] In a fifth aspect, the invention relates to a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0015] (a) determining the expression of one or more genes regulated by PRC1 in CD34+ CD38- cells in a sample obtained from the subject; and

[0016] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0017] In a sixth aspect, the invention relates to a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0018] (a) determining the expression of one or more genes regulated by PRC1 in CD34+ CD38- cells in a sample obtained from the subject; and

[0019] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor. In a seventh aspect, the invention relates to a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0020] (a) determining the expression of one or more genes regulated by PRC1 in CD34+ CD38- cells in a sample obtained from the subject; and

[0021] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0022] In an eighth aspect, the invention relates to a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0023] (a) determining the expression of one or more genes regulated by PRC1 in CD34+ CD38- cells in a sample obtained from the subject; and

[0024] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0025] In some embodiments, the one or more genes is selected from the list consisting of: PRSS21 , ANLN, SCN2A, SPAG6, BMI1 , TFRC, and ALDH1 A1 .

[0026] In some embodiments, the method further comprises: determining the level of immune response activation in CD34+CD38- cells in a sample obtained from the subject; and selecting the subject for treatment if the level of immune response activation is increased relative to a pre-determined average level of immune response activation in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment further comprises an immunomodulator.

[0027] In some embodiments, the one or more genes is selected from the list consisting of PRSS21 , ANLN, SPAG6, BMI1 , and TFRC. In some embodiments, the one or more genes is selected from the list consisting of PRSS21 , SPAG6 and TFRC. In some embodiments, the one or more genes comprise PRSS21 and SPAG6. In some embodiments, the one or more genes consist of PRSS21 and SPAG6.

[0028] In some embodiments, the one or more genes is selected from ALDH1 A1 and SCN2A.

[0029] In some embodiments, the CD34+ CD38- cells are leukaemic stem cells (LSCs). In some embodiments, the treatment for CML is a TKI, a BTT, a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0030] In some embodiments, the treatment for CML is a TKI or a BTT.

[0031] In some embodiments, the sample has been obtained from peripheral blood or bone marrow.

[0032] In some embodiments, the subject is a human.

[0033] In some embodiments, the expression of the one of more genes and / or the level of immune response activation is determined using flow cytometry.

[0034] In some embodiments, the expression of the one of more genes and / or the level of immune response activation is determined using immunohistochemistry (IHC).

[0035] In some embodiments, the expression of the one or more genes is determined using one or more antigen-binding molecules.

[0036] In a ninth aspect, the invention relates to a kit suitable for performing a method disclosed herein.

[0037] In some embodiments, the kit comprises a plurality of antigen-binding molecules capable of detecting one or more genes regulated by PRC1 .

[0038] In some embodiments, the plurality of antigen-binding molecules is capable of detecting one or more of PRSS21 , ANLN, SCN2A, SPAG6, BMI1 , TFRC, and ALDH1 A1 .

[0039] In some embodiments, the plurality of antigen-binding molecules is capable of detecting one or more of PRSS21 , ANLN, SPAG6, BMI1 , and TFRC. In some embodiments, the plurality of antigen-binding molecules is capable of detecting PRSS21 , SPAG6, and / or TFRC. In some embodiments, the plurality of antigen-binding molecules is capable of detecting PRSS21 and / or SPAG6. In some embodiments, the plurality of antigen-binding molecules is capable of detecting PRSS21 and SPAG6.

[0040] In some embodiments, the kit further comprises an antigen-binding molecule capable of detecting a correlate of immune response activation. In some embodiments, the antigen-binding molecules capable of detecting a correlate of immune response activation is capable of detecting TNFa or phosphorylated NFKB.

[0041] In some embodiments, the plurality of antigen-binding molecules is capable of detecting one or more of ALDH1A1 and SCN2A.

[0042] In a tenth aspect, the invention relates to antigen-binding molecule for use in a method disclosed herein. In some embodiments, the antigen-binding molecule is capable of detecting one or more of PRSS21 , ANLN, SCN2A, SPAG6, BMI1 , TFRC, ALDH1A1 , TNFa and phosphorylated NFKB.

[0043] Description

[0044] The present invention is based on the unexpected discovery that biomarkers can be used to predict whether patients with chronic myelogenous leukaemia (CML) will experience treatment-free remission (TFR) or relapse following cessation of a treatment for CML (such as treatment with one or more tyrosine kinase inhibitors (TKIs)). In addition, the present invention is based on the unexpected discovery that leukaemic stem cell (LSC) biology is deterministic for treatment outcomes in CML patients. The inventors have unexpectedly discovered that biology intrinsic to LSCs enables long-term persistence during deep molecular responses (DMR) and that persistent LSCs drive relapse (Figure 3).

[0045] The present invention allows for 1) the identification of patients who are more likely to relapse following treatment discontinuation / cessation, which may lead to a recommendation to not attempt treatment discontinuation / cessation and / or the administration of a therapy to decrease the likelihood of relapse, and 2) the identification of patients who are likely to achieve long-term TFR without further intervention, which may lead to a recommendation to attempt treatment discontinuation / cessation. The identification of patients who are more likely to relapse or achieve long-term TFR can be accurately performed at the time of diagnosis, informing treatment options.

[0046] The ability to segregate CML patients at diagnosis into those likely to achieve TFR versus those who are likely to relapse may be used to 1) increase patient compliance (i.e., to encourage patients to adhere to their medication since they are likely to achieve several decades of TFR), 2) identify patients for inclusion in TFR clinical trials, allowing for more cost-effective clinical trials that are carried out on patients who may benefit from the intervention, and / or 3) avoid futile attempts at treatment discontinuation / cessation, a clinical scenario that provokes significant levels of psychological stress, including depression, in patients.

[0047] Chronic Myelogenous Leukaemia (CML)

[0048] The present disclosure relates to methods for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, methods for predicting whether a subject with CML will achieve TFR, and methods of selecting subjects with CML for treatment.

[0049] Chronic myelogenous leukaemia (CML) is also known as chronic myeloid leukaemia, chronic granulocytic leukaemia, or Philadelphia (Ph*) / BCR-ABL1 -positive chronic myeloid leukaemia. CML is characterised by the chromosomal translocation t(9;22) (q34.1 ;q11 .2), resulting in the BCR-ABL1 fusion gene and formation of the Philadelphia chromosome (Ph*). CML is also characterised by the unregulated proliferation of blast cells in the bone marrow and the accumulation of blast cells in the blood.

[0050] The BCR-ABL1 transcript is translated into a tyrosine kinase containing domains from both the BCR and ABL1 genes. The BCR-ABL1 fusion gene codes for BCR-ABL1 protein (a tyrosine kinase) that is constitutively activated, leading to unregulated cell division. As used herein, “BCR-ABL” or“BCR-ABL1” may refer to the BCR-ABL1 fusion gene or protein. Typically, CML begins in a chronic phase, and then progresses to an accelerated phase, followed by a blast crisis phase.

[0051] In the chronic phase, CML is most stable and develops slowly. Patients with chronic phase CML have predominantly fully functioning blood cells with a low level of blast cells (less than 10% of the cells in the blood and bone marrow). Patients with chronic phase CML may also have a lower level of red blood cells (anaemia) and a higher level of platelets than is normal, but patients may have few or no symptoms.

[0052] As used herein, a “blast cell” refers to an early hematopoietic cell from the lymphoid (lymphocytes) or myeloid (erythrocytes, thrombocytes, monocytes, neutrophils, basophils, eosinophils) cell lines. Blast cells are produced by stem cells in the bone marrow. A blast cell may be a lymphoid blast cell (e.g. lymphoblast) or a myeloid blast cell (e.g., myeloblast). In healthy individuals, the bone marrow contains no more than 5% blast cells, and the peripheral blood contains no blast cells.

[0053] Accordingly, as used herein, “chronic phase CML” refers to CML where less than 10% of the cells in the blood and bone marrow are blast cells. In some embodiments, the chronic phase CML is asymptomatic or associated with mild symptoms. A chronic phase CML patient may be identified according to one or more of these criteria. For example, a chronic phase CML patient may be a patient where less than 10% of the cells in the blood and bone marrow are blast cells, or where the CML is asymptomatic or associated with mild symptoms.

[0054] According to the MD Anderson Cancer Center (MDACC) criteria, accelerated phase may be defined by the following criteria: peripheral blood myeloblasts of >15% and <30%; peripheral blood myeloblasts and promyelocytes combined of >30%; peripheral blood basophils of >20%; a platelet count of <100 x 109 / L unrelated to therapy; or additional clonal cytogenetic abnormalities in Ph+ cells. Accelerated phase may be accompanied by symptoms such as weight loss, fatigue, and an enlarged spleen (also known as splenomegaly). In addition, treatment may be less effective during the accelerated phase, as characterised by persistent or increasing abnormal blood cell counts despite treatment (e.g., leukocytosis (>10x109 / L), thrombocytosis (>1000x109 / L), or thrombocytopenia (<100x109 / L) unrelated to therapy, 20% or more basophils, or 10%-19% blast cells). The accelerated phase may also be accompanied by additional chromosomal abnormalities (in addition to the Philadelphia chromosome) such as second Ph* chromosome, trisomy 8, isochromosome 17q, trisomy 19, complex karyotype, or 3q26.2 abnormalities.

[0055] According to the International Bone Marrow Transplant Registry (IBMTR) criteria, blast crisis phase (also known as the blastic phase, the acute phase, the blast phase, blast crisis, or blast transformation), may be defined as having 30% or more blast cells within the cells in the blood and / or bone marrow, or by the presence of extramedullary infiltrates of leukemic cells. The blast crisis phase is associated with severe symptoms (such as increasing or persistent splenomegaly), increased resistance to treatment, and a poor prognosis. The blast crisis phase may be accompanied by additional chromosomal abnormalities (in addition to the Philadelphia chromosome) such as second Ph* chromosome, trisomy 8, isochromosome 17q , trisomy 19, complex karyotype, or 3q26.2 abnormalities. The blast crisis phase can occur as a myeloid blast crisis (where the blast cells are myeloblasts) or a lymphoid blast crisis (where the blast cells are lymphoblasts). In some cases, the blast cells in the blast crisis phase are a mixture of myeloblasts and lymphoblasts.

[0056] Accordingly, as used herein, “blast crisis phase CML” (also known as “blastic phase CML”, “acute phase CML, “blast phase CML” or “blast transformation CML”) refers to CML where 30% or more of the cells in the blood and / or bone marrow are blast cells, or where extramedullary infiltrates of leukemic cells are present. In some embodiments, blast crisis phase CML is as defined by appropriate diagnostic criteria (e.g. by the International Bone Marrow Transplant Registry (IBMTR) criteria (see for example, Gambacorti-Passerini & le Coutre, Chronic Myelogenous Leukemia in De Vita, Hellman and Rosenberg’s Cancer: Principles & Practice of Oncology (12thEdition) 2022:1773-1784, which is hereby incorporated by reference in its entirety). In some embodiments, extramedullary infiltrates of leukemic cells are present in the lymph node, soft tissue and / or CNS. In some embodiments, the blast crisis phase CML is myeloid blast crisis phase CML. In some embodiments, the blast crisis phase CML is lymphoid blast crisis phase CML. In some embodiments, the blast crisis phase CML is characterised by a mixture of myeloblasts and lymphoblasts.

[0057] A blast phase CML patient (or a chronic phase CML patient that has progressed to the blast crisis phase) may be a patient where 30% or more of the cells in the blood and / or bone marrow are blast cells, or a patient where extramedullary infiltrates of leukemic cells are present. In some embodiments, a blast phase CML patient (or a chronic phase CML patient that has progressed to the blast crisis phase) may be a patient according to an appropriate diagnostic criteria of blast phase CML (e.g. by the International Bone Marrow Transplant Registry (IBMTR) criteria (see for example, Gambacorti-Passerini & le Coutre, Chronic Myelogenous Leukemia in De Vita, Hellman and Rosenberg’s Cancer: Principles & Practice of Oncology (12thEdition) 2022:1773-1784, which is hereby incorporated by reference in its entirety). A blast crisis phase CML patient may have CML that is characterised by a myeloid blast crisis. A blast crisis phase CML patient may have CML characterised by a lymphoid blast crisis. A blast crisis phase CML patient (or a chronic phase CML patient that has progressed to the blast crisis phase) may have increasing or persistent splenomegaly, increased resistance to treatment (such as TKI treatment), and / or a poor prognosis. A blast crisis phase CML patient (or a chronic phase CML patient that has progressed to the blast crisis phase) may be unresponsive to treatment (e.g. TKI treatment, STAMP inhibitor treatment). A blast crisis phase CML patient (or a chronic phase CML patient that has progressed to the blast crisis phase) may have additional chromosomal abnormalities (in addition to the Philadelphia chromosome) such as second Ph* chromosome, trisomy 8, isochromosome 17q, trisomy 19, complex karyotype, or 3q26.2 abnormalities.

[0058] In some embodiments, the subject with CML is a subject with chronic phase CML. In some embodiments, the subject with CML is a subject with accelerated phase CML. In some embodiments, the subject with CML is a subject with blast phase CML. Treatments for CML

[0059] The present disclosure relates to methods for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, methods for predicting whether a subject with CML will achieve TFR, and methods of selecting subjects with CML for treatment.

[0060] A “treatment for CML” or a “therapy for CML” may be any treatment for CML, e.g. a treatment for CML as described herein. In some embodiments, the treatment for CML is a BCR-ABL1 -targeted therapy (BTT). As used herein a “BCR-ABL1 -targeted therapy” (BTT) may refer to any treatment for CML disclosed herein that targets (or is effective against) BCR-ABL1 or cells expressing BCR-ABL1 (e.g. imatinib, dasatinib, nilotinib, ponatinib, bosutinib, asciminib).

[0061] Treatments for CML include TKIs (e.g. imatinib, dasatinib, nilotinib, bosutinib, ponatinib, bafetinib, radotinib, olverembatinib, flumatinib, asciminib, and TERN-701), chemotherapeutic agents (e.g. hydroxycarbamide, fludarabine, cytarabine, idarubicin, busulfan, cyclophosphamide, vincristine, omacetaxine, dexamethasone), interferon therapy (e.g. interferon-alpha), radiation therapy, stem cell therapy (e.g. stem cell transplant (e.g. autologous stem cell transplant, allogeneic stem cell transplant)), immunotherapy, or combinations thereof. Treatments for CML are reviewed in Helhmann, J Clin Med. (2020), 9(11):3671 , which is hereby incorporated by reference in its entirety.

[0062] T reatment of CML may comprise the use of TKIs. TKIs are small molecules that disrupt tyrosine kinases through different modes of action. TKIs may compete with the adenosine triphosphate-binding site of the catalytic domain of a tyrosine kinase, which inhibits autophosphorylation and activation of the tyrosine kinase, and prevents activation of downstream intracellular signalling pathways. Alternatively, TKIs may act as allosteric inhibitors in which the TKI binds to a site other than the active site of the tyrosine kinase, causing a conformational change of the enzyme. TKIs used in the treatment of CML may target BCR- ABL.

[0063] TKIs (e.g. TKIs targeting BCR-ABL) are often the first-line therapy for patients with CML. Imatinib is a first generation BCR-ABL TKI. Second generation BCR-ABL TKIs (which are more potent than imatinib) include nilotinib, dasatinib, bosutinib, and bafetinib. Ponatinib is an example of a third generation BCR- ABL TKI, and is more potent than first and second generation BCR-ABL TKIs. Patients with CML may be treated with a first-line TKI, and then treated with a second-line TKI (for example, if the patient does not respond to treatment with the first-line TKI or becomes resistant to the first-line TKI). Patients with CML may then treated with a third-line TKI (for example, if the patient does not respond to treatment with the second-line TKI or becomes resistant to the second-line TKI). Known TKIs imatinib, dasatinib, nilotinib, bosutinib, ponatinib, bafetinib, radotinib, olverembatinib, flumatinib, asciminib, and TERN-701.

[0064] As used herein, a “first-generation TKI” may include imatinib. A “second-generation TKI” may include dasatinib, nilotinib, bosutinib, bafetinib, radotinib, or flumatinib. A “third-generation TKI” may include ponatinib or olverembatinib. As used herein, a “first-line TKI”, a “second-line TKI”, or a “third line TKI” may refer to a “first-generation TKI”, a “second-generation TKI”, and a “third-generation TKI”, respectively. Treatment of CML may comprise the use of STAMP inhibitors. STAMP inhibitors (Specifically Targeting the ABL Myristoyl Pocket inhibitors) are a type of tyrosine kinase inhibitor. STAMP inhibitors specifically target the ABL myristoyl pocket, and therefore block BCR-ABL1 kinase activity via allosteric binding. Known STAMP inhibitors include asciminib (DrugBank Accession No. DB12597) and TERN-701 (also known as HS-10382, and as described in e.g. Parsons et al., Blood (2023) 142 (Supplement 1): 5757, which is hereby incorporated by reference in its entirety).

[0065] Treatment of CML may comprise the use of chemotherapeutic agents. Chemotherapeutic agents useful in the treatment of CML include hydroxycarbamide, fludarabine, cytarabine, idarubicin, busulfan, cyclophosphamide, vincristine, omacetaxine, dexamethasone. Chemotherapeutic agents may be used alone in combination with other CML treatments, e.g. TKIs.

[0066] Treatment of CML may comprise the use of stem cell therapy or a stem cell transplant. Stem cell transplantation may be used, e.g. for treatment of accelerated or blastic phase CML, or when a patient does not respond to a first line treatment e.g. TKI. The role of stem cell transplantation in the treatment of CML is reviewed e.g. in Barrett & Ito, Blood (2015) 125(21):3230-3235, which is hereby incorporated by reference in its entirety.

[0067] Biomarkers

[0068] The methods disclosed herein comprise the use of biomarkers for predicting or determining whether a subject with CML is at risk of relapse and / or is likely to achieve TFR following cessation of a treatment for CML.

[0069] In some embodiments, the methods disclosed herein comprise determining the expression of one or more genes regulated by polycomb repressive complex 1 (PRC1). PRC1 is a multiprotein complex that can assemble into several distinct variants that retain distinct functional properties. The PRC1 core comprises a heterodimer that includes the E3 ubiquitin ligases RING1A or RING1 B, polycomb complex protein BMI1 (hereafter referred to as “BMI1”), and chromobox (CBX) proteins. The canonical PRC1 complex has been implicated in the maintenance of the stem cell compartment through silencing tumor suppressor genes and regulating stem cells. A non-canonical form of PRC1 (ncPRC1.1) has recently been identified that activates gene expression and regulates genes associated with glycolysis. ncPRC1.1 has been found to be essential for AML leukemogenesis (Maat et al. (2021) and van den Boom et al. (2016)).

[0070] The present inventors have identified a new non-canonical form of PRC1 , which is expressed in primary BCR-ABL1+LSCs and includes BMI1 (also known as polycomb group RING finger protein 4 or RING finger protein 51). This non-canonical form of PRC1 activates gene expression (such as expression of PRSS21 , ANLN, SPAG6, BMI1 , and TFRC). Accordingly, in some embodiments, the methods disclosed herein comprise determining the expression of one or more genes regulated by a non-canonical PRC1 . In some embodiments, the methods disclosed herein comprise determining the expression of one or more genes regulated by a BMI1-containing protein complex. In some embodiments of the methods disclosed herein, the PRC1 , non-canonical PRC1 , or BMI1 -containing complex is sensitive to the BMI1 inhibitor PTC209.

[0071] The present disclosure provides a method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in a sample obtained from the subject. The present disclosure also provides a method for selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising determining the expression of one or more genes regulated by PRC1 in a sample obtained from the subject. The expression of the one or more genes may be determined in CD34+CD38- cells in the sample obtained from the subject. The expression of the one or more genes may be determined relative to a pre-determined average expression obtained from reference samples as described herein. The expression of the one or more genes may be determined relative to the average expression in samples obtained from a reference population as described herein.

[0072] The present disclosure provides a method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject, wherein an increased or decreased expression of the one or more genes regulated by PRC1 relative to a predetermined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the PRC1 is a non-canonical PRC1.

[0073] The present disclosure also provides a method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject, wherein an increased or decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML. In some embodiments, the PRC1 is a non-canonical PRC1 .

[0074] The present disclosure also provides a method for method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by a BMI1 -containing complex in CD34+CD38- cells in a sample obtained from the subject, wherein an increased or decreased expression of the one or more genes regulated by the BMI-containing complex relative to a pre-determined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the BMI1-containing complex is a non-canonical PRC1 .

[0075] The present disclosure also provides a method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by a BMI1-containing complex in CD34+CD38- cells in a sample obtained from the subject, wherein an increased or decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML. In some embodiments, the BMI1-containing complex is a non-canonical PRC1 .

[0076] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising (a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased or decreased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor. In some embodiments, the PRC1 is a non-canonical PRC1 .

[0077] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising (a) determining the expression of one or more genes regulated by PRC1 in CD34+ CD38- cells in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased or decreased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof. In some embodiments, the PRC1 is a non-canonical PRC1 .

[0078] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising (a) determining the expression of one or more genes regulated by a BMI1 -containing complex in CD34+CD38- cells in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes regulated by the BMI1-containing complex is increased or decreased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor. In some embodiments, the BMI1-containing complex is a non-canonical PRC1 .

[0079] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising (a) determining the expression of one or more genes regulated by a BMI1-containing complex in CD34+ CD38- cells in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes regulated by the BMI1 -containing complex is increased or decreased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof. In some embodiments, the BMI1 -containing complex is a non-canonical PRC1 .

[0080] In some embodiments, one or more genes regulated by PRC1 may be selected from PRSS21 , SPAG6, ANLN, SCN2A, BMI1 , TFRC and ALDH1 A1 . In some embodiments, one or more genes regulated by PRC1 may be selected from PRSS21 , SPAG6 and TFRC. In some embodiments, one or more genes regulated by PRC1 may be selected from PRSS21 and SPAG6.

[0081] In some embodiments, the methods disclosed herein comprise determining the expression of one or more genes regulated by BMI1 .

[0082] In some embodiments, the methods disclosed herein comprise determining the expression of one or more of PRSS21 , SPAG6, ANLN, SCN2A, BMI1 , TFRC and / or ALDH1 A1 . In some embodiments, the methods disclosed herein comprise determining the expression of PRSS21 , BMI1 and / or TFRC. In some embodiments, the methods disclosed herein comprise determining the expression of ALDH1A1 and / or SCN2A.

[0083] In some embodiments, the expression of one or more of PRSS21 , ANLN, SPAG6, BMI1 and TFRC is increased in subjects who are at risk of relapse following cessation of a treatment for CML. In some embodiments, the expression of one or more of PRSS21 , BMI1 , and TFRC are increased in subjects who are at risk of relapse following cessation of a treatment for CML. In some embodiments, the expression of one or more of ALDH1A1 and SCN2A is decreased in subjects who are at risk of relapse following cessation of a treatment for CML. In some embodiments, the expression of the one or more genes is increased or decreased with respect to (or relative to) a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML.

[0084] In some embodiments, the expression of one or more of ALDH1A1 and SCN2A is increased in subjects who are likely to achieve TFR following cessation of a treatment for CML. In some embodiments, the expression of one or more of PRSS21 , ANLN, SPAG6, BMI1 and TFRC is decreased in subjects who are likely to achieve TFR following cessation of a treatment for CML. In some embodiments, the expression of the one or more genes is increased or decreased with respect to (or relative to) a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML.

[0085] Accordingly, the present disclosure provides a method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes selected from PRSS21 , ANLN, SPAG6, BMI1 and TFRC in CD34+CD38- cells in a sample obtained from the subject, wherein an increased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. The present disclosure also provides a method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes selected from ALDH1 A1 and SCN2A in CD34+CD38- cells in a sample obtained from the subject, wherein a decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML.

[0086] The present disclosure also provides a method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes selected from ALDH1A1 and SCN2A in CD34+CD38- or within the CD34+ cells in a sample obtained from the subject, wherein an increased expression of the one or more genes relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

[0087] The present disclosure also provides a method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes selected from PRSS21 , ANLN, SPAG6, BMI1 and TFRC in CD34+CD38- or within the CD34+ cells in a sample obtained from the subject, wherein a decreased expression of the one or more genes relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

[0088] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising (a) determining the expression of one or more genes selected from PRSS21 , ANLN, SPAG6, BMI1 and TFRC in CD34+CD38- cells in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0089] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising (a) determining the expression of one or more genes selected from ALDH1 A1 and SCN2A in CD34+CD38- cells in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes is decreased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor. The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising (a) determining the expression of one or more genes selected from PRSS21 , ANLN, SPAG6, BMI1 and TFRC in CD34+ CD38- cells in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes is decreased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a TKI, a BTT, a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0090] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising (a) determining the expression of one or more genes selected from ALDH1 A1 and SCN2A in CD34+ CD38- cells in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes is increased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a TKI, a BTT, a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0091] In some embodiments of the methods disclosed herein, the method comprises determining the expression of PRSS21. In some embodiments of the methods disclosed herein, the method comprises determining the expression of ANLN. In some embodiments of the methods disclosed herein, the method comprises determining the expression of SPAG6. In some embodiments of the methods disclosed herein, the method comprises determining the expression of BMI1. In some embodiments of the methods disclosed herein, the method comprises determining the expression of TFRC. In some embodiments of the methods disclosed herein, the method comprises determining the expression of ALDH1 A1 . In some embodiments of the methods disclosed herein, the method comprises determining the expression of SCN2A.

[0092] In some embodiments of the methods disclosed herein, the method comprises determining the expression of PRSS21 , BMI1 or TFRC. In some embodiments of the methods disclosed herein, the method comprises determining the expression of PRSS21 , BMI1 and TFRC. In some embodiments of the methods disclosed herein, the method comprises determining the expression of ALDH1A1. In some embodiments of the methods disclosed herein, the method comprises determining the expression of PRSS21 , BMI1 , TFRC and ALDH1 A1 . In some embodiments of the methods disclosed herein, the method comprises determining the expression of PRSS21 , SPAG6, and / or TFRC. In some embodiments, the method comprises determining the expression of PRSS21 and / or SPAG6. In some embodiments, the method comprises determining the expression of PRSS21 and SPAG6.

[0093] As used herein, where the expression of the one or more genes is “increased” or “decreased” relative to a pre-determined average as set out herein, this can be taken to mean that the one or more genes is “upregulated” or “downregulated” as appropriate. In this specification “PRSS21 ” (also known as testisin, eosinophil serine protease 1 (ESP-1) or serine protease 21 ; alternative gene names: ESP1 , TEST1) refers to PRSS21 from any species, and includes isoforms, fragments, variants, or homologues from any species. In some embodiments PRSS21 is PRSS21 from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, non-human primate or human)). In some embodiments, the PRSS21 is human PRSS21 , rhesus PRSS21 , mouse PRSS21 , rat PRSS21 or canine PRSS21 . In some embodiments, the PRSS21 is human PRSS21 or mouse PRSS21 . The human PRSS21 amino acid sequence is available under UniProt accession no. Q9Y6M0. As used herein, “PRSS21 ” may refer to the gene or the encoded protein. In some embodiments of the methods disclosed herein, determining the expression of PRSS21 comprises measuring the level of the mRNA or protein. As used herein, “PRSS21” and “PRSS” are used interchangeably.

[0094] In this specification “ANLN” (also known as anillin) refers to ANLN from any species, and includes isoforms, fragments, variants, or homologues from any species. In some embodiments ANLN is ANLN from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, non-human primate or human)). In some embodiments, the ANLN is human ANLN, rhesus ANLN, mouse ANLN, rat ANLN or canine ANLN. In some embodiments, the ANLN is human ANLN or mouse ANLN. The human ANLN amino acid sequence is available under UniProt accession no. Q9NQW6. As used herein, “ANLN” may refer to the gene or the encoded protein. In some embodiments of the methods disclosed herein, determining the expression of ANLN comprises measuring the level of the mRNA or protein.

[0095] In this specification “SPAG6” (also known as sperm-associated antigen 6, protein PF16 homolog, repro- SA-1 , or sperm flagellar protein; alternative gene name: PF16) refers to SPAG6 from any species, and includes isoforms, fragments, variants, or homologues from any species. In some embodiments SPAG6 is SPAG6 from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, non-human primate or human)). In some embodiments, the SPAG6 is human SPAG6, rhesus SPAG6, mouse SPAG6, rat SPAG6 or canine SPAG6. In some embodiments, the SPAG6 is human SPAG6 or mouse SPAG6. The human ANLN amino acid sequence is available under UniProt accession no. 075602. As used herein, “SPAG6” may refer to the gene or the encoded protein. In some embodiments of the methods disclosed herein, determining the expression of SPAG6 comprises measuring the level of the mRNA or protein.

[0096] In this specification “SCN2A” (also known as sodium channel protein type 2 subunit, HBSC II, sodium channel protein brain II subunit alpha, sodium channel protein type II subunit alpha, or voltage-gated sodium channel subunit alpha Nav1.2; alternative gene names: HBA, NAC2, SCN2A1 , and SCN2A2) refers to SCN2A from any species, and includes isoforms, fragments, variants, or homologues from any species. In some embodiments SCN2A is SCN2A from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, non-human primate or human)). In some embodiments, the SCN2A is human SCN2A, rhesus SCN2A, mouse SCN2A, rat SCN2A or canine SCN2A. In some embodiments, the SCN2A is human SCN2A or mouse SCN2A. The human SCN2A amino acid sequence is available under UniProt accession no. Q99250. As used herein, “SCN2A” may refer to the gene or the encoded protein. In some embodiments of the methods disclosed herein, determining the expression of SCN2A comprises measuring the level of the mRNA or protein.

[0097] In this specification “SCN2A” (also known as sodium channel protein type 2 subunit, HBSC II, sodium channel protein brain II subunit alpha, sodium channel protein type II subunit alpha, or voltage-gated sodium channel subunit alpha Nav1.2; alternative gene names: HBA, NAC2, SCN2A1 , and SCN2A2) refers to SCN2A from any species, and includes isoforms, fragments, variants, or homologues from any species. In some embodiments SCN2A is SCN2A from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, non-human primate or human)). In some embodiments, the SCN2A is human SCN2A, rhesus SCN2A, mouse SCN2A, rat SCN2A or canine SCN2A. In some embodiments, the SCN2A is human SCN2A or mouse SCN2A. The human SCN2A amino acid sequence is available under UniProt accession no. Q99250. As used herein, “SCN2A” may refer to the gene or the encoded protein. In some embodiments of the methods disclosed herein, determining the expression of SCN2A comprises measuring the level of the mRNA or protein.

[0098] In this specification “BMI1” (also known as polycomb complex protein BMI1 , polycomb group RING finger protein 4, or RING finger protein 51 ; alternative gene names: PCGF4 or RNF51) refers to BMI1 from any species, and includes isoforms, fragments, variants, or homologues from any species. In some embodiments BMI1 is BMI1 from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, non-human primate or human)). In some embodiments, the BMI1 is human BMI1 , rhesus BMI1 , mouse BMI1 , rat BMI1 or canine BMI1 . In some embodiments, the BMI1 is human BMI1 or mouse BMI1 . The human BMI1 amino acid sequence is available under UniProt accession no. P35226. As used herein, “BMI1” may refer to the gene or the encoded protein. In some embodiments of the methods disclosed herein, determining the expression of BMI1 comprises measuring the level of the mRNA or protein.

[0099] In this specification “TFRC” (also known as transferrin receptor protein 1 , T9 or p90; alternative gene name: CD71) refers to TFRC from any species, and includes isoforms, fragments, variants, or homologues from any species. In some embodiments TFRC is TFRC from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, non-human primate or human)). In some embodiments, the TFRC is human TFRC, rhesus TFRC, mouse TFRC, rat TFRC or canine TFRC. In some embodiments, the TFRC is human TFRC or mouse TFRC. The human TFRC amino acid sequence is available under UniProt accession no. P02786. As used herein, “TFRC” may refer to the gene or the encoded protein. In some embodiments of the methods disclosed herein, determining the expression of TFRC comprises measuring the level of the mRNA or protein.

[0100] In this specification “ALDH1A1” (also known as aldehyde dehydrogenase 1A1 , 3-deoxyglucosone dehydrogenase, ALDH-E1 , ALHDII, aldehyde dehydrogenase family 1 member A1 , aldehyde dehydrogenase, cytosolic, or retinal dehydrogenase 1 ; alternative gene name: ALDC, ALDH1 , or PUMB1) refers to ALDH1 A1 from any species, and includes isoforms, fragments, variants, or homologues from any species. In some embodiments ALDH1 A1 is ALDH1 A1 from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, non-human primate or human)). In some embodiments, the ALDH1 A1 is human ALDH1 A1 , rhesus ALDH1 A1 , mouse ALDH1 A1 , rat ALDH1 A1 or canine ALDH1 A1 . In some embodiments, the ALDH1 A1 is human ALDH1 A1 or mouse ALDH1 A1 . The human ALDH1 A1 amino acid sequence is available under UniProt accession no. P00352. As used herein, “ALDH1A1” may refer to the gene or the encoded protein. In some embodiments of the methods disclosed herein, determining the expression of ALDH1A1 comprises measuring the level of the mRNA or protein.

[0101] In some embodiments, the method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML comprises determining the expression of PRSS21 in a sample obtained from the subject, wherein an increased expression of PRSS21 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the expression of PRSS21 is determined in CD34+CD38- cells in the sample obtained from the subject.

[0102] In some embodiments, the method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML comprises determining the expression of ANLN in a sample obtained from the subject, wherein an increased expression of ANLN relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the expression of ANLN is determined in CD34+CD38- cells in the sample obtained from the subject.

[0103] In some embodiments, the method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML comprises determining the expression of BMI1 in a sample obtained from the subject, wherein an increased expression of BMI1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the expression of BMI1 is determined in CD34+CD38- cells in the sample obtained from the subject.

[0104] In some embodiments, the method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML comprises determining the expression of TFRC in a sample obtained from the subject, wherein an increased expression of TFRC relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the expression of TFRC is determined in CD34+CD38- cells in the sample obtained from the subject.

[0105] In some embodiments, the method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML comprises determining the expression of PRSS21 , BMI1 and / or TFRC in a sample obtained from the subject, wherein an increased expression of PRSS21 , BMI1 and / or TFRC relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the expression of PRSS21 , BMI1 and TFRC is determined in CD34+CD38- cells in the sample obtained from the subject.

[0106] In some embodiments, the method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML comprises determining the expression of PRSS21 , SPAG6 and / or TFRC in a sample obtained from the subject, wherein an increased expression of PRSS21 , SPAG6 and / or TFRC relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the expression of PRSS21 , SPAG6 and TFRC is determined in CD34+CD38- cells in the sample obtained from the subject.

[0107] In some embodiments, the method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML comprises determining the expression of PRSS21 and / or SPAG6 in a sample obtained from the subject, wherein an increased expression of PRSS21 and / or SPAG6 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. In some embodiments, the expression of PRSS21 , and SPAG6 is determined in CD34+CD38- cells in the sample obtained from the subject.

[0108] In some embodiments, the method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML comprises determining the expression of ALDH1A1 in a sample obtained from the subject, wherein an increased expression of ALDH1A1 relative to a pre-determined average in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML. In some embodiments, the expression of ALDH1A1 is determined in CD34+CD38- cells in the sample obtained from the subject.

[0109] In some embodiments, the method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML comprises determining the expression of SCN2A in a sample obtained from the subject, wherein an increased expression of SCN2A relative to a pre-determined average in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML. In some embodiments, the expression of SCN2A is determined in CD34+CD38- cells in the sample obtained from the subject.

[0110] In some embodiments, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, comprises:

[0111] (a) determining the expression of PRSS21 in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of PRSS21 is increased relative to a predetermined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0112] In some embodiments, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, comprises:

[0113] (a) determining the expression of ANLN in a sample obtained from the subject; and

[0114] (b) selecting the subject for treatment if the expression of ANLN is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0115] In some embodiments, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, comprises:

[0116] (a) determining the expression of SPAG6 in a sample obtained from the subject; and

[0117] (b) selecting the subject for treatment if the expression of SPAG6 is increased relative to a predetermined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0118] In some embodiments, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, comprises:

[0119] (a) determining the expression of BMI1 in a sample obtained from the subject; and

[0120] (b) selecting the subject for treatment if the expression of BMI1 is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0121] In some embodiments, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, comprises:

[0122] (a) determining the expression of TFRC in a sample obtained from the subject; and

[0123] (b) selecting the subject for treatment if the expression of TFRC is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0124] In some embodiments, the method selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, comprises:

[0125] (a) determining the expression of ALDH1A1 in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of ALDH1A1 increased relative to a predetermined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0126] In some embodiments, the method selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, comprises:

[0127] (a) determining the expression of SCN2A in a sample obtained from the subject; and

[0128] (b) selecting the subject for treatment if the expression of SCN2A increased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0129] In some embodiments, the method selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, comprises:(a) determining the expression of PRSS21 and / or SPAG6;

[0130] (b) determining the level of immune response activation (e.g. by determining the level / amount of TNFa and / or PNFKB); and

[0131] (c) selecting the subject for treatment if the expression of PRSS21 and / or SPAG6, and the level of immune response activation, are increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a TKI (e.g. a BTT) and an immunomodulator.

[0132] Without being bound by theory, the present inventors identified two biologically distinct relapse subtypes within patients - “Relapse-P” characterised by intrinsic proliferative programs, and “Relapse-I” characterised by extrinsic immune-driven programs. Biomarkers such as PRSS21 , SPAG6 and CD71 are able to identify relapse LSCs of both relapse subtypes ( / .e. they are relapse subtype agnostic biomarkers). The Relapse-I subtype is characterised by increased levels of of immune response activation (e.g. relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML).

[0133] Additional biomarkers such as TNF and pNFkB allow for identification of the Relapse-I subtype v the Relapse-P subtype at diagnosis, with the Relapse-I subtype exhibiting e.g. an increased level / amount of TNF and / or pNFkB relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML). Such patients may benefit from treatment further comprising administration of an immunomodulator.

[0134] Relapse-P subtype may exhibit similar levels of immune response activation (e.g. relative to a predetermined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML). Such patients may not benefit from treatment further comprising administration of an immunomodulator. That is, treatment does not comprise administration of an immunomodulator.

[0135] Accordingly, in some embodiments, the methods disclosed herein further comprise determining the level of immune response activation.

[0136] As used herein “immune response activation” refers to the initiation of signalling pathways that drive immune function / response, e.g. cytokine production and immune cell effector functions. Key immune response signalling pathways include the TNFO-NFKB, JAK-STAT and MAPK pathways. The level of immune response activation can be evaluated by measuring a correlate of the immune response activation. Correlates of immune response activation include cytokines (e.g. TNFa, IL-1 p, IL-6, IFN-y), cell surface markers of activation (e.g. CD69, CD25), and signalling pathway phosphorylation (e.g. PNFKB, pSTAT, p38 MAPK). In some embodiments, determining the level of immune response activation comprises determining the level / amount of TNFa. In some embodiments, determining the level of immune response activation comprises determining the level / amount of phosphorylated NFKB (PNFKB).

[0137] In some embodiments, the methods disclosed herein further comprise determining the level of activation of the NFKB signalling pathway (e.g. the TNFO-NFKB signalling pathway). The NFKB pathway is reviewed in Yu et al., Signal Transduction and Targeted Therapy (2020) 5:209, which is hereby incorporated by reference in its entirety. The level of activation of the NFKB signalling pathway can be evaluated by measuring a correlate of NFKB signalling pathway activation, for example, the level / amount of cytokines known to activate the NFKB signalling pathway (e.g. TNFa, IL1A, IL1 B, IL6, LTA, IL18), the level / amount of PNFKB, p-TAK1 , p-IKKa / p, p-lKBa (Ser32 / 36), p-p65, p-p38 I p-ERK I p-AKT, or p-RIPK1 , the level / amount of cytokines released upon NFKB pathway activation (e.g. TNFa, IL6, CXCL8 and CCL2), and the level of expression of downstream NFKB target genes (e.g. IL-6, TNFa, IL-1 p, ICAM-1). In some embodiments, determining the level of activation of the NFKB pathway comprises determining the amount / level of one or more of TNFa, IL1 A, IL1 B, IL6, LTA, IL18, PNFKB, p-TAK1 , p-IKKa / p, p-lKBa (Ser32 / 36), p-p65, p-p38 I p-ERK I p-AKT, p-RIPK1 , CXCL8 and CCL2. In some embodiments, determining the level of activation of the NFKB signalling pathway (e.g. the TNFO-NFKB signalling pathway) comprises determining the level / amount of TNFa. In some embodiments, determining the level of activation of the NFKB signalling pathway (e.g. the TNFO-NFKB signalling pathway comprises determining the level / amount of phosphorylated NFKB (PNFKB).

[0138] The present disclosure provides a method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, the method comprising: (i) determining the expression of one or more genes regulated by PRC1 ; and (ii) determining the level of immune response activation, in a sample obtained from the subject. The present disclosure also provides a method for selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising: (i) determining the expression of one or more genes regulated by PRC1 ; and (ii) determining the level of immune response activation, in a sample obtained from the subject. The expression of the one or more genes may be determined in CD34+CD38- cells in the sample obtained from the subject. The expression of the one or more genes and / or the level of immune response activation may be determined relative to a pre-determined average expression obtained from reference samples as described herein. The expression of the one or more genes and / or the level of immune response activation may be determined relative to the average expression in samples obtained from a reference population as described herein.

[0139] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising

[0140] (a) (i) determining the expression of one or more genes regulated by PRC1 , and (ii) determining the level of immune response activation in CD34+CD38- cells in a sample obtained from the subject; and

[0141] (b) selecting the subject for treatment if (i) the expression of the one or more genes regulated by PRC1 is increased or decreased, and (ii) the level of immune response activation is increased, relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a tyrosine kinase inhibitor (TKI) and an immunomodulator.

[0142] As used herein, an “immunomodulator” refers to an agent or therapy that modulates the immune response. An immunomodulator may also be referred to as an immunomodulatory agent. An immunomodulator may reduce / inhibit or increase the immune response. In some embodiments, an immunomodulator reduces / inhibits the immune response. Immunomodulatory therapeutics are reviewed e.g. in Nash A et al, Advanced Drug Delivery Reviews 2021 , 176:113896, which is hereby incorporated by reference in its entirety.

[0143] In some embodiments, an immunomodulator is an inhibitor of the TNFO-NFKB signalling pathway. That is, the immunomodulator inhibits / reduces TNFO-NFKB signalling. In some embodiments, an inhibitor of the TNFO-NFKB signalling pathway is a TNFa inhibitor, an IL-1 inhibitor or an NFKB pathway PROTAC degrader. In some embodiments, an immunomodulator may reduce / inhibit / kill immune cells responsible for TNFa and / or IL1 B secretion (e.g. immunosuppressive and tumour associated monocytes, T cells, B cells).

[0144] TNFa inhibitors include Adalimumab (DrugBank: DB00051), Infliximab (DrugBank: DB00065), Etanercept (DrugBank: DB00005), Golimumab (DrugBank: DB06674), Certolizumab (DrugBank: DB08904), Ozoralizumab (DrugBank: DB12014), Senlizumab (DrugBank: DB17894), XPro1595 (DrugBank: DB16454), and Balinatunfib (DrugBank: DB21728). IL-1 inhibitors include anakinra (DrugBank: DB00026), Rilonacept (DrugBank: DB06372), Canakinumab (DrugBank: DB06168), Gevokizumab (DrugBank: DB12119), Bermekimab (DrugBank: DB14947), Goflikicept (DrugBank: DB16396), and Diacerein (DrugBank: DB11994). NFKB pathway PROTAC degraders include JP-163-16 (PROTAC 15d) (as described in Jin et al., RSC Medicinal Chemistry (2025) 16(9):3839-4512), IKK0 / NR4A1 Degrader-1 (A9) (as described in Maharjan et al., bioRxiv 2025.09.24.678355), “dNF-KB” TF-PROTAC (as described in US20240131173A1), KT-474 (IRAK4 Degrader) (as described in Ferguson, Nature Medicine (2023) 29:3006-3007), and JH-XIII-05-1 (as described in Hatcher et al., Blood (2024) 144 (Supplement 1):4359), In some embodiments, the immunomodulator is an antigen-binding molecule (e.g. a therapeutic antibody). In some embodiments, the immunomodulator is a multipspecific antigen-binding molecule, e.g. a bispecific T-cell engager (BiTE) or trispecific killer engager (TriKE). In some embodiments, the antigenbinding molecule is a chimeric antigen receptor (CAR). In some embodiments, the antigen-binding molecule (e.g. the BiTE / TriKE / CAR) targets / binds to LSC markers describer herein (e.g. PRSS21 , SPAG6). In some embodiments, the CAR is expressed by a T cell or an NK cell ( / .e. a CAR-T cell or a CAR-NK cell).

[0145] In some embodiments, the treatment further comprises administration of adaptive NK cells. For example, the treatment may comprise administration of an immunomodulator (e.g. an antigen-binding molecule, e.g. a TriKE) and adaptive NK cells.

[0146] In some embodiments, the one or more genes regulated by PRC1 are selected from PRSS21 , SPAG6 and TFRC. In some embodiments, the one or more genes regulated by PRC1 are PRSS21 and / or SPAG6. In some embodiments, genes regulated by PRC1 are PRSS21 and SPAG6. In each of the embodiments of this paragraph, a subject is selected for treatment if (i) the expression of the one or more genes regulated by PRC1 is increased, and (ii) the level of immune response activation is increased, relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a tyrosine kinase inhibitor (TKI) and an immunomodulator.

[0147] The present disclosure also provides a method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising

[0148] (a) (i) determining the expression of one or more genes regulated by PRC1 , and (ii) determining the level of immune response activation in CD34+CD38- cells in a sample obtained from the subject; and

[0149] (b) selecting the subject for treatment if (i) the expression of the one or more genes regulated by PRC1 is increased or decreased, and (ii) the level of immune response activation is increased, relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML; wherein the treatment comprises a tyrosine kinase inhibitor (TKI) and an immunomodulator.

[0150] In some embodiments, the one or more genes regulated by PRC1 are selected from PRSS21 , SPAG6 and TFRC. In some embodiments, the one or more genes regulated by PRC1 are PRSS21 and / or SPAG6. In some embodiments, genes regulated by PRC1 are PRSS21 and SPAG6. In each of the embodiments of this paragraph, a subject is selected for treatment if (i) the expression of the one or more genes regulated by PRC1 is decreased, and (ii) the level of immune response activation is decreased, relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML; wherein the treatment comprises a tyrosine kinase inhibitor (TKI) and an immunomodulator.

[0151] In some embodiments, the TKI is a BCR-ABL1 -targeted therapy (BTT) (e.g. a BTT as described herein). In some embodiments, the immunomodulator is an inhibitor that targets the TNFa-NFkB pathway (e.g. an inhibitor that targets the TNFa-NFkB pathway described herein). Without being bound by theory, remission rates are improved by targeting of the immune / bone marrow niche cells that release TNFa alongside administration of a TKI therapy (e.g. BTT).

[0152] Diagnostic and prognostic methods, subject selection and reference samples / populations The present disclosure also provides diagnostic, prognostic and predictive methods in connection with the cancers described herein (e.g. CML).

[0153] The methods may be performed for the purpose of diagnosing a cancer (e.g. a cancer described herein). The methods may be performed for the purpose of prognosing / predicting the likely progression of a cancer (e.g. CML). The methods may be performed for the purpose of prognosing / predicting the likely response of a subject to therapeutic / prophylactic intervention as described herein. The methods may be useful to predict the likely response to a given therapeutic / prophylactic intervention, e.g. in terms of efficacy and / or resistance, and may therefore by useful for supporting clinical decision-making. The methods may be performed for the purpose of identifying / selecting a subject for therapeutic / prophylactic intervention as described herein. The methods may comprise evaluating a cancer (e.g. CML) to determine whether it comprises hallmark features indicative of poor patient outcome (e.g. relapse following cessation of a treatment for CML). The methods may comprises evaluating a cancer (e.g. CML) to determine whether it comprises hallmark features indicative of good patient outcome (e.g. TFR following cessation of a treatment for CML). The methods may be performed for the purpose of predicting whether a subject with cancer (e.g. CML) is at risk of relapse following cessation of a treatment (e.g. a treatment for CML). The methods may be performed for the purpose of predicting whether a subject with cancer (e.g. CML) is likely to achieve TFR following cessation of a treatment (e.g. a treatment for CML).

[0154] The present disclosure provides a method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML. The present disclosure also provides a method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML.

[0155] In this context “cessation of a treatment for CML” is where the treatment for CML is discontinued following at least one administration of the treatment to the subject.

[0156] “Treatment-free remission” (TFR), also known as “therapy-free remission” is achieved when a subject who has discontinued a treatment for CML (such as TKI therapy or BCR-ABL1 -targeted therapy) maintains a deep molecular response (DMR) and does not need to restart treatment and / or does not suffer a relapse. In this context “deep molecular response” (DMR) is where <0.01% of cells (compared to a standardised baseline) express the BCR-ABL1 gene, e.g., as measured by quantitative PCR (qPCR) using peripheral blood or bone marrow samples. In other words, DMR is achieved where there is at least a 4-fold reduction in cells expressing the BCR-ABL1 gene following treatment with a treatment for CML, e.g., as measured by qPCR using peripheral blood or bone marrow samples.

[0157] As referred to herein, a subject who achieves TFR is a subject who maintains a DMR following cessation with a CML treatment or therapy without suffering a relapse. In this context a “relapse” (also known as a “molecular relapse”) may be indicated by a loss in molecular response (e.g., a subject having >0.1% of cells that express the BCR-ABL1 gene, e.g., as measured by qPCR using peripheral blood or bone marrow samples). In some embodiments, the subject who achieves TFR is a subject who maintains a DMR following cessation with a TKI therapy or a BCR: :ABL1 -targeted therapy. In some embodiments, the subject maintains a DMR for 1 or more, 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, or 10 or more years following cessation with the CML treatment or therapy.

[0158] A method according to the present disclosure may be performed concurrently or subsequently to a CML diagnosis. That is, a method according to the present disclosure may be performed at the same time a subject is diagnosed with CML or following a subject’s CML diagnosis. A method according to the present disclosure may be performed within 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 weeks following a subject’s CML diagnosis. A method according to the present disclosure may be performed on a sample obtained from a subject before, at the same time as, or after a CML diagnosis. A method according to the present disclosure may be performed on a sample obtained from a subject within 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 weeks following a subject’s CML diagnosis. A method according to the present disclosure may be performed on a sample obtained from a subject who has CML or has been diagnosed with CML and has not received treatment for CML.

[0159] The present disclosure provides a method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in a sample obtained from the subject. The present disclosure also provides a method for selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising determining the expression of one or more genes regulated by PRC1 in a sample obtained from the subject. The expression of the one or more genes may be determined in CD34+CD38- cells in the sample obtained from the subject.

[0160] It will be appreciated that for any of the methods disclosed herein, determining the expression of the one or more genes can be carried out using any known technique. For example, the expression of the one or more genes may be carried out by measuring RNA levels (e.g. mRNA), for example using quantitative PCR or similar. For example, the expression of the one or more genes may be carried out by measuring levels of the (encoded) protein, for example using flow cytometry or immunohistochemistry. In some embodiments of the method disclosed herein, the method comprises determining the expression of one or more genes by measuring the levels of the (encoded) protein by flow cytometry. In some embodiments of the method disclosed herein, the method comprises determining the expression of one or more genes by measuring the levels of the (encoded) protein by immunohistochemistry.

[0161] Immunohistochemistry is a form of immunostaining. It involves the use of antibodies to bind to and identify specific antigens in biological tissue. Immunohistochemistry can be performed on tissue that has been fixed and embedded in paraffin, but also cryopreservated (frozen) tissue. In some embodiments, immunohistochemistry may be performed on formalin-fixed paraffin embedded (FFPE) sections. Principles and methods of immunohistochemistry (including immunohistochemistry of bone marrow tissue) are well known to the skilled person and are reviewed in e.g. Ramos-Vara, J.A. (2011). Principles and Methods of Immunohistochemistry in Gautier, JC. (eds) Drug Safety Evaluation. Methods in Molecular Biology, vol 691 . Humana Press, and Grant, M.L., Zhang, X.M. (2022). Bone Marrow. In: Lin, F., Prichard, J.W., Liu, H., Wilkerson, M.L. (eds) Handbook of Practical Immunohistochemistry. Springer, Cham, which are hereby incorporated by reference in their entirety.

[0162] It will be appreciated that the methods of the present disclosure may be performed using any suitable comparison. In some embodiments, the expression of the one or more genes may be determined relative to a pre-determined average expression obtained from reference samples as disclosed herein. In some embodiments, the level of immune response activation may be determined relative to a pre-determined average level of immune response activation obtained from reference samples as disclosed herein.

[0163] The expression of the one or more genes may be determined relative to a pre-determined average expression obtained from reference samples. The reference samples may be obtained from subjects with CML who relapse following cessation of a treatment for CML. The reference samples may be obtained from subjects with CML who achieve TFR following cessation of a treatment for CML. The reference samples may be obtained from subjects who exhibit (or do not exhibit) molecular relapse following cessation of a treatment for CML. The reference samples may be obtained from subjects who do (or do not) subsequently achieve a DMR following cessation of a treatment for CML. The reference samples may be obtained from healthy individuals. The reference samples may be obtained from subjects who do not have, or have not been diagnosed with, CML.

[0164] An increased expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is at risk of relapse following cessation of the treatment for CML. A decreased expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is at risk of relapse following cessation of the treatment for CML. A similar expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is not at risk of relapse following cessation of the treatment for CML. Optionally, the reference samples are obtained from subjects with CML who achieve TFR following cessation of the treatment for CML.

[0165] By way of example, an increased expression of PRSS21 , ANLN, SPAG6, BMI1 and / or TFRC relative to a pre-determined average expression obtained from reference samples may indicate that the subject is at risk of relapse following cessation of a treatment for CML. A decreased or similar expression of PRSS21 , ANLN, SPAG6, BMI1 and / or TFRC relative to a pre-determined average expression obtained from reference samples may indicate that the subject is not at risk of relapse following cessation of a treatment for CML.

[0166] An increased expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is likely to achieve TFR following cessation of the treatment for CML. A decreased expression of the one or more genes relative to a predetermined average expression obtained from reference samples may indicate that the subject is likely to achieve TFR following cessation of the treatment for CML. A similar expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is not at risk of relapse following cessation of the treatment for CML. Optionally, the reference samples are obtained from subjects with CML who relapse following cessation of the treatment for CML.

[0167] By way of example, an increased expression of ALDH1A1 and / or SCN2A relative to a pre-determined average expression obtained from reference samples may indicate that the subject is likely to achieve TFR following cessation of the treatment for CML. A similar or decreased expression of ALDH1A1 and / or SCN2A relative to a pre-determined average expression obtained from reference samples may indicate that the subject is not likely to achieve TFR following cessation of the treatment for CML.

[0168] An increased expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is suitable for treatment with a hypomethylating agent (HMA) and / or a BMI1 inhibitor. A decreased expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is suitable for treatment with a HMA and / or a BMI1 inhibitor. Optionally, the reference samples are obtained from subjects with CML who achieve TFR following cessation of a treatment for CML.

[0169] By way of example, an increased expression of PRSS21 , ANLN, SPAG6, BMI1 and / or TFRC relative to a pre-determined average expression obtained from reference samples may indicate that the subject is suitable for treatment with a HMA and / or a BMI1 inhibitor. A decreased or similar expression of PRSS21 , ANLN, SPAG6, BMI1 and / or TFRC relative to a pre-determined average expression obtained from reference samples may indicate that the subject is not suitable for treatment with a HMA and / or a BMI1 inhibitor, and / or would be suitable for treatment with one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0170] An increased expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is suitable for treatment with one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof. A decreased expression of the one or more genes relative to a pre-determined average expression obtained from reference samples may indicate that the subject is suitable for treatment with one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof. Optionally, the reference samples are obtained from subjects with CML who relapse following cessation of the treatment for CML.

[0171] By way of example, an increased expression of ALDH1A1 and / or SCN2A relative to a pre-determined average expression obtained from reference samples may indicate that the subject is suitable for treatment with one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof. A decreased or similar expression of ALDH1 A1 and / or SCN2A relative to a pre- determined average expression obtained from reference samples may indicate that the subject is not suitable for treatment with one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof, and / or would be suitable for treatment with a HMA and / or a BMI1 inhibitor.

[0172] In some embodiments, the expression of the one or more genes may be determined relative to the average expression in samples obtained from a reference population.

[0173] Accordingly, in some embodiments, an increased expression of one or more of PRSS21 , ANLN, SPAG6, BMI1 and / or TFRC in a sample obtained from the subject as compared to the average expression in samples obtained from a reference population indicates that the subject is at risk of relapse following cessation of treatment. In some embodiments, a similar or decreased expression of one or more of PRSS21 , ANLN, SPAG6, BMI1 and / or TFRC in a sample obtained from the subject as compared to the average expression in samples obtained from a reference population indicates that the subject is not at risk of relapse following cessation of treatment.

[0174] In some embodiments, the reference population is individuals with CML who subsequently achieve TFR following cessation of a treatment for CML. In some embodiments, the reference population is individuals with CML who subsequently achieve a DMR following cessation of a treatment for CML. In some embodiments, the reference population is individuals who do not exhibit relapse following cessation of a treatment for CML. In some embodiments, the reference population is individuals who do not exhibit molecular relapse following cessation of a treatment for CML. In some embodiments, the reference population is individuals with CML who subsequently have optimal response to treatment with a treatment for CML. In some embodiments, the reference population is healthy individuals. In some embodiments, the reference population is individuals who do not have, or have not been diagnosed with, CML.

[0175] In some embodiments, an increased expression of one or more of ALDH1A1 and / or SCN2A in a sample obtained from the subject as compared to the average expression in samples obtained from a reference population indicates that the subject is likely to achieve TFR following cessation of treatment. In some embodiments, a similar or decreased expression of one or more of ALDH1A1 and / or SCN2A in a sample obtained from the subject as compared to the average expression in samples obtained from a reference population indicates that the subject is not likely to achieve TFR following cessation of treatment.

[0176] In some embodiments, the reference population is individuals with CML who do not subsequently achieve TFR following cessation of a treatment for CML. In some embodiments, the reference population is individuals who subsequently relapse following cessation of a treatment for CML. In some embodiments, the reference population is individuals who do not subsequently achieve a DMR following cessation of a treatment for CML. In some embodiments, the reference population is individuals who exhibit molecular relapse following cessation of a treatment for CML. In some embodiments, the reference population is individuals who do not have an optimal response to treatment with a treatment for CML. In some embodiments, the reference population is healthy individuals. In some embodiments, the reference population is individuals who do not have, or have not been diagnosed with, CML. As used herein, the “average” or “pre-determined average” may refer to the mean or median value (e.g. for a given set of reference samples or samples obtained from a reference population as described herein).

[0177] It will be appreciated that the expression of one or more genes or a number / proportion of a given cell type / subtype in a sample obtained from the subject is similar / increased / reduced relative to the average expression / number / proportion of a given cell type / subtype in a comparable sample (e.g. a sample of the same kind, e.g. obtained from the same fluid, tissue, organ etc., and / or obtained at the same stage of CML progression e.g. newly diagnosed, chronic phase) obtained from individuals in the reference population.

[0178] In some embodiments, a method according to the present disclosure comprises detecting haematopoietic stem cells (HSCs). HSCs are immature cells in the bone marrow and that can develop into multiple cell types, such as white blood cells, red blood cells, and platelets. In some embodiments, a method according to the present disclosure comprises detecting leukaemic stem cells (LSCs). LSCs are cells with the biological potential, intrinsic or acquired, to cause leukaemia. LSCs are CD34+CD38_cells.

[0179] In some embodiments, a method according to the present disclosure comprises detecting megakaryocytic progenitor cells. Megakaryocytic progenitor cells are derived from hematopoietic stem cells in the bone marrow. Megakaryocytic progenitor cells differentiate into megakaryocytes, which produce platelets (also known as thrombocytes).

[0180] Antigen-binding molecules

[0181] In some embodiments of the methods disclosed herein, the expression of the one or more genes is determined using one or more antigen-binding molecules (e.g. one or more antibodies). In some embodiments, the antigen-binding molecule is an antibody that is capable of detecting (or binding to) PRSS21 , ANLN, SPAG6, SCN2A, BMI1 , TFRC and / or ALDH1A1 . In some embodiments, the antigenbinding molecule is an antibody that is capable of detecting (or binding to) PRSS21 , BMI1 , TFRC and / or ALDH1A1. In some embodiments, the antigen-binding molecule is an antibody that is capable of detecting (or binding to) PRSS21 , BMI1 , and / or TFRC. In some embodiments, the antigen-binding molecule is an antibody that is capable of detecting (or binding to) ALDH1A1 .

[0182] In some embodiments of the methods disclosed herein, the expression of the one or more genes is determined using a plurality of antigen-binding molecules (e.g. 2, 3, 4, 5, 6, or 7 or more antigen-binding molecules). In some embodiments, the plurality of antigen-binding molecules (e.g. antibodies) is capable of detecting (or binding to) two or more of: PRSS21 , ANLN, SPAG6, SCN2A, TFRC and / or ALDH1A1 . In some embodiments, the plurality of antigen-binding molecules (e.g. antibodies) is capable of detecting (or binding to) PRSS21 , BMI1 , TFRC and / or ALDH1 A1 . In some embodiments, the plurality of antigenbinding molecules (e.g. antibodies) is capable of detecting (or binding to) PRSS21 , BMI1 and TFRC. In some embodiments, the plurality of antigen-binding molecules (e.g. antibodies) is capable of detecting (or binding to) PRSS21 , SPAG6 and / or TFRC. In some embodiments, the plurality of antigen-binding molecules (e.g. antibodies) is capable of detecting (or binding to) PRSS21 and / or SPAG6.

[0183] It will be appreciated that where an antigen-binding molecule is described in accordance with the methods disclosed herein, the antigen-binding molecule will detect (or bind to) the encoded protein. For example, wherein the expression of the one or more genes is determined using one or more antigenbinding molecules (e.g. one or more antibodies) that are capable or detecting (or binding to) one or more of PRSS21 , ANLN, SPAG6, SCN2A, TFRC and / or ALDH1A1 , this is taken to mean that the antigenbinding molecules bind to one or more of the proteins encoded by the PRSS21 , ANLN, SPAG6, SCN2A, TFRC and / or ALDH1A1 genes.

[0184] In some embodiments of the methods disclosed herein, the level of immune response activation is determined using one or more antigen-binding molecules (e.g. one or more antibodies). In some embodiments, the antigen-binding molecule is an antibody that is capable of detecting (or binding to) a component of the NFKB pathway (e.g. a component of the TNFONFKB pathway). In some embodiments, the antigen-binding molecule is an antibody that is capable of detecting (or binding to) TNFa or PNFKB. In some embodiments of the methods disclosed herein, the the level of immune response activation is determined using a plurality of antigen-binding molecules (e.g. 2, 3, 4, 5, 6, or 7 or more antigen-binding molecules). In some embodiments, the plurality of antigen-binding molecules (e.g. antibodies) is capable of detecting (or binding to) TNFa and PNFKB.

[0185] An “antibody” is used herein in the broadest sense, and encompasses monoclonal antibodies, polyclonal antibodies, monospecific and multispecific antibodies (e.g., bispecific antibodies), and antibody fragments, as long as they display binding to the relevant target molecule.

[0186] In view of today's techniques in relation to monoclonal antibody technology, antibodies can be prepared to most antigens. The antigen-binding portion may be a part of an antibody (for example a Fab fragment) or a synthetic antibody fragment (for example a single chain Fv fragment [ScFv]). Monoclonal antibodies to selected antigens may be prepared by known techniques, for example those disclosed in "Monoclonal Antibodies: A manual of techniques ", H Zola (CRC Press, 1988) and in "Monoclonal Hybridoma Antibodies: Techniques and Applications ", J G R Hurrell (CRC Press, 1982). Chimeric antibodies are discussed by Neuberger et al (1988, 8th International Biotechnology Symposium Part 2, 792-799). Monoclonal antibodies (mAbs) are particularly useful in the methods of the invention and are a homogenous population of antibodies specifically targeting a single epitope on an antigen.

[0187] Polyclonal antibodies are also useful in the methods of the invention. Monospecific polyclonal antibodies are preferred. Suitable polyclonal antibodies can be prepared using methods well known in the art.

[0188] Antigen-binding fragments of antibodies, such as Fab and Fab2 fragments may also be used / provided as can genetically engineered antibodies and antibody fragments. The variable heavy (VH) and variable light (VL) domains of the antibody are involved in antigen recognition, a fact first recognised by early protease digestion experiments. Further confirmation was found by "humanisation" of rodent antibodies. Variable domains of rodent origin may be fused to constant domains of human origin such that the resultant antibody retains the antigenic specificity of the rodent parented antibody (Morrison et al (1984) Proc. Natl. Acad. Sd. USA 81 , 6851-6855).

[0189] Antibodies and antigen-binding fragments according to the present disclosure comprise the complementarity-determining regions (CDRs) of an antibody which is capable of binding to the relevant target molecule (e.g. PRSS21 , BMI1 , TFRC and / or ALDH1A1).

[0190] Accordingly, by way of example, the method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment with CML may comprise determining the expression of PRSS21 , BMI1 and / or TFRC in a sample obtained from the subject, wherein an increased expression of PRSS21 , BMI1 and / or TFRC relative to the corresponding pre-determined average expression in samples obtained from subject with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML, and wherein the expression of PRSS21 , BMI1 and / or TFRC may be determined using one or more antigen-binding molecules that are capable of detecting (or binding to) PRSS21 , BMI1 and / or TFRC.

[0191] By way of a further example, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML may comprise:

[0192] (a) determining the expression of PRSS21 , BMI1 and / or TFRC in a sample obtained from the subject, wherein the expression is determined using one or more antigen-binding molecules that are capable of detecting (or binding to) PRSS21 , BMI1 and / or TFRC; and

[0193] (b) selecting the subject for treatment if the expression of PRSS21 , BMI1 , and / or TFRC is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a HMA and / or a BMI1 inhibitor.

[0194] It will be appreciated that the combination of biomarkers in the above examples may be any combination of biomarkers disclosed herein. For example, PRSS21 , SPAG6 and / or TFRC. As a further example, PRSS21 and / or SPAG6.

[0195] By way of a further example, the method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML may comprise determining the expression of ALDH1A1 in a sample obtained from the subject, wherein an increased expression of ALDH1A1 relative to a predetermined average expression in samples obtained with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML, wherein the expression of ALDH1 A1 may be determined using one or more antigen-binding molecules that are capable of detecting (or binding to) ALDH1 A1 .

[0196] By way of a further example, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML may comprise: (a) determining the expression of ALDH1A1 in a sample obtained from the subject, wherein the expression is determined using one or more antigen-binding molecules that are capable of detecting (or binding to) ALDH1 A1 ; and

[0197] (b) selecting the subject for treatment if the expression of PRSS21 , BMI1 , and / or TFRC is increased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a TKI, a BTT, a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy, and / or a combination thereof.

[0198] Therapeutic and prophylactic intervention

[0199] The present disclosure provides methods for the treatment and / or prevention of chronic myelogenous leukaemia (CML).

[0200] The methods may be effective to reduce the development or progression of CML, alleviate the symptoms of CML or reduce the pathology of CML. The methods may be effective to prevent progression of CML, e.g. to prevent worsening of, or to slow the rate of development of, CML. In some embodiments, the methods may lead to an improvement in CML, e.g. a reduction in the symptoms of CML or reduction in some other correlate of the severity / activity of CML. In some embodiments, the methods may prevent development of CML to a later stage (e.g. blast crisis phase or metastasis).

[0201] In some embodiments, methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises one or more treatments for CML (e.g. one or more treatments for CML as described herein). In some embodiments, methods disclosed herein comprise selecting a subject for treatment, wherein the treatment further comprises administration of an immunomodulator.

[0202] In some embodiments, methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises a TKI. In some embodiments, methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises a TKI selected from imatinib, dasatinib, nilotinib, bosutinib, ponatinib, bafetinib, radotinib, olverembatinib, flumatinib, asciminib, and TERN-701. In some embodiments, methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises a first-generation TKI, a second-generation TKI, and / or a third-generation TKI. In some embodiments, the TKI is a STAMP inhibitor. In some embodiments, the STAMP inhibitor is selected from: asciminib, and TERN-701 . In some embodiments, the TKI is not a STAMP inhibitor.

[0203] In some embodiments, methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises a stem cell transplant.

[0204] In some embodiments, a method comprises selecting a subject for treatment with a conventional first line therapy (e.g. a first-generation or second-generation TKI).

[0205] In some embodiments, the methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises one or more of a BMI1 inhibitor, a PRC1 inhibitor, a HMA, a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0206] In some embodiments, the methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises a BMI1 inhibitor, a PRC1 inhibitor, and / or an HMA. In some embodiments, the BMI inhibitor or PRC1 inhibitor is a small molecule inhibitor. In some embodiments, the BMI1 inhibitor is PTC209 or PTC596.

[0207] In some embodiments, the methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1 -targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0208] In some embodiments, the methods disclosed herein comprise selecting a subject for treatment, wherein the treatment comprises a tyrosine kinase inhibitor (TKI) and an immunomodulator. In some embodiments, the TKI is a BCR-ABL1 -targeted therapy (BTT) (e.g. a BTT as described herein). In some embodiments, the immunomodulator is an inhibitor that targets the TNFa-NFkB pathway (e.g. an inhibitor that targets the TNFa-NFkB pathway described herein).

[0209] In some embodiments, the methods disclosed herein comprise selecting a subject for treatment wherein the treatment further comprises administration of adaptive NK cells. For example, the treatment may comprise administration of an immunomodulator (e.g. an antigen-binding molecule, e.g. a TriKE) and adaptive NK cells.

[0210] It will be appreciated that the selection of treatment will depend on the outcome of determining the expression of one or more genes as disclosed herein. The subject may be administered with the same treatment as the reference population or may be administered with a different treatment as the reference population. Exemplary embodiments are set out below, which may be combined with other embodiments and methods disclosed herein.

[0211] In some embodiments, a method comprises selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0212] (a) determining the expression of one or more genes in a sample obtained from the subject; and

[0213] (b) selecting the subject for treatment if the expression of the one or more genes is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a TKI; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0214] In some embodiments, a method comprises selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0215] (a) determining the expression of one or more genes in a sample obtained from the subject; and (b) selecting the subject for treatment if the expression of the one or more genes is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a TKI; wherein the treatment comprises a TKI.

[0216] In some embodiments, a method comprises selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0217] (a) determining the expression of one or more genes in a sample obtained from the subject; and

[0218] (b) selecting the subject for treatment if the expression of the one or more genes is decreased (or similar) relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a TKI; wherein the treatment comprises a TKI.

[0219] In some embodiments, a method comprises selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0220] (a) determining the expression of one or more genes in a sample obtained from the subject; and

[0221] (b) selecting the subject for treatment if the expression of the one or more genes is decreased (or similar) relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a TKI; wherein the treatment comprises hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0222] In some embodiments, a method comprises selecting a subject for treatment with a TKI, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0223] (a) determining the expression of one or more genes disclosed herein in a sample obtained from the subject; and

[0224] (b) selecting the subject for treatment if the expression of the one or more genes is decreased (or similar) relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a TKI.

[0225] In some embodiments, a method comprises selecting a subject for treatment with a hypomethylating agent (HMA) and / or a BMI1 inhibitor, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0226] (a) determining the expression of one or more genes disclosed herein in a sample obtained from the subject; and

[0227] (b) selecting the subject for treatment if the expression of the one or more genes is decreased (or similar) relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a TKI.

[0228] The present disclosure provides a method for treating or preventing CML in a subject. It will be appreciated that the method may comprise administering any of the treatments / therapies for CML described herein. Also provided is a method for improving the prognosis of CML in a subject, e.g. a subject determined to be at risk of relapse following cessation of treatment.

[0229] In some embodiments of the methods disclosed herein, the method further comprises administering to the subject one of more treatments for CML as disclosed herein.

[0230] For example, the method for predicting whether a subject with chronic myelogenous leukaemia (CML) is at risk of relapse following cessation of a treatment for CML may comprise:

[0231] (a) determining the expression of one or more genes regulated by polycomb repressive complex 1 (PRC1) in CD34+CD38- cells in a sample obtained from the subject, wherein an increased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML; and

[0232] (b) administering to the subject one or more treatments for CML as disclosed herein.

[0233] For example, the method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, may comprise:

[0234] (a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject, wherein a decreased / similar expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML; and

[0235] (b) administering to the subject one or more treatments for CML as disclosed herein.

[0236] For example, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML may comprise:

[0237] (a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject;

[0238] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased / decreased / similar relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor; and

[0239] (c) administering to the subject a HMA and / or BMI1 inhibitor.

[0240] For example, the method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, may comprise:

[0241] (a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject;

[0242] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased / decreased / similar relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof; and

[0243] (c) administering to the subject one or more of a tyrosine kinase inhibitor (TKI) , a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0244] Timing

[0245] A method according to the present disclosure may be performed concurrently or subsequently to a CML diagnosis. That is, a method according to the present disclosure may be performed at the same time a subject is diagnosed with CML or following a subject’s CML diagnosis. A method according to the present disclosure may be performed within 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 weeks following a subject’s CML diagnosis. A method according to the present disclosure may be performed on a sample obtained from a subject before, at the same time as, or after a CML diagnosis. A method according to the present disclosure may be performed on a sample obtained from a subject within 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 weeks following a subject’s CML diagnosis. A method according to the present disclosure may be performed on a sample obtained from a subject who has CML or has been diagnosed with CML and has not received treatment for CML.

[0246] As used herein, a “CML diagnosis” refers to the point at which it is determined that a subject has CML.

[0247] Accordingly, in some embodiments, the method for predicting whether a subject is at risk of relapse following cessation of treatment is performed concurrently or subsequently to a CML diagnosis.

[0248] Accordingly, in some embodiments, the method of selecting a subject for treatment is performed concurrently or subsequently to a CML diagnosis.

[0249] Subject

[0250] The subject in accordance with aspects of the present disclosure may be any animal or human. The subject is preferably mammalian, more preferably human. The subject may be a non-human mammal but is more preferably human. The subject may be male or female. The subject may be a patient (e.g. a CML patient). A subject may have been diagnosed with a disease or condition requiring treatment (e.g. CML), may be suspected of having such a disease / condition, or may be at risk of developing / contracting such a disease / condition.

[0251] In embodiments according to the present disclosure the subject is preferably a human subject. In some embodiments, the subject has CML or has been diagnosed with CML. In some embodiments, the methods of the present disclosure are performed concurrently or subsequently to a CML diagnosis. In some embodiments, the subject has been newly diagnosed with CML (e.g. in the preceding 1 , 2, 3, 4, 5, 6, 7 or 8 weeks) or is concurrently diagnosed with CML. In some embodiments, the subject has CML or has been diagnosed with CML and has not received treatment. Sample

[0252] The methods disclosed herein may be performed in vitro on a sample obtained from a subject, or following processing of a sample obtained from a subject. Once the sample is collected, the subject is not required to be present for the in vitro method to be performed, and therefore the method may be one which is not practised on the human or animal body.

[0253] A sample may be taken from any tissue or bodily fluid. A sample may comprise or may be derived from: a quantity of blood; a quantity of serum derived from a subject’s blood which may comprise the fluid portion of the blood obtained after removal of the fibrin clot and blood cells; a tissue sample or biopsy (e.g. a bone marrow tissue sample or biopsy); pleural fluid; cerebrospinal fluid (CSF); or cells isolated from a subject. In some embodiments, the sample may be obtained or derived from a tissue or tissues which are affected by the disease / condition (e.g. tissue or tissues in which symptoms of the disease manifest, or which are involved in the pathogenesis of the disease / condition). In some embodiments, the sample may be obtained or derived from a cancer, tumor, or cells thereof. In some embodiments, in a method according to the present disclosure the sample comprises or is derived from a quantity of blood (e.g. peripheral blood). In some embodiments, in a method according to the present disclosure the sample comprises or is derived from bone marrow (e.g. a bone marrow tissue sample or biopsy).

[0254] In some embodiments, in a method according to the present disclosure the sample comprises mononuclear cells. For example, a sample (e.g. a blood or bone marrow sample) may be processed to separate / purify the mononuclear cells. In some embodiments, in a method according to the present disclosure the sample comprises bone marrow mononuclear cells (BM-MNCs) or peripheral blood mononuclear cells (PB-MNCs) and / or leukaemia stem cells (LSCs).

[0255] In some embodiments of the methods disclosed herein, the expression of one or more genes is determined in CD34+CD38- cells in the sample obtained from the subject. In some embodiments of the methods disclosed herein, the expression is determined relative to the expression in CD34+CD38- cells in the reference samples described herein (or in samples obtained from a reference population as described herein).

[0256] Determining the number / proportion of a given cell type / sub-type in any of the above embodiments, may comprise analysis using antibodies for the detection of markers for the given cell type / subtype as described herein, permitting the identification of such cells from within heterogeneous cell populations. The number / proportion / presence / absence of cells of a given cell type / subtype described herein can be determined by analysis by an appropriate method, e.g. by flow cytometry (e.g. multicolour flow cytometry) or immunohistochemistry, using antibodies providing for the detection of the given cell type / subtype as described herein.

[0257] The present disclosure provides a method for predicting whether a subject with CML is at risk of relapse following cessation of a treatment for CML. The present disclosure also provides a method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML. The present disclosure also provides methods of selecting a subject for treatment. In the methods disclosed herein, the expression of one or more genes regulated by PRC1 (as disclosed herein) may be determined / measured in a sample obtained from the subject and then compared to a predetermined average expression in reference samples as disclosed herein, or samples obtained from a reference population as disclosed herein.

[0258] By way of an example, an increased expression of the one or more genes regulated by PRC1 in the sample obtained from the subject relative to a pre-determined average expression that has been determined in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. By way of another example, a decreased expression of the one or more genes regulated by PRC1 in the sample obtained from the subject relative to a pre-determined average expression that has been determined in samples obtained from subjects with CML who achieve TFR following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML. By way of another example, a decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression that has been determined in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML. By way of another example, an increased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression that has been determined in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

[0259] By way of another example, a subject may be selected for a treatment disclosed herein if the expression of the one or more genes regulated by PRC1 is increased relative to a pre-determined average expression that has been determined in samples obtained from subjects with CML who achiever TFR following cessation of a treatment for CML. By way of another example, a subject may be selected for a treatment disclosed herein if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression that has been determined in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML. By way of another example, a subject may be selected for a treatment disclosed herein if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression that has been determined in samples obtained from subjects with CML who relapse following cessation of a treatment for CML. By way of another example, a subject may be selected for a treatment disclosed herein if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression that has been determined in samples obtained from subjects with CML who relapse following cessation of a treatment for CML.

[0260] Kits

[0261] Aspects of the disclosure include in vitro diagnostic / prognostic methods and in vitro kits for performing such methods. In some embodiments, the kit provides agents for performing a method according to the present disclosure. The kit may be used for any of the methods described herein. In some embodiments, the kit comprises a plurality of antigen-binding molecules as defined herein. In some embodiments, the plurality of antigen-binding molecules is capable of detecting one or more of PRSS21 , ANLN, SPAG6, SCN2A, BMI1 , ALDH1 A1 and TFRC. In some embodiments, the plurality of antigen-binding molecules is capable of detecting (or binding to) one or more of PRSS21 , ANLN, and SPAG6. In some embodiments, the one or more antigen-binding molecules is capable of detecting (or binding to) ALDH1 A1 . In some embodiments, the plurality of antigen-binding molecules is further capable of detecting (or binding to) one or more components of the NFKB pathway (e.g. a component of the TNFONFKB pathway).

[0262] In some embodiments, the plurality of antigen-binding molecules is capable of detecting one or more of PRSS21 , ANLN, SPAG6, SCN2A, BMI1 , ALDH1 A1 , TFRC, TNFa and PNFKB. In some embodiments, the plurality of antigen-binding molecules is capable of detecting one or more of PRSS21 , SPAG6, TNFa and PNFKB.

[0263] The kit may be suitable for a point-of-care in vitro diagnostic test. It may be a kit for laboratory-based testing. The kit may include instructions for use, such as an instruction booklet or leaflet. The instructions may include a protocol for performing any one or more of the methods described herein.

[0264] Numbered statements

[0265] The following numbered paragraphs (paras) describe particular aspects and embodiments of the present disclosure:

[0266] 1 . A method for predicting whether a subject with chronic myelogenous leukaemia (CML) is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by polycomb repressive complex 1 (PRC1) in CD34+CD38- cells in a sample obtained from the subject, wherein an increased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML.

[0267] 2. A method for predicting whether a subject with chronic myelogenous leukaemia (CML) is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by polycomb repressive complex 1 (PRC1) in CD34+CD38- cells in a sample obtained from the subject, wherein a decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML.

[0268] 3. A method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject, wherein a decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

[0269] 4. A method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject, wherein an increased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

[0270] 5. A method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0271] (a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and

[0272] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0273] 6. A method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0274] (a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and

[0275] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

[0276] 7. A method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:

[0277] (a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and

[0278] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0279] 8. A method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising: (a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and

[0280] (b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0281] 9. The method of any one of the preceding paras, wherein the one or more genes is selected from the list consisting of: PRSS21 , ANLN, SPAG6, SCN2A, BMI1 , TFRC and ALDH1 A1 .

[0282] 10. The method of para 1 , 3, 5 or 7, wherein the one or more genes is selected from PRSS21 , ANLN, SPAG6, BMI1 , and TFRC.

[0283] 11 . The method of para 2, 4, 6 or 8, wherein the one or more genes is selected from ALDH1 A1 and SCN2A.

[0284] 12. The method of any one of the preceding paras, wherein the CD34+CD38- cells are leukaemic stem cells (LSCs).

[0285] 13. The method of any one of the preceding paras, wherein the treatment for CML is a TKI, a BTT, a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

[0286] 14. The method of para 13, wherein the treatment for CML is a TKI or a BTT.

[0287] 15. The method of any one of the preceding paras, wherein the sample has been obtained from peripheral blood or bone marrow.

[0288] 16. The method of any one of the preceding paras, wherein the subject is a human.

[0289] 17. The method of any one of the preceding paras, wherein the expression of the one of more genes is determined using flow cytometry.

[0290] 18. The method of any one of the preceding paras, wherein the expression of the one or more genes is determined using one or more antigen-binding molecules.

[0291] 19. A kit suitable for performing the method according to any one of paras 1-18.

[0292] 20. The kit of para 19 wherein the kit comprises a plurality of antigen-binding molecules capable of detecting the one or more genes regulated by PRC1 . 21 . The kit of para 20, wherein the plurality of antigen-binding molecules is capable of detecting one or more of PRSS21 , ANLN, SCN2A, SPAG6, BMI1 , TFRC, and ALDH1 A1 .

[0293] 22. An antigen-binding molecule for use in a method according to any one of paras 1-18.

[0294] 23. The antigen-binding molecule of para 22, wherein the antigen-binding molecule is capable of detecting one or more of PRSS21 , ANLN, SCN2A, SPAG6, BMI1 , TFRC, and ALDH1 A1 .

[0295] ***

[0296] The present disclosure includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

[0297] The section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.

[0298] Aspects and embodiments of the present disclosure will now be illustrated, by way of example, with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.

[0299] Throughout this specification, including the claims which follow, unless the context requires otherwise, the word ‘comprise,’ and variations such as ‘comprises’ and ‘comprising,’ will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

[0300] It must be noted that, as used in the specification and the appended claims, the singular forms ‘a’, ‘an’, and ‘the’ include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from ‘about’ one particular value, and / or to ‘about’ another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent ‘about’, it will be understood that the particular value forms another embodiment.

[0301] Methods described herein may preferably be performed in vitro. The term ‘in vitro' is intended to encompass procedures performed with cells in culture whereas the term ‘in vivo’ is intended to encompass procedures with / on intact multi-cellular organisms.

[0302] Values may be expressed herein as ‘about’ a particular value. Similarly, ranges may be expressed herein as from ‘about’ a particular value, and / or to ‘about’ another particular value. The term ‘about’ in relation to a numerical value is optional, and means for example + / - 10 %. By way of illustration, reference e.g. to ‘about 10 %’ is to be construed as 9 % to 11 %. In instances herein where ‘about’ is recited, the value it precedes is also specifically contemplated. By way of illustration, reference e.g. to ‘about 10 %’ also specifically contemplates 10 %. Brief Description of the Figures

[0303] Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures.

[0304] Figure 1. Only a minority of CP CML patients achieve TFR. Patients with ELN Optimal responses (green curve) and DMR are eligible for a trial of TKI cessation. Of these patients, ~50% will relapse (purple line), usually within one year, and almost all relapsing patients will regain MMR upon re-treatment.

[0305] Figure 2. Multiple mechanisms of resistance contribute to LSC survival in CML. The mechanisms by which CML cells, including LSCs, evade BCR-ABL1 -targeting therapies are depicted. Boxed region highlights LSC-specific mechanisms. Adapted from Ng and Ong (2022).

[0306] Figure 3. LSC model of TFR. (A) In Optimal responders with TFR, TKI-sensitive LSCs are eradicated over time. Long-lived BCR-ABL1 -positive B cells persist under TKI treatment, but cannot give rise to disease relapse at TKI stop. (B) In relapse, TKI-resistant LSCs persist despite BCR-ABL1 inhibition, give rise to recurrent disease & TKI-sensitive progeny upon TKI stop, but respond well to TKI re-institution.

[0307] Figure 4. A single cell atlas of normal, TFR Remission and Relapse. (A) UMAP combining Normal, Remission, and Relapse BM samples. MKP=megakaryocyte progenitor. NEP=neutrophil progenitor. (B) Individual UMAPs of Normal, Remission, and Relapse BM. (C) Left-hand panel: HSC / LSC proportions in Normal (N), CML BM samples for the current TFR dataset, and for our published dataset. Right-hand panel: ERP proportions for the same datasets described for the left-hand panel. (D) GSEA of BCR::ABL1+ LSCs vs normal HSCs.

[0308] Figure 5. Identification of prognostic cell types and GE signatures. (A) Volcano plot DEGs in BCR- ABL1+ LSCs from Relapse and Remission Groups. (B) A 10 gene expression pattern that distinguishes Relapse from Remission BCR::ABL1+ LSCs. (C) Confusion matrix of model performance for a Relapse vs Remission classifier based on the genes in (B). Accuracy is the ratio of true positives and negatives to all positive and negative observations. SPAG = SPAG6, PRSS = PRSS21 , SCN = SCN2A.

[0309] Figure 6. PRSS21 is expressed in PB and BM CD34+ cells of Relapse samples. (A) UMAP feature plots coloured for PFSS expression in Normal, Remission, and Relapse BM. (B) PFSS expression in BCR-ABL1+ LSCs & BCR-ABL1- HSCs was plotted for Normal, Remission, and Relapse samples. Each dot represents one patient sample. (C) PFSS expression in different CD34+ lineages was plotted as in (B). (D) UMAP of PB and BM MNCs from the same CML patient with Relapse, coloured for PFSS expression. (E) CD34+ cells from 2 Remission and 2 Relapse pre-treatment BM samples. RT-PCR measuring PRSS21 (PRSS) is expressed as fold-change ratio over TBP. MKP / ERP / NEP = megakaryocyte / erythroid / neutrophil progenitors. PRSS = PRSS21.

[0310] Figure 7. PRSS21, ANLN, SPAG6, SCN2A, BMI1, ALDH1A1 and TFRC are differentially expressed in Remission and Relapse samples. Single-cell RNA sequencing (scRNA-seq) data showing expression in Remission and Relapse BM of (A) PRSS21 , (B) ANLN, (C) SPAG6, (D) SCN2A, (E) BMI1 , and (F) ALDH1 A1 . Each dot represents one patient sample. (G) Flow cytometry data showing expression of TFRC in Remission and Relapse BM. Each dot represents one patient sample.

[0311] Figure 8. GSEA validation of the TFR single cell atlas. (A) GSEA comparing BCR-ABL1+ LSCs and normal BM HSCs was performed using Hallmark MSigDB signatures reading out for established LSC features. (B) GSEA analysis was performed as in (A) using gene sets of canonical PRC2 components and PRC2 target genes in CML (Ko gene set) and prostate cancer (Kondo gene set). NES=normalized enrichment score; FDR=false discovery rate. Red / blue font indicates enrichment / depletion in BCR-ABL1 + LSCs vs normal HSCs.

[0312] Figure 9. GSEA & MSigDB Hallmark gene sets. (A) GSEA comparing BCR::ABL1+ Relapse and Remission LSCs was performed using Hallmark signatures. Each gene set was plotted according to its normalized enrichment score and FDR=false discovery rate. (B) Enrichment plots of representative gene sets.

[0313] Figure 10. GSEA-based functional annotation of LSC epigenomes. (A) GSEA comparing BCR-ABL1 .

[0314] LSCs from Relapse and Remission patients was performed using epigenetic gene sets. Each gene set was plotted according to its normalized enrichment score and false discovery rate. (B) PRC1 -related gene sets. (C) PRC2-related gene sets. All gene sets were generated from primary CD34+ CML cells with ChlP-seq (with antibody to BMI1 or EZH2) to identify PRC1 / 2-bound genes, or from in vitro BMI1 (PTC209) or EZH2 (GSK126) inhibitor treated primary CD34+ CML cells (Ko et al. 2020).

[0315] Figure 11. Colony forming assays with normal, Relapse and Remission BM. CD34+ cells were plated in methylcellulose with inhibitors to PRC1 (PTC209) and PRC2 (GSK126). Colonies were counted after 10 days and plotted as a percentage of colonies in untreated cells.

[0316] Figure 12. PRSS21 is dependent on PRC1 activity and contributes to regulation of PRC1- dependent genes. (A) CD34+ CML cells were treated with PTC209 & harvested for PRSS21 RT-PCR after 24 & 48hrs. (B) GE profiles in K562 CML cells were analyzed by GSEA following PRSS21 and scrambled control knockdown. PRSS = PRSS21

[0317] Figure 13. Model of aberrant PRC1 activation in Relapse LSCs. ncPRCI .x activity in Relapse LSCs leads to activation of genes important in mediating cellular features (arrowheads) contributing to LSC persistence during DMR, and relapse. For clarity, cPRC1 is not drawn in Relapse LSCs.

[0318] Figure 14. Schematic defining ncPRCI .x function in Relapse vs Remission LSCs.

[0319] Figure 15. ENCODE validation of CUT&Tag-derived histone marks on CML cells. K562 CML cells were grown in media and processed as per published protocols. H3K27me3 and H3K4me3 marks from CUT&Tag data was compared to ENCODE ChlP-seq tracks in the same cell line. Figure 16. GSEA-based functional annotation of LSC epigenomes. (A) Differentially methylated genes in primary CD34+ CML cells (left-hand panel), or DNMT inhibitor-sensitive genes (right-hand panel). (B) GSEA comparing BCR::ABL1 + LSCs from Relapse and Remission patients was performed using epigenetic gene sets for MLL target genes, and (C) CFAs were performed on CD34+ cells from NBM, Relapse and Remission samples, decitabine treated, and colonies counted at 14 days.

[0320] Figure 17. scATAC-seq of CML LSCs. (A) UMAP of scATAC-seq-defined cell clusters from a cohort of 23 CP BM MNC samples. (B) Browser views of peaks corresponding to open chromatin regions surrounding GATA1 .

[0321] Figure 18. CFSE and CP engraftment assays. (A) CD34+ CP cells were treated with DMSO or imatinib, stained for CFSE, and cultured in vitro. At 4 days, cells were harvested and analyzed as previously described (Graham et al. 2002). As expected, a quiescent non-dividing population of CP cells was observed to increase in IM treated cells. (B) Mice were injected with 7x106CD34+ CP CML cells by tail vein after sub-lethal radiation. PB was sampled at 2 & 8wks post-injection and assessed for human CD34+CD45+ (hCD45) cells by MFC, plotted as a % of murine CD45+ (mCD45) cells. Successful engraftment defined as >0.5% hCD45+ cells.

[0322] Figure 19. Graphs showing CML LSCs GSEA comparison of Relapse vs Remission.

[0323] Figure 20. (A) Plot showing E2F pathway gene expression in CML relapse v CML remission samples. The CML relapse samples showed distinct expression profiles as separated by the dashed line (B) PCA plot with CML LSC group-specific differentially expressed genes (DEGs) as the input. Circles-females, squares-males.

[0324] Figure 21. (A) Gene expression differential of genes in pathways including E2F, MYC, MTOR, lipid metabolism in Relapse-P v healthy; Relapse-P v Remission; and Relapse-I v Remission. (B) Enrichment plots for select genes enriched in Relapse-P samples.

[0325] Figure 22. (A) Gene expression differential of genes linked to immune response, NFkB, epigenetics and E2F / cell cycle / MYC in Relapse-I v healthy; Relapse-I v Remission; and Relapse-P v Remission. (B) Enrichment plots for select genes enriched in Relapse-I samples.

[0326] Figure 23. Schematic of cell-cell communication analysis (Multi-nichenet).

[0327] Figure 24. (A) Chord plots showing the number of cell-cell interactions occurring between all CML BM cell types in Relapse-I, Remission, and Relapse-P samples. (B) Graph showing the number of cell-cell interactions received by CML LSC in Relapse-I, Remission, and Relapse-P samples.

[0328] Figure 25. Heatmap visualizing the regulatory potential of specific ligands in Relapse-I leukemic stem cells (LSCs) by assessing their known downstream gene targets. Each row represents a ligand, and each column represents a differentially expressed gene (DEG) upregulated in Relapse-I compared to Relapse- P and Remission LSCs. The intensity of the grey color indicates the strength of prior evidence that the ligand regulates that gene: darker grey signifies a stronger known regulatory link. The scaled ligand activity score, used to rank ligands, is based on the overlap between known targets of each ligand and the observed DEGs, combining precision (fraction of predicted targets that are true DEGs) and recall (fraction of DEGs that are predicted targets). This score corresponds to the area under the precisionrecall curve (AUPRC), providing a robust measure of ligand relevance in Relapse-1 biology.

[0329] Figure 26. Differential pathway activity within the LSCs of relapse-P and relapse-1 patients is biologically transmitted through the hematopoietic hierarchy. This heatmap displays the top 50 gene sets significantly enriched (FDR <0.25) in the Relapse-1 vs Remission comparison across 12 specified bone marrow cell types. Gene sets (rows) are from three curated GSEA collections. Each cell shows the normalized enrichment score (NES), with red indicating enrichment (NES > 0) and blue indicating depletion (NES < 0), while white cells reflect non-significant results (FDR > 0.25). Gene sets were ranked by the number of significant enriched cell types. (A) Top 50 enriched (FDR<0.25, fixed order). (B) Relapse-I v Remission Top 50 enriched (FDR<0.25).

[0330] Figure 27. Heatmap showing genes that are differentially upregulated in Relapse-I LSCs compared to both Relapse-P and Remission LSCs. Each row represents a ligand, and each column represents a gene target. (B) The intensity of the grey color indicates the strength of prior evidence linking the gene as a known target of the corresponding ligand. The size of each bubble reflects the correlation between the expression of the ligand and its receptor with the expression of the target gene across individual patients. This visualization highlights potential ligand-receptor-target axes active in the Relapse-I subtype

[0331] Examples

[0332] Example 1 : Generation of a treatment-free remission (TFR) single cell atlas

[0333] A single cell atlas (TFR Atlas) of pre-treatment bone marrow mononuclear cells (BM MNCs) from 21 patients who achieved deep molecular responses (DMRs), and subsequently discontinued TKI therapy from the Australian TWISTER (Ross et al. 2013) and Singaporean PRISM studies, was generated (Figures 4A and 4B). Of these, 8 patients had durable TFRs after TKI continuation (Remission Group) while 13 had molecular relapse (Relapse Group). Consistent with previous findings (Krishnan et al. 2023), the TFR Atlas demonstrated accurate cell clustering, with relative decreases in hematopoietic stem cells (HSCs) and leukaemic stem cells (LSCs) and increase in erythroid progenitors (ERP) respectively, compared to normal bone marrow (Figure 4C). Also consistent with previous findings (Krishnan et al. 2023), CML / normal LSC / HSC gene sets were enriched / depleted in the equivalent LSC / HSC clusters in the present atlas (Figure 4D).

[0334] Example 2: Machine learning to identify prognostic cell types and gene expression (GE) signatures The inventors set out to identify differentially expressed genes (DEGs) between Relapse and Remission cell clusters. LSCs and primitive progenitors (LSC / HSCs) had the most DEGs (8 or more), and differentiated cell types, including immune cells, all had 0 DEGs. Focusing on LSCs, and using a BCR::ABL1 -specific gene expression signature (GE) previously validated to identify BCR::ABL1 fusion gene-positive SPCs (Krishnan et al. 2023), 10 DEGs were identified between BCR::ABL1 + LSCs from Relapse and Remission Groups (Figures 5A and 5B). Next, in an independent machine learning (ML) approach which exploits the predictive power of gene combinations (Krishnan et al. 2023), GE profiles that could classify cell clusters to either Relapse or Remission Groups were identified. LSC GE combinations were the most informative and included DEGs that had already been identified (e.g., PRSS21 , SPAG6, and SCN2A, ) (Figure 5B). Impressively, the GE combinations segregated Relapse and Remission LSCs with 95% accuracy (Figure 5C).

[0335] Taken together, it can be concluded that Remission and Relapse LSCs are biologically distinct, and that their DEGs represent potential biomarkers for predicting TFR success or relapse at diagnosis.

[0336] Example 3: GE signatures segregate Relapse and Remission CD34+ cells in bone marrow (BM) and peripheral blood (PB)

[0337] To translate prognostic GE signatures to clinical tests (e.g., flow cytometry), the set of predictive genes from Example 2 (Figures 5A and 5B) were evaluated for consistency of expression across different cell types. PRSS21 expression was found to span the entire population of CD34+ stem and progenitor cells (SPCs) from Relapse but not Remission bone marrow (Figure 6A) and was largely restricted to BCR::ABL1 + SCs (Figure 6B).

[0338] Importantly, significant PRSS21 expression differences persisted between Relapse and Remission samples in multiple lineages (Figure 6C) and could also be detected in peripheral blood MNCs (Figure 6D). gRT-PCR performed on CD34+ cells from two each of pre-treatment Remission and Relapse BM confirmed significantly increased PRSS21 expression in the latter (Figure 6E).

[0339] Single-cell RNA seguencing (scRNA-seg) and flow cytometry were performed in peripheral blood MNCs to study the expression of seven candidate DEGs. Significant differences in the expression of PRSS21 , ANLN, SPAG6, SCN2A, BMI1 , ALDH1A1 , and TFRC were observed (Figure 7). Specifically, the expression of PRSS21 , ANLN, SPAG6, BMI1 and TFRC was shown to be significantly increased in Relapse BM compared with Remission BM. The expression of SCN2A and ALDH1A1 was shown to be significantly increased in Remission BM compared with Relapse BM.

[0340] Taken together, these results demonstrate that diagnostic Relapse and Remission HSPCs are distinguishable by a set of genes, including in peripheral blood. These important findings set the stage for the development of pre-treatment TFR biomarkers.

[0341] Example 4: GSEA-based annotation of hallmark features of BCR::ABL1 + LSCs

[0342] The TFR Atlas of Example 1 was evaluated for its ability to functionally annotate biologic processes known to distinguish CML LSCs and normal HSCs. The dataset correctly reported for previously published features of CML LSCs: enrichment of proliferation, oxidative phosphorylation, and mTOR signalling, and depletion of TNFO-NFKB and KRAS signalling (Kuntz et al. 2017 and Giustacchini et al. 2017) (Figure 8A). The TFR Atlas was also tested for its ability to functionally annotate the LSC epigenome, specifically for increased expression of components of polycomb repressive complex 2 (PRC2) as well as increased PRC2 activity (readout as silencing of target genes) (PRC2 has been extensively characterized for expression and function in primary human CML CD34+CD38+ SPCs (Scott et al. 2016 and Xie et al. 2016). The TFR Atlas reported significant enrichment of PRC2 components and function (Figure 8B).

[0343] Taken together, these results demonstrate that the TFR Atlas of Example 1 faithfully captures multiple CML LSC features, from sternness to epigenetic programming.

[0344] Example 5: Relapse and Remission BCR::ABL1+ LSCs are biologically distinct

[0345] The inventors set out to identify biologic differences between Relapse and Remission LSCs, and interrogated these cells with the 50 Hallmark MSigDB gene sets (Figure 9A). Relapse LSCs were found to be hyperproliferative vs. Remission LSCs, and had diminished interferon-alpha (IFNa) signalling (Figure 9B). These results may explain two TFR-related observations: increased proliferation likely contributes to slower BCR::ABL1 transcript declines in relapsing patients (Shanmuganathan et al. 2021), and enrichment of IFNa signatures (but not IFNy signatures, not shown) in Remission LSCs is consistent with the anti-LSC effects of IFNa (Hsieh et al. 2021 and Jun et al. 2021).

[0346] Example 6: GSEA points to the existence of a ncPRCI that activates BCR::ABL1 + Relapse LSCs Next, and because somatic mutations are uncommon among chronic phase (CP) patients achieving DMRs compared to those who progress (Branford et al. 2018 and Kim et al. 2017) (and therefore unlikely to contribute to differential LSC GE signatures) the inventors focused on identifying potential epigenetic contributors.

[0347] Using GSEA, multiple gene sets associated with PRC1 , PRC2, DNA methylation, MLL / Trithorax, and histone deacetylation were found to be recurrently enriched in Relapse vs Remission LSCs (Figure 10A). Focusing on PRC1 / 2, it was determined whether GE differences between Relapse and Remission LSCs suggested that they were differentially active. Gene sets comprising genes bound by PRC1 in primary CD34+ CML cells were found to be enriched in Relapse LSCs. CML-specific BMI1 inhibitor perturbation gene sets (representing genes induced by BMI1) were also enriched (Figure 10B). These findings suggest that a putative BMI1-containing PRC1 complex is activating GE in Relapse LSCs, which the inventors term ncPRCI .x (since canonical PRC1 silences genes). These observations are reminiscent of another leukaemogenic ncPRCI complex, ncPRCI .1 , recently discovered in primary AML cells that also activates GE but does not contain BMI1 (Maat et al. 2021). In contrast, for PRC2, neither PRC2-bound genes nor PRC2 inhibitor-dependent genes were differentially enriched (Figure 10C).

[0348] Taken together, these results indicate that increased BMI1 activity, likely as part of a ncPRCI complex that activates GE, is a potentially targetable feature of Relapse LSCs. Example 7: Relapse LSCs are preferentially sensitive to BMI1 inhibitors but not EZH2 (PRC2) inhibitors

[0349] To test the hypothesis that Relapse LSCs may rely preferentially on BMI1 , but not PRC2, for survival compared to Remission LSCs, CD34+ cells from Relapse and Remission patients and normal BM were treated with PTC209 and GSK126, to inhibit BMI1 and EZH2 (the enzymatic component of PRC2) respectively. Using colony forming assays (CFA), Relapse samples were found to be 2-fold more sensitive to PTC209, but not to GSK126, than Remission and normal L / HSCs (Figure 11). These data suggest that Relapse LSCs are more reliant on BMI1 activity for survival than Remission LSCs and normal BM HSCs.

[0350] Example 8: BMI1 regulates Relapse DEGs and glycolysis

[0351] It was next determined if Relapse DEGs (Figures 5 and 6) might also contribute to the biology of Relapse. Accordingly, primary CD34+ CML cells were treated with PTC209. CD34+ CML cells were found to downregulate PRSS21 and ANLN (Figure 12A). PRSS21 function was subseguently evaluated. PRSS21 knockdown in CML cells lines resulted in significant changes in known BMI1-bound and -regulated genes, together with changes in glycolytic pathway genes (Figure 12B) and mRNA translation (not shown).

[0352] Taken together, these results suggest that BMI1 contributes to LSC metabolism via activation of PRSS21 . Interestingly, ncPRC1.1 also activates genes involved in glycolysis such as PKM and LDHA6, which were also found to be downregulated in CML cells upon PRSS21 knockdown (not shown).

[0353] In summary, Examples 1-8 suggest the existence of a non-canonical, BM1 -dependent PRC1 complex (ncPRCl .x) which is aberrantly activated in BCR::ABL1+ Relapse vs Remission LSCs, and which activates GE programmes regulating metabolic pathways and survival, including PRSS21. In turn, these GE programmes regulate important cellular programmes such as glycolysis, survival, and proliferation. The above results suggest that these programmes are targetable with BMI1 inhibitors. The inventors hypothesize that, like ncPRCI .1 , ncPRCI .x will bind to transcription start sites (TSSs) as opposed to gene bodies, bind DNA independently of H3K27me3, and activate expression of a unique set of genes (van den Boom et al. 2016) (Figure 13). Thus, ncPRCI .x-regulated genes (such as PRSS21 and ANLN) are potential biomarkers for relapse. Therapeutic targeting of this “Relapse epigenome” may extinguish persistent LSCs and prevent relapse in CML patients.

[0354] Example 9: Validation of diagnostic biomarkers in an independent Japanese cohort

[0355] The preliminary results of Examples 1-9 (Figures 4-10) will be validated in a cohort of 18 TFR, and 11 relapsing patient samples. This will be achieved by testing biomarkers in BM MNCs as follows: qRT-PCR on CD34+ cells for the DEGs already identified (Figures 5 to 7), multi-colour flow cytometry (MFC) with novel antibodies to DEGs, and by scRNA-seq (Figures 6 and 7).

[0356] It is expected that qRT-PCR on CD34+ cells will confirm that the identified DEGs are enriched in Relapse samples (Figures 5 to 7). It is also expected that MFC on CD34+ progenitors, including the relatively abundant CD34+CD71+CD105+ ERPs we previously observed in CML BM, will demonstrate increased protein expression of the seven DEGs (identified in Examples 1-8, i.e. PRSS21 , ANLN, SPAG6, BMI1 , TFRC, SCN2A and ALDH1 A1) in Relapse vs Remission BM MNCs. It is also anticipated that the scRNA- seq will confirm the pre-treatment DEGs, cell clusters, and pathways identified.

[0357] Example 10: Confirmation of the persistence and expansion of Relapse LSCs at TKI-stop and relapse To directly determine if Relapse LSCs persist at TKI-stop, cryopreserved BM from patients who maintained TFR and those who relapsed taken at TKI-stop (-10 Relapse and -10 Remission BM), and at TKI re-start (-10 BM), will be used from the TWISTER study. As done previously (Krishnan et al. 2023), single BCR::ABL1+ LSCs will be identified at these time points, and they will be compared transcriptomically to diagnostic LSCs from Remission and Relapse patients.

[0358] For the samples at TKI-stop, it is anticipated that the Relapse and Remission samples will each more closely approximate the cell cluster ratios found in normal BM, i.e. normalization of HSC / LSC and ERP proportions (Figure 4). It is also anticipated that BCR::ABL1+ LSCs will be detected that retain features of the diagnostic Relapse LSCs in Relapse but not Remission BM at TKI-stop. For the samples assessed at the time of TKI re-start / disease recurrence, it is anticipated that cell types with similar GE signatures to those detected at diagnosis will be rapidly enriched. Given that Relapse DEGs are detectable in CD34+ cells (Figures 6A and 6C and Figure 7), it is anticipated that qRT-PCR will be able to detect the re- emergence of, e.g., PRSS21 / ANLN / SPAG6 / BMI1 / TFRC+ cells at recurrence. Given the rarity of LSCs at DMR, it may be necessary to also pool several samples together and / or sequence much larger numbers of input cells.

[0359] Example 11 : Antibody development

[0360] Out of the identified DEGs, none have viable antibodies that are commercially available. Accordingly, monoclonal antibodies to these antigens will be developed. Sensitivity and specificity will be tested in appropriate cell lines, e.g. the antibody to PRSS21 will be tested in K562 cells (which naturally express PRSS21) and compared to non-PRSS21 -expressing cells (e.g. KCL22) or K562 cells in which PRSS21 has been knocked down. Similar experiments will be performed for the other identified DEGs.

[0361] Example 12: Determining the contribution of epigenetic factors to LSC-driven relapse Preliminary studies (Figures 10-12) are consistent with a BMI1 -dependent ncPRCI activating genes and survival programmes contributing to relapse (Figure 13). This model will be tested in three ways: 1) identification of genes regulated by ncPRCI .x in Relapse vs Remission LSCs by the constellation of H2AK119ub and H3K4me3 marks, and decreased H3K27me3 compared to cPRC1 sites; 2) confirmation that BMI1 contributes to Relapse LSC GE signatures; and 3) Genetically identifying components of ncPRCI (Figure 14).

[0362] ‘Cleavage Under Targets and Tagmentation’ (CUT&Tag™, Active Motif) will be performed using ChlP- grade antibodies to BMI1 , H2AK119, H3K4me3, and H3K27me3 on Relapse and Remission LSCs. This approach can be used on small numbers of cells (5000) and will allow the identification of genomic regions where BMI1 is binding as well as exerting its enzymatic activity (ubiquitinylation of K119 on H2A). As a control, the depletion of PRC2 (EZH2)-binding at these sites will also be confirmed. In order to confirm the set of genes regulated by ncPRCI .x (i.e. bound by BMI1 and associated with H2AK119 marks), Relapse and Remission LSCs will be treated in vitro with BMI1 / EZH2 inhibitors for 48 hours, following which cells will be harvested and analysed for GE changes, as described previously (Ko et al. 2020). The use CUT&Tag™ was verified in CML cell lines (Figure 15), and so no significant technical problems are expected for this assay. Finally, the subset of known canonical and non-canonical PRC1 components will be knocked down, and the effect on GE and cell function will be examined.

[0363] It is expected that genes will be identified in Relapse but not Remission LSCs that exhibit the constellation of histone marks in Figure 14, and that these genes will contribute to proliferation and IFNa signatures (Figure 9B) and include PRSS21 and ANLN (Figure 12A). It is also expected that PTC209 treatment and knockdown of specific c / ncPRC1 components will impair GE expression of ncPRCI .x- bound genes and impair proliferation and / or survival of Relapse compared to Remission LSCs. If ncPRCI .x does not contribute to Relapse LSC GE profiles, other epigenetic processes will be investigated from those identified (Figure 10A), including DNA methylation- and MLL-related signatures (Figures 16A and 16B). Of note, the methylation signatures suggested increased sensitivity of Relapse LSCs to HMA, which will be confirmed (Figure 16C). MLL1 function can be analysed by CUT&Tag with ChIP grade antibodies to MLL1 and H3K4me, while DNA methylation-regulated genes may be annotated using HMA treatment of cells, as we have shown (Ko et al. 2020). Finally, GRNs will be explored as a putative mechanism explaining the LSC observations.

[0364] Example 13: Identification of gene regulatory networks (GRNs) maintaining Relapse and Remission LSC states

[0365] Gene regulatory networks (GRNs), comprising a set of master TFs together with their cognate binding sites on target genes form higher order complexes that maintain cell states, including HSC states (Goode et al. 2016 and Gottgens 2015). About 50 TFs have been shown to affect HSC function (Gottgens 2015), and given that GRNs are often subverted during the process of transformation (Edginton-White and Bonifer 2022, Assi et al. 2019, and Agrawal-Singh et al. 2023), we hypothesize that Relapse LSCs utilize differential GRNs from Remission LSCs to maintain the former’s self-renewal programmes as well as the capacity to survive in a BCR::ABL1-independent manner. With the availability of scATAC-seg, it has now become possible to computationally infer GRNs at the single cell level (Aibar et al. 2017), and using such information, dismantle cancer-specific GRNs (Achinger-Kawecka et al 2024) for therapeutic purposes including in human leukaemias (Zaugg et al. 2022, Claringbould and Zaugg 2021 , and Xu et al. 2022).

[0366] CD34+ and CD34- cells will be purified from the Japanese cohort BM MNCs and subjected to scATAC- seg (10X Genomics). As shown in Figure 17, using a different CML sample set, scATAC-seg was successfully used to confirm that CML LSCs have open chromatin regions around the GATA1 gene that was originally predicted by the scRNA-seg analysis using SCENIC (Giustacchini et al. 2017). SCENIC+ will be used to infer GRNs from the TFR scATAC-seg dataset generated (Bravo Gonzalez-Blas et al 2023). In brief, SCENIC+ identifies putative enhancers, enriched TF binding sites within those enhancers, and by linking TFs to those enhancers and target genes, construct GRNs. Identification of the set of master TFs which maintain those cell type-specific cell states via GRNs will suggest putative regulatory nodes which can be targeted experimentally to functionalise the scATAC-seg analysis. For the current TFR dataset, the inventors expect to identify differential GRNs, including critical TFs and enhancers, that contribute to maintenance of Relapse but not Remission LSC states. The inventors will inactivate these TFs genetically in primary SPCs and determine their effect on viability, proliferation, and TKI sensitivity, as previously described (Ko et al. 2020). If Relapse-specific enhancers are identified, enhancer activity can be confirmed using plasmid reporter assays, and then via CRISPR-based approaches, mutated to inactivate their function. Analogous work in other leukaemias, including CLL and AML, have used such approaches to identify specific TFs critical to their survival, and facilitated identification of novel therapies to dismantle these networks (Assi et al. 2019 and Ott et al. 2018). Although potential drugs to target Relapse LSCs have been identified (Figures 11 and 16), this approach may identify additional targets and drugs, including already FDA-approved ones.

[0367] It is anticipated that the results will identify TFs and enhancers that are critical to Relapse LSC states. While it is challenging to pharmacologically target TFs and enhancers directly, several examples of this approach have been successful in identifying drugs which are capable of dismantling such networks by evaluating either upstream or downstream dependencies. Specific examples include the use of BET inhibitors to target a PAX5 super enhancer in CLL71 , DOT1 L inhibitors in MLL-fusion-driven Chromatin networks (Bernt et al. 2011), and disruption of protein-protein interactions by Menin-MLL interactions (Borkin et al. 2015). Recently, epigenetic modulators have also been successfully targeted with the use of PROteolysis TArgeting Chimeras (PROTACs) (Wu et al. 2020), and these can also be designed to target key TFs maintaining GRNs that are uncovered.

[0368] In addition, it has recently been demonstrated that a network pharmacology approach (Mogrify®) can also be used for drug discovery and repurposing to convert between two leukaemic cell states (Lee et al. 2022). All-trans retinoic acid (ATRA) was correctly identified by the inventors as a top-ranked drug to reverse transcriptional states inferred using RNA-seq data from ATRA-sensitive acute promyelocytic leukaemia (APL) cells, and in addition, identified novel drug combinations to overcome ATRA-resistant APL (Lee et al. 2022). Here, Mogrify® would be used to convert Relapse to Remission LSC states, rendering the former sensitive to standard BTTs.

[0369] Example 14: Determining activity of PRC1 inhibitors in Relapse vs. Remission LSCs Preliminary Studies have identified putative epigenetic processes as being differential active in BCR::ABL1+ Relapse vs Remission LSCs. Initial tests of BMI1 inhibitors and hypomethylating agents (HMAs) have shown preferential activity against Relapse compared to Remission HSPCs (Figures 11 and 16C). This panel of drugs will be tested against LSCs directly in in vitro and in vivo assays. In addition, the ability of putative ncPRCI .x targets, including PRSS21 , to rescue the effects of BMI1 inhibitors will be explored. These experiments will help determine how BMI1 inhibitors work, e.g. via PRSS21 downregulation.

[0370] Three complementary in vitro assays will be performed to compare the effectiveness of the combination therapies vs CML LSCs. These include: 1) Standard LTC-IC using CD34+ CML cells to determine effect on stem cells and primitive progenitors; 2) Serial-replating assays which readout for LSC self-renewal capacity in vitro; and 3) Carboxyfluorescein succinimidyl ester (CFSE) incorporation assays, which track cell division overtime, to assess effects on quiescent CML SPCs (Figure 18A). These experiments will also be repeated with forced expression of PRSS21 to determine PRSS21 downregulation is required for BMI1 inhibitor activity.

[0371] The inventors have recently developed the ability to engraft primary CP cells in immunodeficient mice using the NRG-W41 strain (currently the only model that permits long-term CP engraftment) (Figure 18B) (Chuprin et al. 2023 and Scott et al. 2023). The drugs will be tested in mice engrafted with Relapse and Remission samples. Mice will be treated with the drugs, and persistent engraftment will be measured by serial sampling of peripheral blood and confirmed in BM upon termination.

[0372] It is expected that BMI1 inhibitors and HMAs will preferentially target Relapse LSC. It is also expected that PRSS21 re-expression will rescue some or all the effects of BMI1 inhibitors on GE and cell function.

[0373] Example 15: Identification of Relapse-P and Relapse-I subtypes

[0374] 15.1 Materials and Methods

[0375] Gene-set enrichment analysis (GSEA): Pseudobulk leukemic stem cell (LSC) single-cell transcriptomes were analyzed using GSEA (v4.1 .0) (Subramanian A., et al., Proc Natl Acad Sci U S A.

[0376] 2005;102(43):15545 15550). Hallmark gene-sets as well as gene-sets for stem, leukemic and progenitor functions, CML pathogenesis, and epigenetics were curated from MSigDB v7.2 and other published studies 1000 gene-set permutations were carried out and weighted enrichment statistics were calculated while the signal-to-noise ratio was used as the metric for ranking genes. Enrichment scores were normalized using meandiv and gene-sets with FDR <0.25 were considered as significantly enriched. Normalized enrichment scores (NES) and FDR calculations were performed for the following 6 pair-wise comparisons (1) Relapse vs Remission (2) Relapse-P vs Remission (3) Relapse-I vs Remission (4) Relapse-P vs healthy (5) Relapse-I vs healthy (6) Relapse-P vs Relapse-I.

[0377] Gene Set Variation Analysis (GSVA) was performed to estimate pathway activity variation across samples in an unsupervised manner (Hanzelmann et al., BMC Bioinformatics. 2013 Jan 16:14:7). Expression data were normalized prior to analysis, and genes with low expression were filtered out. GSVA was conducted using the GSI / A R package with default parameters, to transform gene expression data into gene set enrichment scores. The analysis was carried out using the MsSigDB v7.2 Hallmark gene-sets as well as gene-sets for stem, leukemic and progenitor functions, CML pathogenesis, epigenetics and other published studies as input. The resulting GSVA scores were used for downstream analyses and pathways with an p-value < 0.05 were considered significantly enriched.

[0378] MultiNicheNet Analysis: MultiNicheNet was applied to infer condition-specific changes in cell-cell communication from single-cell RNA-seq data (Browaeys et al., BioRxiv 2023.06.13.544751). Cells were annotated into sender and receiver populations based on established markers. Pseudobulk aggregation was performed per cell type per sample to enable robust differential expression analysis using the muscat framework, followed by filtering of ligands and receptors based on expression and specificity.

[0379] Differentially expressed genes in receiver cell types defined the target gene sets. MultiNicheNet integrated ligand-receptor interactions, ligand-target regulatory potential, and expression data across samples to prioritize condition-specific ligand-receptor pairs. Ligand activity scores were computed based on the ability to explain target gene expression changes, and interactions were ranked using a composite score reflecting expression, specificity, and predicted downstream effects. Visualizations were generated using built-in plotting functions.

[0380] 15.2 Results

[0381] GSEA was performed on CML Ph+ LSCs identified from the scRNA-seq of diagnostic bone marrow samples of CML patients who either relapsed or remained in remission after TKI cessation. The input gene expression profiles were used to identify enriched biological processes. Pathway analysis revealed strong upregulation of cell cycle and DNA replication pathways in relapse samples (Figure 19), suggesting an activated LSC at the time of diagnosis of CML patients who relapse after TKI cessation.

[0382] GSEA on the E2F transcription factor target gene set revealed a subset of relapse patients with high E2F activity (Figure 20A). The genes shown in the figure are “leading-edge” genes following the GSEA for the E2F pathway. This group was labeled “Relapse-P” (proliferation-primed, to the right of the dashed line in Figure 20A), while the others, lacking this enrichment, were defined as “Relapse-I” (immune subtype, to the left of the dashed line in Figure 20A). Both GSVA (Gene-set variation analysis, data not shown) as well as a PCA plot (Figure 20B) displaying the CML LSC differentially expressed genes, affirmed the existence of 2 biological subtypes of CML relapse patients at the point of diagnosis.

[0383] Using GSEA, the transcriptional profiles of the Relapse-P subtype were compared to remission LSCs. Enrichments were observed for multiple CML-intrinsic and drug resistance pathways, including E2F, MYC, mTOR, and lipid metabolism (Figure 21 A and Figure 21 B). These findings suggest that Relapse-P LSCs are driven by proliferative, intrinsic programs potentially linked to TKI resistance. In contrast, analysis of the same pathways in the Relapse-I subtype revealed their downregulation relative to the Remission group.

[0384] GSEA was conducted on Relapse-I vs Remission LSCs to characterize the immune subtype. Only a small number of pathways (20 out of 7,950) were significantly upregulated, primarily involving immune signaling and NF-KB activation (Figure 22A and Figure 22B). This suggests that Relapse-I LSCs rely on extrinsic immune-mediated cues rather than intrinsic proliferation signals. In contrast, analysis of the same pathways in the Relapse-P subtype revealed their downregulation relative to the Remission group.

[0385] Normalized Enrichment Scores (NES) from FDR-filtered GSEA (FDR < 0.25) were visualized across 12 bone marrow cell types for the Relapse-I vs Remission comparison. Heatmaps were created using tidy- reshaped NES matrices from GSEA results filtered by a single comparison and cell-type subset. Gene sets were ranked based on the number of significantly enriched cell types, highlighting stable pathway transmission across hematopoiesis (Figure 26A and Figure 26B).

[0386] To test whether Relapse-I enrichment was driven by extrinsic signals, a curated ligand-receptor interaction database was used to infer cell-cell communication networks in bone marrow from the TFR cohort (MultiNicheNet analysis, Figure 23). Communication scores were quantified and compared across Relapse-P, Relapse-1, and Remission samples, showing increased signaling activity in Relapse-1 LSCs.

[0387] Figure 24A shows is a chord plot where each cell type is depicted as a block and the lines are linking a given receptor and ligand interaction. A greater number of lines indicates greater communication between each cell type. The number of signaling interactions was quantified among all bone marrow cell types and within LSCs using a receiver-based model. The Relapse-1 subtype showed significantly higher interaction counts compared to both Remission and Relapse-P, supporting a model where cell-extrinsic cues maintain the Relapse-1 phenotype (Figure 24A and Figure 24B).

[0388] Ligand-receptor pair analysis identified TNFa-TNFRSF1 A / TNFRSF21 / TNFRSF3 signaling as the most significant axis associated with the Relapse-I subtype (Figure 25). This was inferred from differential expression and receptor-ligand prediction models, suggesting TNF signaling as a central driver of the Relapse-I transcriptional state.

[0389] The source of TNF-a ligands in the bone marrow ecosystem was analysed by cell type. CD8+ effector T cells, CD8-GZMK+ T cells and B cells were predicted to be major contributors of TNF-a signals targeting LSCs in Relapse-I patients (Figure 27A and Figure 27B). These findings support a model where immune cells actively maintain the Relapse-I LSC state.

[0390] Overall, the data show that LSC signatures at diagnosis can predict relapse after TKI discontinuation. Two biologically distinct relapse subtypes were identified (Relapse-P and Relapse-I). The relapse subtypes are driven by distinct mechanisms, intrinsic proliferative vs extrinsic immune-driven programs.

[0391] Biomarkers such as PRSS21 , SPAG6 and CD71 are able to identify relapse LSCs of both relapse subtypes ( / .e. they are relapse subtype agnostic biomarkers). Additional biomarkers such as TNF and pNFkB allow for identification of the Relapse-I subtype at diagnosis and the prioritisation of such patients for immune-targeting therapies in combination with TKIs.

[0392] References

[0393] Achinger-Kawecka J, Stirzaker C, Portman N, Campbell E, et al. The potential of epigenetic therapy to target the 3D epigenome in endocrine-resistant breast cancer. Nat Struct Mol Biol 2024 Jan 5.

[0394] Aibar S, Gonzalez-Blas CB, Moerman T, Huynh-Thu VA, et al. SCENIC: single-cell regulatory network inference and clustering. Nat Methods 2017 Nov; 14(11): 1083-1086.

[0395] Assi SA, Imperato MR, Coleman DJL, Pickin A, et al. Subtype-specific regulatory network rewiring in acute myeloid leukemia. Nat Genet 2019 Jan; 51 (1): 151-162.

[0396] Bernt KM, Zhu N, Sinha AU, Vempati S, et al. MLL-rearranged leukemia is dependent on aberrant H3K79 methylation by DOT1 L. Cancer Cell 2011 Jul 12; 20(1): 66-78.

[0397] Borkin D, He S, Miao H, Kempinska K, et al. Pharmacologic inhibition of the Menin-MLL interaction blocks progression of MLL leukemia in vivo. Cancer Cell 2015 Apr 13; 27(4): 589-602. Branford S, Wang P, Yeung DT, Thomson D, et al. Integrative genomic analysis reveals cancer- associated mutations at diagnosis of CML in patients with high-risk disease. Blood 2018 Aug 30; 132(9): 948-961 .

[0398] Bravo Gonzalez-Blas C, De Winter s, Hulselmans G, Hecker N, et al. SCENIC+: single-cell multiomic inference of enhancers and gene regulatory networks. Nat Methods 2023 Sep; 20(9): 1355-1367. Claringbould A, Zaugg JB. Enhancers in disease: molecular basis and emerging treatment strategies. Trends Mol Med 2021 Nov; 27(11): 1060-1073.

[0399] Chuprin J, Buettner H, Seedhom MO, Greiner DL, et al. Humanized mouse models for immuno-oncology research. Nat Rev Clin Oncol 2023 Mar; 20(3): 192-206.

[0400] Edginton-White B, Bonifer C. The transcriptional regulation of normal and malignant blood cell development. FEBS J 2022 Mar; 289(5): 1240-1255.

[0401] Giustacchini A, Thongjuea S, Barkas N, Woll PS, et al. Single-cell transcriptomics uncovers distinct molecular signatures of stem cells in chronic myeloid leukemia. Nat Med 2017 Jun; 23(6): 692-702. Goode DK, Obier N, Vijayabaskar MS, Lie ALM, et al. Dynamic Gene Regulatory Networks Drive Hematopoietic Specification and Differentiation. Dev Cell 2016 Mar 7; 36(5): 572-587.

[0402] Gottgens B. Regulatory network control of blood stem cells. Blood 2015 Apr 23; 125(17): 2614-2620.

[0403] Graham SM, Jorgensen HG, Allan E, Pearson C, et al. Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood 2002 Jan 1 ; 99(1): 319-325.

[0404] Hsieh YC, Kirschner K, Copland M. Improving outcomes in chronic myeloid leukemia through harnessing the immunological landscape. Leukemia 2021 May; 35(5): 1229-1242.

[0405] Jun K, Ya-Zhen Q, Xiao-Su Z, Hong-Xia S, et al. Interferon-alpha may help prevent molecular relapse of chronic myeloid leukemia after the discontinuation of tyrosine kinase inhibitors. Ther Adv Hematol 2021 ; 12: 2040620720986643.

[0406] Kim T, Tyndel MS, Kim HJ, Ahn JS, et al. Spectrum of somatic mutation dynamics in chronic myeloid leukemia following tyrosine kinase inhibitor therapy. Blood 2017 Jan 5; 129(1): 38-47.

[0407] Ko TK, Javed A, Lee KL, Pathiraja TN, et al. An integrative model of pathway convergence in genetically heterogeneous blast crisis chronic myeloid leukemia. Blood 2020 Jun 25; 135(26): 2337-2353.

[0408] Krishnan V, Schmidt F, Nawaz Z, Venkatesh PN, et al. A single-cell atlas identifies pretreatment features of primary imatinib resistance in chronic myeloid leukemia. Blood 2023 Jun 1 ; 141 (22): 2738-2755.

[0409] Kuntz EM, Baquero P, Michie AM, Dunn K, et al. Targeting mitochondrial oxidative phosphorylation eradicates therapy- resista nt chronic myeloid leukemia stem cells. Nat Med 2017 Oct; 23(10): 1234-1240. Lee LM, Christodoulou EG, Shyamsunder P, Chen BJ, et al. A novel network pharmacology approach for leukaemia differentiation therapy using Mogrify®. Oncogene 2022 Nov; 41 (48): 5160-5175.

[0410] Maat H, Atsma TJ, Hogeling SM, Rodriguez Lopez A, et al. The USP7-TRIM27 axis mediates non- canonical PRC1 .1 function and is a druggable target in leukemia. iScience 2021 May 21 ; 24(5): 102435. Ng JJ, Ong ST. Therapy Resistance and Disease Progression in CML: Mechanistic Links and Therapeutic Strategies. Curr Hematol Malig Rep 2022 Dec; 17(6): 181-197.

[0411] Ott CJ, Federation AJ, Schwartz LS, Kasar S, et al. Enhancer Architecture and Essential Core Regulatory Circuitry of Chronic Lymphocytic Leukemia. Cancer Cell 2018 Dec 10; 34(6): 982-995 e987. Patterson SD, Copland M. The Bone Marrow Immune Microenvironment in CML: Treatment Responses, Treatment-Free Remission, and Therapeutic Vulnerabilities. Curr Hematol Malig Rep 2023 Apr; 18(2): 19- 32.

[0412] Ross DM, Branford S, Seymour JF, Schwarer AP, et al. Safety and efficacy of imatinib cessation for CML patients with stable undetectable minimal residual disease: results from the TWISTER study. Blood 2013 Jul 25; 122(4): 515-522.

[0413] Scott MT, Korfi K, Saffrey P, Hopcroft LE, et al. Epigenetic Reprogramming Sensitizes CML Stem Cells to Combined EZH2 and Tyrosine Kinase Inhibition. Cancer Discov 2016 Nov; 6(11): 1248-1257.

[0414] Scott M. Leukemia stem cell molecular phenotypes, lineage potentialties and diversity are influenced by human cytokine expression in PDX models of chronic myeloid leukemia. Annual John Goldman Conference on Chronic Myeloid Leukemia: Biology and Therapy 2023.

[0415] Shanmuganathan N, Pagani IS, Ross DM, Park S, et al. Early BCR-ABL1 kinetics are predictive of subsequent achievement of treatment-free remission in chronic myeloid leukemia. Blood 2021 Mar 4; 137(9): 1196-1207. van den Boom V, Maat H, Geugien M, Rodriguez Lopez A, et al. Non-canonical PRC1.1 Targets Active Genes Independent of H3K27me3 and Is Essential for Leukemogenesis. Cell Rep 2016 Jan 12; 14(2): 332-346.

[0416] Wu T, Yoon H, Xiong Y, Dixon-Clarke SE, et al. Targeted protein degradation as a powerful research tool in basic biology and drug target discovery. Nat Struct Mol Biol 2020 Jul; 27(7): 605-614.

[0417] Xie H, Peng C, Huang J, Li BE, et al. Chronic Myelogenous Leukemia-Initiating Cells Require Polycomb Group Protein EZH2. Cancer Discov 2016 Nov; 6(11): 1237-1247.

[0418] Xu J, Song F, Lyu H, Kobayashi M, et al. Subtype-specific 3D genome alteration in acute myeloid leukaemia. Nature 2022 Nov; 611 (7935): 387-398.

[0419] Zaugg JB, Sahlen P, Andersson R, Alberich-Jorda M, et al. Current challenges in understanding the role of enhancers in disease. Nat Struct Mol Biol 2022 Dec; 29(12): 1148-1158.

Claims

Claims:1 . A method for predicting whether a subject with chronic myelogenous leukaemia (CML) is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by polycomb repressive complex 1 (PRC1) in CD34+CD38- cells in a sample obtained from the subject, wherein an increased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML.

2. A method for predicting whether a subject with chronic myelogenous leukaemia (CML) is at risk of relapse following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by polycomb repressive complex 1 (PRC1) in CD34+CD38- cells in a sample obtained from the subject, wherein a decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who achieve treatment-free remission (TFR) following cessation of the treatment for CML indicates that the subject is at risk of relapse following cessation of the treatment for CML.

3. A method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject, wherein a decreased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

4. A method for predicting whether a subject with CML will achieve TFR following cessation of a treatment for CML, the method comprising determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject, wherein an increased expression of the one or more genes regulated by PRC1 relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of the treatment for CML indicates that the subject is likely to achieve TFR following cessation of the treatment for CML.

5. A method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:(a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and(b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

586. A method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:(a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and(b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment comprises a hypomethylating agent (HMA) and / or a BMI1 inhibitor.

7. A method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:(a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and(b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is decreased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

8. A method of selecting a subject for treatment, wherein the subject has CML or has been diagnosed with CML, the method comprising:(a) determining the expression of one or more genes regulated by PRC1 in CD34+CD38- cells in a sample obtained from the subject; and(b) selecting the subject for treatment if the expression of the one or more genes regulated by PRC1 is increased relative to a pre-determined average expression in samples obtained from subjects with CML who relapse following cessation of a treatment for CML; wherein the treatment comprises one or more of a tyrosine kinase inhibitor (TKI), a BCR-ABL1- targeted therapy (BTT), a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

9. The method of any one of claims 5 to 8, further comprising: determining the level of immune response activation in CD34+CD38- cells in a sample obtained from the subject; and selecting the subject for treatment if the level of immune response activation is increased relative to a pre-determined average level of immune response activation in samples obtained from subjects with CML who achieve TFR following cessation of a treatment for CML; wherein the treatment further comprises an immunomodulator.

10. The method of any one of the preceding claims, wherein the one or more genes is selected from the list consisting of: PRSS21 , ANLN, SPAG6, SCN2A, BMI1 , TFRC and ALDH1 A1 .11 . The method of claim 1 , 3, 5 or 7, wherein the one or more genes is selected from PRSS21 , ANLN, SPAG6, BMI1 , and TFRC.

12. The method of claim 2, 4, 6 or 8, wherein the one or more genes is selected from ALDH1A1 and SCN2A.

13. The method of any one of the preceding claims, wherein the CD34+CD38- cells are leukaemic stem cells (LSCs).

14. The method of any one of the preceding claims, wherein the treatment for CML is a TKI, a BTT, a chemotherapeutic agent, interferon therapy, radiation therapy, stem cell therapy, immunotherapy and / or a combination thereof.

15. The method of claim 14, wherein the treatment for CML is a TKI or a BTT.

16. The method of any one of the preceding claims, wherein the sample has been obtained from peripheral blood or bone marrow.

17. The method of any one of the preceding claims, wherein the subject is a human.

18. The method of any one of the preceding claims, wherein the expression of the one of more genes and / or the level of immune response activation is determined using flow cytometry.

19. The method of any one of the preceding claims, wherein the expression of the one or more genes and / or the level of immune response activation is determined using immunohistochemistry.

20. The method of any one of the preceding claims, wherein the expression of the one or more genes is determined using one or more antigen-binding molecules.21 . A kit suitable for performing the method according to any one of claims 1-20.

22. The kit of claim 21 wherein the kit comprises a plurality of antigen-binding molecules capable of detecting the one or more genes regulated by PRC1 .

23. The kit of claim 22, wherein the plurality of antigen-binding molecules is capable of detecting one or more of PRSS21 , ANLN, SCN2A, SPAG6, BMI1 , TFRC, and ALDH1 A1 .

24. The kit of any one of claims 21 to 23, further comprising an antigen-binding molecule capable of detecting a correlate of immune response activation.

25. The kit of claim 24, wherein the antigen-binding molecules capable of detecting a correlate of immune response activation is capable of detecting TNFa or phosphorylated NFKB.

26. An antigen-binding molecule for use in a method according to any one of claims 1-20.

27. The antigen-binding molecule of claim 22, wherein the antigen-binding molecule is capable of detecting one or more of PRSS21 , ANLN, SCN2A, SPAG6, BMI1 , TFRC, ALDH1 A1 , TNFa and phosphorylated NFKB.61

Images