Methods of reducing the risk of developing hematologic malignancies

Abstract

Disclosed are methods of reducing the risk of a subject developing hematologic malignancies and myelodysplastic syndrome, the methods comprising administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

Description

[0001] METHODS OF REDUCING THE RISK OF DEVELOPING HEMATOLOGIC MALIGNANCIES

[0002] BACKGROUND OF THE INVENTION

[0003] Human hematopoietic stem / progenitor cells are responsible for producing more than 10 billion blood cells daily throughout a person's lifespan. With aging, mutations can occur in these cells conferring a competitive growth advantage, leading to gradual clonal dominance of hematopoiesis. Clonal hematopoiesis involves mutations in hematopoietic stem / progenitor cells. Large cohort studies have demonstrated that clonal hematopoiesis is also associated with an increased risk of developing myeloid malignancies. Accordingly, there remains a need for exploring agents that can help reduce clonal hematopoiesis mutations and therefore, reduce the risk of developing hematologic malignancies.

[0004] SUMMARY OF THE INVENTION

[0005] Colchicine was found to reduce the risk of a subject developing a hematologic malignancy.

[0006] In one aspect, the disclosure features a method of reducing the risk of a subject developing a hematologic malignancy, the method including administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

[0007] In another aspect, the disclosure features a method of reducing the risk of relapse in a subject previously treated for a hematologic malignancy, the method including administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

[0008] In another aspect, the disclosure features a method of reducing the risk of a secondary hematologic malignancy developing in a subject being treated with a chemotherapy or a radiotherapy, the method including administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

[0009] In some embodiments, the hematologic malignancy or secondary hematologic malignancy is a myeloid cancer.

[0010] In some embodiments, the myeloid cancer is acute myeloid leukemia, myeloproliferative neoplasm (MPN), mastocytosis, chronic neutrophilic leukemia, or chronic myelomonocytic leukemia.

[0011] In some embodiments, the myeloproliferative neoplasm is polycythemia vera, essential thrombocythemia, or myelofibrosis.

[0012] In some embodiments, the clonal hematopoiesis mutation is in a ASXL1, CBL, DNMT3A, GNAS, GNB1, JAK2, PPM1D, SF3B1, SRSF2, TET2, or TP53 gene, or a combination thereof.

[0013] In some embodiments, the clonal hematopoiesis mutation is in a ASXL1, DNMT3A, SF3B1, TET2, or TP53 gene, or a combination thereof.

[0014] In some embodiments, the clonal hematopoiesis mutation is in a SF3B1 gene.

[0015] In some embodiments, the clonal hematopoiesis mutation is in a TET2 gene.

[0016] In some embodiments, the clonal hematopoiesis mutation is in a TP53 gene.

[0017] In another aspect, the disclosure features a method of reducing the risk of a subject developing a myelodysplastic syndrome (MDS), the method including administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

[0018] In some embodiments, the clonal hematopoiesis mutation is in a SF3B1 gene.

[0019] In some embodiments, the subject is being treated for a cancer comprising a mutant TP53 gene. In some embodiments, colchicine is administered to the subject at a dose of 0.3 to 0.7 mg, at a dose of 0.4 to 0.6 mg, or at a dose of 0.5 mg once daily.

[0020] In a further aspect, the disclosure features colchicine for use in a method of reducing the risk of a subject developing a hematologic malignancy, where the subject has been determined to have a clonal hematopoiesis mutation.

[0021] In another aspect, the disclosure features colchicine for use in a method of reducing the risk of relapse in a subject previously treated for a hematologic malignancy, where the subject has been determined to have a clonal hematopoiesis mutation.

[0022] In yet another aspect, the disclosure features colchicine for use in a method of reducing the risk of a secondary hematologic malignancy developing in a subject being treated with a chemotherapy or a radiotherapy, where the subject has been determined to have a clonal hematopoiesis mutation.

[0023] In an additional aspect, the disclosure features colchicine for use in a method of reducing the risk of a subject developing a myelodysplastic syndrome (MDS), where the subject has been determined to have a clonal hematopoiesis mutation.

[0024] In a further aspect, the disclosure features a kit comprising colchicine, where the kit further comprises a package insert instructing a user of the kit to administer the colchicine to the patient in accordance with a method or use of any aspect described herein.

[0025] BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0027] FIG. 1 is a flow diagram of study participants.

[0028] FIG. 2A and FIG. 2B show the prevalence of clonal hematopoiesis variant carriers based on the mutation with the highest variant allele fraction (VAF) value in the first sample collected. Percentage of carrier individuals by gene, group, and VAF category (FIG. 2A; the columns for each variant carrier are, left to right, placebo (VAF 0.5-2%), colchicine (VAF 0.5-2%), placebo (VAF>2%), and colchicine (VAF>2%)). Prevalence of variant carriers of any gene mutation by age (FIG. 2B).

[0029] FIG. 3A and FIG. 3B show cumulative incidence curves of the cardiovascular composite endpoint (cardiovascular death, resuscitated cardiac arrest, myocardial infarction, stroke, urgent hospitalization for angina requiring coronary revascularization, or any coronary revascularization) in the colchicine group (red) and the placebo group (black), in a time-to-event analysis, for patients with TET2 CHIP mutations (FIG. 3A) and without TET2 CHIP mutations (FIG. 3B).

[0030] FIG. 4 is a graph showing observed VAF versus expected VAF for control sample dilutions in which 8 clonal hematopoiesis variants of 5% VAF in the Myeloid sample (Horizon) are considered.

[0031] DEFINITIONS

[0032] As used herein, “reducing the risk” refers to lowering or decreasing an individual’s risk or chance of developing or contracting a disease, disorder, or condition, such that the individual does not develop or contract the disease, disorder, or condition, or does not exhibit any symptoms (e.g. , clinically significant symptoms) of the disease, disorder, or condition. As used herein, “reducing the risk” can refer to lowering the chance of a disease, disorder, or condition developing in an individual carrying a disease-causing mutation.

[0033] As used herein, “hematologic malignancy” refers to a blood cancer or any type of cancer that begins in the blood-forming tissues (e.g., bone marrow) or cells of the immune system. Blood cancer can affect the blood, bone marrow, lymph, and lymphatic system. Examples of hematologic malignancies include leukemia, myeloid cancers, acute myeloid leukemia (AML), lymphoma, myeloma, multiple myeloma, Langerhans cell histiocytosis, myelodysplastic syndrome (MDS), myeloproliferative neoplasm (MPN), acute lymphoblastic / lymphocytic leukemia (ALL), chronic myeloid leukemia (CML), and myeloproliferative disease, among others.

[0034] As used herein, “secondary hematologic malignancy (SHM)” refers to a new cancer that develops in an individual because of a prior cancer treatment, where the individual underwent chemotherapy, radiotherapy, or immunotherapy (e.g., using immunomodulators). SHMs can occur because of prior treatment with chemotherapy (e.g., platinum-based drugs such as carboplatin or cisplatin), radiotherapy, or immunomodulators, which can cause DNA damage and mutations. SHMs can develop months or even decades after the initial treatment and are generally non-curable. Examples of SHMs include secondary acute myeloid leukemia (sAML), secondary acute lymphoblastic / lymphocytic leukemia (s-ALL), and secondary chronic myeloid leukemia (sCML).

[0035] As used herein, “clonal hematopoiesis mutation” refers to a somatic mutation in a hematopoietic stem cell such that the hematopoietic stem cell starts producing cells with that somatic mutation. Hematopoietic stem cells give rise to different types of blood cells in the body but when clonal hematopoiesis occurs, the blood cells produced are mutated.

[0036] As used herein, “relapse” refers to the return or recurrence of a disease or condition or its symptoms after a period of improvement, for example, a cancer coming back after a period of remission. During relapse, the symptoms of a disease return or may become even worse. Cancer relapse may occur if the some of the original cells were not targeted earlier or survived the initial treatment.

[0037] As used herein, “myeloid cancer” refers to a cancer or malignancy that affects hematopoietic stem cells. Hematopoietic stem cells are immature cells which can give rise to different types of blood cells in the body and are also called blood stem cells. Examples of myeloid cancers include Acute Myeloid Leukemia (AML), Myelodysplastic Syndromes (MDS), Myeloproliferative Neoplasm (MPN), mastocytosis, chronic neutrophilic leukemia, and chronic myelomonocytic leukemia.

[0038] As used herein, “myeloproliferative neoplasms (MPN)” refers to a group of diseases or rare blood cancers in which the bone marrow produces excess red blood cells, white blood cells, or platelets. In MPN, the blood cell production process is disrupted, and blood cells are overproduced, hence, it can be potentially life-threatening. MPN may occur because of a mutation in a stem cell in the bone marrow. Examples of MPNs include polycythemia vera, essential thrombocythemia, myelofibrosis, primary myelofibrosis, chronic neutrophilic leukemia, and chronic eosinophilic leukemia.

[0039] As used herein, “myelodysplastic syndromes (MDS)” refers to a group of disorders or cancers that are caused by disruption of blood cell production, such that there is a dearth of healthy blood cells, the blood cells are not formed properly, or don’t function properly. These disorders are a result of a scarcity of healthy blood cells (e.g., red blood cells, white blood cells, or platelets) because the bone marrow produces immature blood cells that do not mature into healthy blood cells. Ineffective hematopoiesis, marrow dysplasia, and peripheral blood cytopenia can occur in MDS. MDS also has a high propensity for transformation into Acute Myeloid Leukemia (AML). MDS may occur in response to cancer treatments or chemical exposure.

[0040] DETAILED DESCRIPTION

[0041] The disclosure is directed towards assessing the impact of colchicine, a potent anti-inflammatory medication, on clonal hematopoiesis mutant cell fractions.

[0042] Human hematopoietic stem / progenitor cells are responsible for producing more than 10 billion blood cells daily throughout a person's lifespan. With aging, mutations can occur in these cells conferring a competitive growth advantage, leading to gradual clonal dominance of hematopoiesis (Busque et al., Blood, 1996;88:59-65; Busque et a / ., Nat Genet., 2012;44:1179-1181 ). Large cohort studies have demonstrated that clonal hematopoiesis (CH) is associated with an increased risk of developing myeloid malignancies and cardiovascular disease (Jaiswal et al., N Engl J Med., 2014;371 :2488-2498; Jaiswal et al., N Engl J Med., 2017;377:111-121 ; Genovese et a!., N Engl J Med., 2014;371:2477-2487) and may be a biomarker of unhealthy aging. The prevalence of CH is a function of age, depth of sequencing, and number of genes analyzed (Jaiswal and Ebert, Science, 2019;366). The World Health Organisation (Busque, Chapter 2, WHO classification of tumours haematolymphoid tumours, (volume 11) 5th edition, [S.l.]: IARC; 2024:19) distinguishes age-related CH, which has no specific quantitative threshold for mutation burden, from “Clonal hematopoiesis of indeterminate potential (CHIP) and its distinction from myelodysplastic syndromes” (Steensma et al., Blood, 2015;126:9-16), where the variant allele frequency (VAF) has to be > 2%. Somatic mutations in over 100 genes have been identified as candidate drivers of CH, with the most common mutations being in the three epigenetic regulators: DNA methyltransferase 3a (DNMT3A), ten-eleven translocation-2 (TET2), and additional sex combs-like 1 (ASXL1). Risk factors for adverse events in CHIP carriers include clone size, driver gene identity, and number of mutations (Weeks et al., NEJM Evid., 2023;2).

[0043] Carriers of CHIP exhibit a pro-inflammatory response, with higher levels of C-reactive protein, IL- 6, IL-1 p, IL-18, and TNF-a compared to non-carriers (Busque et al., Blood Advances, 2020;4:2430-2438; Cook et a / ., Blood Adv., 2019;3:2482-2486; Bick et a / ., Nature, 2020;586:763-768). In a pro-inflammatory environment, mutant hematopoietic stem / progenitor cells display an attenuated inflammatory response relative to wild-type hematopoietic stem / progenitor cells, providing them with a relative fitness advantage over non-mutated cells (Avagyan et al., Science, 2021;374:768-772; Jakobsen et al., Cell stem cell, 2024). The proliferation of these mutated cells can be mitigated by the experimental neutralization of pro- inflammatory interleukin-6 and IL-1 signaling (Meisel et al., Nature, 2018;557:580-584; Kovtonyuk et a / ., Blood, 2022;139:44-58; Caiado et al., Blood, 2023;141 :886-903). A post-hoc analysis of the CANTOS trial showed that carriers of CH mutations in the TET2 gene experienced fewer cardiovascular events when treated with the anti-inflammatory monoclonal antibody canakinumab, which targets interleukin-1 (Svensson et al., JAMA Cardiol., 2022;7:521-528). Inflammation plays an important role in atherosclerosis (Hansson, New England Journal of Medicine, 2005;352:1685-1695), and large randomized clinical trials have shown the benefits of inflammation reduction on cardiovascular outcomes in patients with atherosclerotic cardiovascular disease, including the COLchicine Cardiovascular Outcomes T rial (COLCOT) and LoDoCo2 studies with low-dose colchicine (Ridker et al., N Engl J Med., 2017;377:1119- 1131; Tardif et a / ., N Engl J Med., 2019;381 :2497-2505; Nidorf et al., New England Journal of Medicine, 2020;383:1838-1847). Colchicine is a potent and widely used anti-inflammatory medication that acts by inhibiting tubulin polymerization and microtubule formation, leading to effects on the inflammasome, inflammatory chemokines, and adhesion molecules (Cook et al., Blood Adv., 2019;3:2482-2486; Bick et al., Nature, 2020;586:763-768; Avagyan et al., Science, 2021 ;374:768-772).

[0044] The currently disclosed study was performed to assess the impact of colchicine, a potent antiinflammatory medication, on clonal hematopoiesis mutant cell fractions. In this study, error-corrected sequencing was used to investigate the effect of colchicine compared to placebo on changes in CH variant allele frequency (VAF) over time in patients who were recruited following a recent myocardial infarction, as part of the COLCOT study. Following that, an assessment was done to determine whether individuals with CHIP derived greater benefits in the reduction of clinical events as compared to non-CHIP carriers. The study provides evidence supporting colchicine’s ability to contribute to the reduction of CH mutations in some of the most common and clinically relevant genes involved in CH, TET2, TP53, and SF3B1, possibly intercepting clonal progression and reducing disease risks in carriers. In line with recent findings, the current results support the potential for using therapeutic strategies targeting inflammation and inflammasomes to reduce the risk of cardiovascular events in patients with TET2 CH mutations (Yalcinkaya, Atherosclerosis, 2024;396:118541).

[0045] EXAMPLES

[0046] The following example is put forth so as to provide those of ordinary skill in the art with a description of how the compositions and methods described herein may be used and evaluated and is intended to be purely exemplary and is not intended to limit the scope of the disclosure.

[0047] Example 1. Assessment of the impact of colchicine, a potent anti-inflammatory medication, on clonal hematopoiesis mutant cell fractions.

[0048] The aim of this study was to assess the impact of colchicine, a potent anti-inflammatory medication, on clonal hematopoiesis mutant cell fractions.

[0049] METHODS

[0050] STUDY DESIGN AND POPULATION

[0051] COLchicine Cardiovascular Outcomes Trial (COLCOT) was a randomized, double-blind, placebo- controlled, multinational clinical trial of 4745 patients who were randomly assigned to receive either colchicine 0.5 mg once daily or placebo in a 1 :1 ratio on top of standard of care and followed for a median of 22.6 months (NCT02551094). The COLCOT trial design and main clinical study results were previously published (Tardif et al., N Engl J Med., 2019;381 :2497-2505). Briefly, adult patients were eligible if they had experienced a myocardial infarction within 30 days before enrollment, had completed any planned percutaneous revascularization procedures, and were treated according to guidelines that included the intensive use of statins. Patients were excluded if they had severe heart failure, a left ventricular ejection fraction of less than 35%, stroke within the past 3 months, coronary bypass surgery within the past 3 years or planned, a history of non-cutaneous cancer within the last 3 years, significant non-transient hematological abnormalities, severe renal disease with serum creatinine greater than twice the upper limit of normal, or severe hepatic disease. Clinical evaluations occurred at 1 month and 3 months following randomization and every 3 months thereafter. Potential study endpoints were adjudicated by an independent clinical endpoint committee who were unaware of the trial-group assignments. The COLCOT primary endpoint was a composite of cardiovascular death, resuscitated cardiac arrest, myocardial infarction, stroke, or urgent hospitalization for angina requiring coronary revascularization in a time-to- event analysis. In the present study, any coronary revascularization was added to the composite clinical endpoint to increase event occurrence and improve statistical power.

[0052] In total, 53 sites agreed to take part in the genetic sub-study from 7 countries compared to 167 sites from 12 countries in the main COLCOT trial. Blood samples were centrally collected and processed at the Pharmacogenomics Centre in Montreal as previously described (Dube et al., Circ Genom Precis Med., 2021 ;14:e003183). An initial blood draw was performed at the time of enrollment or a subsequent visit after randomization, and a second blood draw was performed at the end-of-study visit in a subset of patients. 1656 participants consented to take part in the sub-study, exclusions included poor DNA quality and technological failures, leaving 1610 patients for analysis (FIG. 1 , which is a flow diagram of study participants). Clonal hematopoiesis was evaluated in 1610 patients with a recent myocardial infarction from the COLCOT study, who were randomly assigned to receive low-dose colchicine (0.5 mg once daily) or placebo. Informed consent was obtained from all study participants. The study was approved by the Montreal Heart Institute research ethics committee and complied with the Declaration of Helsinki.

[0053] LIBRARY PREPARATION AND TARGETED SEQUENCING

[0054] All samples were sequenced using a custom designed targeted assay using multiplex PCR- based enrichment by CleanPlex technology (Paragon Genomics, Inc., Hayward, CA, USA). The panel captured the exomic regions of 11 of the most commonly mutated genes linked to CH (ASXL1, CBL, DNMT3A, GNAS, GNB1, JAK2, PPM1D, SF3B1, SRSF2, TET2, and TP 53), covering 35,840 bases (97.64%) of the coding regions using 322 amplicons (Tables 1 a-b). The assay covered the full exonic regions of 6 genes (PPM1D, TP53, SF3B1, TET2, GNB1, and JAK2) and partial exonic regions of 5 genes (ASXL _cdr_1 , CBL_cdr_1, DNMT3A_cdr_6, DNMT3A_cdr_7 G / VAS_cdr_3, and SRSF2_cdr_1 ). Genomic DNA was isolated from 5-10 mL of frozen whole blood (Autopure LS V5.0) and quantified by fluorescence (Tecan Infinite F200 Plate Reader). Libraries were prepared from 60 ng of genomic DNA following recommendations of the CleanPlex protocol. A multiplex PCR reaction was performed using unique molecular indices (UMI)-labeled target-specific primers to barcode and amplify the targeted regions. This was followed by post-barcoding purification (CleanMag Magnetic Beads) and an enzymatic background cleaning step was used to remove non-specific PCR product (primer dimers and PCR artifacts), followed by a bead purification. Unique dual indexes (UDI) were used in a final PCR amplification to generate unique libraries per sample. Samples were bead-purified and the integrity, quality, and quantity was verified using the TapeStation 4200 D1000 assay. Each library was normalized to 4 nM, pooled, and sequenced on Illumina MiSeq Sequencer as a quality control check prior to sequencing. Five sequencing runs were performed on the Illumina NovaSeq 6000 instrument (Illumina Inc., San Diego, CA, USA) using the NovaSeq Control Software (Recipe Fragment v1.7.5, Analysis viewer v2.4.7) with paired-end sequencing 2x150 bp (NovaSeq S4 Reagent Kit v1.5 and NovaSeq XP 4- Lane Kit v1.5), according to the manufacturer’s instructions.

[0055] Tables 1a-b are shown below:

[0056]

[0057]

[0058] The recommended Paragon UMI 12 step bioinformatics analysis pipeline was implemented. After demultiplexing the sequencing data, the trailing sequencing adapters in the R1 and R2 reads were trimmed using cutadapt v.4.6. Trimmed fastq reads were then converted to unmapped alignment (BAM) files. The 16 bp UMI were removed from BAM files to be stored as an alignment tag. Template reads were reconverted to FASTQ format and mapped against the human genome reference GRCh37 using bwa-mem v.0.7.17. The mapped and unmapped tagged reads were merged and the alignment file was created. The resulting alignments were processed using the Fulcrum Genomics consensus reads calling workflow (fgbio v2.2.1) (Nils et al., fulcrumgenomics / fgbio; Release 2.2.1, Zenodo; 2024). Aligned reads were grouped as read families presumed to originate from the same molecule by adjacency and allowing a maximum of one edit between UMIs and keeping only bases with mapping quality > 20. The CallMolecularConsensusReads from fgbio was used to create consensus reads based on > 3 identical UMI reads to produce a consensus read. The consensus reads were aligned against the human genome reference GRCh37 using bwa-mem v.0.7.17 again to improve alignment accuracy. We implemented the bioinformatics workflow in GenPipes (Bourgey et al., GigaScience, 2019;8) as per Paragon’s instructions. Source code and guidelines can be found in the code repository directory on Gitlab (github.com / johanna- sa nd ova l / ch i p_co I cot_scri pts ) .

[0059] VARIANT CALLING, ANNOTATION, AND FILTERING

[0060] Variant calling was performed using vardict (v4.2.6.1) (Lai et al., Nucleic Acids Research, 2016;44:e108-e108) using the default cuttoffs to all but the following arguments: -a, meaning amplicon mode, where reads that do not map to the amplicon are skipped. Read pairs were assigned to an amplicon with a minimal overlap fraction of 95% and the PHRED score threshold was set to 20. Variant consequence prediction was performed using Annovar (v.2020-06-08) (Wang and Hakonarson, Nucleic Acids Research., 2010;38:e164-e164) and variants were annotated using the refGene hg19 assembly reference and COSMIC v70 hg19 resource (Tate et al., Nucleic Acids Res., 2019;47:D941-D947). Variants were identified as putative drivers of CH according to gene-specific criteria and if they matched a pre-specified list of previously reported variants (Bick et al., Nature, 2020;586:763-768; Vlasschaert et al., Blood, 2023;141 :2214-2223) (Table 2). Artifacts were removed as per recommended post-variant call filtering (Vlasschaert et al., Blood, 2023;141 :2214-2223) implemented in the script whitelist_filter_rscript.R described by Vlasschaert et al. (Vlasschaert et al., Blood, 2023;141 :2214-2223) with required resources extracted from Terra Workspaces (app.terra.bio / #workspaces / terra-outreach / CHIP-Detection-Mutect2), including NEJM_2017_genes_01262020.txt, CHIP_splice_vars_agb_01262020.txt, CHIP_nonsense_FS_vars_agb_01262020.txt, and CHIP_missense_vars_agb_01262020.txt. The script was cloned and modified to account for minor script errors, including the detection of nucleotide duplication introducing an immediate stop codon without a frameshift (denoted by A python script was used to perform modifications including for synonymous variants comprising a stop codon which were wrongly tagged by the R script (e.g.: TET2 X2003X, PPM1D X606X), and JAK2 variant F537_K539delinsL was manually added to the list of CHIP mutations. To minimize artifacts, the following were excluded: (a) variants with a consensus sequencing depth (DP) < 20; (b) insertions or deletions in sites within homopolymer runs (a sequence of 5 identical bases) unless there was a total of 10 or more supporting reads and a VAF>8% for these variants; (c) previously detected sequencing artifacts (TP53:NM_000546:c.1129A>C, TP53:NM_000546:c.215C>G, ASXL1:NM_015338.6:c.708G>C, ASXL1:NM_015338.6:c.2444T>C, and ASXL1:NM_015338.6:c.1934dupG variants with VAF < 0.06); and

[0061] (d) variants present in 10 individuals or more (>0.6%) which were not associated to TERT variant rs7705526 or (at p-value < 0.10 by Firth’s logistic regression) (Table 3).

[0062] Table 2 is shown below:

[0063] Table 3 is shown below:

[0064] VARIANT DETECTION ASSESSMENT To assess the performance metrics of the multiplex targeted sequencing panel and bioinformatics pipeline, two DNA reference standard samples were used from Horizon (Horizon Discovery Ltd.), Myeloid DNA (5%), and Tru-Q 0 (100% Wildtype), which were combined at varying dilution ratios to provide expected VAF of 2.5%, 1%, 0.5%, and 0.1% for the myeloid sample validated variants at an initial VAF of 5 5%. The Horizon reference standards are well-characterized, cell line-derived, and control material manufactured under ISO 13485. The Myeloid DNA reference standard contains 22 variants across 19 genes relevant to myeloid cancer (horizondiscovery.com / en / reference-standards / products / myeloid-dna- reference-standard) and the Tru-Q 0 reference standard is a well-characterized quality control sample for sequencing assays performance assessment. From horizondiscovery.com, the whole exome sequencing 10 VCF files of these samples was obtained based on the Agilent SureSelect Clinical Research Exome v3 (CREV3) with NEXTFLEX™ Pre- & Post- Capture Combo Kit and sequenced on an Illumina NovaSeq 6000 instrument at more than 500x. The variants present in either the Myeloid or the Tru-Q 0 sample within the targeted region of the assay were identified, and CH variants were annotated and identified according to the study’s bioinformatics pipeline. The Myeloid reference sample and the Tru-Q sample had 15 12 and 5 CH variants respectively, providing 15 unique variants of which 7 corresponded to the Myeloid reference sample’s verified allele frequency variants of 5% VAF (Table 4). Library preparation and sequencing of the reference samples was performed using the targeted multiplex sequencing panel, and applied variant calling and annotation as described above for 5 replicates for dilutions at VAF 2.5% and 1%, and 4 replicates for dilutions at VAF 0.1% and 0.5%. To assess the positive predictive value (PPV;

[0065] 20 TP / (TP+FP)) and positive percent agreement (PPA; TP / (TP+FN)) of the assay, the 5 replicate samples with the dilution providing expected VAF of 2.5% was relied on with all 15 variants. When considering all variants at a variant read depth threshold (VD) > 1, the mean PPV and PPA were 8% and 96% respectively, however, restricting to variants with VD > 5 increased the PPV to 64% (Table 5a). For a VAF > 2% and a variant depth threshold VD > 5 as filtering criteria, the mean PPV and PPA were 100% and

[0066] 25 75% respectively for all 15 variants (T able 5a). The limit of detection was assessed based on the validated mutations in the HORIZON sample (Table 5b and FIG. 4). FIG. 4 is a graph showing observed VAF versus expected VAF for control sample dilutions in which 8 clonal hematopoiesis variants of 5% VAF in the Myeloid sample (Horizon) are considered. Based on 8 validated mutations in the HORIZON sample (includes the 7 mutations listed in Table 4, plus ASXL1_W796C which was not whitelisted). There 30 were 5 repeat samples per dilution ratio for VAF 0.025 and 0.01, and 4 repeat samples for VAF 0.001 and 0.005.

[0067] For a variant depth threshold VD > 1 as filtering criteria, the PPA was 52% for VAF of 0.1% and 89% for VAF of 0.5%. When restricting to VD > 5, the PPA was 19% for VAF of 0.1% and 61% for VAF of 0.5%.

[0068] 35 Table 4 is shown below:

[0069]

[0070] Tables 5a-b are shown below:

[0071] STATISTICAL ANALYSES

[0072] Changes in VAF were assessed by using the proportion of variant read counts (VD) over the

[0073] 5 position-specific read depth (DP) modeled using a generalized mixed regression model fitting a betabinomial distribution using a logit link function. The beta-binomial model was selected for its ability to directly model VAF while accounting for variability in read depth and overdispersion of the data, and the logit link function ensures that VAF values are bounded in [0,1], Any variant observed in either the first or end-of-study timepoint sample with a VD of 5 or more was included, along with its matching observation

[0074] 10 at the other timepoint, including when the matching observation had a VD of 0 or below 5. Multiple variants per patient were accounted for by including the patient as a random factor in the model. The timepoint variable representing the first and end-of-study samples was the exposure of interest and estimates were obtained by contrasting the two timepoint observations using maximum likelihood estimation. Analyses were adjusted for age and the time interval between the two timepoint collection dates. The model’s regression parameters are interpretable as log odds of VAF (VD / DP). The timepoint regression coefficient estimate represents the ratio of the odds of the VAF at end-of-study sample over the odds of the VAF in the first sample. Differences in VAF change between the colchicine and placebo groups were tested by including an interaction term between the timepoint and treatment group variables. The exponent of the interaction term coefficient estimate represents the increase or decrease of the odds of the VAF in the colchicine end-of-study sample compared to the rest. In sensitivity analyses, further adjustment for the delay in time between study drug initiation and the time of the first DN A sample provided consistent results. Sensitivity analyses considering variants with read counts of 1 or above (VD>1 ) at one or both timepoints were conducted using a generalized mixed model fitting a negative binomial distribution to model VD and adjusting for DP as an offset. The change in VD was tested by contrasting the end-of-study sample to the first sample observations using Laplace approximation. The expected mean percentage change in mutation counts from the first to the end-of-study sample was obtained by subtracting one from the exponentiated estimate of the timepoint variable. The negative binomial model was selected as it can better account for the zero-inflation that occurs at this lowered VD threshold than the beta-binomial distribution.

[0075] For the analysis of clinical outcomes, positively adjudicated data was used according to the intention-to-treat principle. The composite endpoint consisted of the combination of the COLCOT trial primary endpoint (cardiovascular death, resuscitated cardiac arrest, myocardial infarction, stroke, or urgent hospitalization for angina requiring coronary revascularization) to which any coronary revascularization event was added, to increase the number of observed events and potentially increase statistical power. Time to the first occurrence of the composite endpoint was compared between the two treatment groups using the hazard ratio from a Cox proportional hazard model with adjustment for age and sex. Event dates and censoring dates were complete and event-free patients were censored at the date of study completion or at the date of last contact, whichever was the latest. The proportionality of hazards was verified. To account for the multiple occurrences of composite endpoint events within patients, a recurrent event analysis was performed using the Andersen and Gill (AG) model adjusted for age and sex. Individuals were categorized by considering the mutation with the highest VAF on a per- gene basis, allowing the assignation of individuals to VAF groups (VAF > 0; > 0.5%; > 2%; > 10%). All statistical tests were two-sided and conducted at the 0.05 significance level. Statistical analyses were performed using SAS version 9.4.

[0076] RESULTS

[0077] PARTICIPANTS

[0078] Of the 4745 patients enrolled in the COLCOT trial, 1610 were evaluated for CH. Table 6 presents the baseline characteristics of these patients. The average age of participants was 60.6 years, 18.1% were female, and the mean BMI was 28.8 kg / m2For 25% of the participants, the first DNA sample was collected within 24 days of initiating the study drug. The median number of days between the baseline visit and the first DNA sample collection was 187 (6.1 months). Of the 1610 participants, 29 had only one DNA sample collected at the end-of-study visit. A total of 848 participants provided two DNA samples for longitudinal assessment. For the 848 participants with two DNA samples, the median interval between the first and second DNA sample was 19.5 months. The primary endpoint in the COLCOT trial consisted of cardiovascular death, resuscitated cardiac arrest, myocardial infarction, stroke, or urgent hospitalization for angina requiring coronary revascularization, to which was added any coronary revascularization event for the purpose of the present study. This composite endpoint occurred in 8.7% and 7.4% of participants in the placebo and colchicine arms respectively in the present study, compared to 9.3% and 7.6% in the placebo and colchicine arms respectively in the main COLCOT trial.

[0079] Table 6 is shown below:

[0080] Table 6. Characteristics of study participants

[0081] AE, adverse event; BMI, body mass index; CRP, C-reactive protein; TIA, transient ischemic attack.

[0082] CHANGE IN VAF OF CLONAL HEMATOPOIESIS MUTATIONS

[0083] Targeted error-corrected deep sequencing of 11 of the most common genes driving clonal expansion was performed. The sequencing libraries were constructed with unique molecular indices to label both DNA strands, enabling the creation of a consensus sequence and enhancing the error detection capability. Sequencing was conducted at an average depth of 64,232x, resulting in an average consensus read depth of 3,896x after accounting for unique molecular tags. Changes in variant allele fraction (VAF) of 15,919 mutations over a mean period of 19.5 months were analyzed and the benefit of colchicine in reducing mutation VAF and cardiovascular events in mutation carriers was assessed. When categorizing individuals according to specific VAF groups based on the mutation with the highest VAF per gene, a high prevalence of mutation carriers with low VAF was observed. Specifically, for VAF < 0.5% in the first DNA sample, 77.4% of individuals had TET2 mutations, 68.7% had DNMT3A mutations, and 64.8% had TP53 mutations (Table 7a). The prevalence of CH mutation carriers within the 0.5% to 2% VAF range was 14.2% for TET2, 12.7% for DNMT3A, and 2.0% for TP53. As expected, the prevalence of mutation carriers for mutations of VAF > 0.5% increased with age (FIG. 2B; Table 7b). However, for individuals with mutations of VAF < 0.5%, the prevalence of DNMT3A and TET2 mutation carriers significantly decreased with age (p < 0.01 ), suggesting possible selective clonal dynamics favoring higher VAF mutations in aging individuals.

[0084] CHIPs (VAF > 2%) were detected in 294 out of 1610 participants (18.3%) across the 11 tested genes, considering their occurrence at either the first or second DNA sample (FIG. 2A; Table 7b). The most frequently observed CHIP genes were DNMT3A (9.1%), TET2 (6.0%), and ASXL1 (1.7%), consistent with previous studies (Bick et al., Nature, 2020;586:763-768).

[0085] FIG. 2 shows the prevalence of clonal hematopoiesis variant carriers, based on the mutation with the highest variant allele fraction (VAF) value in the first sample collected. In particular, FIG. 2A shows the percentage of carrier individuals by gene, group, and VAF category. FIG. 2B shows the prevalence of variant carriers of any gene mutation by age.

[0086] Tables 7a-b are shown below:

[0087] In the subgroup of 848 individuals with two timepoint samples, 1954 unique CH variants were identified among 15,919 longitudinal observations (Table 9a). There were 7,878 longitudinally assessed mutations in the placebo group, with an observed mean increase in VAF of 0.00011 , and 8,041 mutations in the colchicine group with an observed mean increase in VAF of 0.00007. The VAFs of mutations in the end-of-study sample were compared to those of the first sample by using a generalized mixed model that fits a beta-binomial distribution with a logit link function. The VAFs of mutations in the end-of-study

[0088] 5 samples in the colchicine group were found to be significantly lower compared to those of the placebo group (pint = 0.03). Notably, the VAF of TET2 variants increased by 9.1% in the end-of-study samples compared to the first sample in the placebo group (p = 0.05) whereas the VAF of TET2 variants decreased by 10.3% in the colchicine group (p = 0.007), with a significant difference between colchicine and placebo groups (pint = 0.001 ) (Table 8). Significant reductions of 11.8% in the VAF of TP53 mutations 10 (p = 0.001 ; pint = 0.03) and 19.9% in the VAF of SF3B1 mutations (p = 0.006; pint = 0.005) with colchicine (Table 8 and Table 9a) was also observed. Surprisingly, a significant reduction in VAFs for JAK2 mutations was observed in the placebo group (-28.1%, p = 3.4x1 O'5) but not in the colchicine group (Pint = 0.003). However, this difference was largely due to patients with VAF> 2% (n=5 / 185) with only 1 patient in the colchicine group. This difference was not seen in smaller clones.

[0089] 15 Table 8 is shown below:

[0090] Table 8. Results of the analysis of change in clonal hematopoiesis variant allele frequency between two sample timepoints

[0091] * Generalized mixed model fitting a beta binomial with logit link function for the VAF defined as VD / DP, adjusted for the interval in days between sample collections and patient age. The reported estimate is for the timepoint variable contrasting end-of-study to 1st sample. The % change in VAF was obtained from the ratio of the estimated VAF at end-of-study over the estimated VAF in the 1stsample.

[0092] DP, sequencing read depth; SE, standard error; VAF, variant allele fraction; VD, variant counts.

[0093] Overall, the VAFs of mutations at the end of follow-up were significantly lower in the colchicine group compared to placebo (pint=0.03). Significant reductions were observed in the VAF of mutations in the colchicine group over time and as compared to the placebo group for TET2 mutations (10.3% reduction, p=0.007; pint=0.001 ), TP53 mutations (11.8% reduction, p=0.001 ; pint=0.03), and SF3B1 mutations (19.9% reduction, p=0.006; pint=0.005). In those with TET2 mutations of VAF > 2%, the composite cardiovascular endpoint occurred in 2 patients (4.2%) in the colchicine group and 4 (8.3%) patients in the placebo group (HR=0.62; 95% Cl, 0.11-3.42). Thus, the proportion of clonal hematopoiesis mutations was significantly reduced in patients treated with colchicine compared to placebo.

[0094] The analysis of patients with CHIP (VAF > 2%) was limited by the smaller number of observations, which was 100 times less than the overall cohort, and did not reach statistical significance when comparing the placebo and colchicine groups (Table 9a). However, it was observed that clones grew more rapidly in the placebo group when restricting to VAFs > 2% compared to no restriction (15.5% vs. 8.2%), suggesting that clonal expansion may be more pronounced as clones reach a certain proportion of hematopoiesis.

[0095] In sensitivity analyses, all variants observed were considered without restriction on VD (VD > 1 ). While lowering the VD threshold from 5 to 1 increased the number of false positive variants, it also increased the total number of observations detected, including true positive mutations, and improved the detection of mutations with very low VAF (Table 5). With a VD threshold of 1 , 206,015 pairwise observations were included into the analysis. Using a mixed regression model fitting a binomial negative distribution to account for the predominance of null VAF, the results obtained were consistent with those obtained at the threshold of VD > 5. Over all CH mutations, a significant difference was found in the counts of mutant variants between the colchicine and placebo groups (pint=8.0x1 O'4), with an increase in counts of 5.3% in the placebo group (p=1 .1x10'11) compared to an increase in counts of 1.8% in colchicine group ( p=0.009) (Table 9b). Notably, for TET2 mutations, an increase in counts of 5.7% was observed in the placebo group ( p=1 .2x1 O'10) compared to an increase of 2.1% in the colchicine group (p=0.01 ; pint=0.003).

[0096] Tables 9a-b are shown below:

[0097]

[0098] IMPACT ON CARDIOVASCULAR ENDPOINTS

[0099] In patients with TET2 CHIP mutations (VAF > 2%), the composite endpoint occurred in 2 (4.2%) patients in the colchicine group and 4 (8.3%) patients in the placebo group (hazard ratio (HR) = 0.62; 95% confidence interval (Cl): 0.11-3.42; p = 0.58). Among patients without TET2 CHIP mutations, the composite endpoint occurred in 58 (7.6%) patients in the colchicine group and 65 (8.7%) patients in the placebo group (HR = 0.85; 95% Cl: 0.60-1.21; p = 0.36). FIG. 3 illustrates cumulative incidence curves of the cardiovascular composite endpoint (cardiovascular death, resuscitated cardiac arrest, myocardial infarction, stroke, urgent hospitalization for angina requiring coronary revascularization, or any coronary revascularization) in the colchicine group (red) and the placebo group (black), in a time-to-event analysis, for patients with TET2 CHIP mutations (FIG. 3A) and without TET2 CHIP mutations (FIG. 3B).

[0100] For patients with TET2 CHIP mutations, the total number of endpoint events (first and subsequent) was 2 in the colchicine group and 5 in the placebo (HR = 0.49; 95% Cl: 0.09-2.83; p = 0.43, Table 10). In patients without TET2 CHIP mutations, the total number of events was 84 in the colchicine and 113 in the placebo group (HR = 0.71 ; 95% Cl: 0.48-1.05; p = 0.09, Table 10). Sensitivity analyses of TET2 mutation carriers at VAF > 0.5% and > 1% were concordant, showing lower event rates in TET2 mutation carriers compared to non-TET2 carriers and to individuals without CH mutations, although these differences did not reach statistical significance (T ables 11 a-b). The low number of carriers of TP53 and SF3B1 mutations precluded the conduct of stratified analyses with the composite endpoint at VAF > 0.5%. For VAF > 0, when testing for the reduction in composite endpoint events with colchicine compared to placebo in mutation carriers and non-carriers of TP53 and SF3B1 mutation carriers, the difference did not reach statistical significance (Tables 11 a-b). Table 10 is shown below:

[0101] Table 10. Hazards ratios of the effect of colchicine versus placebo on cardiovascular clinical events in patients with TET2 CHIP mutations.

[0102] 5

[0103] CHIP, clonal hematopoiesis of indeterminate potential; HR, hazard ratio. fThe composite primary endpoint is composed of cardiovascular death, resuscitated cardiac arrest, myocardial infarction, stroke, urgent hospitalization for angina requiring coronary revascularization, or any coronary revascularization. jzCox proportional hazards model adjusted for age and sex. 1] Andersen and Gill (AG) model adjusted for 0 age and sex.

[0104] T ables 11 a-b are shown below:

[0105] The CH-related mutations detected via error-corrected genomic DNA sequencing are provided in Table 12 (not included due to size constraints). CH has emerged as a candidate biomarker of disease prediction and progression, with a complex relationship with the body’s inflammatory response mechanisms. In this study, the impact of colchicine, a potent anti-inflammatory medication, on CH was explored. By using error-corrected targeted sequencing with unique molecular tags at a mean sequencing depth of 64,232x, a high level of precision was reached that enabled the detection of subtle changes in variant allele fraction (VAF) that are typically missed with whole exome or whole genome sequencing. Significant changes were observed in the proportion of mutant cells over a mean period of 1.5 years in individuals randomly assigned to receive low-dose colchicine (0.5 mg once daily) or placebo. The results show that over this period, individuals in the colchicine group had significantly lower clonal expansion as compared to those in the placebo group. When broken down into specific genes, it was found that the proportion of mutant cells was significantly decreased for TET2, TP53, and SF3B1 mutations in the colchicine group, whereas the proportion increased in the placebo group, with a significant difference between the two groups. Specifically, in the colchicine group, the VAF of TET2 mutations decreased by 10%, that of TP53 mutations decreased by 12%, and the VAF of SF3B1 mutations decreased by 20%. These results suggest that colchicine can attenuate the expansion of these clones.

[0106] This study had several notable strengths. The longitudinal design, conducted within the context of a randomized clinical trial, allowed for robust assessment of changes in CH over time under different exposures, a feature that had not yet been evaluated in this context before. By focussing on 11 of the most commonly mutated genes linked to CH, the study provides a thorough assessment of mutations that are likely to be clinically meaningful for patient populations. Furthermore, the use of high-resolution, error- corrected targeted sequencing with unique molecular tags ensured high accuracy and precision in detecting somatic mutations, enhancing the reliability of the current findings, especially in a longitudinal context where the detection of small changes is critical. Finally, the statistical approach employed, using a beta-binomial distribution to model the proportion of variant reads, effectively accounted for the inherent variability in read depth across different genomic positions, thereby further enhancing the precision and reliability of these VAF estimates and their changes over time.

[0107] In conclusion, this study demonstrated that in patients with a recent coronary event treated with a low dose of colchicine daily, the proportion of CH mutations in the TET2, TP53, and SF3B1 genes significantly decreased over the study period, and this effect was not observed in patients in the placebo group. As CH is a risk factor for several diseases, these results have potentially far-reaching implications beyond cardiovascular populations.

[0108] Other Embodiments

[0109] Various modifications and variations of the described compositions, methods, and uses of the invention will be apparent to those skilled in the art without departing from the scope and spirit of the invention. Although the invention has been described in connection with specific embodiments, it should be understood that the invention as claimed should not be unduly limited to such specific embodiments. Indeed, various modifications of the described modes for carrying out the invention that are obvious to those skilled in the art are intended to be within the scope of the invention.

[0110] All publications, patents, and patent applications are herein incorporated by reference in their entirety to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference in its entirety.

Claims

CLAIMSWhat is claimed is:1 . A method of reducing the risk of a subject developing a hematologic malignancy, the method comprising administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

2. A method of reducing the risk of relapse in a subject previously treated for a hematologic malignancy, the method comprising administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

3. A method of reducing the risk of a secondary hematologic malignancy developing in a subject being treated with a chemotherapy or a radiotherapy, the method comprising administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

4. The method of any one of claims 1-3, wherein the hematologic malignancy or secondary hematologic malignancy is a myeloid cancer.

5. The method of claim 4, wherein the myeloid cancer is acute myeloid leukemia, myeloproliferative neoplasm (MPN), mastocytosis, chronic neutrophilic leukemia, or chronic myelomonocytic leukemia.

6. The method of claim 5, wherein the myeloproliferative neoplasm is polycythemia vera, essential thrombocythemia, or myelofibrosis.

7. The method of any one of claims 1-6, wherein the clonal hematopoiesis mutation is in a ASXL1, CBL, DNMT3A, GNAS, GNB1, JAK2, PPM1D, SF3B1, SRSF2, TET2, or TP53 gene, or a combination thereof.

8. The method of claim 7, wherein the clonal hematopoiesis mutation is in a ASXL1, DNMT3A, SF3B1, TET2, or TP53 gene, or a combination thereof.

9. The method of any one of claims 1-6, wherein the clonal hematopoiesis mutation is in a SF3B1 gene.

10. The method of any one of claims 1-6, wherein the clonal hematopoiesis mutation is in a TET2 gene.

11. The method of any one of claims 1-6, wherein the clonal hematopoiesis mutation is in a TP53 gene.

12. A method of reducing the risk of a subject developing a myelodysplastic syndrome (MDS), the method comprising administering an effective amount of colchicine to the subject, wherein the subject has been determined to have a clonal hematopoiesis mutation.

13. The method of claim 12, wherein the clonal hematopoiesis mutation is in a SF3B1 gene.

14. The method of claim 3, wherein the subject is being treated for a cancer comprising a mutant TP53 gene.

15. The method of any one of claims 1-14, wherein the colchicine is administered to the subject at a dose of 0.3 to 0.7 mg.

16. The method of claim 15, wherein the colchicine is administered to the subject at a dose of 0.4 to 0.6 mg.

17. The method of claim 16, wherein the colchicine is adminstered to the subject at a dose of 0.5 mg once daily.

18. Colchicine for use in a method of reducing the risk of a subject developing a hematologic malignancy, wherein the subject has been determined to have a clonal hematopoiesis mutation.

19. Colchicine for use in a method of reducing the risk of relapse in a subject previously treated for a hematologic malignancy, wherein the subject has been determined to have a clonal hematopoiesis mutation.

20. Colchicine for use in a method of reducing the risk of a secondary hematologic malignancy developing in a subject being treated with a chemotherapy or a radiotherapy, wherein the subject has been determined to have a clonal hematopoiesis mutation.

21. Colchicine for use in a method of reducing the risk of a subject developing a myelodysplastic syndrome (MDS), wherein the subject has been determined to have a clonal hematopoiesis mutation.

22. Colchicine for use of any one of claims 18-21 , wherein the colchicine is to be administered to the subject at a dose of 0.3 to 0.7 mg.

23. Colchicine for use of claim 22, wherein the colchicine is to be administered to the subject at a dose of 0.4 to 0.6 mg.

24. Colchicine for use of claim 23, wherein the colchicine is to be adminstered to the subject at a dose of 0.5 mg once daily.

25. A kit comprising colchicine, wherein the kit further comprises a package insert instructing a user of the kit to administer the colchicine to the patient in accordance with the method of any one of claims 1-17.

Images