Methods and compositions for predicting and treating triple-negative breast cancer

Abstract

Biomarkers that can be used for the detection or diagnosis of disease states, preferably cancer (e.g., triple-negative breast cancer (TNBC)), for predicting disease prognosis and / or treatment outcomes, for identifying treatment regimens for cancer (e.g., TNBC), and / or for indicating responsiveness to cancer (e.g., TNBC) treatment regimens in a subject are described. Probes capable of detecting biomarkers, as well as related methods and kits for determining cancer (e.g., TNBC) disease states and / or identifying treatment regimens for cancer (e.g., TNBC) disease states are also described.

Description

Technical Field

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 382,702, filed Nov. 7, 2022, the entire disclosure of which is incorporated herein by reference.

[0002] The present invention generally relates to the detection or diagnosis of a disease state, preferably a cancer (e.g., triple-negative breast cancer (TNBC)) disease state, prediction of disease prognosis and / or treatment outcome, identification of a treatment method for cancer (e.g., TNBC), and / or indication of responsiveness to a treatment method and / or surgical treatment for cancer (e.g., TNBC) in a subject. Further, the present invention provides methods, systems, computer programs, reagents, and / or kits useful for these purposes. As used herein, a panel of biomarkers that indicates a cancer (e.g., triple-negative breast cancer) state, diagnoses it, is useful for predicting disease prognosis and identifying treatment methods, and / or indicates responsiveness to its treatment method and / or surgical treatment, a probe capable of detecting the panel of biomarkers, and related methods and kits thereof are provided. Methods, systems, and computer programs for predicting the disease prognosis of cancer (e.g., triple-negative breast cancer) in a subject and / or the response to treatment are also provided herein.

Background Art

[0003] Breast cancer is one of the most frequently diagnosed cancers and is increasing worldwide, including in Korea (References 1-3). Thanks to recent research, the clinical application of new treatment methods, and precision medicine approaches, the prognosis of patients with breast cancer has been greatly improved over time (Reference 4).

[0004] For better treatment outcomes in breast cancer, intensive research has enabled the development of a multidisciplinary approach that leverages the rational application of surgery, radiation, systemic chemotherapy, and endocrine therapy. Notably, intensification or attenuation of postoperative systemic chemotherapy based on multigene assays has become the standard of care in patients with hormone receptor-positive cancer (References 5, 6). However, triple-negative breast cancer remains the most challenging clinical subtype of breast cancer, possessing a variety of biological characteristics. Traditionally, it has been identified immunohistochemically as the absence of estrogen receptor (ER) / progesterone receptor (PR) / human epidermal growth factor receptor-2 (HER2) (ER / PR / HER2) expression. Compared to other subtypes with hormone receptors or HER2, the prognosis for triple-negative breast cancer is poor due to its aggressive biology and high metastatic potential, even after a good response to standard systemic chemotherapy (Reference 7). With advances in new technologies, triple-negative breast cancer has been most extensively studied using the PAM-50 subtype, classifying it according to its molecular characteristics in multi-omics analysis (References 8-11). However, no biomarkers have been identified to predict high-risk patients, guide treatment, or discover new therapeutic targets. Therefore, considerable effort should be made to address the unmet need for biomarkers that accurately predict prognosis and response to treatment in order to establish more sophisticated treatment strategies in patients with this subtype. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] (Reference 1) Collaboration, GB o. DC Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-Years for 29 Cancer Groups, 1990 to 2016: A Systematic Analysis for the Global Burden of Disease Study. JAMA Oncology 4, 1553-1568, doi:10.1001 / jamaoncol.2018.2706 (2018). [Non-Patent Document 2] (Reference 2) Sung, H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: a cancer journal for mobs 71, 209-249, doi:https: / / doi.org / 10.3322 / caac.21660 (2021). [Non-Patent Document 3] (Reference 3) Hong, S. et al. Cancer Statistics in Korea: Incidence, Mortality, Survival, and Prevalence in 2017. Cancer Res Treat 52, 335-350, doi:10.4143 / crt.2020.206 (2020). [Non-Patent Document 4] (Reference 4) Allemani, C. et al. Global surveillance of trends in cancer survival 2000-14 (CONCORD-3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population-based registries in 71 countries. Lancet (London, England) 391, 1023-1075, doi:10.1016 / s0140-6736(17)33326-3 (2018). [Non-Patent Document 5] (Reference 5) Sparano, JA et al. Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. The New England journal of medicine 379, 111-121, doi:10.1056 / NEJMoa1804710 (2018). [Non-Patent Document 6] (Reference 6) Henry, NL et al. Role of Patient and Disease Factors in Adjuvant Systemic Therapy Decision Making for Early-Stage, Operable Breast Cancer: Update of the ASCO Endorsement of the Cancer Care Ontario Guideline. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 37, 1965-1977, doi:10.1200 / jco.19.00948 (2019). [Non-Patent Document 7] (Reference 7) Carey, LA et al. The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clinical cancer research : an official journal of the American Association for Cancer Research 13, 2329-2334, doi:10.1158 / 1078-0432.Ccr-06-1109 (2007). [Non-licensed document 8] (Reference 8) Perou, CM et al. Molecular portraits of human breast tumours. Nature 406, 747-752, doi:10.1038 / 35021093 (2000).

Non-licensed literature 9

Non-licensed literature 10

Non-licensed Document 11

[0006] In a general embodiment, the present invention includes an ankyrin repeat domain 36 (ANKRD36), an ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), a B-box and SPRY domain (BSPRY), a chromosome 12 open reading frame 65 (C12 or f65), a chromosome 2 open reading frame 49 (C2 or f49), a chromosome 1 open reading frame 198 (C1 or f198), and a coiled-coil domain 114 (CCDC1). 14) Claudine 4 (CLDN4), CUE domain-containing 1 (CUEDC1), outer dynein arm docking complex subunit 1 (ODAD1), cell growth regulator with EF hand domain 1 (CGREF1), DEP domain-containing 7 (DEPDC7), double cortin-like kinase 2 (DCLK2), diacylglycerol kinase (DGKH), Disco interaction protein 2 homolog B (DIP2B), schizophrenia disruption 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH mRNA splicing and mitotic factor (ERH), growth arrest / DNA damage-inducing β (GADD45B), glutaminase (GLS), Grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member X (H2AFX), HRas proto-oncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium potential-opening channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor 7 (KLF7), keratin 17 (KRT17), LON pept Tididase N-terminal domain and ring finger protein 2 (LONRF2), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, U7 micronucleus ribonucleic acid-related (LSM11), lysosome transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisense ribonucleic acid 1 (MCM3AP_AS1), MICAL-like 2 (MICALL2), MHC class I polypeptide-related sequence B (MICB),Metallothionein 2A (MT2A), Myelin Expression Factor 2 (MYEF2), NEDD4-binding protein 3 (N4BP3), Neuroblastoma Breakpoint Family Member 20 (NBPF20), NADH: Ubiquinone Oxidoreductase Core Subunit V2 (NDUFV2), Neurogenic Locus Notch Homolog Protein 2 (NOTCH2), NADPH Oxidase Activator 1 (NOXA1), Natriuretic Peptide Receptor 3 (NPR3), Nuclear Receptor Subfamily 6 Group A Member 1 (NR6A1), P21 (RAC1) Activated Kinase 3 (PAK3), Pantothenate Kinase 3 (PANK3), Par-6 Family Cell Polarity Regulators Beta (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyBac transposable element 4 (PGBD4), Praja ring finger ubiquitin ligase 1 (PJA1), plekstrin homology and FYVE domain-containing 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K), prickle planar cell polarity protein 1 (PRICKLE1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), PYD and CARD domain-containing (PYCARD), RAS p21 protein activator 1 (RASA1), RAS family member 2 (RASD2), RAS guanyl-releasing protein 1 (RASGRP1), rhophyllin RHO GTPase-binding protein 2 (RHPN2), Rab-interacting lysosomal protein-like 2 (RILPL2), roundabout guidance receptor 1 (ROBO1), RAR orphan receptor A (RORA), stromal cell-derived factor 4 (SDF4), SERTA domain-containing 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-induced proliferation-related 1-like 2 (SIPA1L2), solute carrier family 22 member 20 (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), solute carrier family 2 member 12 (SLC2A12), solute carrier family 39 member 10 (SLC39A10),Solute carrier family 25 member 40 (SLC25A40), solute carrier family 43 member 1 (SLC43A1), solute carrier family 45 member 4 (SLC45A4), solute carrier family 6 member 20 (SLC6A20), spectrin beta, erythrocyte (SPTB), interstitial antigen 3-like 3 (STAG3L3), sucrose domain-containing 3 (SUSD3), TATA box-binding protein-related factor 10 (TAF10), t complex-associated testicular expression 3 (TCTE3), transmembrane channel-like 7 (TMC7), transmembrane and coiled-coil domain family 2 (TMCC2), transcriptional repressor GATA binding 1 (TRPS1), tubulin tyrosine ligase-like 4 (TTLL4), tRNA-yW synthesis protein 5 (TYW5), ubiquitin-conjugating enzyme E2 This relates to an isolated set of probes capable of detecting a panel of biomarkers comprising at least two biomarkers selected from the group consisting of W(UBE2W), WD repeat, sterile alpha motif and U-box domain-containing 1 (WDSUB1), N-protein N-terminal glutamine amidohydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED-type-containing 6 (ZBED6), zinc finger and BTB domain-containing 46 (ZBTB46), zinc finger CCCH-type-containing 13 (ZC3H13), zinc finger and homeobox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785).

[0007] In a particular embodiment, the biomarker panel includes DGKH, KLF7, NR6A1, PYCARD, ROBO1, SLC22A20P, SLC24A3, DIP2B, EMP1, NOTCH2, RORA, NOXA1, CUEDC1, PRICKLE1, DCLK2, C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, TYW5, ODAD1, DEPDC7, MICALL2, SLC43A1, S The biomarkers include at least two selected from the group consisting of LC6A20, RASA1, SLC45A4, NTAQ1, CGREF1, MICB, LSM11, PJA1, C2orf49, HRAS, KCNC3, MT2A, LRIT3, SHISA5, SLC25A40, H2AFX, PTPRA, RILPL2, ZC3H13, ZHX2, CLDN4, ERH, GYPC, MT2A, NDUFV2, SDF4, and UBE2W.

[0008] In certain embodiments, the biomarker panel includes at least three biomarkers, at least four biomarkers, at least five biomarkers, at least six biomarkers, at least seven biomarkers, at least eight biomarkers, at least nine biomarkers, or ten biomarkers selected from the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A 4;(b)DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3;(c)GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SLC24A3, and NTAQ1;(d)DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20;(e)DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SLC6A20;(f)CUEDC1, D GKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l) LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m) C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n) TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o) LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49;(p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q)DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICALL2, SLC43A1, RASA1, KCNC 3, and LRIT3; (t)GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0009] In certain embodiments, the probe is selected from the group consisting of aptamers, antibodies, aphibodies, peptides, proteins, organic molecules, and nucleic acids.

[0010] In a particular embodiment, a computer implementation method is provided for predicting the prognosis and / or response to treatment of cancer (e.g., breast cancer, i.e., TNBC), the computer implementation method comprising: (a) receiving computer-readable data of a panel of biomarkers for a sample from a subject; (b) analyzing the computer-readable data; (c) generating a risk score based on the analysis; (d) predicting the prognosis of cancer (e.g., breast cancer, i.e., TNBC) in the subject based on the analysis of the risk score and / or the computer-readable data; and (e) classifying the subject as high or low risk for disease progression, relapse, recurrence, and / or death based on the risk score. In one embodiment, analyzing the computer-readable data includes identifying patterns in the panel of biomarkers in the received computer-readable data that predict and / or determine the prognosis of cancer (e.g., breast cancer, i.e., TNBC). In a particular embodiment, computer-readable data may include, for example, ankyrin repeat domain 36 (ANKRD36), ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), B-box and SPRY domain-containing (BSPRY), chromosome 12 open reading frame 65 (C12 or f65), chromosome 2 open reading frame 49 (C2 or f49), chromosome 1 open reading frame 198 (C1 or f198), and coiled-coil domain-containing 114 (CCD). C114), claudin 4 (CLDN4), CUE domain-containing 1 (CUEDC1), outer dynein arm docking complex subunit 1 (ODAD1), cell growth regulator with EF-hand domain 1 (CGREF1), DEP domain-containing 7 (DEPDC7), double cortin-like kinase 2 (DCLK2), diacylglycerol kinase (DGKH), Disco interaction protein 2 homolog B (DIP2B), schizophrenia interruption 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH mRNA splicing and mitotic factor (ERH), growth arrest / DNA damage-inducible β (GADD45B), glutaminase (GLS),Grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member X (H2AFX), HRas protooncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium voltage-gated channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor 7 (KLF7), keratin 17 (KRT17), LON peptidase N-terminal domain and RING finger protein 2 (LONRF2) ), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, related to U7 micronuclear RNA (LSM11), lysosomal transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisense RNA 1 (MCM3AP_AS1), MICAL-like 2 (MICALL2), MHC class I polypeptide-related sequence B (MICB), metallothionein 2A (MT2A), Myelin expression factor 2 (MYEF2), NEDD4 binding protein 3 (N4BP3), Neuroblastoma breakpoint family member 20 (NBPF20), NADH: Ubiquinone oxidoreductase core subunit V2 (NDUFV2), Neurogenic gene locus Notch homolog protein 2 (NOTCH2), NADPH oxidase activator 1 (NOXA1), Natriuretic peptide receptor 3 (NPR3), Nuclear receptor subfamily 6 group A member 1 (NR6A1), P21 (RAC1) activating kinase 3 (PA K3), pantothenate kinase 3 (PANK3), par-6 family cell polarity regulator beta (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyback transfer factor 4 (PGBD4), Praja ring finger ubiquitin ligase 1 (PJA1), plekstrin homology and FYVE domain-containing 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K),Prickle Planar Cell Polarity Protein 1 (PRICKLE1), Protein Kinase AMP Activated Non-Catalytic Subunit Beta 2 (PRKAB2), PYD and CARD Domain-containing (PYCARD), RAS p21 Protein Activator 1 (RASA1), RAS Family Member 2 (RASD2), RAS Guanyl-Releasing Protein 1 (RASGRP1), Rhofilin RHO GTPase-binding protein 2 (RHPN2), Rab-interacting lysosomal protein-like 2 (RILPL2), roundabout guidance receptor 1 (ROBO1), RAR orphan receptor A (RORA), stroma cell-derived factor 4 (SDF4), SERTA domain-containing 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-induced proliferation-related 1-like 2 (SIPA1L2), solute carrier family 22 member 20 (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), solute carrier family 2 member 12 (SLC2A12), solute carrier family 39 member 10 (SLC39A10), solute carrier family 25 member 40 (SLC2 5A40), Solute carrier family 43 member 1 (SLC43A1), Solute carrier family 45 member 4 (SLC45A4), Solute carrier family 6 member 20 (SLC6A20), Spectrin beta, erythrocyte (SPTB), Interstitial antigen 3-like 3 (STAG3L3), Sucrose domain-containing 3 (SUSD3), TATA box-binding protein-related factor 10 (TAF10), t-complex-associated testicular expression 3 (TCTE3), Transmembrane channel-like 7 (TMC7), Transmembrane and coiled-coil domain family 2 (TMCC2), Transcriptional repressor GATA binding 1 (TRPS1), Tubulin tyrosine ligase-like 4 (TTLL4), tRNA-yW synthesis protein 5 (TYW5), Ubiquitin-conjugating enzyme E2 W (UBE2W), containing WD repeat, sterile alpha motif and U-box domain 1 (WDSUB1), N-protein N-terminal glutamine amidohydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED type 6 (ZBED6), zinc finger and BTB domain 46 (ZBTB46),The data may include a set of isolated probes capable of detecting a panel of biomarkers comprising at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or ten biomarkers selected from the group consisting of zinc finger CCCH type 13 (ZC3H13), zinc finger and homeobox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785). In certain embodiments, the method further includes a step of treating cancer in a subject (e.g., breast cancer, i.e., TNBC) based on the classification of the subject.

[0011] In a particular embodiment, a system is provided for predicting the prognosis and / or response to treatment of cancer (e.g., breast cancer, i.e., TNBC), the system comprising: (a) a receiver configured to receive computer-readable data of a panel of biomarkers for a sample from a subject; and (b) a system configured to (i) analyze the computer-readable data, (ii) generate a risk score based on the analysis thereof, (iii) predict the prognosis of TNBC in the subject based on the risk score and / or the analysis of the computer-readable data, and (iv) classify the subject as high or low risk for disease progression, relapse, recurrence, and / or death based on the risk score. In certain embodiments, computer-readable data may include, for example, ANKRD36, ANKRD36BP2, BSPRY, C12orf65, C2orf49, C1orf198, CCDC114, CLDN4, CUEDC1, ODAD1, CGREF1, DEPDC7, DCLK2, DGKH, DIP2B, DISC1, EMP1, ERH, KCNC3, KLF6, KLF7, KRT17, LONRF2, LRBA, LRIT3, LRRC37B, LSM11, LYST, These are MALAT1, MCM3AP_AS1, MICALL2, MICB, MT2A, MYEF2, N4BP3, NBPF20, NDUFV2, NOTCH2, NOXA1, NPR3, NR6A1, PAK3, PANK3, PARD6B, PEX1, PGBD4, PJA1, PLEKHF1, PNP, PPM1K, PRICKLE1, PRKAB2, PTPRA, PYCARD, RASA1, RAS, ZNF233, ZNF248, ZNF469, and ZNF785. In one embodiment, the system is configured to analyze computer-readable data and identify patterns in a panel of biomarkers in the received computer-readable data to predict and / or determine the prognosis and / or treatment outcomes of cancer (e.g., TNBC). In one particular embodiment, the system includes formulating a treatment regimen for treating cancer (e.g., TNBC) in a subject based on the subject's classification and outputting it via a display or other user interface device.

[0012] Furthermore, a method is provided for predicting the disease prognosis and / or treatment outcomes of subjects diagnosed with cancer (e.g., breast cancer, i.e., TNBC), the method comprising the steps of (a) obtaining a sample from the subject; and (b) contacting the sample with a set of isolated probes to detect a panel of biomarkers in the sample (this panel of biomarkers includes ANKRD36, ANKRD36BP2, BSPRY, C12orf65, C2orf49, C1orf198, CCDC114, CLDN4, CUE domain-containing 1, ODAD1, CGREF1, D EPDC7, DCLK2, DGKH, DIP2B, DISC1, EMP1, ERH, growth arrest / DNA damage-induced β, GLS, GRHL1, GYPC, H2AFX, HRAS, ICAM1, IMPG2, KCNC3, KLF6, KLF7, KRT17, LONRF2, L RBA, LRIT3, LRRC37B, LSM11, lysosomal transport regulator, MALAT1, MCM3AP_AS1, MICALL2, MICB, MT2A, MYEF2, N4BP3, NBPF20, NDUFV2, NOTCH2, NOXA1, NPR3, NR6A1 , containing PAK3, PANK3, PARD6B, PEX1, PGBD4, PJA1, PLEKHF1, PNP, PPM1K, PRICKLE1, PRKAB2, PTPRA, PYD and CARD domains, RASA1, RASD2, RASGRP1, RHPN2, RILPL2, ROBO1, RAR orphan receptor A, SDF4, SERTAD4, SHISA5, SIPA1L2, SLC22A20P, SLC24A3, SLC2A12, SLC39A10, SLC25A40, SLC43A1, SLC45A4, SLC6A20, S (c) comprising at least two biomarkers selected from the group consisting of PTB, STAG3L3, SUSD3, TAF10, TCTE3, TMC7, TMCC2, TRPS1, TTLL4, TYW5, UBE2W, WDSUB1, WDYHV1, NTAQ1, ZBED6, ZBTB46, ZC3H13, ZHX2, ZNF217, ZNF233, ZNF248, ZNF469, and ZNF785 (zinc finger protein 785); (c) comprising the step of analyzing the pattern of the panel of biomarkers to determine the risk score of the subject.In certain embodiments, the method further includes the step of (d) classifying subjects as high-risk or low-risk for disease progression, relapse, recurrence, and / or death based on a risk score. In certain embodiments, the method further includes the step of treating cancer (e.g., TNBC) in the subjects based on the classification of the subjects.

[0013] In certain embodiments, the biomarker panel includes biomarkers with the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (c) GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20 P, SLC24A3, and NTAQ1; (d) DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20; (e) DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, L YST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l) LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m )C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49; (p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q)DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A;(r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICALL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t)GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AF, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0014] In certain embodiments, the sample is a tissue sample, a blood sample, or a urine sample. The tissue sample is, for example, a fresh frozen tumor tissue sample or a formalin-fixed paraffin-embedded tumor tissue sample.

[0015] In certain embodiments, the subject is at high risk of disease progression, relapse, recurrence, and / or death, and the method further includes the step of administering advanced, enhanced, or standard treatment to the subject to treat cancer (e.g., TNBC). Advanced, enhanced, or standard treatment includes, for example, in the case of early TNBC, surgery and administration of chemotherapeutic agents, radiotherapy, immunotherapy agents, any novel treatment, or a combination of these treatments as neoadjuvant, adjuvant, and / or maintenance therapy. Chemotherapy agents may be selected from, for example, capecitabine, doxorubicin, cyclophosphamide, docetaxel, olaparib, carboplatin, paclitaxel, epirubicin, methotrexate, and / or fluorouracil. Immunotherapy agents may be, for example, immune checkpoint inhibitors. Immune checkpoint inhibitors may be selected from pembrolizumab, atezolizumab, nivolumab, ipilimumab, durvalumab, and / or avelumab.

[0016] In certain embodiments, the subjects are at low risk of disease progression, relapse, recurrence, and / or death, and the method further includes administering standard or attenuated treatment for cancer (e.g., TNBC). Standard or attenuated treatment may include, for example, surgery alone, or surgery and administration of chemotherapeutic agents, radiotherapy, immunotherapy agents, any novel treatment, or a combination of those treatments as neoadjuvant, adjuvant, and / or maintenance therapy, in the case of early TNBC. Chemotherapy agents may be selected from, for example, capecitabine, doxorubicin, cyclophosphamide, docetaxel, olaparib, carboplatin, paclitaxel, epirubicin, methotrexate, and / or fluorouracil. Immunotherapy agents may be, for example, immune checkpoint inhibitors. Immune checkpoint inhibitors may be selected from pembrolizumab, atezolizumab, nivolumab, ipilimumab, durvalumab, and / or avelumab.

[0017] Furthermore, a kit is provided to predict the prognosis and / or treatment outcomes of subjects diagnosed with TNBC, which includes (a) ankyrin repeat domain 36 (ANKRD36), ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), B-box and SPRY domain (BSPRY), chromosome 12 open reading frame 65 (C12orf65), chromosome 2 open reading frame 49 (C2orf49), chromosome 1 open reading frame 198 (C1orf198), and coiled doco Ill-domain-containing protein 114 (CCDC114), claudin 4 (CLDN4), CUE-domain-containing protein 1 (CUEDC1), outer dynein arm docking complex subunit 1 (ODAD1), cell growth regulator with EF-hand domain 1 (CGREF1), DEP-domain-containing protein 7 (DEPDC7), double cortin-like kinase 2 (DCLK2), diacylglycerol kinase (DGKH), Disco interaction protein 2 homolog B (DIP2B), schizophrenia disruptor 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH mRNA splicing and mitotic factors (ERH), growth arrest / DNA damage-inducing β (GADD45B), glutaminase (GLS), Grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member X (H2AFX), HRas proto-oncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium voltage-gated channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor 7 (KLF7), keratin 17 (KR T17), LON peptidase N-terminal domain and RING finger protein 2 (LONRF2), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, related to U7 micronuclear RNA (LSM11), lysosome transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisense RNA 1 (MCM3AP_AS1), MICAL-like 2 (MICALL2),MHC class I polypeptide-related sequence B (MICB), metallothionein 2A (MT2A), myelin expression factor 2 (MYEF2), NEDD4-binding protein 3 (N4BP3), neuroblastoma breakpoint family member 20 (NBPF20), NADH: ubiquinone oxidoreductase core subunit V2 (NDUFV2), neurogenic locus Notch homolog protein 2 (NOTCH2), NADPH oxidase activator 1 (NOXA1), natriuretic peptide receptor 3 (NPR3), nuclear receptor subfamily 6 group A member 1 (NR6A1), P21 (RAC1)-activated kinase 3 (PAK3), pantothenate kinase 3 (PANK3), par -6 family cell polarity regulator beta (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyBac transposable element 4 (PGBD4), Praja ring finger ubiquitin ligase 1 (PJA1), plekstrin homology and FYVE domain-containing 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K), spiky planar cell polarity protein 1 (PRICKLE1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), PYD and CARD domain-containing (PYCARD), RAS p21 protein activator 1 (RASA1), RASD family member 2 (RASD2), RAS guanylate-releasing protein 1 (RASGRP1), rhophyllin RHO GTPase-binding protein 2 (RHPN2), Rab-interacting lysosomal protein-like 2 (RILPL2), roundabout-inducing receptor 1 (ROBO1), RAR orphan receptor A (RORA), stromal cell-derived factor 4 (SDF4), SERTA domain-containing 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-inducing proliferation-related 1-like 2 (SIPA1L2), solute carrier family 22 member 20 (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), solute carrier family 2 member 12 (SLC2A12), solute carrier family 39 member 10 (SLC39A10),Solute carrier family 25 member 40 (SLC25A40), solute carrier family 43 member 1 (SLC43A1), solute carrier family 45 member 4 (SLC45A4), solute carrier family 6 member 20 (SLC6A20), spectrin beta, erythrocyte (SPTB), interstitial antigen 3-like 3 (STAG3L3), sucrose domain-containing 3 (SUSD3), TATA box-binding protein-related factor 10 (TAF10), t-complex-associated testicular expression 3 (TCTE3), transmembrane channel-like 7 (TMC7), transmembrane and coiled-coil domain family 2 (TMCC2), transcriptional repressor GATA binding 1 (TRPS1), tubulin tyrosine ligase-like 4 (TTLL4), tRNA-yW synthase 5 (TYW5), ubiquitin-conjugating enzyme E2 (b) an isolated probe set capable of detecting a panel of biomarkers comprising at least two biomarkers selected from the group consisting of W (UBE2W), WD repeat, steroid alpha motif and U-box domain-containing 1 (WDSUB1), N-protein N-terminal glutamine amide hydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED-type-containing 6 (ZBED6), zinc finger and BTB domain-containing 46 (ZBTB46), zinc finger CCCH-type-containing 13 (ZC3H13), zinc finger and homeobox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785), and (b) instructions for use. ,

[0018] In certain embodiments, an isolated set of probes capable of detecting a panel of biomarkers includes at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or ten biomarkers selected from the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD (c) 45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (d) GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SLC24A3, and NTAQ1; (e) DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20; (f) DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOX A1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65 GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l)LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m)C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49;(p) GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q) DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r) LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s) PEX1, MICALL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t) GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u) GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v) CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0019] The probe can be selected from the group consisting of, for example, aptamers, antibodies, affibodies, proteins, organic molecules, and nucleic acids.

[0020] Further aspects, features, and advantages of the present invention will be better understood upon reading the following detailed description of the invention and the claims.

Brief Description of the Drawings

[0021] The foregoing summary, as well as the following detailed description of the preferred embodiments of the present application, will be better understood when read in conjunction with the accompanying drawings. However, it should be understood that the present application is not limited to the exact embodiments shown in the drawings. [Figure 1] Figures 1A - 1B show the clinical performance evaluation of selected prognostic gene signatures. The clinical performance of the prognostic gene signatures was evaluated using receiver operating characteristic (ROC) analysis, cross - validation, and logistic regression analysis. (Figure 1A) ROC analysis of a prognostic gene signature for predicting recurrence in triple - negative breast cancer. (Figure 1B) Clinical performance of gene signatures in logistic regression analysis, cross - validation, and ROC analysis. [Figure 2]Figures 2A-2C show invasive disease survival rates in high-risk and low-risk groups. Invasive disease survival rates were analyzed in various cases. (Figure 2A) Kaplan-Meier curve for all patients. (Figure 2B) Kaplan-Meier curve for patients who received adjuvant chemotherapy. (Figure 2C) Kaplan-Meier curve for patients who received neoadjuvant chemotherapy. [Figure 3] Figures 3A-3E show the validation of the prognostic predictive ability of gene signatures in the validation cohort. Prognostic gene signatures were validated in various cases by invasive disease survival analysis. (Figure 3A) Kaplan-Meier curves for all patients. (Figure 3B) Kaplan-Meier curves for surgical specimens (primary tumors in adjuvant patients and residual tumors in neoadjuvant patients). (Figure 3C) Kaplan-Meier curves for patients treated with adjuvant chemotherapy. (Figure 3D) Kaplan-Meier curves for biopsies in patients treated with neoadjuvant chemotherapy. (Figure 3E) Kaplan-Meier curves for invasive disease survival in residual tumors of patients treated with neoadjuvant chemotherapy. [Figure 4] Figure 4 shows a flowchart of biomarker development. In order to develop biomarkers for the detection or diagnosis of triple-negative breast cancer disease status, to predict treatment outcomes, to identify treatment regimens for triple-negative breast cancer, and / or to demonstrate the responsiveness of treatment regimens for triple-negative breast cancer in a subject, in the first step, RNA sequencing was performed on the target sample to provide expression data for correlation analysis. From this analysis, differentially expressed genes were identified and candidate gene signatures were generated. The candidate gene signatures were validated through cross-validation, univariate / multivariate analysis, and meta-analysis to generate specific gene signatures, which were then validated in separate cohorts of the sample. [Figure 5]Figures 5A-5G show the PAM50 classification analysis. (Figure 5A) PCA analysis of triple-negative breast cancer patients using the PAM50 classification. (Figure 5B) Kaplan-Meier curve for the basal subtype. (Figure 5C) Kaplan-Meier curve for the Her-2 subtype. (Figure 5D) Kaplan-Meier curve for the LumA subtype. (Figure 5E) Kaplan-Meier curve for the LumB subtype. (Figure 5F) Kaplan-Meier curve for the normal subtype. (Figure 5G) Kaplan-Meier curve for the PAM50 subtype. [Figure 6] Figures 6A-6C show T cell receptor repertoire analysis. (Figure 6A) Diversity index of T cell receptor β in triple-negative breast cancer patients with or without recurrence. (Figure 6B) ROC analysis of TRB diversity to predict recurrence in triple-negative breast cancer. (Figure 6C) Kaplan-Meier curve with TRB diversity. [Figure 7] Figure 7 shows a non-limiting embodiment of a system constructed according to the principles of the present invention. [Figure 8] Figure 8 shows a non-limiting embodiment of a computer-implemented process based on the principles of the present invention. [Figure 9] Figures 9A-9B show the clinical performance evaluation of the selected prognostic gene signatures. The clinical performance of the prognostic gene signatures was evaluated using receiver operating characteristic (ROC) analysis, cross-validation, and logistic regression analysis. (Figure 9A) ROC analysis of prognostic gene signatures for predicting recurrence in triple-negative breast cancer. (Figure 9B) Clinical performance of gene signatures in logistic regression analysis, cross-validation, and ROC analysis. [Figure 10] Figures 10A-10C show invasive disease survival rates in high-risk and low-risk groups. Invasive disease survival rates were analyzed in various cases. (Figure 10A) Kaplan-Meier curves for all patients. (Figure 10B) Kaplan-Meier curves for patients treated with adjuvant chemotherapy. (Figure 10C) Kaplan-Meier curves for patients treated with neoadjuvant chemotherapy. [Figure 11]Figures 11A-11E show the prognostic validation of gene signatures in the validation cohort. Prognostic gene signatures were validated by invasive disease survival analysis in various cases. (Figure 11A) Kaplan-Meier curves for all patients. (Figure 11B) Kaplan-Meier curves for surgical specimens (primary tumors in adjuvant patients and residual tumors in neoadjuvant patients). (Figure 11C) Kaplan-Meier curves for patients treated with adjuvant chemotherapy. (Figure 11D) Kaplan-Meier curves for biopsies in patients treated with neoadjuvant chemotherapy. (Figure 11E) Kaplan-Meier curves for invasive disease survival in residual tumors in patients treated with neoadjuvant chemotherapy. [Figure 12] Figures 12A-12G show PAM50 classification analysis. (Figure 12A) PCA analysis of triple-negative breast cancer patients using the PAM50 classification. (Figure 12B) Kaplan-Meier curve for the basal subtype. (Figure 12C) Kaplan-Meier curve for the Her-2 subtype. (Figure 12D) Kaplan-Meier curve for the LumA subtype. (Figure 12E) Kaplan-Meier curve for the LumB subtype. (Figure 12F) Kaplan-Meier curve for the normal subtype. (Figure 12G) Kaplan-Meier curve for the PAM50 subtype. [Figure 13] Figures 13A-13C show T cell receptor repertoire analysis. (Figure 13A) Diversity index of T cell receptor β in triple-negative breast cancer patients with or without recurrence. (Figure 13B) ROC analysis of TRB diversity to predict recurrence in triple-negative breast cancer. (Figure 13C) Kaplan-Meier curve with TRB diversity. [Figure 14] Figures 14A–14C show immune cell clustering. Immune cells of the same type were closely clustered (Figure 14A). Further subclusters were analyzed within each T cell cluster and myelocyte cluster from the previous clustering (Figures 14B, 14C). [Figure 15]Figures 15A-15E show gene set signatures associated with CD8-positive T cells. The gene set signatures associated with CD8-positive T cells were identified as follows: growth arrest / DNA damage-inducible β, H2AFX, PTPRA, TILPL2, RAR orphan receptor A, zinc finger CCCH-type 13, and ZHX2 (AUC = 0.927, sensitivity = 92.31%, specificity = 93.65%, precision = 93.42%). (Figures 15A, 15B) In Kaplan-Meier (KM) analysis, patients with tumors possessing high-risk gene signatures (n=16, median iDFS = 42.7 months) showed a significantly shorter iDFS than patients with low-risk gene signatures (n=60, median iDFS not reached). (Figures 15C, 15D) CD8-positive T cell-associated gene signatures were marked on CD8-positive T cells near CD4-positive T cells in t-SNE (Figure 15E). [Figure 16] Figures 16A-16E show macrophage-related gene set signatures. Macrophage-related gene signatures were identified as follows: CLDN4, ERH, growth arrest / DNA damage-inducible β, glycophorin C, H2AFX, MT2A, NDUFV2, SDF4, UBE2W (AUC=0.963, sensitivity=92.31%; specificity=93.65%; precision=93.42%) (Figures 16A, 16B). In Kaplan-Meier (KM) analysis, patients with tumors possessing high-risk gene signatures (n=16, median iDFS=42.7 months) showed a significantly shorter iDFS than patients with low-risk gene signatures (n=60, median iDFS not reached) (Figures 16C, 16D). Macrophage-related gene signatures were marked on macrophages, monocytes, and dendritic cells in t-SNE (Figure 16E). [Modes for carrying out the invention]

[0022] Various publications, articles, and patents are cited or referenced throughout the background and this specification, and each of these references is incorporated herein by reference in its entirety. The consideration of documents, acts, materials, apparatus, articles, etc., included herein is for the purpose of providing context for the invention. Such consideration does not constitute an admission that any or all of these matters form part of the prior art with respect to the disclosed or claimed invention.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art in which the present invention pertains. Otherwise, any particular terms used herein have the meanings set forth in the specification.

[0024] When used herein and in the appended claims, the singular forms "a," "an," and "the" should be noted as including plural references unless the context clearly indicates otherwise.

[0025] Unless otherwise stated, any numerical values, such as concentrations or concentration ranges, described herein should be understood in all cases as being modified by the term “approximately.” Thus, numerical values ​​typically include ±10% of the listed value. For example, a concentration of 1 mg / mL includes 0.9 mg / mL to 1.1 mg / mL. Similarly, a concentration range of 1% to 10% (w / v) includes 0.9% (w / v) to 11% (w / v). Where used herein, the use of numerical ranges explicitly includes all possible subranges, all individual numerical values ​​within that range (including integers and fractions of values ​​within such ranges), unless the context clearly indicates otherwise.

[0026] Unless otherwise indicated, the term “at least” preceding a set of elements should be understood to refer to all elements of the set. Those skilled in the art will recognize, or can verify by mere conventional experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the invention.

[0027] As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having,” “contains,” or “containing,” or any other variation thereof, are understood to mean the inclusion of the integer or group of integers listed, but not the exclusion of any other integer or group of integers, and are intended to be non-exclusive or open-ended. For example, a composition, mixture, process, method, system, article, or apparatus containing a list of elements is not necessarily limited to those elements alone and may include other elements not expressly enumerated or specific to such composition, mixture, process, method, article, system, or apparatus. Furthermore, unless expressly stated otherwise, “or” refers to an inclusive “or” and not an exclusive “or.” For example, condition A or B is satisfied by one of the following: A is true (or exists) and B is false (or does not exist), A is false (or does not exist) and B is true (or exists), and both A and B are true (or exist).

[0028] As used herein, the connecting term “and / or” between multiple enumerated elements is understood to encompass both individual and combined options. For example, when two elements are joined by “and / or,” the first option refers to the applicability of the first element without the second element; the second option refers to the applicability of the second element without the first element; and the third option concerns the application of the first and second elements together. Any one of these options is understood to be included in its meaning and therefore satisfy the requirements of the term “and / or” as used herein. The simultaneous applicability of two or more of the options is also understood to fall within the scope of its meaning and therefore satisfy the requirements of the term “and / or.”

[0029] As used herein and in the claims, the term "consists of," or variations such as "consist of" or "consisting of," indicates the inclusion of any enumerated integer or group of integers, but that no further integers or groups of integers may be added to the specified method, structure, or composition.

[0030] As used herein and throughout the claims, the terms “consists essentially of,” or variations such as “consist essentially of” or “consisting essentially of,” indicate the inclusion of any enumerated integer or group of integers, and the inclusion of any selection of any enumerated integer or group of integers that does not substantially alter the basic or novel properties of the specified method, structure, or composition. See MPEP § 2111.03.

[0031] Furthermore, as will be understood by those skilled in the art, the terms “about,” “approximately,” “generally,” “substantially,” and similar terms used herein when referring to the dimensions or characteristics of preferred components of an invention should be understood to indicate that the described dimensions / characteristics are not strict boundaries or parameters and do not exclude slight variations from those that are functionally the same or similar. At the very least, such references involving numerical parameters include variations that do not change the least significant digit, using mathematical and industrial principles accepted in the art (e.g., rounding, measurement or other systematic errors, manufacturing tolerances, etc.).

[0032] As used herein, “biomarker” refers to a gene or protein whose level of expression or concentration in a sample is altered or indicates a state compared to that of a normal or healthy sample. Biomarkers disclosed herein are genes and / or proteins whose level of expression or concentration or timing of expression or concentration correlates with the ability to determine whether a subject is at high or low risk for a cancerous (e.g., breast cancer, i.e., triple-negative breast cancer) disease state. As used herein, the term “high risk” means having a high risk of developing cancer, a high risk of further progression of cancer, a high risk of cancer recurrence or recurrence, and / or death. As used herein, the term “low risk” means having a low risk of developing cancer, a low risk of further progression of cancer, a low risk of cancer recurrence or recurrence, and / or death. Determining whether a subject is at high or low risk for cancer (e.g., breast cancer, i.e., triple-negative breast cancer) can lead to determining, for example, the outcomes of biological therapy in subjects diagnosed with cancer (e.g., breast cancer, i.e., triple-negative breast cancer), and / or determining a biological therapy treatment program for subjects with cancer (e.g., breast cancer, i.e., triple-negative breast cancer).

[0033] As used herein, “probe” means any molecule or drug capable of selectively binding to an intended target biomolecule. The target molecule may be a biomarker, for example, a nucleotide transcript or protein encoded by or corresponding to a biomarker. Probes are synthesized by those skilled in the art or derived from appropriate biological preparations in consideration of this disclosure. Probes are specifically designed to be labeled. Examples of molecules used as probes include, but are not limited to, ribonucleic acids, deoxyribonucleic acids, proteins, peptides, antibodies, aptamers, aphibodies, and organic molecules.

[0034] As used herein, “predicting treatment outcomes” when referring to subjects having cancer (e.g., breast cancer such as TNBC) means that a panel of biomarkers can, for example, determine and / or determine which subjects will respond to which particular treatment for cancer (e.g., breast cancer, i.e., TNBC). As an example, a method disclosed herein can predict and / or determine whether a subject is at high risk for further progression of TNBC, and if a subject is predicted and / or determined to be at high risk, the subject is treated with advanced, intensified, or standard forms of treatment, including, for example, surgery and further chemotherapy, radiotherapy, immunotherapy, any novel therapeutic agent, or a combination of those therapies as neoadjuvant, adjuvant, and / or maintenance therapy in the case of early TNBC. As another example, the methods disclosed herein can predict and / or determine whether a subject is at low risk of further progression of TNBC, and if a subject is predicted and / or determined to be at low risk, the subject may be treated with standard or reduced treatment, for example, in the case of early TNBC, with surgery alone, or with surgery and chemotherapeutic agents, radiotherapy, immunotherapy agents, any novel therapeutic agents, or a combination thereof as neoadjuvant, adjuvant, and / or maintenance therapy.

[0035] As used herein, “prognosis of triple-negative breast cancer” or “prognosis of TNBC” refers to predicting various conditions of a subject with TNBC, such as complete recovery from TNBC, the likelihood of TNBC recurrence, and / or the likelihood of survival after being diagnosed with TNBC. This may vary based on various factors, such as the severity of TNBC, the time of diagnosis, and / or the progress of treatment. TNBC can be effectively treated when various treatment methods are appropriately applied according to the prognosis.

[0036] As used herein, “subject” means any animal, preferably a mammal, most preferably a human. As used herein, the term “mammal” encompasses any mammal. Examples of mammals include, but are not limited to, cattle, horses, sheep, pigs, cats, dogs, mice, rats, rabbits, guinea pigs, monkeys, and humans. More preferably, humans.

[0037] As used herein, “sample” is intended to include any sample of cells, tissues, or bodily fluids in which the expression of a biomarker can be detected. Examples of such samples include, but are not limited to, biopsies, smears, blood, lymph, urine, saliva, or any other bodily secretions or derivatives thereof. Blood may include, for example, whole blood, plasma, serum, or any derivative of blood. Samples can be obtained from a subject by various techniques known to those skilled in the art. A sample may be, for example, a fresh frozen tumor sample. A sample may be, for example, a formalin-fixed paraffin-embedded (FFPE) tumor tissue sample.

[0038] With respect to the methods of the present invention, the term "administering" means a method for therapeutically or preventively preventing, treating or improving a syndrome, disorder, or disease described herein (e.g., cancer such as breast cancer, i.e., triple-negative breast cancer). Such a method involves administering an effective amount of the therapeutic agent (e.g., chemotherapy) at different points in time during the course of therapy, or simultaneously in combination. The methods of the present invention should be understood to encompass all known therapeutic treatment regimens.

[0039] The term "effective dose" means the amount of an active compound or drug that elicits a biological or medical response in a tissue system, animal or human, as sought by researchers, veterinarians, physicians, or other clinicians, including preventing, treating, or improving the symptoms of a syndrome, disorder or disease being treated, or a syndrome, disorder or disease being treated (e.g., cancer such as breast cancer, i.e., triple-negative breast cancer).

[0040] Biomarker panel and probes for detecting biomarkers The present invention generally relates to predicting prognosis and / or treatment outcomes for treatment regimens for cancer (e.g., breast cancer, i.e., triple-negative breast cancer) in a subject, and provides methods, reagents, systems, and kits useful for this purpose. Biomarkers that predict the prognosis and / or response to treatment regimens for cancer (e.g., breast cancer such as triple-negative breast cancer) in a subject are provided herein. In certain embodiments, the present invention provides a panel of biomarkers (e.g., genes or proteins expressed in a subject at a specific point in time) that can be used to predict and / or determine the prognosis of cancer (e.g., breast cancer such as triple-negative breast cancer), and / or predict and / or determine treatments, or to indicate responsiveness to treatments for cancer (e.g., breast cancer such as triple-negative breast cancer).

[0041] Any method available in the art for detecting the expression of a biomarker is incorporated herein. The expression, presence, or amount of the biomarkers of the present invention can be detected at the nucleic acid level (e.g., as a ribonucleic acid transcript) or at the protein level. "Detecting or determining the expression of a biomarker" is intended to include determining the amount or presence of a protein or its ribonucleic acid transcript for the biomarker disclosed herein. Thus, "detecting expression" includes cases where it is determined that the biomarker is not expressed, not detected, expressed at a low level, expressed at a normal level, or overexpressed.

[0042] In some embodiments, the Specified herein provides diagnostic methods based on deoxyribonucleic acid, ribonucleic acid, and proteins for directly or indirectly detecting biomarkers described herein. The Invention also provides compositions, reagents, systems, and kits for such diagnostic purposes. The diagnostic methods described herein may be qualitative or quantitative. Quantitative diagnostic methods may be used, for example, to compare detected biomarker levels with cutoff or threshold levels. Where applicable, qualitative or quantitative diagnostic methods may also include amplification of a target, signal, or intermediary.

[0043] In certain embodiments, biomarkers are detected at the nucleic acid (e.g., ribonucleic acid) level. For example, the amount of biomarker ribonucleic acid (e.g., mRNA) present in a sample is determined (e.g., to determine the level of biomarker expression). Biomarker nucleic acids (e.g., ribonucleic acid, amplified cDNA, etc.) can be detected / quantified using a variety of nucleic acid techniques known to those skilled in the art, including, but not limited to, nucleic acid hybridization and nucleic acid amplification, such as RNA-Seq (RNA sequencing), reverse transcription (RT)-polymerase chain reaction (PCR), RT-quantitative PCR (RT-qPCR), competitive RT-PCR, RNase protection assays, Northern blotting, and deoxyribonucleic acid chips.

[0044] In certain embodiments, microarrays are used to detect biomarkers. Microarrays can include, for example, deoxyribonucleic acid microarrays, protein microarrays, tissue microarrays, cell microarrays, compound microarrays, and antibody microarrays. Deoxyribonucleic acid microarrays, commonly called gene chips, can be used to simultaneously monitor the expression levels of thousands of genes. Microarrays can be used to identify disease genes by comparing their expression in diseased and normal states. Microarrays can also be used for diagnostic purposes. That is, patterns of gene expression levels can be studied in a sample before or after the diagnosis of a disease (e.g., triple-negative breast cancer), and these patterns can be used later to predict the prognosis and / or treatment of the disease in subjects at risk of or diagnosed with the disease, or the response to a specific treatment for the disease in subjects at risk of or diagnosed with the disease.

[0045] In certain embodiments, the expression product is a protein corresponding to a biomarker in the panel. In certain embodiments, detecting the level of the expression product involves exposing the sample to an antibody against the protein corresponding to the biomarker in the panel. In certain embodiments, the antibody is covalently bound to a solid surface. In certain embodiments, detecting the level of the expression product involves exposing the sample to a mass spectrometry technique (e.g., mass spectrometry).

[0046] Methods for detecting protein expression levels and / or patterns using antibodies include, but are not limited to, Western blotting, ELISA (enzyme-linked immunosorbent assay), radioimmunoassay, radioimmunodiffusion, Octarony immunodiffusion analysis, rocket immunoelectrophoresis, immunohistochemistry, immunoprecipitation assay, complement fixation assay, fluorescence-activated cell sequencing (FACS), and protein chips.

[0047] In certain embodiments, reagents for the detection and / or quantification of biomarker proteins are provided. These reagents include, but are not limited to, primary antibodies that bind to protein biomarkers, secondary antibodies that bind to primary antibodies, affibodies that bind to protein biomarkers, aptamers that bind to protein or nucleic acid biomarkers (e.g., ribonucleic acid or deoxyribonucleic acid), and / or nucleic acids that bind to nucleic acid biomarkers (e.g., ribonucleic acid or deoxyribonucleic acid). The detection reagents may be labeled (e.g., fluorescently) or unlabeled. Furthermore, the detection reagents may be free in solution or immobilized.

[0048] In certain embodiments, when quantifying the level of a biomarker(s) present in a sample, the level can be determined on an absolute or relative basis. If determined absolutely or relatively, a comparison with a control may be made, which may include, but is not limited to, past samples from the same patient (e.g., a series of samples over a specific period), or levels, thresholds, and acceptable ranges found in subjects or populations of subjects without the disease or disorder (e.g., triple-negative breast cancer).

[0049] Accordingly, the present invention provides an isolated probe set capable of detecting a biomarker panel that serves as an indicator for predicting treatment outcomes in TNBC patients. In a particular embodiment, an isolated probe set is provided that is capable of detecting a biomarker panel comprising at least two biomarkers selected from the group consisting of the following biomarkers. Ankyrin repeat domain 36 (ANKRD36), Ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), containing B-box and SPRY domain (BSPRY), chromosome 12 open reading frame 65 (C12orf65), chromosome 2 open reading frame 49 (C2orf49), chromosome 1 open reading frame 198 (C1orf198), containing coiled-coil domain 114 (CCDC114), claw Din 4 (CLDN4), CUE domain-containing protein 1 (CUEDC1), outer dynein arm docking complex subunit 1 (ODAD1), EF hand domain-containing cell growth regulator (CGREF1), DEP domain-containing protein 7 (DEPDC7), double cortin-like kinase 2 (DCLK2), diacylglycerol kinase (DGKH), Disco interaction protein 2 homolog B (DIP2B), schizophrenia-related factor 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH mRNA splicing and mitotic factors (ERH), growth arrest / DNA damage-inducing β (GADD45B), glutaminase (GLS), grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member X (H2AFX), HRas proto-oncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium potential-opening channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor Child 7 (KLF7), keratin 17 (KRT17), LON peptidase N-terminal domain and RING finger protein 2 (LONRF2), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, U7 nuclear small RNA-related (LSM11), lysosomal transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisense RNA1 (MCM3AP_AS1), MICAL-like 2 (MICALL2), MHC class I polypeptide-related sequence B (MICB), metallothionein 2A (MT2A), myelin expression factor 2 (MYEF2), NEDD4 binding protein 3 (N4BP3), neuroblastocyte breakpoint family member 20 (NBPF20), NADH: ubiquinone oxidoreductase core subunit V2 (NDUFV2), neurogenic locus Notch homolog protein 2 (NOTCH2), NADPH oxidase activator 1 (NOXA1), natriuretic peptide receptor 3 (NPR3), nuclear receptor subfamily 6 group A member 1 (NR6A1), P21 (RAC1) activating kinase 3 (PAK3), pantothenate kinase 3 (PANK3), par-6 family cell polarity regulator beta (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyBac transposable element-derived 4 (PGBD4), Praja ring finger ubiquitin ligase 1 (PJA1), plekstrin homology and FYVE domain-containing 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K), prickle planar cell polarity protein 1 (PRICKLE1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), PYD and CARD domain-containing (PYCARD), RAS p21 protein activator 1 (RASA1), RAS family member 2 (RASD2), RAS guanylate-releasing protein 1 (RASGRP1), rhophyllin RHOGTPase-binding protein 2 (RHPN2), Rab-interacting lysosomal protein-like protein 2 (RILPL2), roundabout guidance receptor 1 (ROBO1), RAR orphan receptor A (RORA), stromal cell-derived factor 4 (SDF4), SERTA domain-containing protein 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-induced proliferation-related protein 1-like protein 2 (SIPA1L2), solute carrier family 22 member 20 (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), solute carrier family 2 member 12 (SLC2A12), solute carrier family 39 member 10 (SLC39A10), solute carrier family 25 member 40 (SLC25 A40), Solute Carrier Family 43 Member 1 (SLC43A1), Solute Carrier Family 45 Member 4 (SLC45A4), Solute Carrier Family 6 Member 20 (SLC6A20), Spectrin Beta, Erythrocyte (SPTB), Interstitial Antigen 3-like 3 (STAG3L3), Sucrose Domain-containing 3 (SUSD3), TATA Box-binding Protein-related Factor 10 (TAF10), t-complex-associated Testicular Expression 3 (TCTE3), Transmembrane Channel-like 7 (TMC7), Transmembrane and Coiled-Coil Domain Family 2 (TMCC2), Transcriptional Repressor GATA-binding 1 (TRPS1), Tubulin Tyrosine Ligase-like 4 (TTLL4), tRNA-YW Synthesizing Protein 5 (TYW5), Ubiquitin-conjugating Enzyme E2 This group consists of W (UBE2W), WD repeat, sterile alpha motif and U-box domain-containing 1 (WDSUB1), N-protein N-terminal glutamine amide hydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED type-containing 6 (ZBED6), zinc finger and BTB domain-containing 46 (ZBTB46), zinc finger CCCH type-containing 13 (ZC3H13), zinc finger and homeobox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785).

[0050] In a particular embodiment, the biomarker panel includes DGKH, KLF7, NR6A1, PYCARD, ROBO1, SLC22A20P, SLC24A3, DIP2B, EMP1, NOTCH2, RORA, NOXA1, CUEDC1, PRICKLE1, DCLK2, C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, TYW5, ODAD1, DEPDC7, MICALL2, SLC43A1, S The biomarkers include at least two selected from the group consisting of LC6A20, RASA1, SLC45A4, NTAQ1, CGREF1, MICB, LSM11, PJA1, C2orf49, HRAS, KCNC3, MT2A, LRIT3, SHISA5, SLC25A40, H2AFX, PTPRA, RILPL2, ZC3H13, ZHX2, CLDN4, ERH, GYPC, MT2A, NDUFV2, SDF4, and UBE2W.

[0051] In certain embodiments, an isolated set of probes can detect a panel of biomarkers containing three or more, four or more, five or more, six or more, seven or more, eight or more, nine or more, or ten biomarkers.

[0052] In a particular embodiment, the biomarker panel includes at least two biomarkers, at least three biomarkers, at least four biomarkers, at least five biomarkers, at least six biomarkers, at least seven biomarkers, at least eight biomarkers, at least nine biomarkers, or ten biomarkers selected from the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SL (b) C22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (c) GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SLC24A3, and NTAQ1; (d) DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20; (e) DGKH, EMP1, GADD45B, LYST, SLC22A20 P, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k)C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l)LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m)C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A;(o) LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49; (p) GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q) DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r) LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s) PEX1, MI CALL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t)GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0053] In a particular embodiment, the biomarker panel includes the following biomarkers: (a), (b), (c), or (d). (a) Diacylglycerol kinase (DGKH), growth arrest and DNA damage-inducible beta (GADD45B), Kruppel-like factor (KLF7), lysosome transport regulator (LYST), nuclear receptor subfamily 6 group A member 1 (NR6A1), PYD and CARD domain-containing (PYCARD), roundabout guidance receptor 1 (ROBO1), solute carrier family 22 member 20 pseudogene (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), and solute carrier family 45 member 4 (SLC45A4) (b) Chromosome 12 open reading frame 65 (C12orf65), growth arrest and DNA damage-inducible beta (GADD45B), LPS-responsive beige-like anchor protein (LRBA), lysosome transport regulator (LYST), peroxisome biosynthesis factor 1 (PEX1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), and tRNA-yW synthesis protein 5 (TYW5). (c) Growth arrest and DNA damage-inducible beta (GADD45B), H2A histone family member X (H2AFX), protein tyrosine phosphatase receptor type A (PTPRA), Rab-interacting lysosomal protein-like 2 (RILPL2), RAR-associated orphan receptor A (RORA), zinc finger CCCH-containing 13 (ZC3H13), and zinc finger and homeobox 2 (ZHX2). (d) Claudin 4 (CLDN4), ERH mRNA splicing and mitotic factor (ERH), growth arrest and DNA damage-inducing beta (GADD45B), glycophorin C (GYPC), H2A histone family member X (H2AFX), metallothionein 2A (MT2A), NADH: ubiquinone oxidoreductase core subunit V2 (NDUFV2), stromal cell-derived factor 4 (SDF4), and ubiquitin-conjugating enzyme E2 W (UBE2W).

[0054] A probe is any molecule or drug that specifically detects a biomarker. In certain embodiments, the probe is selected from the group consisting of aptamers, antibodies, aphibodies, peptides, and nucleic acids (such as oligonucleotides that hybridize to the biomarker gene or mRNA). An aptamer is an oligonucleotide or peptide that specifically binds to a target molecule. Aptamers are typically prepared by selection from a large random sequence pool. Examples of aptamers useful in the present invention include oligonucleotides such as deoxyribonucleic acid, ribonucleic acid or nucleic acid analogs, or peptides that bind to the biomarker of the present invention.

[0055] How to use A method is provided for predicting the prognosis and / or treatment outcomes of subjects with cancer or candidates at risk of developing cancer. The method comprises (a) obtaining a sample from a subject; (b) contacting the sample with a set of isolated probes capable of detecting a panel of biomarkers in the sample; and (c) analyzing the pattern of the panel of biomarkers to determine a risk score for the subject. In certain embodiments, the method further includes (d) classifying the subject as high-risk or low-risk for disease progression, relapse, recurrence, and / or death based on the risk score. In certain embodiments, the cancer is breast cancer. In certain embodiments, the breast cancer is triple-negative breast cancer (TNBC). In certain embodiments, TNBC is early-stage triple-negative breast cancer. In certain embodiments, the method further includes treating the triple-negative breast cancer in the subject based on the classification of the subject.

[0056] In certain embodiments, the subject is classified as high-risk, and the method may be an advanced, enhanced, or standard form of treatment, for example in the case of early TNBC, including surgery and / or chemotherapy, radiotherapy, immunotherapy, any novel therapeutic agent, or a combination of these treatments as neoadjuvant therapy, adjuvant therapy, and / or maintenance therapy. Chemotherapy agents may include, for example, capecitabine, doxorubicin, cyclophosphamide, docetaxel, olaparib, carboplatin, paclitaxel, epirubicin, methotrexate, and / or fluorouracil. Immunotherapy agents may be, for example, immune checkpoint inhibitors. Immune checkpoint inhibitors may include, for example, pembrolizumab, atezolizumab, nivolumab, ipilimumab, durvalumab, and / or avelumab.

[0057] In certain embodiments, the subjects are classified as low-risk, and the treatment may be standard or attenuated treatment, for example, in the case of early TNBC, including surgery alone, or surgery and chemotherapy, radiotherapy, immunotherapy, any novel treatment, or a combination of these treatments as neoadjuvant therapy, adjuvant therapy, and / or maintenance therapy. Chemotherapy agents may include, for example, capecitabine, doxorubicin, cyclophosphamide, docetaxel, olaparib, carboplatin, paclitaxel, epirubicin, methotrexate, and / or fluorouracil. Immunotherapy agents may be, for example, immune checkpoint inhibitors. Immune checkpoint inhibitors may include, for example, pembrolizumab, atezolizumab, nivolumab, ipilimumab, durvalumab, and / or avelumab.

[0058] The sample may be, for example, a tissue sample, a blood sample, or a urine sample. Preferably, the sample is a tissue sample from the subject. The tissue sample may be, for example, a tumor tissue sample embedded in fixed formalin and paraffin.

[0059] In certain embodiments, the biomarker panel includes the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (c) GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SLC2 4A3, and NTAQ1; (d) DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20; (e) DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l) LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m) C 12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49; (p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q)DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A;(r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICALL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t)GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0060] kit Kits are also provided for predicting the response to treatment regimens for cancer (e.g., TNBC such as breast cancer) in the target population. The kits include, for example, (a) an isolated set of probes capable of detecting a panel of biomarkers containing at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or ten biomarkers selected from (a) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SL24A3; (a) DGKH, GA (b) DD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (c) GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SLC24A3, and NTAQ1; (d) DGKH (e) DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5;(l)LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m)C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49; (p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q)DEPDC7, SLC43A1, MICB, C2orf49, HR AS, KCNC3, and MT2A; (r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICALL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t)GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W, and (b) Instructions for Use.

[0061] In certain embodiments, a set of isolated probes capable of detecting a panel of biomarkers includes: (a) diacylglycerol kinase (DGKH), growth arrest / DNA damage-inducing β (GADD45B), Kruppel-like factor (KLF7), lysosome transport regulator (LYST), nuclear receptor subfamily 6 group A member 1 (NR6A1), PYD and CARD domain-containing (PYCARD), roundabout guidance receptor 1 (ROBO1), solute transporter family 22 member 20 pseudogene (SLC22A20P), solute transporter family 24 member 3 (SLC24A3), and solute transporter family 45 member 4 (SLC45A4); (b) chromosome 12 open reading frame 65 (C12orf65), growth arrest / DNA damage-inducing (c) Conductive β (GADD45B), LPS-responsive beige-like anchor protein (LRBA), lysosome transport regulator (LYST), peroxisome biosynthesis factor 1 (PEX1), protein kinase AMP-activated non-catalytic subunit β2 (PRKAB2), and tRNA-yW synthesis protein 5 (TYW5); (c) Growth arrest / DNA damage-inducible β (GADD45B), H2A histone family member X (H2AFX), protein tyrosine phosphatase receptor type A (PTPRA), Rab-interacting lysosomal protein-like 2 (RILPL2), RAR orphan receptor A (RORA), zinc finger CCCH-containing 13 (ZC3H13), and zinc finger and homeobox 2 (ZHX2); or (d) Claudin 4 (CLDN4), ERH This includes biomarker signatures for mRNA splicing and mitotic factor (ERH), growth arrest / DNA damage-inducible β (GADD45B), glycophorin C (GYPC), H2A histone family member X (H2AFX), metallothionein 2A (MT2A), NADH:ubiquinone oxidoreductase core subunit V2 (NDUFV2), stromal cell-derived factor 4 (SDF4), and ubiquitin-conjugating enzyme E2 W (UBE2W).

[0062] Compositions for use in the methods disclosed herein include, but are not limited to, probes, antibodies, affibodies, nucleic acids, and / or aptamers. Preferred compositions can detect the expression levels (e.g., ribonucleic acid or protein levels) of a panel of biomarkers from a biological sample.

[0063] Any of the compositions may be provided in the form of a kit or reagent mixture. For example, labeled probes may be provided in a kit for the detection of a panel of biomarkers. The kit may include, but is not limited to, detection reagents (e.g., probes), buffers, control reagents (e.g., positive and negative controls), amplification reagents, solid supports, labels, instructions for use, etc., and may include all components necessary or sufficient for the assay. In a particular embodiment, the kit includes a set of probes for a panel of biomarkers, solid supports, and reagents for processing the sample to be tested (e.g., reagents for isolating proteins or nucleic acids from the sample).

[0064] Computer implementation method A computer implementation method is provided for predicting the prognosis and / or treatment outcomes of subjects diagnosed with cancer (e.g., breast cancer such as triple-negative breast cancer) or candidates at risk of developing cancer (e.g., breast cancer such as triple-negative breast cancer). The computer implementation method includes the steps of (a) receiving computer-readable data of a panel of biomarkers for a sample derived from the subject; (b) generating a risk score based on the analysis; (c) predicting the prognosis of triple-negative breast cancer in the subject based on the analysis of the risk score and / or computer-readable data; and (d) classifying the subject as high-risk or low-risk for disease progression, recurrence, relapse, and / or death based on the risk score. In certain embodiments, the computer-readable data may include data from an isolated probe set capable of detecting a panel of biomarkers comprising at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or ten biomarkers selected from, for example, the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A 3;(c)GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SLC24A3, and NTAQ1;(d)DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20;(e)DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SL C6A20; (f)CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g)DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h)DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1;(i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l) LYST, PEX1, ODAD1, DEPDC 7, MICALL2, SLC43A1, and SLC6A20; (m) C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n) TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o) LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2or f49;(p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1;(q)DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A;(r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS;(s)PEX1, MICALL2, SLC43A1, RASA1, KC NC3, and LRIT3; (t)GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0065] In one embodiment, the analysis of computer-readable data includes identifying patterns in a panel of biomarkers in the received computer-readable data that predict and / or determine the prognosis of cancer (e.g., breast cancer such as TNBC). In a particular embodiment, the method further includes the step of treating cancer (e.g., breast cancer such as TNBC) in a subject based on the classification of the subject.

[0066] system A system is provided for predicting the prognosis and / or treatment outcomes of subjects diagnosed with cancer (e.g., breast cancer such as TNBC) or subjects at risk of developing cancer (e.g., breast cancer such as TNBC). The system includes (a) a receiver configured to receive computer-readable data of a panel of biomarkers for a sample from a subject, and (b) a system configured to (i) analyze the computer-readable data, (ii) generate a risk score based on the analysis, (iii) predict the prognosis for cancer (e.g., breast cancer such as TNBC) in the subject based on the risk score and / or the analysis of the computer-readable data, and (iv) classify the subject as high-risk or low-risk for disease progression, relapse, recurrence, and / or death based on the risk score. In certain embodiments, the computer-readable data may include data from an isolated probe set capable of detecting a panel of biomarkers comprising at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or ten biomarkers selected from, for example, the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6 A1, RORA, SLC22A20P, and SLC24A3; (c) GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SLC24A3, and NTAQ1; (d) DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20; (e) DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3;(h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65, GADD45B, LRBA, LYST, PEX 1, PRKAB2, and TYW5; (l)LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m)C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49; (p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q)DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICALL2, SLC 43A1, RASA1, KCNC3, and LRIT3; (t) GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u) GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v) CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W. In certain embodiments, the system formulates a treatment regimen for treating cancer in the subject (e.g., breast cancer such as TNBC) based on the subject's classification and outputs it via a display or other user interface device.

[0067] The present invention also provides the following non-limiting embodiments.

[0068] Embodiment 1 includes an ankyrin repeat domain 36 (ANKRD36), an ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), a B-box and SPRY domain (BSPRY), a chromosome 12 open reading frame 65 (C12orf65), a chromosome 2 open reading frame 49 (C2orf49), a chromosome 1 open reading frame 198 (C1orf198), a coiled-coil domain 114 (CCDC114), claudin 4 (CLDN4), and a CUE domain 1 (CUEDC1). External dynein arm docking complex subunit 1 (ODAD1), cell growth regulator with EF hand domain 1 (CGREF1), DEP domain-containing 7 (DEPDC7), double cortin-like kinase 2 (DCLK2), diacylglycerol kinase (DGKH), Disco interaction protein 2 homolog B (DIP2B), schizophrenia-related 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH ribonucleic acid splicing and mitotic factor (ERH), growth arrest / deoxyribonucleic acid damage-inducing β (GADD45B), glutaminase (GLS), grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member x (H2AFX), HRas proto-oncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium voltage-gated channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor 7 (KLF7), keratin 17 (KRT17), LON peptidase N-terminal domain and RING finger protein 2 (LONRF2), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, U7 nuclear small ribonucleic acid-related (LSM11), lysosome transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisense ribonucleic acid 1 (MCM3AP_AS1), MICAL-like 2 (MICALL2), MHC class I polypeptide-related sequence B (MICB),Metallothionein 2A (MT2A), Myelin Expression Factor 2 (MYEF2), NEDD4-binding protein 3 (N4BP3), Neuroblastoma Breakpoint Family Member 20 (NBPF20), NADH: Ubiquinone Oxidoreductase Core Subunit V2 (NDUFV2), Neurogenic Locus Notch Homogen Protein 2 (NOTCH2), NADPH Oxidase Activator 1 (NOXA1), Natriuretic Peptide Receptor 3 (NPR3), Nuclear Receptor Subfamily 6 Group A Member 1 (NR6A1), P21 (RAC1) Activated Kinase 3 (PAK3), Pantothenate Kinase 3 (PANK3), Par-6 Family Cell Polarity Regulator (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyBac transposable element 4 (PGBD4), Praja ring finger ubiquitin ligase 1 (PJA1), plekstrin homology and FYVE domain-containing 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K), prickle planar cell polarity protein 1 (PRICKLE1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), PYD and CARD domain-containing (PYCARD), RAS p21 protein activator 1 (RASA1), RAS family member 2 (RASD2), RAS guanylate-releasing protein 1 (RASGRP1), rhophyllin RHO GTPase-binding protein 2 (RHPN2), Rab-interacting lysosomal protein-like 2 (RILPL2), roundabout-inducing receptor 1 (ROBO1), RAR orphan receptor A (RORA), stromal cell-derived factor 4 (SDF4), SERTA domain-containing 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-inducing proliferation-related 1-like 2 (SIPA1L2), solute transporter family 22 member 20 (SLC22A20P), solute transporter family 24 member 3 (SLC24A3), solute transporter family 2 member 12 (SLC2A12), solute transporter family 39 member 10 (SLC39A10), solute transporter family 25 member 40 (SLC25A40),Solute transporter family 43 member 1 (SLC43A1), solute transporter family 45 member 4 (SLC45A4), solute transporter family 6 member 20 (SLC6A20), spectrin beta, erythrocyte (SPTB), interstitial antigen 3-like 3 (STAG3L3), sucrose domain-containing 3 (SUSD3), TATA box-binding protein-related factor 10 (TAF10), t-complex-associated testicular expression 3 (TCTE3), transmembrane channel-like 7 (TMC7), transmembrane and coiled-coil domain family 2 (TMCC2), transcriptional repressor GATA binding 1 (TRPS1), tubulin tyrosine ligase-like 4 (TTLL4), tRNA-yW synthesis protein 5 (TYW5), ubiquitin-conjugating enzyme E2 This is an isolated set of probes capable of detecting a panel of biomarkers containing at least two biomarkers selected from the group consisting of W (UBE2W), WD repeat, sterile alpha motif and U-box domain-containing 1 (WDSUB1), N-protein N-terminal glutamine amide hydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED type-containing 6 (ZBED6), zinc finger and BTB domain-containing 46 (ZBTB46), zinc finger CCCH type-containing 13 (ZC3H13), zinc finger and homeobox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785).

[0069] Embodiment 2 is a panel of biomarkers including DGKH, KLF7, NR6A1, PYCARD, ROBO1, SLC22A20P, SLC24A3, DIP2B, EMP1, NOTCH2, RORA, NOXA1, CUEDC1, PRICKLE1, DCLK2, C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, TYW5, ODAD1, DEPDC7, MICALL2, SLC43A1, SLC6A20, RASA1, SLC This is a set of isolated probes according to Embodiment 1, comprising at least two biomarkers selected from the group consisting of 45A4, NTAQ1, CGREF1, MICB, LSM11, PJA1, C2orf49, HRAS, KCNC3, MT2A, LRIT3, SHISA5, SLC25A40, H2AFX, PTPRA, RILPL2, ZC3H13, ZHX2, CLDN4, ERH, GYPC, MT2A, NDUFV2, SDF4, and UBE2W.

[0070] Embodiment 3 is a set of isolated probes according to Embodiment 1 or 2, wherein the biomarker panel includes at least three biomarkers, at least four biomarkers, at least five biomarkers, at least six biomarkers, at least seven biomarkers, at least eight biomarkers, at least nine biomarkers, or ten biomarkers selected from the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SLC24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (c) Growth arrest / DNA damage-inducing β, lysosomal transport regulator, containing NOXA1, NR6A1, PYD and CARD domains, SLC22A20P, SLC24A3, and NTAQ1; (d) DGKH, KLF7, lysosomal transport regulators, NR6A1, ROBO1, SLC24A3, and SLC6A20; (e) DGKH, EMP1, growth arrest / DNA damage-inducing β, lysosome transport regulator, SLC22A20P, SLC24A3, and SLC6A20; (f) CUE domain-containing 1, DGKH, EMP1, lysosomal transport regulator, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, lysosomal transport regulator, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, growth arrest / DNA damage-inducing β, lysosome transport regulator, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) Growth arrest / DNA damage-inducing β, KLF7, lysosomal transport regulator, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, growth arrest / DNA damage-inducing β, lysosome transport regulator, NR6A1, SLC22A20P, and NTAQ1; (k)C12orf65, growth arrest / DNA damage-inducing β, LRBA, lysosomal transport regulator, PEX1, PRKAB2, and TYW5; (l) Lysosomal transport regulators, PEX1, ODAD1, DEP domain-containing 7, MICAL-like 2, SLC43A1, and SLC6A20; (m)C12orf65, growth arrest / DNA damage-induced β, lysosomal transport regulator, PEX1, RASA1, SLC45A4, NTAQ1; (n)TYW5, DEP domain-containing 7, SLC43A1, CGREF1, MICB, HRAS, MT2A; (o)LRBA, lysosomal transport regulator, PEX1, DEP domain-containing 7, SLC43A1, MICB, C2orf49; (p) Growth arrest / DNA damage-induced β, PEX1, DEP domain-containing 7, SLC43A1, LSM11, PJA1; (q) DEP domain-containing 7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, MT2A; (r) Lysosomal transport regulators, DEP domain-containing 7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICAL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t) Growth arrest / DNA damage-inducing β, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AF, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v) CLDN4, ERH, growth arrest / DNA damage-inducing β, glycophorin C, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0071] Embodiment 4 is an isolated probe set according to any one of Embodiments 1 to 3, wherein the probe is selected from the group consisting of aptamers, antibodies, aphibodies, proteins, organic molecules, and nucleic acids.

[0072] Embodiment 5 is a method for predicting the disease prognosis and / or treatment outcomes of a subject diagnosed with cancer, wherein: (a) Obtain a sample from the subject; (b) Contacting the sample with an isolated probe set described in any one of Embodiments 1 to 4 to detect a panel of biomarkers in the sample; and (c) Analyze the pattern of the biomarker panel to determine the risk score of the subject.

[0073] Embodiment 6 is the method of Embodiment 5, which further includes: d) A step of classifying the subject as high-risk or low-risk based on the risk score.

[0074] Embodiment 7 is the method of Embodiment 5 or 6, wherein the cancer is breast cancer.

[0075] Embodiment 8 is the method of Embodiment 7, wherein the breast cancer is triple-negative breast cancer (TNBC).

[0076] Embodiment 9 is the method of Embodiment 8, wherein the TNBC is an early-stage TNBC.

[0077] Embodiment 10 is the method of Embodiment 8 or 9, further comprising treating triple-negative breast cancer in a subject based on the classification of the subject.

[0078] Embodiment 11 is a method according to any one of Embodiments 5 to 10, the subject being high-risk, and the method further includes advanced, enhanced, or standard forms of treatment for triple-negative breast cancer, including surgery and / or administration of chemotherapy agents, radiotherapy, immunotherapy agents, any novel therapeutic agents, or combinations of treatments.

[0079] Embodiment 11a is one of the methods of Embodiments 5 to 10, relating to high-risk cases, and further comprising advanced, enhanced, or standard forms of treatment for triple-negative breast cancer, including surgery and / or administration of additional chemotherapeutic agents, radiotherapy, immunotherapy agents, any novel therapeutic agents, or combinations of treatments as neoadjuvant, adjuvant, and / or maintenance therapy.

[0080] Embodiment 12 is the method according to Embodiment 11 or 11a, wherein the immunotherapy agent is an immune checkpoint inhibitor.

[0081] Embodiment 13 is the method according to Embodiment 12, wherein the immune checkpoint inhibitor comprises pembrolizumab, atezolizumab, nivolumab, ipilimumab, durvalumab, and / or avelumab.

[0082] Embodiment 14 is the method according to Embodiment 11 or 11a, wherein the chemotherapeutic agent comprises capecitabine, doxorubicin, cyclophosphamide, docetaxel, olaparib, carboplatin, paclitaxel, epirubicin, methotrexate, and / or fluorouracil.

[0083] Embodiment 15 is a method from any one of Embodiments 5 to 10, the subject being low risk, and the method further comprises administering standard or attenuated treatment for triple-negative breast cancer, either surgery alone or including surgery, and / or administering chemotherapy, radiotherapy, immunotherapy, any novel treatment, or a combination of treatments.

[0084] Embodiment 15a is one of the methods from Embodiments 5 to 10, which is low-risk, and the method further includes administering standard or attenuated treatment for triple-negative breast cancer, which includes surgery alone, or surgery and administration of further chemotherapeutic agents, radiotherapy, immunotherapy, any novel treatment, or a combination of treatments as neoadjuvant, adjuvant, and / or maintenance therapy.

[0085] Embodiment 16 is the method according to Embodiment 15 or 15a, wherein the immunotherapy agent is an immune checkpoint inhibitor.

[0086] Embodiment 17 is the method according to Embodiment 16, wherein the immune checkpoint inhibitor comprises pembrolizumab, atezolizumab, nivolumab, ipilimumab, durvalumab, and / or avelumab.

[0087] Embodiment 18 is the method of Embodiment 15 or 15a, wherein the chemotherapeutic agent comprises capecitabine, doxorubicin, cyclophosphamide, docetaxel, olaparib, carboplatin, paclitaxel, epirubicin, methotrexate, and / or fluorouracil.

[0088] Embodiment 19 is one of the methods described in Embodiments 5 to 18, wherein the sample is a tissue sample, a blood sample, or a urine sample.

[0089] Embodiment 20 is the method of Embodiment 19, wherein the tissue sample is a fresh frozen tumor tissue sample or a fixed formalin paraffin-embedded tumor tissue sample.

[0090] Embodiment 21 is a kit for predicting disease prognosis and / or treatment outcomes for subjects diagnosed with triple-negative breast cancer (TNBC), which: (a) Ankyrin repeat domain 36 (ANKRD36), ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), containing B box and SPRY domain (BSPRY), chromosome 12 open reading frame 65 (C12orf65), chromosome 2 open reading frame 49 (C2orf49), chromosome 1 open reading frame 198 (C1orf198), containing coiled-coil domain 114 (CCDC114), claudin 4 (CLDN4), CUE domain-containing 1 (CUEDC1), outer dynein arm docking complex subunit 1 (ODAD1), cell growth regulator with EF hand domain 1 (CGREF1), DEP domain-containing 7 (DEPDC7), double cortin-like kinase 2 (DCLK2), diacylglycerol kinase (DGKH), Disco interaction protein 2 homolog B (DIP2B), schizophrenia-related 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH mRNA splicing and mitotic factors (ERH), growth arrest / DNA damage-inducing β (GADD45B), glutaminase (GLS), grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member X (H2AFX), HRas proto-oncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium voltage-gated channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor Child 7 (KLF7), keratin 17 (KRT17), LON peptidase N-terminal domain and RING finger protein 2 (LONRF2), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, U7 nuclear small RNA-related (LSM11), lysosomal transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisense RNA1 (MCM3AP_AS1), MICAL-like 2 (MICALL2), MHC class I polypeptide-related sequence B (MICB), metallothionein 2A (MT2A), myelin expression factor 2 (MYEF2), NEDD4 binding protein 3 (N4BP3), neuroblastoma breakpoint family member 20 (NBPF20), NADH: ubiquinone oxidoreductase core subunit V2 (NDUFV2), neurogenic gene locus Notch homolog protein 2 (NOTCH2), NADPH oxidase activator 1 (NOXA1), natriuretic peptide receptor 3 (NPR3), nuclear receptor subfamily 6 group A member 1 (NR6A1), P21 (RAC1) activating kinase 3 (PAK3), pantothenate kinase 3 (P ANK3), par-6 family cell polarity regulator beta (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyBac transposable element 4 (PGBD4), Praja ring finger ubiquitin ligase 1 (PJA1), pleckstrin homology and FYVE domain-containing 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K), prickle planar cell polarity protein 1 (PRICKLE1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), PYD and CARD domain-containing (PYCARD), RAS p21 protein activator 1 (RASA1), RASAD family member 2 (RASD2), RAS guanylate-releasing protein 1 (RASGRP1), rhophilin RHOGTPase-binding protein 2 (RHPN2), Rab-interacting lysosomal protein-like protein 2 (RILPL2), roundabout guidance receptor 1 (ROBO1), RAR-related orphan receptor A (RORA), stromal cell-derived factor 4 (SDF4), SERTA domain-containing protein 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-induced proliferation-related protein 1-like protein 2 (SIPA1L2), solute carrier family 22 member 20 (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), solute carrier family 2 member 12 (SLC2A12), solute carrier family 39 member 10 (SLC39A10), solute carrier family 25 member 40 (SLC2 5A40), Solute carrier family 43 member 1 (SLC43A1), Solute carrier family 45 member 4 (SLC45A4), Solute carrier family 6 member 20 (SLC6A20), Spectrin beta, erythrocyte (SPTB), Interstitial antigen 3-like 3 (STAG3L3), Sucrose domain-containing 3 (SUSD3), TATA box-binding protein-related factor 10 (TAF10), t-complex-associated testicular expression 3 (TCTE3), Transmembrane channel-like 7 (TMC7), Transmembrane and coiled-coil domain family 2 (TMCC2), Transcriptional repressor GATA binding 1 (TRPS1), Tubulin tyrosine ligase-like 4 (TTLL4), tRNA-yW synthesis protein 5 (TYW5), Ubiquitin-conjugating enzyme E2An isolated set of probes capable of detecting a panel of biomarkers comprising at least two biomarkers selected from the group consisting of W(UBE2W), WD repeat, sterile alpha motif and U-box domain-containing 1 (WDSUB1), N-protein N-terminal glutamine amidohydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED-type-containing 6 (ZBED6), zinc finger and BTB domain-containing 46 (ZBTB46), zinc finger CCCH-type-containing 13 (ZC3H13), zinc finger and homeobox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785); and (b) Instructions for use.

[0091] Embodiment 22 is a kit of Embodiment 21, wherein an isolated set of probes capable of detecting a panel of biomarkers comprises at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or ten biomarkers selected from the following biomarker signatures: (a) DGKH, growth arrest / DNA damage-inducing β, KLF7, lysosome transport regulator, NR6A1, PYD and CARD domain-containing, ROBO1, SLC22A20P, SLC24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, growth arrest / DNA damage-inducible β, MT2A, NOTCH2, NR6A1, RAR orphan receptor A, SLC22A20P, and SLC24A3; (c) Growth arrest / DNA damage-inducing β, lysosomal transport regulator, containing NOXA1, NR6A1, PYD and CARD domains, SLC22A20P, SLC24A3, and NTAQ1; (d) DGKH, KLF7, lysosomal transport regulators, NR6A1, ROBO1, SLC24A3, and SLC6A20; (e) DGKH, EMP1, growth arrest / DNA damage-inducing β, lysosome transport regulator, SLC22A20P, SLC24A3, and SLC6A20; (f) CUE domain-containing 1, DGKH, EMP1, lysosomal transport regulator, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, lysosomal transport regulator, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, growth arrest / DNA damage-inducing β, lysosome transport regulator, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) Growth arrest / DNA damage-inducing β, KLF7, lysosomal transport regulator, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, growth arrest / DNA damage-inducing β, lysosome transport regulator, NR6A1, SLC22A20P, and NTAQ1; (k)C12orf65, growth arrest / DNA damage-inducing β, LRBA, lysosomal transport regulator, PEX1, PRKAB2, and TYW5; (l) Lysosomal transport regulators, PEX1, ODAD1, DEP domain-containing 7, MICAL-like 2, SLC43A1, and SLC6A20; (m)C12orf65, growth arrest / DNA damage-inducing β, lysosomal transport regulator, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEP domain-containing 7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, lysosomal transport regulator, PEX1, DEP domain-containing 7, SLC43A1, MICB, and C2orf49; (p) Growth arrest / DNA damage-inducing β, PEX1, DEP domain-containing 7, SLC43A1, LSM11, and PJA1; (q) DEP domain-containing 7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r) Lysosomal transport regulators, DEP domain-containing 7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICAL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t) Growth arrest / DNA damage-inducing β, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AF, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v) CLDN4, ERH, growth arrest / DNA damage-inducible β, glycophorin C, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0092] Embodiment 23 is the kit described in Embodiment 21 or 22, in which an isolated set of probes capable of detecting a panel of biomarkers includes the biomarker signatures of DGKH, growth arrest / DNA damage-inducible β, KLF7, lysosomal transport regulator, NR6A1, PYD and CARD domain-containing, ROBO1, SLC22A20P, SLC24A3, and SLC45A4.

[0093] Embodiment 24 is the kit described in Embodiment 21 or 22, in which an isolated set of probes capable of detecting a panel of biomarkers includes the biomarker signatures of C12orf65, growth arrest / DNA damage-inducible β, LRBA, lysosomal transport regulator, PEX1, PRKAB2, and TYW5.

[0094] Embodiment 25 is the kit described in Embodiment 21 or 22, wherein the isolated set of probes capable of detecting a panel of biomarkers includes biomarker signatures for growth arrest / DNA damage-inducible β, H2AF, PTPRA, RILPL2, RAR orphan receptor A, zinc finger CCCH type 13, and ZHX2.

[0095] Embodiment 26 is the kit described in Embodiment 21 or 22, wherein the isolated set of probes capable of detecting a panel of biomarkers includes the biomarker signatures of CLDN4, ERH, growth arrest / DNA damage-inducible β, glycophorin C, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

[0096] Embodiment 27 is a kit according to any one of Embodiments 21 to 26, wherein the probe is selected from the group consisting of aptamers, antibodies, aphibodies, peptides, proteins, organic molecules, and nucleic acids. [Examples]

[0097] material and method This study included 184 patients with early-stage triple-negative breast cancer (TNBC). The 76 patients in the training cohort were from the National Cancer Center Korea (NCC), and the 108 patients in the validation cohort were from Samsung Medical Center (SMC). All patients were eligible if they were 18 years of age or older with early-stage TNBC (stages I-III), were able to safely obtain histological biopsies, and had received standard systemic chemotherapy, including surgery and radiotherapy, as well as local treatment. Tumor samples were identified as TNBC according to the American Society of Clinical Oncology / College of American Pathologists (ASCO / CAP) guidelines for evaluation of ER, PR, and HER2 (References 12, 13). The training cohort consisted of 15 patients who received neoadjuvant chemotherapy and 61 patients who received adjuvant chemotherapy after primary surgery for early-stage TNBC between March 2002 and August 2018. The validation cohort included 73 patients who received neoadjuvant chemotherapy between July 2011 and November 2017, and 35 patients who received adjuvant chemotherapy after surgery. In the validation cohort, 42 specimens in the neoadjuvant chemotherapy group were biopsy tissues taken before neoadjuvant chemotherapy, and the other specimens were surgical tissues. All specimens were fresh-frozen. All patients provided written informed consent, and the study protocol was approved by the Institutional Review Boards of National Cancer Center Korea and Samsung Medical Center (NCC IRB# 2012-08-065, SMC IRB# 2014-11-015).

[0098] Complete clinical information and outcomes Clinical data, including date of diagnosis, clinical and surgical stage, response to neoadjuvant chemotherapy, recurrence, and survival, were collected from medical records. Disease-free survival (IDFS) is defined as the time from diagnosis of primary breast cancer to invasive breast cancer recurrence or death from any cause.

[0099] Ribonucleic acid extraction and cDNA library preparation Total ribonucleic acids were extracted using TRIzol reagent (Invitrogen, Thermo Fisher Scientific, CA, USA) and the AllPrep Deoxyribonucleic Acid / Ribonucleic Acid Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol. Deoxyribonucleic acid contamination was eliminated using DNase. RNA quality control was assessed by RNA integrity number (RIN) using the Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbronn, Germany), and RNA with a RIN greater than 8 passed quality control. cDNA libraries were prepared using the TruSeq Stranded mRNA LT Sample Prep Kit (Illumina, CA, USA) according to the manufacturer's protocol (TruSeq Stranded mRNA Sample Preparation Guide, Part #15031047 Rev. E).

[0100] RNA sequencing and quality control For the training cohort, paired-end sequencing was performed using an Illumina HiSeq 4000 sequencer (Illumina, CA, USA) with a cDNA library prepared for RNA sequencing. During RNA sequencing quality control, artifacts including adapter sequences, contaminating DNA, and PCR replicas were removed to reduce bias in the sequencing data. After quality control of the sequencing data, aligned reads were generated by mapping the sequencing data onto a reference genome using the HISAT2 program (GitHub, http: / / daehwankimlab.github.io / hisat2 / ). Transcript assembly was performed using StringTie (https: / / ccb.jhu.edu / software / stringtie / ) with the generated aligned reads. Based on transcript quantification for each sample, expression levels were normalized to transcript length and coverage depth. Through normalization, the expression profile was extracted as fragments (FPKMs) per kilobase of transcript per million mapped reads.

[0101] For the validation cohort, sequencing libraries were prepared using fresh frozen tissue with the TruSeq RNA Sample Preparation Kit v2 (Illumina Inc.) according to the manufacturer's protocol. RNA library sequencing was performed on the HiSeq 2500 sequencing platform (Illumina Inc.). After trimming low-quality bases from the FASTQ files, reads were aligned against the human reference genome (hg19) using STAR v.2.5, and gene expression was estimated in terms of fragments per kilobase per million exons (FPKM) using RSEM v.1.3. Quality control of sequencing results was evaluated using RNA-SeQC (v1.1.8).

[0102] For comparison between tumor and non-tumor data, non-tumor data were collected from the Gene Expression Omnibus (GEO, ncbi.nlm.nih.gov / geo). GSE58135 (GEO accession number) was selected as the non-tumor group in GEO, which had non-tumor RNA sequencing data from 21 patients with TNBC. In the non-tumor data, the results were either in a failed state or 1.0 × 10⁶. -6 Data with values ​​less than a certain amount were excluded.

[0103] Gene combination analysis By matching tumor and non-tumor genes, 10,856 genes were identified in both groups. These 10,856 genes were used to analyze differentially expressed genes (DEGs). DEGs were screened to meet one of the following criteria: 1) a statistically significant difference between tumor and non-tumor individuals, and 2) a statistically significant difference between patients who showed recurrence / metastasis after surgical resection and those who did not. Previously screened DEGs were further selected using Cox regression analysis for recurrence / metastasis. Before combining DEGs, the Cox regression coefficients for each gene were identified, and gene expression was weighted by the corresponding coefficient values. Gene signature was calculated using the formula (Reference 14):

number

[0104] The number of candidate DEGs analyzed in combination and the total number of gene combinations are given by the formula (Reference 14):

number

[0105] Here, n is the total number of selected DEGs, and k is the number of genes included in the combination.

[0106] Pre-validation of candidate gene signatures using machine learning cross-validation The optimal gene combination was identified by ranking candidate gene signatures (achieving p-value < 0.05, area under the curve (AUC) > 0.90, sensitivity > 90%, and specificity > 90%) using k-fold cross-validation. Patients were randomly divided into two sets (training set and test set) 300 times (14).

[0107] Meta-analysis-based signaling pathway analysis Signal transduction analysis is performed using CBS Probe PINGS, which consists of five modules: PPI module, Path-Finder module, Path-Linker module, Path-maker module, and Path-Lister module. TM The study was conducted using (Reg. No.2008-01-129-000568; CbsBioscience, Daejeon, Korea) (14). To validate the gene signature, signaling was analyzed for pathways associated with DEG in each patient, comparing recurrent / metastatic and non-recurrent / non-metastatic, as well as for pathways associated with the gene signature. Genes were mapped to signaling pathways obtained from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. The top 10 signaling pathways for each patient's DEG and gene signature were selected according to the weight of interactions and the number of interacting genes. The 10 pathways associated with DEG and the gene signature-associated pathways for each patient were compared. For each selected signaling pathway, the gene interaction frequency ratio, which is the score of genes interacting with the signaling gene in gene signature validation, was calculated in the signaling pathway analysis. The top 10 high-interaction-frequency genes were selected by applying a 100% gene interaction frequency to the highest probability of gene interaction within each signaling pathway. In addition, we compared 10 high-interaction-frequency genes associated with DEG in each patient, as well as the gene signatures associated with these high-interaction-frequency genes.

[0108] Molecular subtype classification using PAM 50 analysis PAM 50 call analysis was performed using R v.3.4.3 (R Development Core Team, r-project.org) with publicly available R scripts (References 15, 16). The median of FPKM data and PAM50 centroid data was used as the library. Patient-specific subtypes and relapse risk (ROR) scores were analyzed by R using the above library configuration by inputting patient RNA sequencing data. The prognostic potential of PAM 50 calls was analyzed using ROR-S with KM analysis.

[0109] T-cell receptor diversity analysis Using MiXCR 2.1.3 (GitHub, github.com / milaboratory / mixcr), T cell receptor (TCR) profiles were obtained using RNA sequencing data (References 17, 18). The RNA sequencing data was aligned to all IG / TCR loci. After two contig assemblies, the V / J junctions of the TCRs were extended. The assembled chronotypes were exported. TCR diversity was analyzed using the Shannon index in T cell receptor β-rox (TCRB). The Shannon index is given by the following formula:

number

[0110] Here, s is the number of different chronotypes, ni is the clonal size of the i-th chronotype, and N is the total number of TCRB sequences analyzed. KM analysis was used to analyze the prognostic predictive power of TCRB diversity.

[0111] statistical analysis Clinicopathological variables between the training and validation cohorts were assessed using the chi-square test or Fisher's exact test. Gene expression data were tested for normality using the Shapiro-Wilk test. Since the data did not meet the normality assumption, the Wilcoxon test was used to assess significant differences between responders and non-responders. Receiver operating characteristic (ROC) curve analysis was used to determine the precision of thresholds for classifying relapse / metastasis and non-relapse / non-metastasis using gene signatures. Kaplan-Meier survival (KM) curves were calculated using death and invasive disease as endpoints in iDFS. Differences in KM curves were examined using the log-rank test, and differences in hazard ratios were examined using Cox regression analysis. Candidate gene signatures were analyzed using Cox regression to understand the relationships between relapse / metastasis, classification, and clinicopathological variables. Significance was set at p < 0.05. All statistical analyses were performed using R v.3.4.3 software (R Development Core Team, r-project.org / ).

[0112] Example 1: Results of Analysis 1 Patient's Clinical Characteristics Of the 184 patients, 76 were in the training cohort and 108 in the validation cohort. Table 1 summarizes the clinical characteristics of patients in the training and validation cohorts. Overall, there were no significant differences between the cohorts in patients in the very young age group. Patients in the training cohort were more likely to have earlier disease stages than those in the validation cohort. However, there was no difference between the two groups in the stage distribution among patients who underwent surgery or who had residual tumor after neoadjuvant chemotherapy. The TAC (Taxotere, Adriamycin, and Cyclophosphamide) regimen was used more frequently in the validation cohort for postoperative chemotherapy, likely due to the higher proportion of patients in more advanced stages of the disease. Schematic diagrams of patients and samples are shown in Figures 5A-5G.

[0113] Table 1: Pathological baselines of the training cohort and validation cohort. [Table 1-1] [Table 1-2] [Table 1-3]

[0114] TNM, tumor-lymph node-metastasis (AJCC stage); pCR, pathological complete response; Event, recurrence or metastasis; AC, Adriamycin, cyclophosphamide; AC-D, AC followed by docetaxel; TC, Taxotere and cyclophosphamide; FAC, 5-FU, Adriamycin, cyclophosphamide; AC-wP, AC followed by weekly paclitaxel; AC-PC, AC followed by paclitaxel and carboplatin; PCarbo, paclitaxel and carboplatin; DA, docetaxel and Adriamycin; TAC, Taxotere, and AC. * The p-value was calculated using Fisher's exact test.

[0115] DEG analysis of tumor vs. non-tumor and recurrent / metastatic vs. non-recurrent / non-metastatic In tumor-nontumor-specific DNA (DEG) analysis, 9,741 out of 10,856 genes were significantly differentially expressed with changes exceeding 1.5-fold. DEG analysis of primary tumors between recurrent and non-recurrent patients revealed that 141 out of 10,856 genes showed a significant 1.5-fold difference in expression. Subsequently, 587 out of 10,856 genes were statistically significant in single-Cox analysis. 70 DEGs overlapped across the three different DEG analyses (Table 2).

[0116] Table 2: DEGS correlation with prognosis of triple-negative breast cancer [Table 2-1] [Table 2-2] [Table 2-3] [Table 2-4] [Table 2-5]

[0117] Candidate gene signatures based on gene combinations and selected gene signatures through cross-validation. The top 10 candidate gene signatures were ranked by AUC. The 10 candidates had equal values ​​for sensitivity (90.91), specificity (100.00), and precision (98.68), but differed in their AUC. Prognostic gene signatures were selected using two-fold cross-validation precision. The selected gene signatures were C12orf65; growth arrest / DNA damage-inducible β; LRBA; lysosomal transport regulator; PEX1; PRKAB2; and TYW5, which demonstrated a cross-validation precision of 94.67% and were statistically significant in discrete Cox analysis. The risk score was calculated using a cutoff value of 4.043659 as follows: (-0.334912 × C12orf65) + (0.018572 × growth arrest / DNA damage-inducible β) + (0.124030 × LRBA) + (0.257051 × lysosomal transport regulator) + (0.046903 × PEX1) + (0.220736 × PRKAB2) + (0.083961 × TYW5) (Figures 1A-1B, Table 3).

[0118] Table 3: Candidate gene signatures as prognostic biomarkers for triple-negative breast cancer [Table 3-1] [Table 3-2]

[0119] AUC, Area under the curve; C12orf65, Chromosome 12 open reading frame 65; GADD45B, Growth arrest / DNA damage-inducible β; LRBA, LPS-responsive beige-like anchor protein; LYST, Lysosome transport regulator; PEX1, Peroxisome biosynthesis factor 1; PRKAB2, Protein kinase AMP-activated non-catalytic subunit β2; TYW5, tRNA-yW synthesis protein 5; ODAD1, Outer dynein arm docking complex subunit 1; DEPDC7, DEP domain-containing 7; MICALL2, MICAL-like 2; SLC43A1, Solute carrier family 43 member 1; SLC6A20, Solute carrier family 6 member 20; RASA1, RAS p21 protein activator 1, SLC45A4, solute carrier family 45 member 4, NTAQ1, N-terminal glutamine amidase 1, CGREF1, cell growth regulator with EF hand domain 1, HRAS, HRas proto-oncogene, GTPase, MICB, MHC class I polypeptide-related sequence B, MT2A, metallothionein 2A, C2orf49, chromosome 2 open reading frame 49, LSM11, LSM11, U7 nuclear small RNA-related, PJA1, plugging finger ubiquitin ligase 1, KCNC3, potassium potential open channel subfamily C member 3, LRIT3, leucine-rich repeat, Ig-like and transmembrand main 3, SHISA5, SHISA family member 5, SLC25A40, solute carrier family 25 member 40

[0120] Prognostic significance of genetic signatures in training cohorts During a median follow-up of 29.5 months (range: 4.6–185.9), patients with tumors possessing a high-risk genetic signature (n=10) were more likely to have a low-risk signature (n=66, median not reached, p=6.35×10) than patients with tumors possessing a high-risk genetic signature in the overall population. -12The high-risk group showed significantly shorter invasive disease survival (iDFS) (median, 95% confidence interval [CI]: 21.9, 11.5 to not reached) than the low-risk group (Figure 2A). Further analyses in a separate group of patients who underwent primary surgery and in patients with residual tumor after neoadjuvant chemotherapy showed similar results. Among patients who underwent primary surgery and patients with residual tumor after neoadjuvant chemotherapy, the median iDFS in the high-risk group was 43.4 months (95% CI: 15.5 to not reached, p=0.00000923) and 18.2 months (95% CI: 4.6 to not reached, p=0.000020), respectively, while the median iDFS in the low-risk group was not reached (Figure 2B, C).

[0121] Prognostic significance of gene signatures in validation cohorts The median follow-up period for the validation cohort was 45.6 months (range; 6.6–74.5). Across the entire validation cohort, patients with high-risk genetic signatures did not reach a median iDFS, but their risk of recurrence or metastasis was significantly higher than those with low-risk genetic signatures (median iDFS not reached, p = 0.0055 by log-rank test). When patients were segmented according to treatment order, the prognostic significance of genetic signatures in surgical tissue from patients who underwent primary surgery was consistent with the training cohort (p = 0.0084). However, median iDFS was not yet reached in the high-risk and low-risk groups. High-risk genetic signatures remain effective in predicting prognosis in patients with residual tumor after neoadjuvant chemotherapy. The median iDFS was 15.9 months (95% CI, 10.1–not reached) in the high-risk group, but not in the low-risk group (p = 0.053). However, when genetic signatures were examined in tissue obtained by central biopsy in the neoadjuvant chemotherapy group, the trend of prognostic significance in iDFS did not reach statistical significance (p = 0.702). (Figures 3A-3E).

[0122] Investigation of other potential prognostic factors To compare genetic signature with other prognostic methods, we investigated the prognostic values ​​of PAM 50 classification and TCRB diversity. In the PAM 50 call analysis of the training cohort, 76 patients with triple-negative breast cancer were classified as follows: 31 patients were basal (40.8%), 7 patients were HER-2 type (9.2%), 22 patients were luminal A type (28.9%), 12 patients were luminal B type (15.8%), and 4 patients were typical (5.3%). There were no significant differences in iDFS in KM analysis by type and ROR-S. In the TCRB diversity analysis, using the highest point of the Yoden index in the ROC analysis as the cutoff (cutoff: 5.26), 35 patients had high TCRB diversity, and the remaining patients had low TCRB diversity (n=41). However, TCRB diversity did not show any significant effect on iDFS (Figures 6A-6C).

[0123] Cox regression analysis of selected gene signatures The independence of selected gene signatures (C12orf65; growth arrest / DNA damage-inducible β; LRBA; lysosomal transport regulator; PEX1; PRKAB2; TYW5) was investigated using Cox regression analysis. In univariate Cox regression analysis, gene signatures were significantly different and positively correlated with prognosis. TNM stage and TRB diversity were not statistically significant but showed trends. In multivariate Cox regression analysis using gene signatures, TNM stage, and TRB diversity, only gene signatures were statistically significant (Table 4).

[0124] Table 4: Cox regression analysis of prognostic gene signatures and variables. [Table 4]

[0125] RC, regression coefficient; HR, hazard ratio; C12orf65, chromosome 12 open reading frame 65; GADD45B, growth arrest / DNA damage-inducible β; LRBA, LPS-responsive beige-like anchor protein; LYST, lysosome transport regulator; PEX1, peroxisome biosynthesis factor 1; PRKAB2, protein kinase AMP-activated non-catalytic subunit beta 2; TYW5, tRNA-yW synthesis protein 5; ROR-S, risk of relapse based on subtype; TRB, T cell receptor beta locus; TNM, tumor-nodule-metastasis (AJCC stage); HR, hazard ratio; CI, confidence interval.

[0126] Signaling pathway analysis and high-interaction-frequency gene analysis for prognostic gene signatures Through biological meta-analysis, gene signatures and prognostic-associated KEGG signaling pathways, as well as high-interaction-frequency genes, were identified. Signaling pathway analysis revealed that Epstein-Barr virus infection pathways, cancer pathways, cell cycle pathways, and viral oncogenesis pathways are associated with prognostic gene signatures and prognosis. In these pathways, CDK2 and TP53 are genes with high interaction frequencies associated with gene signatures and prognosis (Table 5).

[0127] Table 5: Gene signatures and pathways and interacting genes related to prognosis. [Table 5]

[0128] Example 1: Results of Analysis 2 Patient's Clinical Characteristics Of the 184 patients, 76 were in the training cohort and 108 were in the validation cohort. The clinical characteristics of patients in the training and validation cohorts are summarized in Table 6. Overall, there were no significant differences between the cohorts in patients in the very young age group. Patients in the training cohort were more likely to have earlier disease stages than those in the validation cohort. However, the stage distribution among patients who underwent primary surgery or had residual tumor after neoadjuvant chemotherapy did not differ between the two groups. The TAC (Taxotere, Adriamycin, and Cyclophosphamide) regimen was used more frequently in the validation cohort for adjuvant chemotherapy, likely due to the presence of patients with more advanced stages. Schematic diagrams of patients and samples are shown in Figures 1A and 1B. 13A–13G.

[0129] Table 6: Pathological baselines of the training cohort and validation cohort. [Table 6-1] [Table 6-2] [Table 6-3]

[0130] TNM, tumor-lymph node-metastasis (AJCC stage); pCR, complete pathological response; event, recurrence or metastasis; AC, Adriamycin, cyclophosphamide; AC-D, AC followed by docetaxel; TC, Taxotere and cyclophosphamide; FAC, 5-FU, Adriamycin, cyclophosphamide; AC-wP, AC followed by weekly paclitaxel; AC-PC, AC followed by paclitaxel and carboplatin; PCarbo, paclitaxel and carboplatin; DA, docetaxel and Adriamycin; TAC, Taxotere and AC. *The p-value was calculated using Fisher's exact test.

[0131] DEG analysis of tumor vs. non-tumor, and recurrent / metastatic vs. non-recurrent / non-metastatic cells. In tumor-nontumor-specific DNA (DEG) analysis, 9,741 out of 10,856 genes showed significantly different expression levels, exceeding a 1.5-fold change. DEG analysis of primary tumors between recurrent and non-recurrent patients revealed that 136 out of 10,856 genes showed a significant 1.5-fold difference in expression. Subsequently, 584 out of the 10,856 genes were statistically significant in single-Cox analysis. 59 DEGs overlapped across the three DEG analyses (Table 7).

[0132] Table 7: DEG correlation with prognosis of triple-negative breast cancer [Table 7-1] [Table 7-2] [Table 7-3] [Table 7-4]

[0133] Candidate gene signatures based on gene combinations and selected gene signatures through cross-validation. The top 10 candidate gene signatures were ranked using consecutive Cox p-values. All 10 candidates showed values ​​of 80 or higher in sensitivity, specificity, and precision. Prognostic gene signatures were selected by meeting the requirement of being statistically significant in subgroups of the cohort. The selected gene signatures, DGKH_GADD45B_KLF7_LYST_NR6A1_PYCARD_ROBO1_SLC22A20P_SLC24A3_SLC45A4, showed 99.00% cross-validation precision and were statistically significant in discrete Cox analysis. The risk score was calculated as follows with a cutoff value of 5.959715: (0.818636 × DGKH) + (0.018069 × growth arrest / DNA damage-inducing β) + (0.605352 × KLF7) + (0.231666 × lysosome transport regulator) + (1.305352 × NR6A1) + (-0.052086 × PYD and CARD domain-containing) + (-0.196973 × ROBO1) + (0.968759 × SLC22A20P) + (0.098331 × SLC24A3) + (0.311646 × SLC45A4) (Figures 9A-9B, Table 8).

[0134] Table 8: Candidate gene signatures as prognostic biomarkers for triple-negative breast cancer [Table 8-1] [Table 8-2]

[0135] AUC, Area under the curve; DGKH, Diacylglycerol kinase eta; GADD45B, Growth arrest / deoxyribonucleic acid damage-inducing β; KLF7, Kruppel-like factor 7; LYST, Lysosome transport regulator; NR6A1, Nuclear receptor subfamily 6 group A member 1; PYCARD, Contains PYD and CARD domains; ROBO1, Roundabout guidance receptor 1; SLC22A20P, Solute carrier family 22 member 20, Pseudogene; SLC24A3, Solute carrier family 24 member 3; SLC45A4, Solute carrier family 45 member 4; DIP2B, Disco interaction protein 2 homolog B; EMP1, Epithelial membrane protein 1; MT2A, Metallothionein 2A; NOTCH2, Notch receptor 2; RORA, RAR orphan receptor A; NOXA1, NADPH oxidase activator 1; NTAQ1, N-terminal glutamine amidase 1; SLC6A20, solute carrier family 6 member 20; CUEDC1, CUE domain-containing 1; PRICKLE1, spiny planar cell polarity protein 1; DCLK2, double cortin-like kinase 2

[0136] Prognostic significance of genetic signatures in training cohorts During a median follow-up period of 51.5 months (range: 4.6–230.8), patients with tumors possessing high-risk gene signatures (n=17) were more likely to have a lower incidence rate than patients with low-risk gene signatures (n=59, median not reached, p=1.32×10) in the overall population. -11The high-risk group showed a significantly shorter iDFS (median, 95% confidence interval [CI]; 58.5, 25.8–not reached) than the low-risk group (Figure 10A). Further analyses in another group of patients who underwent primary surgery and in patients with residual tumor after neoadjuvant chemotherapy showed similar results. Among patients who underwent primary surgery and those with residual tumor after neoadjuvant chemotherapy, the median iDFS in the high-risk group was 68.9 months (95% CI: 58.5–not reached; p=0.0000112) and 25.8 months (95% CI: 10.6–not reached, p=0.0000183), respectively, while the median iDFS in the low-risk group was not reached (Figures 10B, 10C).

[0137] Prognostic significance of genetic signatures in validation cohorts The median follow-up period for the validation cohort was 58.3 months (range; 6.6–99.8). Across the entire validation cohort, the median iDFS for patients with high-risk genetic signatures was not achieved, but the risk of recurrence or metastasis was significantly higher than for patients with low-risk genetic signatures (median iDFS not achieved, p = 0.00000584 by log-rank test). When patients were segmented according to treatment order, the prognostic significance of genetic signatures in surgical tissue from patients who underwent primary surgery was consistent with the training cohort (p = 0.0379). However, median iDFS for the high-risk and low-risk groups has not yet been achieved. High-risk genetic signatures remain effective in predicting prognosis in patients with residual tumor after neoadjuvant chemotherapy. The median iDFS was 13.6 months (95% CI, 12.2–not achieved) in the high-risk group, but not achieved in the low-risk group (p = 0.00338). Furthermore, when gene signatures were examined in tissue obtained by core biopsy in the neoadjuvant chemotherapy group, the prognostic significance in iDFS was statistically significant (p = 0.0224). (Figures 11A-11E).

[0138] Investigation of other potential prognostic factors To compare genetic signatures and other prognostic methods, we investigated the prognostic values ​​of PAM 50 call and TCRB diversity. In the PAM 50 call analysis of the training cohort, 76 patients with triple-negative breast cancer were classified as follows: 31 patients were basal (40.8%), 7 patients were HER-2 type (9.2%), 22 patients were luminal A type (28.9%), 12 patients were luminal B type (15.8%), and 4 patients were typical (5.3%). There were no significant differences in iDFS in KM analysis by type and ROR-S. In the TCRB diversity analysis, using the highest point of the Yoden index in the ROC analysis as the cutoff (cutoff: 5.26), 35 patients had high TCRB diversity, and the remaining patients had low TCRB diversity (n=41). However, TCRB diversity did not show any significant effect on iDFS (Figures 13A-13C).

[0139] Cox regression analysis of selected gene signatures The independence of selected gene signatures (DGKH_growth arrest / DNA damage-inducible β_KLF7_lysosome transport regulator_NR6A1_PYD and CARD domain-containing_ROBO1_SLC22A20P_SLC24A3_SLC45A4) was investigated using Cox regression analysis. In univariate Cox regression analysis, gene signatures were significantly different and positively correlated with prognosis. TNM stage was not statistically significant but showed a trend. In multivariate Cox regression analysis using gene signatures, TNM stage, and TRB diversity, only gene signatures were statistically significant (Table 9).

[0140] Table 9: Cox regression analysis of prognostic gene signatures and variables. [Table 9]

[0141] RC, regression coefficient; HR, hazard ratio; CI, confidence interval; DGKH, diacylglycerol kinase eta; GADD45B, growth arrest / DNA damage-inducible β; KLF7, Kruppel-like factor 7; LYST, lysosome transport regulator; NR6A1, nuclear receptor subfamily 6 group A member 1; PYCARD, containing PYD and CARD domains; ROBO1, roundabout guidance receptor 1; SLC22A20P, solute carrier family 22 member 20, pseudogene; SLC24A3, solute carrier family 24 member 3; SLC45A4, solute carrier family 45 member 4; TNM, tumor-nodule-metastasis (AJCC stage); ROR-S, risk of recurrence based on subtype; TCRB, T cell receptor β-roxen

[0142] Signaling pathway analysis and high-interaction-frequency gene analysis for prognostic gene signatures Through biological meta-analysis, gene signatures and prognostic-associated KEGG signaling pathways, as well as high-interaction-frequency genes, were identified. Signaling pathway analysis revealed that the following pathways—cancer, PI3K-Akt signaling, Alzheimer's disease, human cytomegalovirus infection, hepatitis C, breast cancer, and MAPK signaling—are associated with prognostic gene signatures and prognosis. In these pathways, KRAS, HRAS, and APP were high-interaction-frequency genes associated with gene signatures and prognosis (Table 10).

[0143] Table 10: Pathways and interacting genes associated with gene signatures and prognostic features. [Table 10]

[0144] Example 2: Prognostic gene signature reflecting CD8-positive T cell enrichment in early-stage triple-negative breast cancer. Methods: Seventy-six patients with triple-negative breast cancer were enrolled for gene expression profiling, and GSE 169246 data were collected for single-cell ribonucleic acid (scRNA) profiling. The median follow-up period for enrolled patients was 51.5 months (range: 4.6–230.8). Of the enrolled patients, 13 had relapse or metastasis. RNA sequencing was performed using a HiSeq 4000 sequencer to analyze gene expression profiles of tumor samples from triple-negative breast cancer patients. Single-cell ribonucleic acid analysis was performed using the Seurat package (v.4.0.5). Differential expression genes (DEGs) were defined as those that satisfy both the conditions for Cox regression and Wilcoxon analysis in gene expression profiling, and the conditions for logistic regression analysis in CD8-positive T cell profiling in scRNA analysis. Gene signatures were analyzed by combinations of the above DEGs. Gene signatures were marked on t-SNEs in the scRNA profile. Statistical analysis will be performed using the R language (v.3.4.3). Results: Gene signatures reflecting CD8-positive T cell enrichment characteristics were identified to stratify triple-negative breast cancer patients by risk score. These signatures were GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2 (sensitivity = 92.31%; specificity = 93.65%; precision = 93.42%). Kaplan-Meier (KM) analysis showed that patients with tumors possessing high-risk gene signatures (n=16, median iDFS = 42.7) had significantly shorter iDFS than patients with low-risk gene signatures (n=60, median iDFS not reached) (Figure 15A-15B). CD8-positive T cell-related gene signatures were marked on CD8-positive T cells near CD4-positive T cells on t-SNEs (Figure 15C-15E).

[0145] Furthermore, gene signatures reflecting macrophage enrichment features stratified from patients with triple-negative breast cancer based on risk scores were identified. The macrophage-related gene set signatures identified were as follows: CLDN4, ERH, growth arrest / DNA damage-inducible β, glycophorin C, H2AFX, MT2A, NDUFV2, SDF4, UBE2W (AUC = 0.963, sensitivity = 92.31%, specificity = 93.65%, precision = 93.42%). In the Kaplan-Meier (KM) analysis, patients with tumors bearing high-risk gene signatures (n=16, median iDFS = 42.7 months) showed significantly shorter iDFS than patients with low-risk gene signatures (n=60, median iDFS not reached) (Figure 16A-16B). Macrophage-related gene signatures were marked on macrophages, monocytes, and dendritic cells in t-SNE (Figure 16C-16E).

[0146] Non-limiting embodiments of a system and computer implementation method for predicting the prognosis and / or treatment outcomes of subjects diagnosed with TNBC or candidates at risk of developing TNBC. Figure 7 shows a non-limiting embodiment of System 100 configured according to the principles of the present invention. System 100 is configured to receive computer-readable data of multiple biomarkers (e.g., including a panel of biomarkers) about a sample from a subject, which can be obtained with respect to certain embodiments provided in this disclosure, and to predict the prognosis and / or treatment outcomes of subjects diagnosed with cancer (e.g., breast cancer such as TNBC) or candidates at risk of developing cancer (e.g., breast cancer such as TNBC). In certain embodiments, the multiple biomarkers include ankyrin repeat domain 36 (ANKRD36), ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), B-box and SPRY domain-containing (BSPRY), chromosome 12 open reading frame 65 (C12orf65), chromosome 2 open reading frame 49 (C2orf49), chromosome 1 open reading frame 198 (C1orf198), coiled-coil domain-containing 114 (CCDC114), claudin 4 (CLDN4), CUE domain-containing 1 (CUEDC1), outer dynein arm docking complex subunit 1 (ODAD1), cell growth regulator with EF-hand domain 1 (CGREF1), DEP domain-containing 7 (DEPDC7), and double cortin-like kinase 2. (DCLK2), diacylglycerol kinase (DGKH), Disco interaction protein 2 homolog B (DIP2B), schizophrenia-related protein 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH ribonucleic acid splicing and mitotic factor (ERH), growth arrest / DNA damage-inducible β (GADD45B), glutaminase (GLS), Grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member X (H2AFX), HRas proto-oncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium voltage-gated channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor 7 (KLF7), keratin 17 (KRT17),LON peptidase N-terminal domain and RING finger protein 2 (LONRF2), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, U7 micronucleus ribonucleic acid-related (LSM11), lysosome transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisene Slibonucleic acid 1 (MCM3AP_AS1), MICAL-like 2 (MICALL2), MHC class I polypeptide-related sequence B (MICB), metallothionein 2A (MT2A), myelin expression factor 2 (MYEF2), NEDD4-binding protein 3 (N4BP3), neuroblastoma breakpoint family member 20 (NBPF20), NADH: ubiquinone oxidoreductate zecore subunit V2 (NDUFV2), neurogenic gene locus Notch homolog protein 2 (NOTCH2), NADPH oxidase activator 1 (NOXA1), natriuretic peptide receptor 3 (NPR3), nuclear receptor subfamily 6 group A member 1 (NR6A1), P21 (RAC1) activating kinase 3 (PAK3), pantothenate kinase 3 (PANK3), par-6 family cell polarity regulator beta (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyBac transposable element 4 (PGBD4), praja ring finger ubiquitin rigger Plexin homology and FYVE domain-containing protein 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K), spiny planar cell polarity protein 1 (PRICKLE1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), PYD and CARD domain-containing protein (PYCARD), RAS p21 protein activator 1 (RASA1), RASD family member 2 (RASD2), RAS guanylate-releasing protein 1 (RASGRP1), rhophylline RHO GTPase-binding protein 2 (RHPN2),Rab-interacting lysosomal protein-like 2 (RILPL2), roundabout guidance receptor 1 (ROBO1), RAR orphan receptor A (RORA), stromal cell-derived factor 4 (SDF4), SERTA domain-containing 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-induced proliferation-related 1-like 2 (SIPA1L2), solute carrier family 22 member 20 (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), solute carrier family 2 member 12 (SLC2A12), solute carrier family 39 member 10 (SLC39A10), solute carrier family 25 member 40 (SLC25A40), solute carrier Family 43 member 1 (SLC43A1), Solute carrier family 45 member 4 (SLC45A4), Solute carrier family 6 member 20 (SLC6A20), Spectrin beta, Red blood cell (SPTB), Interstitial antigen 3-like 3 (STAG3L3), Suci domain-containing 3 (SUSD3), TATA box-binding protein-related factor 10 (TAF10), t-complex-associated testicular expression 3 (TCTE3), Transmembrane channel-like 7 (TMC7), Transmembrane and coiled-coil domain family 2 (TMCC2), Transcriptional repressor GATA binding 1 (TRPS1), Tubulin tyrosine ligase-like 4 (TTLL4), tRNA-YW synthase 5 (TYW5), Ubiquitin-conjugating enzyme E2 At least two biomarkers selected from the group consisting of W(UBE2W), WD repeat, steryl alpha motif and U-box domain-containing 1 (WDSUB1), N-protein N-terminal glutamine amide hydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED type-containing 6 (ZBED6), zinc finger and BTB domain-containing 46 (ZBTB46), zinc finger CCCH type-containing 13 (ZC3H13), zinc finger and homeobox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785), at least three biomarkers,The system includes at least four biomarkers, at least five biomarkers, at least six biomarkers, or seven biomarkers. In a particular embodiment, system 100 is configured to calculate multiple biomarkers using ribonucleic acid sequencing data from tumor tissue of triple-negative breast cancer patients. In a particular embodiment, system 100 is configured to calculate multiple biomarkers (e.g., 6, 7, 8, 9, or 10 gene signatures - (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (c) GADD45B (d) DGKH, KLF7, LYST, NR6A1, PYCARD, SLC22A20P, SLC24A3, and NTAQ1; (e) DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g) D GKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf6 5, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l) LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m) C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n) TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o) LRBA, LYST, PEX1,DEPDC7, SLC43A1, MICB, and C2orf49; (p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q)DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICALL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t)GADD45B, SLC4 The system is configured to determine a score based on (u)GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W) and predict cancer (e.g., TNBC) patients to classify them into high-risk or low-risk groups for disease progression, relapse, recurrence, and / or death. In one embodiment, the system 100 is configured to analyze multiple biomarkers to predict prognosis and provide guidance for accurate treatment strategies.

[0147] System 100 comprises a processor 110, memory 120, network interface 130, input-output (IO) interface 140, driver suite 150, biomarker analyzer and cancer (e.g., TNBC) predictor 160, and communication unit 170, all of which are arranged to be connected to bus 105. In a non-limiting embodiment, system 100 is configured to perform the process shown in Figure 4. System 100 may include a machine learning platform that includes supervised machine learning, unsupervised machine learning, or a combination of supervised and unsupervised machine learning, capable of performing one or more machine learning processes.

[0148] In one embodiment, System 100 is configured to perform combination gene analysis and predict the n-gene signatures of optimal genes that can be used as biomarkers in large-scale analysis, where n is a positive non-zero integer. System 100 can be configured to perform pre-validation by a machine learning process (e.g., 6, 7, 8, 9, or 10 signature cross-validation, where n = 6, 7, 8, 9, or 10) and then perform validation, such as 10-gene signature validation in separate validation cohorts. System 100 can be configured to perform a meta-analysis of n-genes (e.g., 10 genes) by a machine learning process to determine or confirm the biological association between n-genes (e.g., 10 genes) and triple-negative breast cancer (e.g., TNBC). System 100 can be configured to update parametric model values ​​of the machine learning platform during operation.

[0149] As a non-limiting application of System 100, patients with early-stage triple-negative breast cancer classified as high-risk for prognosis and / or treatment outcomes by System 100 may be candidates for further systemic therapy in addition to standard care.

[0150] In one embodiment, system 100 can be used as a tool for selecting patients for escalation or deescalation trials. In this regard, patients can benefit from risk-based care and are not required to receive standardized treatment.

[0151] The biomarker analyzer and cancer (e.g., triple-negative breast cancer) predictor 160 may include or be contained within a computing device. The cancer (e.g., triple-negative breast cancer) biomarker analyzer and cancer (e.g., triple-negative breast cancer) predictor 160 may include a machine learning platform that includes supervised machine learning, unsupervised machine learning, or a combination of supervised and unsupervised machine learning. A machine learning platform can include, for example, artificial neural networks (ANNs), convolutional neural networks (CNNs), time-series convolutional networks (TCNs), deep CNNs (DCNNs), RCNNs, Mask-RCNNs, deep convolutional encoder-decoders (DCEDs), recurrent neural networks (RNNs), neural Turing machines (NTMs), differential neural computers (DNCs), support vector machines (SVMs), deep learning neural networks (DLNNs), long short-term memory (LSTMs), naive Bayes, decision trees, linear regression, Q-learning, time-lag (TD), deep adversarial networks, fuzzy logic, or any other machine intelligence platform capable of supervised or unsupervised machine learning. A machine learning platform can include machine learning models. The biomarker analyzer and cancer (e.g., triple-negative breast cancer) predictor 160 may include statistical prediction techniques such as standard regression (SR), support vector regression (SVR), ridge regression (Ridge), random forest (RF), autoregressive integrated moving average (ARIMA), vector autoregression (VAR), predictive expert arbitrage (AFE), extra-tree regression (ETR), multilayer perceptron (MLPR), or vector error correction model (VECM).

[0152] In certain embodiments of the biomarker analyzer and cancer (e.g., triple-negative breast cancer) predictor 160, a machine learning platform including a machine learning model is trained using training datasets created based on the various embodiments / examples provided herein. For example, a portion of the datasets created based on the various embodiments / examples provided herein can be prepared for training a machine learning model by processes such as deduplication, error correction, providing missing values, normalization, data type conversion, data randomization, and annotation, as will be understood by those skilled in the art. In addition, the remaining portion of the datasets created based on the various embodiments / examples provided herein can be used to create validation datasets for validating the machine learning model. In certain embodiments, the original dataset can be split so that 50% of the dataset is used to construct a training dataset and the remaining 50% is used to construct a validation dataset. Other ratios are also possible for training / validation, such as 90 / 10, 80 / 20, 70 / 30, or 60 / 40, for example, but not limited to these. The training dataset can be used to train a machine learning model to make consistently correct predictions. The model can then be validated using the validation dataset. Once trained, the model can be tuned for performance improvement, for example, by adjusting hyperparameters.

[0153] The biomarker analyzer and cancer (e.g., triple-negative breast cancer) predictor 160 may include a biomarker analysis unit 160A, a triple-negative breast cancer prediction unit 160B, and a subject classification unit 160C. Each of units 160A, 160B, and 160C may include (or be contained within) a machine learning platform. In one embodiment, the biomarker analysis unit 160A is configured to analyze computer-readable data corresponding to a sample from a subject, which may include one or more biomarkers. In one particular embodiment, the cancer (e.g., triple-negative breast cancer) prediction unit 160B is configured to predict the prognosis and / or treatment outcome of a subject and generate a risk score based on the results of the analysis of the computer-readable data. In one particular embodiment, the subject classification unit 160C is configured to determine the classification of a subject, including whether the subject is at high or low risk for disease progression, recurrence, and / or death, based on the prediction and risk score.

[0154] In a particular embodiment, the biomarker analysis unit 160A may include, for example, ANKRD36, ANKRD36BP2, BSPRY, C12orf65, C2orf49, C1orf198, CCDC114, CLDN4, CUEDC1, ODAD1, CGREF1, DEPDC7, DCLK2, DGKH, DIP2B, DISC1, EMP1, ERH, KCNC3, KLF6, KLF7, KRT17, LONRF2, LRBA, LRIT3, LRRC37B, LSM11, LYS These are T, MALAT1, MCM3AP_AS1, MICALL2, MICB, MT2A, MYEF2, N4BP3, NBPF20, NDUFV2, NOTCH2, NOXA1, NPR3, NR6A1, PAK3, PANK3, PARD6B, PEX1, PGBD4, PJA1, PLEKHF1, PNP, PPM1K, PRICKLE1, PRKAB2, PTPRA, PYCARD, RASA1, ZNF217, ZNF233, ZNF248, ZNF469, and ZNF785. The RILPL WDYHV biomarker analysis unit 160A can be configured to analyze ribonucleic acid and / or protein sequencing data from tumor tissue of cancer (e.g., TNBC) patients.

[0155] In a particular embodiment, the cancer (e.g., TNBC) prediction unit 160B includes multiple biomarkers (e.g., 6, 7, 8, 9, or 10 gene signatures - (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (c) GADD45B, LYST, NOXA1, NR6A1, PY (d) CARD, SLC22A20P, SLC24A3, and NTAQ1; (e) DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l) LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A 20;(m)C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1;(n)TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A;(o)LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49;(p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1;(q)DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A;(r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICALL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t)GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W) are configured to determine the score. This is based, for example, on scoring based on ribonucleic acid and / or protein sequence data, mRNA expression values ​​of 6, 7, 8, 9, or 10 gene signatures from RT-PCR, mRNA expression values ​​of 6, 7, 8, 9, or 10 gene signatures in fresh-frozen tumor tissue, and / or mRNA expression values ​​of 6, 7, 8, 9, or 10 gene signatures in FFPE tumor tissue.

[0156] In a non-limiting embodiment, the machine learning platform in the biomarker analyzer and cancer (e.g., TNBC) predictor 160 was trained based on the analysis of gene expression profiles by ribonucleic acid sequencing using tumor samples from 184 cancer (e.g., TNBC) patients, including a training cohort (n = 76) and a validation cohort (n = 108). By combining weighted gene expressions, the biomarker analyzer and cancer (e.g., TNBC) predictor 160 uses 6, 7, 8, 9, or 10 gene signatures ((a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, S (c) LC22A20P, and SLC24A3; (d) GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SLC24A3, and NTAQ1; (e) DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3, and SLC6A20; (f) DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3, and SLC6A20; (f) CUEDC1, DGKH (g) DGKH, KLF7, LYST, NOXA1, SLC22A20P, and SLC6A20; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k)C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l)LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m)C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1;(n)TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49; (p)GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q)DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r)LYST, DEPDC7, SLC43A1, SLC6A20, MICB The models were trained based on HRAS;(s)PEX1, MICALL2, SLC43A1, RASA1, KCNC3, and LRIT3;(t)GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40;(u)GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2;or(v)CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W). This included stratification of cancer (e.g., TNBC) patients by risk score (e.g., sensitivity = 90.91%, specificity = 100.00%, precision = 98.68%), which was validated in a validation cohort. The biomarker analyzer and cancer (e.g., triple-negative breast cancer) predictor 160 is configured to identify and / or validate a set of 6, 7, 8, 9, or 10 gene signatures and predict the prognosis of a patient with early-stage cancer (e.g., triple-negative breast cancer) based, for example, the transcriptome of the primary tumor.

[0157] The communication unit 170 may include one or more devices, such as a transmitter 170A, a receiver 170B, a transceiver (not shown), a modulator (not shown), a demodulator (not shown), a modem (not shown), an encoder (not shown), or a decoder (not shown). The communication unit 170 may be configured to communicate with one or more communication devices (not shown), such as a smartphone, a tablet, or a computer. The communication unit 170 may be configured to transmit cancer (e.g., triple-negative breast cancer) results, including a risk score, a predicted prognosis for cancer (e.g., triple-negative breast cancer), and a target classification, to one or more communication devices (not shown) for each patient.

[0158] The processor 110 may include computing devices such as, for example, one of various commercially available graphics processing unit devices. Dual microprocessors and other multiprocessor architectures can be included in the processor 110. The processor 110 may include a CPU (Central Processing Unit), a GPU (Graphic Processing Unit), a GPGPU (General-Purpose GPU), an FPGA (Field Programmable Gate Array), an ASIC (Application-Specific Integrated Circuit), or a many-core processor.

[0159] The processor 110 can be configured to process instructions for execution within the system 100, including instructions stored in memory 120. The processor 110 can also process instructions for displaying graphical information for a GUI on an external input / output device, such as a display device coupled to an I / O interface 140 or a high-speed interface (not shown). In other implementations, multiple processors and / or multiple buses may be used, along with multiple memories and multiple types of memory, as needed.

[0160] System 100 may include a non-transient computer-readable medium that, when executed by the processor 110, can hold executable or interpretable computer program code or instructions that can be used to perform the steps, processes, and methods of this disclosure. The computer-readable medium may be contained in memory 120.

[0161] Memory 120 may include read-only memory (ROM) 120A, random access memory (RAM) 120B, and a hard disk drive (HDD) 120C. The basic input / output system (BIOS) may be stored in non-volatile memory, for example, which may include ROM 120A. ROM 120A may include erasable programmable read-only memory (EPROM) or electrically erasable programmable read-only memory (EEPROM).

[0162] The hard disk drive interface (not shown) may include a Universal Serial Bus (USB) (not shown), an IEEE 1394 interface (not shown), or other suitable interface for external applications.

[0163] Memory 120 can provide non-volatile storage for data, data structures, and computer executable code or instructions. Memory 120 can accommodate the storage of any data in a suitable digital format. Memory 120 may include one or more computer applications that can be used to perform aspects of the architecture described herein. Memory 120 may include, for example, flash memory or NVRAM memory.

[0164] Memory 120 may contain one or more computer resources, including, for example, an operating system (not shown), one or more application programs (not shown), one or more APIs, and program data (not shown). In certain embodiments, a machine learning platform and / or machine learning model may be contained in memory 120. APIs may include, for example, a JSON API, XML API, Web API, SOAP API, RPC API, REST API, or other utility or service APIs. Any (or all) of the computer programs may be cached in RAM 120B as an executable section of computer program code.

[0165] The network interface 130 can be connected to a network (not shown). System 100 can be connected to a communication device (not shown) via the network interface 130, for example, which communicates with the communication device over a communication link. The network interface 130 can be connected to a network via one or more communication links (not shown). The network interface 130 may include a wired or wireless network interface (not shown) or a modem (not shown). When used in a private network, system 100 can be connected to the private network via a wired or wireless network interface, and when used in a wide-area network, system 100 can be connected to the wide-area network via a modem. The network may include a private network, a wide-area network, the Internet, or any other network. The modem (not shown) may be internal or external, and may be wired or wireless. The modem may be connected to bus 105 via a serial port interface (not shown), for example.

[0166] The IO interface 140 can be configured to receive commands and data from a user. The IO interface 140 can be configured to connect to or communicate with one or more input / output devices (not shown), including, for example, a keyboard (not shown), a mouse (not shown), a pointer (not shown), a microphone (not shown), a speaker (not shown), or a display (not shown). The received commands and data can be transferred from the IO interface 140 as instruction and data signals via the bus 105 to any computer asset in the system 100.

[0167] The driver suite 150 may include an audio driver 150A and a video driver 150B. The audio driver 150A may include a sound card, a sound driver (not shown), an IVR unit, or any other device necessary to render sound signals onto a sound generating device (not shown), such as a speaker (not shown). The video driver 150B may include a video card (not shown), a graphics driver (not shown), a video adapter (not shown), or any other device necessary to render image signals onto a display device (not shown).

[0168] Figure 8 shows a non-limiting embodiment of a computer implementation process that may be performed by system 100. Refer to the figure. Referring together to Figures 7 and 8, system 100 can receive a panel of biomarker data for a subject / patient (step 210). System 100 can receive the biomarker panel data via network interface 130, I / O interface 140, or receiver (RX) 170B. The received data can be input to a biomarker analyzer and cancer (e.g., TNBC) predictor 160, which can analyze the data by a biomarker analysis unit 160A (step 220). Biomarker panel data can be analyzed for patterns including the presence of at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or ten of the following 6, 7, 8, 9, or 10 gene signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SL24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, GADD45B, MT2A, NOTCH2, NR6A1, RORA, SLC22A20P, and SLC24A3; (c) GADD45B, LYST, NOXA1, NR6A1, PYCARD, SLC22A20P, SL C24A3 and NTAQ1; (d) DGKH, KLF7, LYST, NR6A1, ROBO1, SLC24A3 and SLC6A20; (e) DGKH, EMP1, GADD45B, LYST, SLC22A20P, SLC24A3 and SLC6A20; (f) CUEDC1, DGKH, EMP1, LYST, NOXA1, SLC22A20P and SL C6A20; (g) DGKH, KLF7, LYST, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) GADD45B, KLF7, LYST, NR6A1, ROBO1, SLC22A20P, and SLC6A20;(j) DCLK2, GADD45B, LYST, NR6A1, SLC22A20P, and NTAQ1; (k) C12orf65, GADD45B, LRBA, LYST, PEX1, PRKAB2, and TYW5; (l) LYST, PEX1, ODAD1, DEPDC7, MICALL2, SLC43A1, and SLC6A20; (m) C12orf65, GADD45B, LYST, PEX1, RASA1, SLC45A4, and NTAQ1; (n) TYW5, DEPDC7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o) LRBA, LYST, PEX1, DEPDC7, SLC43A1, MICB, and C2orf49; (p) GADD45B, PEX1, DEPDC7, SLC43A1, LSM11, and PJA1; (q) DEPDC7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r) LYST, DEPDC7, SLC43A1, SLC6A20, MICB, and HRAS; (s) PEX1, MICALL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t) GADD45B, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u) GADD45B, H2AFX, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v) CLDN4, ERH, GADD45B, GYPC, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W. ;

[0169] Based on the analysis, the cancer (e.g., TNBC) prediction unit 160B can determine a risk score (step 230) and predict the prognosis for cancer (e.g., TNBC) in a particular subject / patient (step 240). Based on the risk score and prediction, the subject classification unit 160C can classify the subject / patient as high-risk or low-risk (step 250). In certain embodiments, the classification may include other values ​​besides high-risk and low-risk, such as a numerical value representing the predicted probability of cancer (e.g., triple-negative breast cancer) prognosis for the subject / patient. The cancer (e.g., triple-negative breast cancer) prediction results can be packaged and transmitted to one or more communication devices (not shown), such as a smartphone, tablet, or computer (step 270). The cancer (e.g., triple-negative breast cancer) prediction results can then be rendered on a display and / or used to create a treatment plan for a particular subject / patient.

[0170] In certain embodiments, the biomarker analyzer and cancer (e.g., triple-negative breast cancer) predictor 160 may be configured to provide a treatment based on the cancer (e.g., triple-negative breast cancer) results.

[0171] In a particular embodiment, the parametric values ​​of the model can be adjusted using predictions, risk scores, and / or classifications determined by a machine learning model in the biomarker analyzer and cancer (e.g., triple-negative breast cancer) predictor 160 (step 260).

[0172] As used in this disclosure, the term “backbone” means a transmission medium or infrastructure that interconnects one or more computing or communication devices to provide a path for carrying data packets or instructions between computing or communication devices. A backbone may include a network. A backbone may include Ethernet TCP / IP. A backbone may include a distributed backbone, an aggregated backbone, a parallel backbone, or a serial backbone.

[0173] As used in this disclosure, the term “bus” means any of several types of bus structures that can be further interconnected to a memory bus, peripheral bus, or local bus (with or without a memory controller) using any of the various commercially available bus architectures. The term “bus” may include a backbone.

[0174] As used in this disclosure, the terms “communicating device” or “communication device” mean any computing device, hardware, or computing resource capable of transmitting or receiving digital or analog signals or data packets, or command signals or data signals, over a communication link. The device may be portable or stationary.

[0175] As used in this disclosure, the term “communication link” means a wired and / or wireless medium that carries data or information between at least two points. The wired or wireless medium may include, for example, a metal conductor link, a radio frequency (RF) communication link, an infrared (IR) communication link, or an optical communication link. An RF communication link may include, for example, GSM voice calls, SMS, EMS, MMS messaging, CDMA, TDMA, PDC, WCDMA®, CDMA2000, GPRS, WiFi, WiMAX, IEEE 802.11, DECT, 0G, 1G, 2G, 3G, 4G, or 5G cellular standards, or Bluetooth. A communication link may include, for example, RS-232, RS-422, RS-485, or any other suitable interface.

[0176] As used in this disclosure, the terms “computer” or “computing device” mean any machine, device, circuit, component, or module, or any system of machines, devices, circuits, components, or modules that may be capable of manipulating data according to one or more instructions, such as, for example, a processor, application-specific integrated circuit (ASIC), programmable gate array (FPGA), microprocessor (μP), central processing unit (CPU), graphics processing unit (GPU), general-purpose computer, supercomputer, personal computer, laptop computer, palmtop computer, notebook computer, smartphone, mobile phone, tablet, desktop computer, workstation computer, server, server farm, computer cloud, or array of processors, ASICs, FPGAs, μPs, CPUs, GPUs, general-purpose computers, supercomputers, personal computers, laptop computers, palmtop computers, notebook computers, desktop computers, workstation computers, or servers. A computer or computing device may include hardware, firmware, or software that can send or receive data packets or instructions over a communication link. A computer or computing device may be portable or stationary.

[0177] As used in this disclosure, the term “computer asset” means a computer resource, computing device, communication device, or computer-readable medium.

[0178] As used in this disclosure, the term “Computer Resources” means software, software applications, web applications, web pages, documents, files, records, application programming interfaces (APIs), web content, computer applications, computer programs, computer code, machine-executable instructions, or firmware. Computer resources may include information resources. Computer resources may include machine instructions for programmable computing devices and may be implemented in high-level procedural or object-oriented programming languages ​​or in assembly / machine languages.

[0179] As used in this disclosure, the term “computer-readable medium” means any storage medium involved in providing data (e.g., instructions) that can be read by a computer. Such mediums can take many forms, including non-volatile and volatile media. Non-volatile media may include, for example, optical or magnetic disks and other persistent memory. Volatile media may include dynamic random-access memory (DRAM). Common forms of computer-readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tapes, any other magnetic media, CD-ROMs, DVDs, any other optical media, punch cards, paper tapes, any other physical media having a pattern of holes, RAM, PROMs, EPROMs, FLASH-EEPROMs, any other memory chips or cartridges, carriers as described below, or any other media that can be read by a computer. Computer-readable media may include “clouds” that include the distribution of files across multiple (e.g., thousands) memory caches on multiple (e.g., thousands) computers. Computer-readable media may include magnetic disks, optical disks, memory, or programmable logic devices (PLDs).

[0180] Various forms of computer-readable media can be involved in transporting sequences of instructions to a computer. For example, a sequence of instructions may (i) be delivered from RAM to the processor, (ii) be transported via a wireless transmission medium, and / or (iii) be formatted according to a number of formats, standards, or protocols, including, for example, WiFi, WiMAX, IEEE 802.11, DECT, 0G, 1G, 2G, 3G, 4G, or 5G cellular standards, or Bluetooth.

[0181] As used in this disclosure, the term “database” means any combination of software and / or hardware, including at least one application and / or at least one computer. A database may include a structured collection of records or data organized according to a database model, such as, but not limited to, at least one of the following: a relational model, a hierarchical model, or a network model. A database may include a database management system application (DBMS). At least one application may include, but is not limited to, an application program that can accept connections to service requests from clients by sending responses back to the client. A database may often be configured to run at least one application unattended and with minimal human intervention for extended periods, often under heavy workloads.

[0182] As used in this disclosure, the term “Network” means, but is not limited to, a personal area network (PAN), a local area network (LAN), a wireless local area network (WLAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), a global area network (GAN), a broadband area network (BAN), a mobile network, a storage area network (SAN), a systems area network, a passive optical local area network (POLAN), an enterprise private network (EPN), a virtual private network (VPN), the Internet, or at least one of any combination thereof, any of which may be configured to communicate data over wireless and / or wired communication media. These networks may implement a variety of protocols, including, but not limited to, Ethernet®, IP, IPX, TCP, UDP, SPX, IP, IRC, HTTP, FTP, Telnet, SMTP, DNS, ARP, and ICMP.

[0183] As used in this disclosure, the term “server” means any combination of software and / or hardware, including at least one application and / or at least one computer for performing services for connected clients as part of a client-server architecture. The at least one server application may, but is not limited to, include an application program capable of accepting connections to service requests from clients by, for example, sending a response back to the client. A server can often be configured to run at least one application unattended and with minimal human intervention for extended periods, under heavy workloads. A server may include multiple computers configured such that at least one application is distributed among the computers depending on the workload. For example, under light loads, at least one application may run on a single computer. However, under heavy loads, multiple computers may be required to run at least one application. A server, or any of its computers, can be used as a workstation.

[0184] Devices communicating with each other do not need to communicate continuously unless explicitly specified otherwise. In addition, devices communicating with each other may communicate directly or indirectly through one or more intermediate devices.

[0185] Process steps, method steps, algorithms, etc., may be described in a sequential or parallel order, but such processes, methods, and algorithms may be configured to function in an alternative order. In other words, any sequence or order of steps that can be described in a sequential order does not necessarily indicate a requirement that the steps be performed in that order, and some steps may be performed simultaneously. Similarly, if a sequence or order of steps is described in a parallel (or simultaneous) order, such steps may be performed in a sequential order. The steps of the processes, methods, or algorithms described herein may be performed in any practical order.

[0186] Where a single device or article is described herein, it is readily apparent that two or more devices or articles may be used instead of that single device or article. Similarly, where two or more devices or articles are described herein, it is readily apparent that a single device or article may be used instead of two or more devices or articles. The function or features of a device may be embodied by one or more other devices not expressly described as having such a function or features.

[0187] The subject matter described herein is provided for illustrative purposes only and should not be construed as limiting. Various modifications and changes can be made to the subject matter described herein without adhering to the exemplary embodiments and uses, and without departing from the true spirit and scope of the invention as encompassed in this disclosure, as defined by the series of descriptions and equivalent structures, functions or steps in the following claims.

[0188] (Reference 1) Collaboration, G. B. o. D. C. Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-Years for 29 Cancer Groups, 1990 to 2016: A Systematic Analysis for the Global Burden of Disease Study. JAMA Oncology 4, 1553-1568, doi:10.1001 / jamaoncol.2018.2706 (2018). (Reference 2) Sung, H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: a cancer journal for clinicians 71, 209-249, doi:https: / / doi.org / 10.3322 / caac.21660 (2021). (Reference 3) Hong, S. et al. Cancer Statistics in Korea: Incidence, Mortality, Survival, and Prevalence in 2017. Cancer Res Treat 52, 335-350, doi:10.4143 / crt.2020.206 (2020). (Reference 4) Allemani, C. et al. Global surveillance of trends in cancer survival 2000-14 (CONCORD-3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population-based registries in 71 countries. Lancet (London, England) 391, 1023-1075, doi:10.1016 / s0140-6736(17)33326-3 (2018). (Reference 5) Sparano, J. A. et al. Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. The New England journal of medicine 379, 111-121, doi:10.1056 / NEJMoa1804710 (2018). (Reference 6) Henry, N. L. et al. Role of Patient and Disease Factors in Adjuvant Systemic Therapy Decision Making for Early-Stage, Operable Breast Cancer: Update of the ASCO Endorsement of the Cancer Care Ontario Guideline. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 37, 1965-1977, doi:10.1200 / jco.19.00948 (2019). (Reference 7) Carey, L. A. et al. The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clinical cancer research : an official journal of the American Association for Cancer Research 13, 2329-2334, doi:10.1158 / 1078-0432.Ccr-06-1109 (2007). (Reference 8) Perou, C. M. et al. Molecular portraits of human breast tumours. Nature 406, 747-752, doi:10.1038 / 35021093 (2000). (Reference 9) Park, J. et al. Nonphosphorylatable PEA15 mutant inhibits epithelial-mesenchymal transition in triple-negative breast cancer partly through the regulation of IL-8 expression. Breast cancer research and treatment, doi:10.1007 / s10549-021-06316-2 (2021). (Reference 10) Prat, A. et al. Phenotypic and molecular characterization of the claudin-low intrinsic subtype of breast cancer. Breast cancer research : BCR 12, R68, doi:10.1186 / bcr2635 (2010). (Reference 11) Tutt, A. et al. Carboplatin in BRCA1 / 2-mutated and triple-negative breast cancer BRCAness subgroups: the TNT Trial. Nature Medicine 24, 628-637, doi:10.1038 / s41591-018-0009-7 (2018). (Reference 12) Wolff, A. C. et al. Recommendations for Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: American Society of Clinical Oncology / College of American Pathologists Clinical Practice Guideline Update. Journal of Clinical Oncology 31, 3997-4013, doi:10.1200 / jco.2013.50.9984 (2013). (Reference 13) Allison, K. H. et al. Estrogen and Progesterone Receptor Testing in Breast Cancer: American Society of Clinical Oncology / College of American Pathologists Guideline Update. Archives of Pathology & Laboratory Medicine 144, 545-563, doi:10.5858 / arpa.2019-0904-SA (2020). (Reference 14) Park, I. J., Yu, Y. S. & Mustafa, B. A Nine-Gene Signature for Predicting the Response to Preoperative Chemoradiotherapy in Patients with Locally Advanced Rectal Cancer. 12, doi:10.3390 / cancers12040800 (2020). (Reference 15) Picornell, A. C. et al. Breast cancer PAM50 signature: correlation and concordance between RNA-Seq and digital multiplexed gene expression technologies in a triple negative breast cancer series. BMC genomics 20, 452, doi:10.1186 / s12864-019-5849-0 (2019). (Reference 16) Raj-Kumar, P. K. et al. PCA-PAM50 improves consistency between breast cancer intrinsic and clinical subtyping reclassifying a subset of luminal A tumors as luminal B. Scientific reports 9, 7956, doi:10.1038 / s4l598-019-44339-4 (2019). (Reference 17) Bolotin, D. A. et al. Antigen receptor repertoire profiling from RNA-seq data. 35, 908-911, doi:10.1038 / nbt.3979 (2017). (Reference 18) Han, J. et al. TCR Repertoire Diversity of Peripheral PD-1(+)CD8(+) T Cells Predicts Clinical Outcomes after Immunotherapy in Patients with Non-Small Cell Lung Cancer. Cancer immunology research 8, 146-154, doi:10.1158 / 2326-6066.cir-19-0398 (2020). (Reference 19) Lerebours, F. et al. NF-kappa B genes have a major role in inflammatory breast cancer. BMC cancer 8, 41, doi:10.1186 / 1471-2407-8-41 (2008). (Reference 20) Bjarnadottir, O. et al. Global Transcriptional Changes Following Statin Treatment in Breast Cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 21, 3402-3411, doi:10.1158 / 1078-0432.ccr-14-1403 (2015). (Reference 21) Ouhtit, A. et al. Simultaneous inhibition of cell-cycle, proliferation, survival, metastatic pathways and induction of apoptosis in breast cancer cells by a phytochemical super-cocktail: genes that underpin its mode of action. Journal of Cancer 4, 703-715, doi: (Reference 22) Wang, J.-W. et al. Deregulated expression of LRBA facilitates cancer cell growth. Oncogene 23, 4089-4097, doi:10.1038 / sj.onc.1207567 (2004). (Reference 23) Tozlu, S. et al. Identification of novel genes that co-cluster with estrogen receptor alpha in breast tumor biopsy specimens, using a large-scale real-time reverse transcription-PCR approach. Endocrine-related cancer 13, 1109-1120, doi:10.1677 / erc.1.01120 (2006). (Reference 24) Andres, S. A., Brock, G. N. & Wittliff, J. L. Interrogating differences in expression of targeted gene sets to predict breast cancer outcome. BMC cancer 13, 326, doi:10.1186 / 1471-2407-13-326 (2013). (Reference 25) Wang, J. W. et al. Deregulated expression of LRBA facilitates cancer cell growth. Oncogene 23, 4089-4097, doi:10.1038 / sj.onc.1207567 (2004). (Reference 26) Andres, S. A., Smolenkova, I. A. & Wittliff, J. L. Gender-associated expression of tumor markers and a small gene set in breast carcinoma. Breast 23, 226-233, doi:10.1016 / j.breast.2014.02.007 (2014). (Reference 27) Holt, O. J., Gallo, F. & Griffiths, G. M. Regulating Secretory Lysosomes. The Journal of Biochemistry 140, 7-12, doi:10.1093 / jb / mvj126 (2006). (Reference 28) Bong, I. P., Ng, C. C., Fakiruddin, S. K., Lim, M. N. & Zakaria, Z. Small interfering RNA-mediated silencing of nicotinamide phosphoribosyltransferase (NAMPT) and lysosomal trafficking regulator (LYST) induce growth inhibition and apoptosis in human multiple myeloma cells: A preliminary study. Bosnian journal of basic medical sciences 16, 268-275, doi:10.17305 / bjbms.2016.1568 (2016). (Reference 29) Sara, M. PEX1 and PEX7 Genes are Necessary for Completion of Immune Pathways and Survival Post-Infection. STEM Fellowship Journal 6, 5-11, doi:10.17975 / sfj-2020-002 (2020). (Figure 30)Pang, EY et al. Identification of PFTAIRE protein kinase 1, a novel cell division cycle-2 related gene, in the motile phenotype of hepatocellular carcinoma cells. Hepatology 46, 436–445, doi:10.1002 / hep.21691 (2007). (Table 31)Chen, Y. et al. Identification of druggable cancer driver genes amplified across TCGA datasets. PLoS One 9, e98293, doi:10.1371 / journal.pone.0098293 (2014). (Introduction 32)Del Rizzo, PA, Krishnan, S. & Trievel, RC Crystal structure and functional analysis of JMJD5 indicate an alternate specificity and function. Molecular and cellular biology 32, 4044–4052. (Publication 33)Oh, S., Shin, S. & Janknecht, R. The small members of the JMJD protein family: Enzymatic jewels or jinxes? Biochemistry and Biophysics Acta (BBA) - Reviews on Cancer 1871, 406-418. (Reference 34) Tucci, A. et al. Novel C12orf65 mutations in patients with axonal neuropathy and optic atrophy. Journal of neurology, neurosurgery, and psychiatry 85, 486 - 492, doi:10.1136 / jnnp-2013-306387 (2014). (Reference 35) Pai, T. et al. Evidence for the association of Epstein-Barr Virus in breast cancer in Indian patients using in-situ hybridization technique. The breast journal 24, 16 - 22, doi:10.1111 / tbj.12828 (2018). (Reference 36) Horakova, D. et al. Risks and protective factors for triple negative breast cancer with a focus on micronutrients and infections. Biomedical papers of the Medical Faculty of the University Palacky, Olomouc, Czechoslovakia 162, 83 - 89, doi:10.5507 / bp.2018.014 (2018). (Reference 37) Chen, X. et al. Cyclin E Overexpression Sensitizes Triple-Negative Breast Cancer to Wee1 Kinase Inhibition. Clinical cancer research : an official journal of the American Association for Cancer Research 24, 6594 - 6610, doi:10.1158 / 1078-0432.CCR-18-1446 (2018). (Reference 38) Sharma, P. Biology and Management of Patients With Triple-Negative Breast Cancer. The oncologist 21, 1050-1062, doi:10.1634 / theoncologist.2016-0067 (2016). (Reference 39) Masuda, N. et al. Adjuvant Capecitabine for Breast Cancer after Preoperative Chemotherapy. The New England journal of medicine 376, 2147-2159, doi:10.1056 / NEJMoa1612645 (2017).

Claims

1. Ankyrin repeat domain 36 (ANKRD36), Ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), containing B-box and SPRY domain (BSPRY), chromosome 12 open reading frame 65 (C12orf65), chromosome 2 open reading frame 49 (C2orf49), chromosome 1 open reading frame 198 (C1orf198), containing coiled-coil domain 114 (CCDC114), claw Din 4 (CLDN4), CUE domain-containing protein 1 (CUEDC1), outer dynein arm docking complex subunit 1 (ODAD1), EF hand domain-containing cell growth regulator (CGREF1), DEP domain-containing protein 7 (DEPDC7), double cortin-like kinase 2 (DCLK2), diacylglycerol kinase (DGKH), Disco interaction protein 2 homolog B (DIP2B), schizophrenia-related factor 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH mRNA splicing and mitotic factors (ERH), growth arrest / DNA damage-inducing β (GADD45B), glutaminase (GLS), grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member X (H2AFX), HRas proto-oncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium potential-opening channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor Child 7 (KLF7), keratin 17 (KRT17), LON peptidase N-terminal domain and RING finger protein 2 (LONRF2), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, U7 nuclear small RNA-related (LSM11), lysosomal transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisense RNA1 (MCM3AP_AS1), MICAL-like 2 (MICALL2), MHC class I polypeptide-related sequence B (MICB), metallothionein 2A (MT2A), myelin expression factor 2 (MYEF2), NEDD4 binding protein 3 (N4BP3), neuroblastocyte breakpoint family member 20 (NBPF20), NADH: ubiquinone oxidoreductase core subunit V2 (NDUFV2), neurogenic locus Notch homolog protein 2 (NOTCH2), NADPH oxidase activator 1 (NOXA1), natriuretic peptide receptor 3 (NPR3), nuclear receptor subfamily 6 group A member 1 (NR6A1), P21 (RAC1) activating kinase 3 (PAK3), pantothenate kinase 3 (PANK3), par-6 family cell polarity regulator beta (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyBac transposable element-derived 4 (PGBD4), Praja ring finger ubiquitin ligase 1 (PJA1), plekstrin homology and FYVE domain-containing 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K), prickle planar cell polarity protein 1 (PRICKLE1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), PYD and CARD domain-containing (PYCARD), RAS p21 protein activator 1 (RASA1), RAS family member 2 (RASD2), RAS guanylate-releasing protein 1 (RASGRP1), rhophyllin RHOGTPase-binding protein 2 (RHPN2), Rab-interacting lysosomal protein-like protein 2 (RILPL2), roundabout guidance receptor 1 (ROBO1), RAR orphan receptor A (RORA), stromal cell-derived factor 4 (SDF4), SERTA domain-containing protein 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-induced proliferation-related protein 1-like protein 2 (SIPA1L2), solute carrier family 22 member 20 (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), solute carrier family 2 member 12 (SLC2A12), solute carrier family 39 member 10 (SLC39A10), solute carrier family 25 member 40 (SLC25 A40), Solute Carrier Family 43 Member 1 (SLC43A1), Solute Carrier Family 45 Member 4 (SLC45A4), Solute Carrier Family 6 Member 20 (SLC6A20), Spectrin Beta, Erythrocyte (SPTB), Interstitial Antigen 3-like 3 (STAG3L3), Sucrose Domain-containing 3 (SUSD3), TATA Box-binding Protein-related Factor 10 (TAF10), t-complex-associated Testicular Expression 3 (TCTE3), Transmembrane Channel-like 7 (TMC7), Transmembrane and Coiled-Coil Domain Family 2 (TMCC2), Transcriptional Repressor GATA-binding 1 (TRPS1), Tubulin Tyrosine Ligase-like 4 (TTLL4), tRNA-YW Synthesizing Protein 5 (TYW5), Ubiquitin-conjugating Enzyme E2An isolated set of probes capable of detecting a panel of biomarkers comprising at least two biomarkers selected from the group consisting of W (UBE2W), WD repeat, sterile alpha motif and U-box domain-containing 1 (WDSUB1), N-protein N-terminal glutamine amide hydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED type-containing 6 (ZBED6), zinc finger and BTB domain-containing 46 (ZBTB46), zinc finger CCCH type-containing 13 (ZC3H13), zinc finger and homeobox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785).

2. The biomarker panel includes DGKH, KLF7, NR6A1, PYD and CARD domain-containing, ROBO1, SLC22A20P, SLC24A3, DIP2B, EMP1, NOTCH2, RAR orphan receptor A, NOXA1, CUE domain-containing 1, PRICKLE1, DCKL2, C12orf65, growth arrest / DNA damage-inducing β, LRBA, lysosomal transport regulator, PEX1, PRKAB2, TYW5, ODAD1, DEP domain-containing 7, MICAL-like 2, SLC43A1, S A set of isolated probes according to claim 1, comprising at least two biomarkers selected from the group consisting of LC6A20, RASA1, SLC45A4, NTAQ1, CGREF1, MICB, LSM11, PJA1, C2orf49, HRAS, KCNC3, MT2A, LRIT3, SHISA5, SLC25A40, H2AFX, PTPRA, RILPL2, ZC3CH13, ZHX2, CDLN4, ERH, glycophorin C, MTA2, NDUFV2, SDF4, and UBE2W.

3. A set of isolated probes according to claim 1 or 2, wherein the biomarker panel comprises at least three biomarkers, at least four biomarkers, at least five biomarkers, at least six biomarkers, at least seven biomarkers, at least eight biomarkers, at least nine biomarkers, or ten biomarkers selected from the following biomarker signatures: (a) DGKH, growth arrest / DNA damage-inducing β, KLF7, lysosome transport regulator, NR6A1, PYD and CARD domain-containing, ROBO1, SLC22A20P, SLC24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, growth arrest / DNA damage-inducible β, MT2A, NOTCH2, NR6A1, RAR orphan receptor A, SLC22A20P, and SLC24A3; (c) Growth arrest / DNA damage-inducing β, lysosomal transport regulator, NOXA1, NR6A1, PYD and CARD domain-containing, SLC22A20P, SLC24A3, and NTAQ1; (d) DGKH, KLF7, lysosomal transport regulators, NR6A1, ROBO1, SLC24A3, and SLC6A20; (e) DGKH, EMP1, growth arrest / DNA damage-inducing β, lysosome transport regulator, SLC22A20P, SLC24A3, and SLC6A20; (f) CUE domain-containing 1, DGKH, EMP1, lysosomal transport regulator, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, lysosomal transport regulator, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, growth arrest / DNA damage-inducing β, lysosome transport regulator, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) Growth arrest / DNA damage-inducing β, KLF7, lysosomal transport regulator, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, growth arrest / DNA damage-inducing β, lysosome transport regulator, NR6A1, SLC22A20P, and NTAQ1; (k)C12orf65, growth arrest / DNA damage-induced β, LRBA, lysosomal transport regulator, PEX1, PRKAB2, and TYW5; (l) Lysosomal transport regulators, PEX1, ODAD1, DEP domain-containing 7, MICAL-like 2, SLC43A1, and SLC6A20; (m)C12orf65, growth arrest / DNA damage-inducing β, lysosomal transport regulator, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEP domain-containing 7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, lysosomal transport regulator, PEX1, DEP domain-containing 7, SLC43A1, MICB, and C2orf49; (p) Growth arrest / DNA damage-inducing β, PEX1, DEP domain-containing 7, SLC43A1, LSM11, and PJA1; (q) DEP domain-containing 7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r) Lysosomal transport regulators, DEP domain-containing 7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICAL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t) Growth arrest / DNA damage-inducing β, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AF, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v) CLDN4, ERH, growth arrest / DNA damage-inducing β, glycophorin C, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

4. An isolated probe set according to any one of claims 1 to 3, wherein the probe is selected from the group consisting of aptamers, antibodies, aphibodies, peptides, proteins, organic molecules, and nucleic acids.

5. A method for predicting the disease prognosis and / or treatment outcomes of a subject diagnosed with cancer, comprising: (a) Obtaining a sample from the subject, (b) Contacting the sample with the isolated probe set described in any one of claims 1 to 4 to detect a panel of biomarkers in the sample, (c) Analyze the patterns of the biomarkers in the panel to determine the risk score of the subject.

6. A method according to claim 5, wherein the method further comprises the following: (d) Classify subjects as high-risk or low-risk based on their risk score.

7. A method according to claim 5 or 6, wherein the cancer is breast cancer.

8. A method according to claim 7, wherein the breast cancer is triple-negative breast cancer (TNBC).

9. A method according to claim 8, wherein the triple-negative breast cancer is an early-stage triple-negative breast cancer.

10. A method according to claim 8 or 9, further comprising treating triple-negative breast cancer in a subject based on the classification of the subject.

11. The method according to any one of claims 5 to 10, wherein the subject is high-risk, and the method further comprises an advanced, enhanced, or standard form of treatment for triple-negative breast cancer, comprising surgery and / or administration of a chemotherapeutic agent, radiotherapy, an immunotherapy agent, any novel therapeutic agent, or a combination of treatments.

12. The method according to claim 11, wherein the immunotherapy agent is an immune checkpoint inhibitor.

13. The method according to claim 12, wherein the immune checkpoint inhibitor comprises pembrolizumab, atezolizumab, nivolumab, ipilimumab, durvalumab, and / or avelumab.

14. The method according to claim 11, wherein the chemotherapeutic agent comprises capecitabine, doxorubicin, cyclophosphamide, docetaxel, olaparib, carboplatin, paclitaxel, epirubicin, methotrexate, and / or fluorouracil.

15. The method according to any one of claims 5 to 10, wherein the subject is low risk, and the method further comprises administering standard or attenuated treatment for triple-negative breast cancer, either surgery alone or including surgery, and / or administering a chemotherapeutic agent, radiotherapy, immunotherapy, any novel treatment, or a combination of treatments.

16. The method according to claim 15, wherein the immunotherapy agent is an immune checkpoint inhibitor.

17. The method according to claim 16, wherein the immune checkpoint inhibitor comprises pembrolizumab, atezolizumab, nivolumab, ipilimumab, durvalumab, and / or avelumab.

18. The method according to claim 15, wherein the chemotherapeutic agent comprises capecitabine, doxorubicin, cyclophosphamide, docetaxel, olaparib, carboplatin, paclitaxel, epirubicin, methotrexate, and / or fluorouracil.

19. The method according to any one of claims 5 to 18, wherein the sample is a tissue sample, a blood sample, or a urine sample.

20. The method according to claim 19, wherein the tissue sample is a fresh frozen tumor tissue sample or a fixed formalin paraffin-embedded tumor tissue sample.

21. A kit for predicting disease prognosis and / or treatment outcomes in patients diagnosed with triple-negative breast cancer (TNBC), comprising: (a) Ankyrin repeat domain 36 (ANKRD36), ankyrin repeat domain 36B pseudogene 2 (ANKRD36BP2), containing B box and SPRY domain (BSPRY), chromosome 12 open reading frame 65 (C12orf65), chromosome 2 open reading frame 49 (C2orf49), chromosome 1 open reading frame 198 (C1orf198), containing coiled-coil domain 114 (CCDC114), clode Cell growth regulator containing 4 (CLDN4), CUE domain 1 (CUEDC1), outer dynein arm docking complex subunit 1 (ODAD1), EF hand domain 1 (CGREF1), DEP domain 7 (DEPDC7), double cortin-like kinase 2 (DCLK2), diacylglycerol kinase (DGKH), discone-acting protein 2 homolog B (DIP2B), schizophrenia-related factor 1 (DISC1), epithelial membrane protein 1 (EMP1), ERH mRNA splicing and mitotic factors (ERH), growth arrest / DNA damage-inducing β (GADD45B), glutaminase (GLS), Grainyhead-like transcription factor 1 (GRHL1), glycophorin C (GYPC), H2A histone family member X (H2AFX), HRas proto-oncogene (HRAS), intracellular adhesion molecule 1 (ICAM1), photoreceptor matrix proteoglycan 2 (IMPG2), potassium potential-opening channel subfamily C member 3 (KCNC3), Kruppel-like factor 6 (KLF6), Kruppel-like factor 7 (KLF7), keratin 17 (KRT17), LON peptidase N-terminal domain and RING finger protein 2 (LONRF2), LPS-responsive beige-like anchor protein (LRBA), leucine-rich repeat, Ig-like and transmembrane domain 3 (LRIT3), leucine-rich repeat-containing 37B (LRRC37B), LSM11, U7 nuclear small RNA-related (LSM11), lysosomal transport regulator (LYST), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), minichromosome maintenance complex component 3-related protein antisense RNA1 (MCM3AP_AS1), MICAL-like 2 (MICALL2), MHC class I polypeptide-related sequence B (MICB), metallothionein 2A (MT2A), myelin expression factor 2 (MYEF2), NEDD4 binding protein 3 (N4BP3), neuroblastocyte breakpoint family member 20 (NBPF20), NADH: ubiquinone oxidoreductase core subunit V2 (NDUFV2), neurogenic locus Notch homolog protein 2 (NOTCH2), NADPH oxidase activator 1 (NOXA1), natriuretic peptide receptor 3 (NPR3), nuclear receptor subfamily 6 group A member 1 (NR6A1), P21 (RAC1) activating kinase 3 (PAK3), pantothenate kinase 3 (PANK3), par-6 family cell polarity regulator beta (PARD6B), peroxisome biosynthesis factor 1 (PEX1), piggyBac transposable element-derived 4 (PGBD4), Praja ring finger ubiquitin ligase 1 (PJA1), plekstrin homology and FYVE domain-containing 1 (PLEKHF1), purine nucleoside phosphorylase (PNP), protein tyrosine phosphatase receptor type A (PTPRA), protein phosphatase, Mg2+ / Mn2+ dependent 1K (PPM1K), prickle planar cell polarity protein 1 (PRICKLE1), protein kinase AMP-activated non-catalytic subunit beta 2 (PRKAB2), PYD and CARD domain-containing (PYCARD), RAS p21 protein activator 1 (RASA1), RASD family member 2 (RASD2), RAS guanylate-releasing protein 1 (RASGRP1), rhophyllin RHOGTPase-binding protein 2 (RHPN2), Rab-interacting lysosomal protein-like protein 2 (RILPL2), roundabout-inducing receptor 1 (ROBO1), RAR orphan receptor A (RORA), stromal cell-derived factor 4 (SDF4), SERTA domain-containing protein 4 (SERTAD4), SHISA family member 5 (SHISA5), signal-inducing proliferation-related protein 1-like protein 2 (SIPA1L2), solute carrier family 22 member 20 (SLC22A20P), solute carrier family 24 member 3 (SLC24A3), solute carrier family 2 member 12 (SLC2A12), solute carrier family 39 member 10 (SLC39A10), solute carrier family 25 member 40 (SLC25A4) 0), Solute carrier family 43 member 1 (SLC43A1), Solute carrier family 45 member 4 (SLC45A4), Solute carrier family 6 member 20 (SLC6A20), Spectrin beta, Red blood cell (SPTB), Interstitial antigen 3-like 3 (STAG3L3), Sucrose domain-containing 3 (SUSD3), TATA box-binding protein-related factor 10 (TAF10), t-complex-associated testicular expression 3 (TCTE3), Transmembrane channel-like 7 (TMC7), Transmembrane and coiled-coil domain family 2 (TMCC2), Transcriptional repressor GATA binding 1 (TRPS1), Tubulin tyrosine ligase-like 4 (TTLL4), tRNA-yW synthesis protein 5 (TYW5), Ubiquitin-conjugating enzyme E2W (UBE2W), WD repeat, sterile alpha motif and U-box domain contained 1 (WDSUB1), N-protein N-terminal glutamine amidohydrolase (WDYHV1), N-terminal glutamine amidase 1 (NTAQ1), zinc finger BED type contained 6 (ZBED6), zinc finger and BTB domain contained 46 (ZBTB46), zinc finger CCCH type contained 13 (ZC3H13), zinc finger and home An isolated set of probes capable of detecting a panel of biomarkers comprising at least two biomarkers selected from the group consisting of Ovox 2 (ZHX2), zinc finger protein 217 (ZNF217), zinc finger protein 233 (ZNF233), zinc finger protein 248 (ZNF248), zinc finger protein 469 (ZNF469), and zinc finger protein 785 (ZNF785); and (b) Instructions for use.

22. The kit according to claim 21, wherein the isolated set of probes capable of detecting a panel of biomarkers comprises at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or ten biomarkers selected from the following biomarker signatures: (a) DGKH, GADD45B, KLF7, LYST, NR6A1, PYCARD, ROBO1, SLC22A20P, SLC24A3, and SLC45A4; (b) DGKH, DIP2B, EMP1, growth arrest / DNA damage-inducing β, MT2A, NOTCH2, NR6A1, RAR orphan receptor A, SLC22A20P, and SLC24A3; (c) Growth arrest / DNA damage-inducing β, lysosomal transport regulator, NOXA1, NR6A1, PYD and CARD domain-containing, SLC22A20P, SLC24A3, and NTAQ1; (d) DGKH, KLF7, lysosomal transport regulators, NR6A1, ROBO1, SLC24A3, and SLC6A20; (e) DGKH, EMP1, growth arrest / DNA damage-inducing β, lysosome transport regulator, SLC22A20P, SLC24A3, and SLC6A20; (f) CUE domain-containing 1, DGKH, EMP1, lysosomal transport regulator, NOXA1, SLC22A20P, and SLC6A20; (g) DGKH, KLF7, lysosomal transport regulator, NR6A1, PRICKLE1, ROBO1, and SLC24A3; (h) DCLK2, growth arrest / DNA damage-inducing β, lysosome transport regulator, NR6A1, SLC22A20P, SLC24A3, and NTAQ1; (i) Growth arrest / DNA damage-inducing β, KLF7, lysosomal transport regulator, NR6A1, ROBO1, SLC22A20P, and SLC6A20; (j) DCLK2, growth arrest / DNA damage-inducing β, lysosome transport regulator, NR6A1, SLC22A20P, and NTAQ1; (k)C12orf65, growth arrest / DNA damage-inducing β, LRBA, lysosomal transport regulator, PEX1, PRKAB2, and TYW5; (l) Lysosomal transport regulators, PEX1, ODAD1, DEP domain-containing 7, MICAL-like 2, SLC43A1, and SLC6A20; (m)C12orf65, growth arrest / DNA damage-inducing β, lysosomal transport regulator, PEX1, RASA1, SLC45A4, and NTAQ1; (n)TYW5, DEP domain-containing 7, SLC43A1, CGREF1, MICB, HRAS, and MT2A; (o)LRBA, lysosomal transport regulator, PEX1, DEP domain-containing 7, SLC43A1, MICB, and C2orf49; (p) Growth arrest / DNA damage-inducing β, PEX1, DEP domain-containing 7, SLC43A1, LSM11, and PJA1; (q) DEP domain-containing 7, SLC43A1, MICB, C2orf49, HRAS, KCNC3, and MT2A; (r) Lysosomal transport regulators, DEP domain-containing 7, SLC43A1, SLC6A20, MICB, and HRAS; (s)PEX1, MICAL2, SLC43A1, RASA1, KCNC3, and LRIT3; (t) Growth arrest / DNA damage-inducing β, SLC43A1, SLC45A4, KCNC3, SHISA5, and SLC25A40; (u)GADD45B, H2AF, PTPRA, RILPL2, RORA, ZC3H13, and ZHX2; or (v) CLDN4, ERH, growth arrest / DNA damage-inducing β, glycophorin C, H2AFX, MT2A, NDUFV2, SDF4, and UBE2W.

23. The kit according to claim 21 or 22, wherein the probe is selected from the group consisting of aptamers, antibodies, affibodies, peptides, proteins, organic molecules, and nucleic acids.

Images