Diagnostic signature

JP2025516374A5Pending Publication Date: 2026-05-21ビーカルダイアグノスティックスリミテッド

View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office: JP · JP
Patent Type: Applications
Current Assignee / Owner: ビーカルダイアグノスティックスリミテッド
Filing Date: 2023-05-10
Publication Date: 2026-05-21

Application Information

Patent Timeline

10 May 2023

Application

21 May 2026

Publication

JP2025516374A5

IPC: G01N33/92

CPC: G01N33/92; G01N2570/00; G01N2405/00; G01N2800/7028; G01N33/57515; G01N2405/02; G01N2405/06; G01N2405/10

AI Tagging

Application Domain

OmicsDisease diagnosis

Explore More Agents

Novelty Search
Search existing technologies and assess novelty
↗
FTO
Analyze whether a product may infringe others' patents
↗
Design FTO
Check prior-design risk for exterior design
↗
Drafting
Draft patent application text based on a technical solution
↗
Find Solutions with TRIZ
Generate feasible solution to solve your technical challenge
↗

Similar Technology Patents

Marker for early detection of proliferative disorders of colonic cells
JP2025521164A5Microbiological testing/measurementOmics
Methods, compositions, and systems for mapping locations of single molecules in multi-dimensional space
US12655473B2Microbiological testing/measurementOmics
An alveolar microresponse system for PM2.5 carrying non-water-soluble particles and a model for assessing its relative bioavailability.
CN119385108BClear degradation and transformation microreaction systemOmicsBiological testingLipidomeActive agent
Method of ultrasound microbubble-assisted liquid biopsy
US20260165689A1Microbiological testing/measurement Preparing sample for investigation
Systems and methods for biomolecular assays
JP2025522439A5Microbiological testing/measurementOmics

Get free access to AI patent search and analysis

Check patentability, review prior art and ask IP Agent with full patent context.

AI Technical Summary

Technical Problem

Current methods for diagnosing breast cancer, such as mammograms, are less effective in younger ages and lack accuracy for detecting cancer at the cellular level before metastasis.

Method used

The use of specific lipid biomarkers, including Cer(d36:1), Cer(d42:0), DG(34:2), and others, to diagnose breast cancer by measuring their levels in biological samples through methods like mass spectrometry.

Benefits of technology

This approach enables early and accurate detection of breast cancer, potentially improving treatment outcomes by identifying cancer at a cellular level before metastasis.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure 00000098_0000
Figure 00000098_0001
Figure 00000098_0002

Patent Text Reader

Abstract

The present disclosure relates to a method for diagnosing and treating breast cancer in a subject, including determining the levels of one or more lipid biomarkers in a biological sample obtained from the subject. Systems and kits for use in such methods are also provided.

Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority from Australian Provisional Patent Application No. 2022 / 901245, filed on May 10, 2022, and Australian Provisional Patent Application No. 2022 / 903097, filed on October 20, 2022, the contents of which are incorporated herein by reference in their entireties.

[0002] The present disclosure relates to methods for diagnosing and treating breast cancer. [Background technology]

[0003] In 2020, 2.3 million women were diagnosed with breast cancer and 685,000 died worldwide. As of the end of 2020, 7.8 million women had been diagnosed with breast cancer in the previous five years, making it the most common cancer in the world (World Health Organization). Although breast cancer is primarily a disease of women, 1 in 1,100 men will also develop the disease (Society, 2016).

[0004] The key to overcoming breast cancer is early detection and treatment. The current gold standard for detection is via mammography, however, it is known to be less effective at younger ages. Therefore, there remains a need for more accurate screening tests for breast cancer in women of all ages, such as those that detect cancer at the cellular level and before metastasis (Mistry and French, 2016). Summary of the Invention

[0005] The present disclosure is based on the surprising discovery of several lipid biomarkers that can be easily detected to diagnose breast cancer in women. By extension, these lipid biomarkers (or lipidomic signatures) show promise in identifying patients in need of treatment for breast cancer.

[0006] In a first aspect, the disclosure provides a method of diagnosing a subject with breast cancer, the method comprising measuring a level of one or more lipid biomarkers in a biological sample from the subject, the one or more lipid biomarkers being Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), and / or LPC(16:0e). ), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4) ), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34: 1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36) :2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG (54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof, wherein the level of one or more lipid biomarkers is diagnostic or indicative of the subject having breast cancer.

[0007] In some examples of the method, Cer(d36:1), Cer(d42:0), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PI(32:1), PI(34:1), increased levels of PS(38:4), PS(40:6), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), and / or SM(d44:4), or fragments, variants, or derivatives thereof; and / or Cer(42:1), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE( 38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), T Decreased levels of G(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and / or TG(62:2), or fragments, variants, or derivatives thereof, are diagnostic or indicative of a subject having breast cancer.

[0008] Preferably, the method further comprises the step of administering to the subject a treatment for breast cancer.

[0009] In a second aspect, the present disclosure provides a method for measuring the level of one or more lipid biomarkers in a biological sample from a subject, the method comprising: (a) providing a biological sample; (b) measuring the level of one or more lipid biomarkers in the biological sample, wherein the one or more lipid biomarkers are Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPC(19:0), LPC(20:0), LPC(21:0), LPC(22:0), LPC(23:0), LPC(24:0), LPC(25:0), LPC(26:0), LPC(27:0), LPC(28:0), LPC(29:0), LPC(30:0), LPC(31:0), LPC(32:0), LPC(33:0), LPC(34:0), LPC(35:0), LPC(36:0), LPC(37:0), LPC(38:0), LPC(39 ... E(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34 :1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36 :1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36) :2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG( and measuring a nucleotide sequence selected from the group consisting of TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0010] Preferably, the subject is suspected of having or has previously been diagnosed with breast cancer.

[0011] In some examples, the measuring step comprises determining the presence or absence of: (i) Cer(d36:1), Cer(d42:0), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PI(3 2:1), PI(34:1), PS(38:4), PS(40:6), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), and / or SM(d44:4), or fragments, variants, or derivatives thereof , and / or (ii) Cer(42:1), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), PE(34:1), PE(34:2p), PE(36:4), PE (38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS (40:7), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and / or TG(62:2), or fragments, variants, or derivatives thereof.

[0012] In a third aspect, the disclosure provides a method of treating breast cancer in a subject, the method comprising administering the treatment to a subject in which levels of one or more lipid biomarkers diagnostic of or indicative of the subject having breast cancer have been measured in a biological sample from the subject, wherein the one or more lipid biomarkers are Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34: 1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34: 1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2) , SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM( d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0013] Preferably, the treatment comprises administering to the subject a therapeutically effective amount of an anti-cancer treatment.

[0014] In some examples of the method, Cer(d36:1), Cer(d42:0), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PI(32:1), PI(34:1), , PS(38:4), PS(40:6), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), and / or SM(d44:4), or fragments, variants, or derivatives thereof, and / or Yes, Cer(42:1), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7) ), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and / or TG(62:2), or fragments, variants, or derivatives thereof, are measured from a biological sample of the subject.

[0015] With reference to the method of the previous embodiment, the one or more lipid biomarkers are preferably LPC(14:0), or a fragment, variant, or derivative thereof, as well as Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(16:0e), LPC(17:1), LPC(18:3), , LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE (34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI (36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM( d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), and one or more other lipid biomarkers selected from the group consisting of TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0016] For the above aspects, the one or more lipid biomarkers are preferably PI(38:6), or a fragment, variant, or derivative thereof, as well as Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC( 18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38: 5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34 :1), PI(36:1), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d3 6:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), T and one or more other lipid biomarkers selected from the group consisting of G(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0017] In some examples of the methods of the aforementioned aspects, the one or more lipid biomarkers are LPC(14:0) and PI(38:6), or fragments, variants, or derivatives thereof, and optionally Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(16:0e), LPC( 17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37: 4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1) ), PI(34:1), PI(36:1), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e) ), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0018] In certain examples of the methods of the above embodiments, the one or more lipid biomarkers are selected from the group consisting of LPC(14:0), PC(32:1), PC(38:5), PE(34:1), PS(38:4), SM(d36:2), SM(d38:4), TG(54:4), TG(56:1), and TG(58:2), or fragments, variants, or derivatives thereof.

[0019] In other examples of the methods of the above aspects, the one or more lipid biomarkers are selected from the group consisting of LPC(14:0), PC(32:1), PC(36:2), PC(38:5), PE(34:1), PI(34:1), PS(38:4), SM(d36:2), SM(d38:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(58:2), and TG(58:3), or fragments, variants, or derivatives thereof.

[0020] In various examples of the methods of the above aspects, the one or more lipid biomarkers are selected from the group consisting of LPC(14:0), PE(34:1), PS(38:4), SM(d36:2), TG(52:3e), and TG(58:3), or fragments, variants, or derivatives thereof.

[0021] In certain examples of the methods of the above embodiments, the one or more lipid biomarkers are selected from the group consisting of LPC(14:0), LPC(16:0e), LPE(22:6), PE(34:2p), PE(38:3p), PI(36:1), PI(38:6), PS(38:4), PS(40:6), SM(d33:1), SM(d35:1), and TG(57:1), or fragments, variants, or derivatives thereof. To this end, the one or more lipid biomarkers are selected from the group consisting of: (a) LPC(14:0), PS(38:4), and TG(57:1), or fragments, variants, or derivatives thereof; (b) LPC(16:0e), LPE(22:6), and PI(38:6), or fragments, variants, or derivatives thereof; (c) PE(34:2p), PE(38:3p), PI(36:1), PI(38:6), and PS(40:6), or fragments, variants, or derivatives thereof; (d) LPC(14:0), PI(38:6), and SM(d33:1), or fragments, variants, or derivatives thereof; or (e) LPC(14:0), PI(38:6), and SM(d35:1), or fragments, variants, or derivatives thereof; and optionally one or more other lipid biomarkers selected from the group consisting of LPC(14:0), LPC(16:0e), LPE(22:6), PE(34:2p), PE(38:3p), PI(36:1), PI(38:6), PS(38:4), PS(40:6), SM(d33:1), SM(d35:1), and TG(57:1), or fragments, variants, or derivatives thereof.

[0022] In various examples of the methods of the above aspects, the one or more lipid biomarkers described herein are selected from the group consisting of LPC(14:0), LPE(22:6), PI(38:6), PE(34:2p), SM(d35:1), PS(38:4), and PS(38:4), or fragments, variants, or derivatives thereof.

[0023] Suitably, for the above embodiments, the levels of one or more lipid biomarkers are or have been determined, at least in part, by mass spectrometry.

[0024] Preferably, the predictive accuracy of the methods of the above-described embodiments, as determined by ROC AUC values, is at least about 0.65, at least about 0.70, at least about 0.75, or at least about 0.80.

[0025] In a fourth aspect, the present disclosure provides a system for determining the presence or absence of breast cancer in a subject, the system comprising: 1. A mass spectrometry unit configured to determine a level of one or more lipid biomarkers in a biological sample obtained from a subject, wherein the one or more lipid biomarkers are Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(1 7:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4) ), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1) , PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM (d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), a mass spectrometry unit selected from the group consisting of TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof; and a processing unit configured to use or analyze the levels of one or more lipid biomarkers to determine the presence or absence of breast cancer in a subject.

[0026] In a fifth aspect, the present disclosure provides a kit for determining the presence or absence of breast cancer in a subject, the kit comprising one or more reagents for determining the level of one or more lipid biomarkers in a biological sample obtained from the subject, the one or more lipid biomarkers being Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d 42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4) ), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38: 5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d3 3:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4) ), SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0027] Preferably, the one or more reagents include one or more probes, each probe being specific or selective for one of the one or more lipid biomarkers.

[0028] With reference to the above embodiment, the one or more lipid biomarkers are preferably Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC (34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p) , PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS( 40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM (d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54 :6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0029] In some examples of the above aspects, the one or more lipid biomarkers comprise two or more lipid biomarkers, three or more lipid biomarkers, four or more lipid biomarkers, or five or more lipid biomarkers selected from the group consisting of LPC(14:0), PC(32:1), PC(38:5), PE(34:1), PS(38:4), SM(d36:2), SM(d38:4), TG(54:4), TG(56:1), and TG(58:2), or fragments, variants, or derivatives thereof.

[0030] In other examples of the above aspects, the one or more lipid biomarkers comprise two or more lipid biomarkers, three or more lipid biomarkers, four or more lipid biomarkers, or five or more lipid biomarkers selected from the group consisting of LPC(14:0), PC(32:1), PC(36:2), PC(38:5), PE(34:1), PI(34:1), PS(38:4), SM(d36:2), SM(d38:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(58:2), and TG(58:3), or fragments, variants, or derivatives thereof.

[0031] In particular examples of the above embodiments, the one or more lipid biomarkers include two or more lipid biomarkers, three or more lipid biomarkers, four or more lipid biomarkers, or five or more lipid biomarkers selected from the group consisting of LPC(14:0), PE(34:1), PS(38:4), SM(d36:2), TG(52:3e), and TG(58:3), or fragments, variants, or derivatives thereof.

[0032] In various examples of the above aspects, the one or more lipid biomarkers comprise two or more lipid biomarkers, three or more lipid biomarkers, four or more lipid biomarkers, or five or more lipid biomarkers selected from the group consisting of LPC(14:0), LPC(16:0e), LPE(22:6), PE(34:2p), PE(38:3p), PI(36:1), PI(38:6), PS(38:4), PS(40:6), SM(d33:1), SM(d35:1), and TG(57:1), or fragments, variants, or derivatives thereof.

[0033] With regard to the above-mentioned aspects, the biological sample preferably is or comprises a blood sample, a plasma sample, and / or a serum sample.

[0034] Suitably the system of the fourth aspect or the kit of the fifth aspect is for use in the method of the first, second or third aspect. [Brief explanation of the drawings]

[0035] The following figures form part of the present specification and are included to further demonstrate certain aspects of the present disclosure. The present disclosure may be better understood by reference to one or more of these figures in combination with the detailed description of specific embodiments presented herein. It will be understood by those skilled in the art that numerous variations and / or modifications may be made to the above-described embodiments without departing from the broad general scope of the present disclosure. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

[0036] [Figure 1-1] Violin plot of lipids with feature importance >0.05 for (a) LPC(14:0), (b) TG(16:0_18:1_23:0), (c) LPC(18:3), and (d) PS(18:0_20:4). [Figure 1-2] (As mentioned above.) [Figure 2-1] Violin plot of lipids with feature importance >0.05 for (a) PI(18:2_20:4), (b) LPE(22:6), (c) LPC(16:0e), (d) Cer(d18:1_24:0), and (e) PE(16:0_20:4). [Figure 2-2] (As mentioned above.) [Figure 2-3] (As mentioned above.) [Figure 3-1] Violin plot of lipids with feature importance >0.05 for (a) PI(18:2_20:4), (b) PE(16:0p_18:2), (c) PI(18:0_18:1), (d) PS(18:0_22:6), and (e) PE(20:1p_18:2). [Figure 3-2] (As mentioned above.) [Figure 3-3] (As mentioned above.) [Figure 4-1] Violin plot of lipids with feature importance >0.05 for (a) LPC(14:0), (b) SM(d18:1_15:0), (c) LPC(17:1), (d) PI(18:2_20:4), and (e) SM(d41:2). [Figure 4-2] (As mentioned above.) [Figure 4-3] (As mentioned above.) [Figure 5-1] Violin plot of lipids with feature importance >0.05 for (a) PI (18:2_20:4), (b) LPC (14:0), and (c) SM (d35:1). [Figure 5-2] (As mentioned above.) [Figure 6] Overlaid ROC for candidate signatures arising from dataset 1 [Figure 7] Overlaid ROC for candidate signatures arising from dataset 2 [Figure 8] Overlaid ROC for candidate signatures arising from dataset 3 [Figure 9] Overlaid ROC for candidate signatures arising from dataset 4 [Figure 10] Overlaid ROC for candidate signatures arising from dataset 234 (validation) [Figure 11] Volcano plot showing fold change and log p-value - more significant changes are at the top. The initial set of markers was selected based on significance values (p-value < 0.05) after initial filtering; lipid ions were shown at earlier stages because perfect discriminators were not available. [Figure 12] Filtering based on fold change and p-value was used to select an additional set of 18 markers from Cohort 3, and the combined set was then condensed using stepwise regression and significance in a logistic model. [Figure 13]In this approach, Cohort 4 was used primarily for validation as described in Example 1. Repeating the process of identifying differentially expressed lipids for Cohort 4 showed that four of the biomarkers identified in Example 1 were present in the top 10 selections, and these were included in a small panel of six common markers for both approaches, "Common6-ML_GBM." [Figure 14] An initial panel of 18 established using cohorts 1 and 2 - performance estimated on the next available cohort (cohort 3) via leave-one-out cross-validation. [Figure 15] Performance estimated in the next available cohort (cohort 4) via a second panel of 18—leave-one-out cross-validation established using cohorts 1–3. [Figure 16] Panel 6 - Boxplot of lipid 1 (order from left to right for each dataset pair: control, breast cancer) [Figure 17] Panel 6 - Lipid 2 boxplot (order from left to right for each dataset pair: control, breast cancer) [Figure 18] Panel 6 - Box plot of lipid 2, same m / z (order from left to right for each dataset pair: control, breast cancer) [Figure 19] Panel 6 - Lipid 3 boxplot (order from left to right for each dataset pair: control, breast cancer) [Figure 20] Panel 6 - Box plot of lipid 3, same m / z (order from left to right for each dataset pair: control, breast cancer) [Figure 21] Panel 6 - Lipid 4 boxplot (order from left to right for each dataset pair: control, breast cancer) [Figure 22] Panel 6 - Lipid 5 boxplot (order from left to right for each dataset pair: control, breast cancer) [Figure 23] Panel 6 - Lipid 6 boxplot (order from left to right for each dataset pair: control, breast cancer) [Figure 24]ROC curve performance of the panel of 18 markers estimated in cohort 2 via leave-one-out cross-validation. [Figure 25] ROC curve performance of the panel of 18 markers estimated in cohort 3 via leave-one-out cross-validation. [Figure 26] ROC curve performance of the 400 reoptimized panel estimated in cohort 4 via leave-one-out cross-validation. [Figure 27] ROC curve performance of the 400 re-optimized panel estimated in cohorts 3 and 4 via leave-one-out cross-validation. [Figure 28] ROC curve performance of the restricted nine-lipid panel estimated in cohort 2 via leave-one-out cross-validation. [Figure 29] ROC curve performance of the optimized top 10 panels estimated in cohorts 2 and 3 via leave-one-out cross-validation. [Figure 30] ROC curve performance of a panel of four markers for a panel of six estimated in cohorts 2–4 via leave-one-out cross-validation. [Figure 31] ROC curve performance of a 4-marker panel of 6 panels estimated in cohorts 3 and 4 via leave-one-out cross-validation. [Figure 32] ROC curve performance of a panel of six common ML-GBMs estimated in cohorts 3 and 4 via leave-one-out cross-validation. [Figure 33] ROC curve performance of the Cohort 2 optimized panel estimated in Cohort 2 via leave-one-out cross-validation. [Figure 34-1] Lipids from the three BCAL panels are overlaid on a volcano plot (fold change is fasted / fed). Several of the markers are differentially expressed between fed and fasted states. None of the ML-GBM panel lipids (red) are differentially expressed (i.e., present at significantly different concentrations) between fasted and fed. [Figure 34-2] (As mentioned above.) [Figure 35-1] Correlation of ML-GBM signature scores obtained with different combinations of sample types. [Figure 35-2] (As mentioned above.) [Figure 36] ML-GBM score boxplot showing median and interquartile range for all sample categories; NEV = non-fasting EV, FEV = fasting EV, FP = fasting plasma, NV = non-fasting plasma. Low scores represent correct "control" classification. DETAILED DESCRIPTION OF THE INVENTION

[0037] General Techniques and Definitions Unless specifically defined otherwise, all technical and scientific terms used herein shall be construed to have the same meaning as commonly understood by one of ordinary skill in the art (e.g., in genomics, immunology, molecular biology, immunohistochemistry, biochemistry, oncology, and pharmacology).

[0038] The present disclosure will be practiced using, unless otherwise indicated, conventional techniques of molecular biology, microbiology, recombinant DNA technology, and immunology. Such procedures are described, for example, in Sambrook, Fritsch & Maniatis, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratories, New York, Fourth Edition (2012), Vols. I, II, and III, in their entirety; DNA Cloning: A Practical Approach, Vols. I and II (D.N. Glover, Second Edition, 1995), IRL Press, Oxford, in their entirety; Oligonucleotide Synthesis: A Practical Approach (M.J. Gait, ed., 1984), IRL Press, Oxford, in their entirety, and in particular in the articles therein by Gait, p. 22; Atkinson et al., pp. 35-81; Sproat et al., pp. 83-115; and Wu et al., pp. 135-151; Nucleic Acid Hybridization: A Practical Approach (B.D. Hames & S.J. Higgins, eds., 1985), IRL Press, Oxford, in their entirety. Press, Oxford, entire text; Immobilized Cells and Enzymes: A Practical Approach (1986) IRL Press, Oxford, entire text; Perbal, B., A Practical Guide to Molecular Cloning (1984), and Methods In Enzymology (S. Colowick and N. Kaplan, eds., Academic Press, Inc.), entire series.

[0039] Those skilled in the art will understand that the present disclosure is susceptible to variations and modifications other than those specifically described. The present disclosure should be understood to include all such variations and modifications. The present disclosure also includes, individually or collectively, all of the steps, features, compositions, and compounds referred to or shown in this specification, and any and all combinations of any two or more of such steps or features.

[0040] The present disclosure is not to be limited in scope by the specific embodiments described herein, which are intended for the purpose of illustration only. Functionally equivalent products, compositions, and methods are clearly within the scope of the present disclosure as described herein.

[0041] Each feature of any particular aspect or embodiment of the present disclosure may be applied mutatis mutandis to any other aspect or embodiment of the present disclosure.

[0042] Throughout this specification, unless specifically stated otherwise or the context requires otherwise, references to a single step, composition of matter, group of steps, or group of compositions of matter shall be interpreted to encompass one and more (i.e., one or more) of that step, composition of matter, group of steps, or group of compositions of matter.

[0043] As used herein, the singular forms "a," "and," and "the" include the plural forms of these words unless the context clearly dictates otherwise.

[0044] The term "and / or," e.g., "X and / or Y," shall be understood to mean either "X and Y" or "X or Y," and shall be interpreted as providing explicit support for both meanings or for either meaning.

[0045] Throughout this specification the word "comprise" or variations such as "comprises" or "comprising" will be understood to mean the inclusion of the specified element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

[0046] It will be understood by those skilled in the art that numerous variations and / or modifications may be made to the above-described embodiments without departing from the broad general scope of the present disclosure, and the present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

[0047] Diagnostic methods The inventors have surprisingly shown that the concentration levels of specific lipids from a lipidomic signature in a biological sample, such as a blood sample, can be used to diagnose a subject as having breast cancer. Advantageously, such a method may enable physicians to make appropriate, informed, and timely follow-up and treatment decisions based on this information.

[0048] Therefore, the present inventors have developed a method for diagnosing breast cancer.

[0049] Thus, in one broad aspect, the disclosure provides a method for measuring the level of one or more lipid biomarkers in a biological sample from a subject, the method comprising: (a) providing a biological sample; (b) measuring the level of one or more lipid biomarkers in the biological sample, wherein the one or more lipid biomarkers are Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPC(19:0), LPC(20:0), LPC(21:0), LPC(22:0), LPC(23:0), LPC(24:0), LPC(25:0), LPC(26:0), LPC(27:0), LPC(28:0), LPC(29:0), LPC(30:0), LPC(31:0), LPC(32:0), LPC(33:0), LPC(34:0), LPC(35:0), LPC(36:0), LPC(37:0), LPC(38:0), LPC(39 ... E(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34 :1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36 :1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36) :2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG( and measuring a nucleotide sequence selected from the group consisting of TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0050] In a related broad aspect, the disclosure provides a method of diagnosing a subject with breast cancer, the method comprising measuring a level of one or more lipid biomarkers in a biological sample from the subject, the one or more lipid biomarkers being Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16: 0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37: 4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34 :1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36) :2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG (54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof, wherein the level of one or more lipid biomarkers is diagnostic or indicative of the subject having breast cancer.

[0051] In reference to the embodiments described herein, the term "subject" includes, but is not limited to, mammals, including humans, performance animals (such as horses, camels, and greyhounds), livestock (such as cattle, sheep, and horses), and companion animals (such as cats and dogs). In one example, the subject is a human. In certain examples, the subject is a female human. In other examples, the subject is a male human.

[0052] As generally used herein, the terms "cancer," "tumor," "malignant," and "malignant" refer to a disease or condition, or a cell or tissue associated with a disease or condition, characterized by aberrant or abnormal cell proliferation, differentiation, and / or migration, expression of tumor markers, loss of tumor suppressor gene expression or activity, and / or aberrant or abnormal cell surface marker expression, often accompanied by an aberrant or abnormal molecular phenotype, including one or more genetic mutations or other genetic alterations associated with tumorigenesis.

[0053] The term "breast cancer" refers to a condition characterized by abnormally rapid growth of abnormal cells in one or both breasts of a subject. Breast cancer can include, but is not limited to, ductal carcinoma in situ (DCIS), invasive breast cancer (e.g., carcinoma in situ), inflammatory breast cancer, angiosarcoma of the breast, phyllodes tumor of the breast, and / or Paget's disease of the nipple. As used herein, "invasive cancer" or "invasive breast cancer" refers to a type of cancer that can include, but is not limited to, invasive ductal carcinoma (IDC), infiltrating ductal carcinoma, invasive lobular carcinoma (ILC), adenoid cystic (or adenoid cystic) carcinoma, low-grade adenosquamous carcinoma, medullary carcinoma, mucinous (or colloid) carcinoma, papillary carcinoma, tubular carcinoma, metaplastic carcinoma, micropapillary carcinoma, and / or mixed carcinoma with features of both invasive ductal and lobular breast cancer. Preferably, the breast cancer to be diagnosed in the subject is selected from IDC, DCIS, and ILC. In certain examples, the breast cancer to be diagnosed in the subject is IDC. In certain examples, the breast cancer to be diagnosed in the subject is DCIS. In other examples, the breast cancer to be diagnosed in the subject is ILC.

[0054] Those skilled in the art will understand that breast cancer may be classified as triple-negative breast cancer, lymph node-positive (LN+) breast cancer, HER2-positive (HER2+) breast cancer, PR-negative (PR-negative) breast cancer, or HER2-positive (HER2+) breast cancer. - ) Breast cancer, PR positive (PR + ) Breast cancer, ER negative (ER - It will be further understood that the term "breast cancer" may include any aggressive breast cancer and cancer subtype known in the art, such as ER-positive (ER+) breast cancer, and ER-positive (ER+) breast cancer.

[0055] The breast cancer may also be of any stage or grade (e.g., stage I, II, III, or IV), and as such may include metastatic breast cancer. Preferably, the breast cancer to be diagnosed is an early stage cancer (e.g., stage 1 or stage 2 breast cancer). In a particular example, the breast cancer to be diagnosed is stage 1 breast cancer. In another example, the breast cancer to be diagnosed is stage 2 breast cancer. Alternatively, the breast cancer to be diagnosed can be a later stage cancer (e.g., stage 3 or stage 4 breast cancer). In a particular example, the breast cancer to be diagnosed is stage 3 breast cancer. In another example, the breast cancer to be diagnosed is stage 4 breast cancer.

[0056] As used herein, the terms "diagnosis" and "diagnosing" refer to methods by which a skilled artisan can evaluate and / or determine whether a patient or subject is suffering from a given disease or condition, such as determining the presence or absence of breast cancer. Those skilled in the art often base their diagnosis on one or more diagnostic indicators or markers, the presence, absence, or amount of which indicates the presence or absence of a disease, disorder, or condition. It will be further understood that these terms do not indicate the ability to determine the presence or absence of a particular disease with 100% accuracy, nor do they indicate that a given course or outcome is more likely to occur. Rather, those skilled in the art will understand that the terms "diagnosis" and "diagnosing" refer to an increased probability that a subject will have a particular disease, disorder, or condition, such as breast cancer.

[0057] Preferably, the methods described herein are performed in conjunction with (e.g., before and / or after) one or more additional diagnostic tests as known in the art (e.g., breast examination; breast imaging such as mammogram, ultrasound, and MRI; biopsy). In this regard, the methods may be utilized as preliminary screening tests to identify subjects who may benefit from further diagnostic tests. Additionally or alternatively, the methods may be utilized to confirm the presence or absence of breast cancer as indicated by a previous diagnostic test. Thus, in some examples, the methods may include an initial or earlier and / or subsequent step of administering one or more further diagnostic tests to the subject in question. In alternative examples, the methods described herein are performed as primary diagnostic tests for breast cancer, without any further diagnostic tests.

[0058] Preferably, if the level, such as the concentration level or expression level, of one or more lipid biomarkers is altered or regulated in a biological sample from a subject, this may be diagnostic of breast cancer in the subject. In one example, an increased level of expression or concentration of a first subset of one or more lipid biomarkers and / or a decreased level of expression or concentration of a second subset of one or more lipid biomarkers is diagnostic or indicative of a subject having breast cancer. As shown in the examples herein, certain lipid biomarkers (e.g., PE(34:1), PI(32:1), PI(34:1), PS(38:4), and PS(40:6)) may exhibit increased expression or concentration levels in a first cohort of subjects, which is diagnostic of breast cancer, while also exhibiting decreased expression or concentration levels in a second cohort of subjects, which is diagnostic of breast cancer. Thus, for example, a decreased or increased level of such lipid biomarkers relative to a threshold or reference level may be indicative or diagnostic of breast cancer in a subject.

[0059] Specific examples include Cer(d36:1), Cer(d42:0), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PI(32:1), PI(34:1), PS(3 8:4), PS(40:6), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), and / or SM(d44:4), or fragments, variants, or derivatives thereof, and / or Cer (42:1), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38 :4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), TG( Decreased levels of TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and / or TG(62:2), or fragments, variants, or derivatives thereof, are diagnostic or indicative of a subject having breast cancer.

[0060] In certain examples, the measuring step comprises determining the presence or absence of: (i) Cer(d36:1), Cer(d42:0), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PI(3 2:1), PI(34:1), PS(38:4), PS(40:6), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), and / or SM(d44:4), or fragments, variants, or derivatives thereof , and / or (ii) Cer(42:1), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), PE(34:1), PE(34:2p), PE(36:4), PE (38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS (40:7), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and / or TG(62:2), or fragments, variants, or derivatives thereof.

[0061] According to certain examples, the measuring step includes determining the presence or absence of reduced levels of LPC(14:0) and / or PI(38:6), or fragments, variants, or derivatives thereof, in the subject's biological sample. More particularly, the measuring step may include determining the presence or absence of reduced levels of LPC(14:0), or fragments, variants, or derivatives thereof, in the subject's biological sample. Similarly, the measuring step may include determining the presence or absence of reduced levels of PI(38:6), or fragments, variants, or derivatives thereof, in the subject's biological sample. Even more particularly, the measuring step may include determining the presence or absence of reduced levels of LPC(14:0) and PI(38:6), or fragments, variants, or derivatives thereof, in the subject's biological sample.

[0062] Preferably, the method further comprises determining or calculating a risk or diagnostic score from the levels of one or more lipid biomarkers, e.g., by a method described herein. In some examples, a diagnostic score of 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or greater is determined for the subject.

[0063] Suitably, the diagnostic methods described herein may include the step of administering a treatment to the subject. By way of example, this may include administering to the subject a therapeutically effective amount of a treatment, such as an anti-cancer treatment described herein, where the level of one or more lipid biomarkers (and / or risk or diagnostic score derived therefrom) is diagnostic or indicative of the subject having breast cancer.

[0064] Treatment methods Further to the above, the methods described herein may improve outcomes for patients who could potentially benefit from treatment by diagnosing subjects with breast cancer.

[0065] Accordingly, the present inventors have developed a method of treating breast cancer in a subject.

[0066] In one broad aspect, the disclosure provides a method of treating breast cancer in a subject, the method comprising administering a treatment, e.g., surgery, and / or a therapeutically effective amount of an anti-cancer treatment, to a subject in which levels of one or more lipid biomarkers that are diagnostic or indicative of the subject having breast cancer have been measured in a biological sample from the subject, wherein the one or more lipid biomarkers are Cer(d36:1), Cer(d42:0), Cer(d4 2:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI (20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), P E(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(4 0:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d 42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0067] Suitably, the method includes an initial step of measuring the level of one or more lipid biomarkers in a biological sample from the subject.

[0068] Preferably, for this method, the risk or diagnostic score is determined using a level, such as the concentration level or expression level, of one or more lipid biomarkers. The risk or diagnostic score can be a diagnosis or indicator of a subject having breast cancer. In certain examples, the diagnostic score is generated at least in part through a logistic model. For this purpose, the diagnostic score can be in the form of a probability of the subject having breast cancer, such that, in the absence of additional information, a score of 50% or more indicates that the subject has a higher probability of having breast cancer than the probability of not having breast cancer. In some examples, a diagnostic score of 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or more is determined for the subject.

[0069] As used herein, the term "therapeutically effective amount" describes an amount of a particular agent (e.g., an anti-cancer agent) or treatment, such as chemotherapy, radiation therapy, molecular targeted therapy, and immunotherapy, that is sufficient to achieve a desired effect in a subject being treated with the agent. For example, this can be the amount of a composition containing one or more agents necessary to reduce, alleviate, and / or prevent cancer or a cancer-related disease, disorder, or condition. In some instances, a "therapeutically effective amount" is sufficient to reduce or eliminate symptoms of cancer. In other instances, a "therapeutically effective amount" is an amount sufficient to achieve a desired biological effect, e.g., an amount that is effective in reducing or preventing cancer growth and / or metastasis.

[0070] Ideally, a therapeutically effective amount of an agent is an amount sufficient to induce a desired result without causing substantial cytotoxic effects in the subject. The effective amount of an agent useful for reducing, alleviating, and / or preventing breast cancer will depend on the subject being treated, the type and severity of any associated disease, disorder, and / or condition (e.g., the number and location of any associated metastases), and the manner of administration of the therapeutic composition.

[0071] Preferably, the various drugs, anti-cancer agents, or cancer treatments described herein are administered to a subject as a pharmaceutical composition comprising a pharmaceutically acceptable carrier, diluent, or excipient. In this regard, any dosage form and route of administration, such as those provided therein, may be used to provide the compositions of the present disclosure to a subject.

[0072] "Pharmaceutically acceptable carrier, diluent, or excipient" refers to a solid or liquid filler, diluent, or encapsulating substance that can be safely used for systemic administration. Various carriers known in the art can be used depending on the particular route of administration. These carriers can be selected from the group including sugars, starches, cellulose and its derivatives, malt, gelatin, talc, calcium sulfate, liposomes and other lipid-based carriers, vegetable oils, synthetic oils, polyols, alginic acid, phosphate buffer solutions, emulsifiers, isotonic saline, inorganic acid salts including hydrochlorides, bromides, and sulfates, organic acids such as acetates, propionates, and malonates, and pyrogen-free water.

[0073] A useful reference describing pharmaceutically acceptable carriers, diluents, and excipients is Remington's Pharmaceutical Sciences (Mack Publishing Co. NJUSA, 1991), which is incorporated herein by reference.

[0074] Any safe route of administration may be used to provide the compositions of the present disclosure to a patient, such as oral, rectal, parenteral, sublingual, buccal, intravenous, intraarticular, intramuscular, intradermal, subcutaneous, inhalation, intraocular, intraperitoneal, intracerebroventricular, transdermal, etc.

[0075] Dosage forms include tablets, dispersions, suspensions, injectables, solutions, syrups, troches, capsules, suppositories, aerosols, transdermal patches, and the like. These dosage forms may also include injection or implantation of controlled-release devices specifically designed for this purpose, or other forms of implants modified to additionally act in this manner. Controlled release of therapeutic agents can be achieved, for example, by coating the same with hydrophobic polymers including acrylic resins, waxes, higher aliphatic alcohols, polylactic acid, and polyglycolic acid, and certain cellulose derivatives such as hydroxypropylmethylcellulose. Additionally, controlled release can be achieved by using other polymer matrices, liposomes, and / or microspheres.

[0076] Compositions of the present disclosure suitable for oral or parenteral administration may be presented as discrete units, such as capsules, sachets, or tablets, each containing a predetermined amount of one or more therapeutic agents of the present disclosure as a powder or granules, or as a solution or suspension in an aqueous liquid, a non-aqueous liquid, an oil-in-water emulsion, or a water-in-oil liquid emulsion. Such compositions may be prepared by any of the methods of pharmacy, which may include the step of bringing into association one or more of the agents described above with the carrier, which constitutes one or more necessary ingredients. In general, the compositions are prepared by uniformly and intimately admixing an agent of the present disclosure with a liquid carrier, or a finely divided solid carrier, or both, and then, if necessary, shaping the product into the desired presentation.

[0077] The compositions can be administered in a manner compatible with the dosage formulation and in an amount that is pharmaceutically effective.In the context of the present disclosure, the dosage administered to a patient should be sufficient to bring about a beneficial response in the patient over a reasonable period of time.The amount of the drug to be administered can depend on the subject to be treated, including the age, sex, weight, and general health of the subject to be treated, factors that depend on the judgment of the practitioner.

[0078] It is contemplated that the various drugs, anti-cancer agents, or cancer treatments described herein can be formulated as separate doses, such as in the form of a kit. Such kits may further include a package insert containing printed instructions for the simultaneous, combined, sequential, consecutive, alternating, or separate use of the drugs in treating, ameliorating, and / or preventing cancer as described herein in a patient in need thereof. Thus, the aforementioned kits are preferably for use in methods for treating, ameliorating, and / or preventing breast cancer, including one or more symptoms, consequences, sequelae, or complications thereof, as described herein.

[0079] Alternatively, the various therapeutic agents described herein can be formulated together in a composition that optionally includes a pharmaceutically acceptable carrier, excipient, or diluent.

[0080] Methods of treating breast cancer can be prophylactic, preventative, or therapeutic and can be suitable for treating breast cancer in mammals, particularly humans. As used herein, "treating," "treat," or "treatment" refers to a therapeutic intervention, course of action, or protocol that ameliorates at least the symptoms of cancer after the cancer and / or its symptoms have at least begun to develop. As used herein, "preventing," "prevent," or "prevention" refers to a therapeutic intervention, course of action, or protocol initiated prior to the onset of cancer and / or the onset of symptoms of cancer, such that the onset or progression of cancer or symptoms is prevented, inhibited, or delayed.

[0081] Anti-cancer treatment Those skilled in the art will understand that cancer treatments for use in the methods described herein can include, but are not limited to, drug therapy, chemotherapy, antibody, nucleic acid and other biomolecular therapy, radiation therapy, surgery, nutritional therapy, relaxation or meditation therapy, and other natural or holistic therapies. Generally, drugs, biomolecules (e.g., antibodies, inhibitory nucleic acids such as siRNA), or chemotherapeutic agents are referred to herein as "anti-cancer therapeutic agents" or "anti-cancer agents."

[0082] Preferably, the treatment is or includes one or more of surgery (eg, lumpectomy or mastectomy), chemotherapy, radiation therapy, molecular targeted therapy, and immunotherapy.

[0083] As generally used herein, the terms "chemotherapy" or "chemotherapeutic agent" refer broadly to treatments or drugs involving cytostatic or cytotoxic agents (i.e., compounds) to reduce or eliminate the growth or proliferation of unwanted cells, such as cancer cells. Thus, these terms can refer to cytotoxic or cytostatic agents used to treat proliferative disorders, e.g., cancer. The cytotoxic effect of a drug can be, but is not required to be, the result of one or more of nucleic acid intercalation or binding, DNA or RNA alkylation, inhibition of RNA or DNA synthesis, inhibition of another nucleic acid-associated activity (e.g., protein synthesis), or any other cytotoxic effect.

[0084] Exemplary chemotherapeutic agents include alkylating agents (e.g., nitrogen mustards such as chlorambucil, cyclophosphamide, isofamide, mechlorethamine, melphalan, and uracil mustard); aziridines such as thiotepa; methanesulfonate esters such as busulfan; nitrosoureas such as carmustine, lomustine, and streptozocin; platinum complexes such as cisplatin and carboplatin, oxaliplatin, nedaplatin, triplatin tetranitrate, phenanthriplatin, picoplatin, satraplatin, and lipoplatin; mitomycin, protease inhibitors, ribozymes ... bioreductive alkylating agents such as rocarbazine, dacarbazine, and altretamine; DNA strand breakers (e.g., bleomycin); topoisomerase II inhibitors (e.g., amsacrine, dactinomycin, daunorubicin, idarubicin, mitoxantrone, doxorubicin, etoposide, and teniposide); DNA minor groove binders (e.g., plicamidine); antimetabolites (e.g., folate antagonists such as methotrexate and trimetrexate; fluorouracil, fluorodeoxyuridine, CB3717, azacitidine, cytarabine, and flocc pyrimidine antagonists such as uridine; purine antagonists such as mercaptopurine, 6-thioguanine, fludarabine, pentostatin; asparginase; and ribonucleotide reductase inhibitors such as hydroxyurea; tubulin interacting agents (e.g., vincristine, vinblastine, and paclitaxel (Taxol)); hormonal agents (e.g., estrogens; conjugated estrogens; ethinyl estradiol; diethylstilbestrol; chlortrianisene; idenestrol; hydroxyprogesterone caproate, medicamentol, etc.) progestins such as roxyprogesterone, and megestrol; and androgens such as testosterone, testosterone propionate, fluoxymesterone, and methyltestosterone; adrenal corticosteroids (e.g., prednisone, dexamethasone, methylprednisolone, and prednisolone); leutinizing hormone-releasing agents or gonadotropin-releasing hormone agonists (e.g., leuprolide acetate and goserelin acetate); and antihormonal agents (e.g., antiandrogens such as tamoxifen, flutamide;These include, but are not limited to, aromatase inhibitors such as anastrozole, exemestane, and letrozole; and antiadrenal agents such as mitotane and aminoglutethimide.

[0085] As used herein, the term "radiation therapy" or "radiotherapy" refers to the medical use of ionizing radiation to control or destroy malignant cells, generally as part of cancer treatment. It can also be used as part of adjuvant therapy to prevent tumor recurrence after surgery to remove a primary malignant tumor.

[0086] Radiation therapy can be delivered by devices placed outside the patient's body (external radiation therapy) or sources placed inside the patient's body (internal radiation therapy or brachytherapy), or intravenously or orally. It can also be delivered by systemically delivered radioisotopes. Radiation therapy can be planned and administered in conjunction with imaging-based techniques such as computed tomography (CT) or magnetic resonance imaging (MRI) to precisely determine the dose and location of radiation to be administered. In various embodiments, radiation therapy includes systemic radiation therapy, conventional external beam radiation therapy, stereotactic radiosurgery, stereotactic radiation therapy, three-dimensional conformal radiation therapy, intensity-modulated radiation therapy (IMRT), image-guided radiation therapy, tomotherapy, and / or brachytherapy. In some examples, radiation therapy includes stereotactic radiation therapy or intensity-modulated radiation therapy (IMRT).

[0087] As used herein, "molecularly targeted therapy" or "molecularly targeted therapeutic agent" refers to a therapy that targets a specific class of proteins involved in cancer growth or signal transduction. In some examples, the additional anticancer agents described herein are or include inhibitors of tyrosine kinases. The term "tyrosine kinase" refers to an enzyme that can transfer a phosphate group from ATP to a tyrosine residue in a protein. Protein phosphorylation by tyrosine kinases is an important mechanism in signal transduction for the regulation of enzyme activity and cellular events, such as cell survival or proliferation. In particular examples, the molecular targeted therapy includes one or more of human epidermal growth factor receptor 2 (HER2; also referred to as ErbB-2, NEU, HER-2, and CD340) inhibitors (e.g., trastuzumab, pertuzumab, neratinib, tucatinib), PARP inhibitors (e.g., olaparib, talazoparib), CDK4 / 6 inhibitors (e.g., abemaciclib), PI3K inhibitors (e.g., alpelisib), dual HER2 / EGFR inhibitors (e.g., lapatinib), and neurotrophic T receptor kinase (NTRK) inhibitors (e.g., entrectinib, larotrectinib).

[0088] In the context of cancer, immunotherapy or immunotherapeutic agents utilize or modify the subject's immune system to promote or facilitate cancer treatment. In this context, immunotherapy or immunotherapeutic agents used to treat cancer include cell-based therapy, antibody therapy (e.g., anti-PD1, anti-PDL1, or anti-CTLA4 antibodies), and cytokine therapy. All of these therapies take advantage of the phenomenon that cancer cells often have subtly different molecules, called cancer antigens, on their surface that can be detected by the immune system of a cancer subject. Therefore, immunotherapy is used to induce the cancer patient's immune system to attack cancer cells by using these cancer antigens as targets.

[0089] Non-limiting examples of immunotherapies or immunotherapeutic agents include adalimumab, alemtuzumab, basiliximab, belimumab, bevacizumab, BMS-936559, brentuximab, certolizumab, cituximab, daclizumab, eculizumab, ibritumomab, infliximab, ipilimumab, lambro lkizumab), mepolizumab, MPDL3280A, muromonab, natalizumab, nivolumab, ofatumumab, omalizumab, pembrolizumab, pexelizumab, pidilizumab, rituximab, tocilizumab, tositumomab, trastuzumab, ustekinumab, abatacept, alefacept, and denileukin diftitox. In particular examples, the immunotherapeutic agent is an immune checkpoint inhibitor, for example, an anti-PD1 antibody (e.g., pidilizumab, nivolumab, lambrolkizumab, pembrolizumab), an anti-PDL1 antibody (e.g., BMS-936559, MPDL3280A), and / or an anti-CTLA4 antibody (e.g., ipilimumab).

[0090] Lipid biomarkers As described herein, the inventors have discovered that the concentration level of a specific lipid biomarker in a blood or plasma sample from a subject can be diagnostic for breast cancer. Without being bound by any theory, it is believed that the lipid biomarker is derived from extracellular vesicles, such as exosomes, present in the biological sample. It is also possible that other sources of lipid biomarkers, such as apolipoproteins or lipid droplets, co-isolated with extracellular vesicles may contribute.

[0091] As used herein, the term "lipid" includes bis(monoacylglycero)phosphate (BMP), cholesterol ester (CE), ceramide (Cer), diacylglycerol (DG or DAG), dihydroleukotriene B4 (DH-LTB4), fatty acid (FA), ganglioside A2 (GA2), ganglioside M3 (GM3), hexoseceramide (HexCer), dihexosylceramide (Hex2Cer), hexosyldihydroceramide (HexDHCer), lactosylceramide (LacCer), lysophosphatidic acid (LysoPA or LPA), lysophosphatidylcholine (LysoPC or LPC), lysophosphatidylcholine-plasmalogen (LysoPC-pmg), lysophosphatidylethanolamine (LysoPE or LPE), lysophosphatidylethanolamine-plasmalogen (LysoPC-pmg), lysophosphatidylethanolamine-plasmalogen (LysoPE ... It refers to a group of organic compounds with lipophilic or amphiphilic properties, including, but not limited to, smalogen (LysoPE-pmg), lysophosphatidylserine (LysoPS or LPS), lysophosphatidylinositol (LPI), monoacylglycerol (MAG), phosphatidylcholine (PC), phosphatidylcholine-plasmalogen (PC-pmg), phosphatidylethanolamine (PE), phosphatidylethanolamine-plasmalogen (PE-pmg), prostaglandin A1 (PGA1), prostaglandin B1 (PGB1), phosphatidylinositol (PI), phosphatidylserine (PS), sphingomyelin (SM), sphingosine, triacylglycerol (TG or TAG), and tetrahydro-12-keto-leukotriene B4 (TH-12-keto-LTB4).

[0092] As used herein, the term "biomarker" refers to a lipid molecule whose level is indicative or diagnostic of a subject having breast cancer. It will be understood that the term "biomarker" is intended to encompass all classes, forms (e.g., phosphorylated or oxidized forms), fragments (e.g., lipid head groups, lipid acyl chains), and variants of lipid biomarkers known in the art, including those provided herein. Also, unless otherwise specified, the ether-linked lipids (e.g., PC, PE, PS, etc.) described herein are intended to encompass both their alkyl ether and alkenyl ether forms (e.g., the alkyl and alkenyl ethers of PE(40:7e) include PE(O-40:7) and PE(P-40:6), respectively). For this purpose, the "O-" prefix is used to indicate the presence of an alkyl ether substituent, while the "P-" prefix or "p" suffix is used for alkenyl ether substituents.

[0093] For the methods, systems, and kits described herein, the one or more lipid biomarkers described herein can be selected from one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, etc.) classes of lipids, such as Cer, DG, Hex2Cer, LPC, LPE, LPI, PC, PE, PI, PS, SM, and TG.

[0094] In certain examples, the one or more lipid biomarkers include LPC(14:0) and / or PI(38:6). More particularly, the one or more lipid biomarkers may include LPC(14:0) and / or PI(18:2_20:4). In other examples, the one or more lipid biomarkers include LPC(14:0) and PI(38:6). More particularly, the one or more lipid biomarkers may include LPC(14:0) and PI(18:2_20:4).

[0095] In some examples, the one or more lipid biomarkers include LPC(14:0), or a fragment, variant, or derivative thereof. In such examples, the one or more lipid biomarkers are optionally selected from the group consisting of Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), P C(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), P E(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4) ), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM (d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54: The lipid biomarkers may further comprise one or more other lipid biomarkers selected from the group consisting of TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0096] In other examples, the one or more lipid biomarkers include PI(38:6), or a fragment, variant, or derivative thereof. For such examples, the one or more lipid biomarkers are optionally selected from the group consisting of Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), , PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PS(38: 4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM (d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54: The lipid biomarkers may further comprise one or more other lipid biomarkers selected from the group consisting of TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0097] In certain examples, the one or more lipid biomarkers include LPC(14:0) and PI(38:6), or fragments, variants, or derivatives thereof. In such examples, the one or more lipid biomarkers are optionally selected from the group consisting of Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1). , PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4 ), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PS(38:4), PS( 40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40 :3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), The composition may further comprise one or more other lipid biomarkers selected from the group consisting of TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0098] Preferably, the one or more lipid biomarkers are selected from the group consisting of LPC(14:0), PC(32:1), PC(38:5), PE(34:1), PS(38:4), SM(d36:2), SM(d38:4), TG(54:4), TG(56:1), and TG(58:2), or fragments, variants, or derivatives thereof. In various examples, the one or more lipid biomarkers comprise or consist of LPC(14:0), PC(32:1), PC(38:5), PE(34:1), PS(38:4), SM(d36:2), SM(d38:4), TG(54:4), TG(56:1), and TG(58:2), or fragments, variants, or derivatives thereof.

[0099] Suitably, the one or more lipid biomarkers described herein are selected from the group consisting of LPC(14:0), PC(32:1), PC(36:2), PC(38:5), PE(34:1), PI(34:1), PS(38:4), SM(d36:2), SM(d38:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(58:2), and TG(58:3), or fragments, variants, or derivatives thereof. In certain examples, the one or more lipid biomarkers comprise or consist of LPC(14:0), PC(32:1), PC(36:2), PC(38:5), PE(34:1), PI(34:1), PS(38:4), SM(d36:2), SM(d38:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(58:2), and TG(58:3), or fragments, variants, or derivatives thereof. More particularly, the one or more lipid biomarkers can be selected from the group consisting of LPC(14:0), PE(34:1), PS(38:4), SM(d36:2), TG(52:3e), and TG(58:3), or fragments, variants, or derivatives thereof. In some examples, the one or more lipid biomarkers comprise or consist of LPC(14:0), PE(34:1), PS(38:4), SM(d36:2), TG(52:3e), and TG(58:3), or fragments, variants, or derivatives thereof.

[0100] Preferably, the one or more lipid biomarkers described herein are selected from the group consisting of LPC(14:0), LPE(22:6), PI(38:6), PE(34:2p), SM(d35:1), PS(38:4), and PS(38:4), or fragments, variants, or derivatives thereof. In certain examples, the one or more lipid biomarkers comprise or consist of LPC(14:0), LPE(22:6), PI(38:6), PE(34:2p), SM(d35:1), PS(38:4), and PS(38:4), or fragments, variants, or derivatives thereof.

[0101] Suitably, the one or more lipid biomarkers are selected from the group consisting of Cer(d42:1), LPC(14:0), LPC(16:0e), LPC(17:1), LPE(22:6), PE(34:2p), PE(36:4), PE(38:3p), PI(36:1), PI(38:6), PS(38:4), PS(40:6), SM(d33:1), SM(d35:1), SM(d41:2), and TG(57:1), or fragments, variants, or derivatives thereof. In particular examples, the one or more lipid biomarkers comprise or consist of Cer(d42:1), LPC(14:0), LPC(16:0e), LPC(17:1), LPE(22:6), PE(34:2p), PE(36:4), PE(38:3p), PI(36:1), PI(38:6), PS(38:4), PS(40:6), SM(d33:1), SM(d35:1), SM(d41:2), and TG(57:1), or fragments, variants, or derivatives thereof. In some examples, the one or more lipid biomarkers are selected from the group consisting of LPC(14:0), LPC(16:0e), LPE(22:6), PE(34:2p), PE(38:3p), PI(36:1), PI(38:6), PS(38:4), PS(40:6), SM(d33:1), SM(d35:1), and TG(57:1), or fragments, variants, or derivatives thereof. In certain examples, the one or more lipid biomarkers comprise or consist of LPC(14:0), LPC(16:0e), LPE(22:6), PE(34:2p), PE(38:3p), PI(36:1), PI(38:6), PS(38:4), PS(40:6), SM(d33:1), SM(d35:1), and TG(57:1), or fragments, variants, or derivatives thereof.

[0102] In some examples, the one or more lipid biomarkers comprise or consist of LPC (14:0), PS (38:4), and TG (57:1), or fragments, variants, or derivatives thereof. In other examples, the one or more lipid biomarkers comprise or consist of LPC (16:0e), LPE (22:6), and PI (38:6), or fragments, variants, or derivatives thereof. In particular examples, the one or more lipid biomarkers comprise or consist of PE (34:2p), PE (38:3p), PI (36:1), PI (38:6), and PS (40:6), or fragments, variants, or derivatives thereof. In various examples, the one or more lipid biomarkers comprise or consist of LPC (14:0), PI (38:6), and SM (d33:1), or fragments, variants, or derivatives thereof. In certain examples, the one or more lipid biomarkers comprise or consist of LPC(14:0), PI(38:6), and SM(d35:1), or fragments, variants, or derivatives thereof. For the above example, the one or more lipid biomarkers can optionally further comprise one or more other lipid biomarkers provided herein, such as those selected from the group consisting of LPC(14:0), LPC(16:0e), LPE(22:6), PE(34:2p), PE(38:3p), PI(36:1), PI(38:6), PS(38:4), PS(40:6), SM(d33:1), SM(d35:1), and TG(57:1), or fragments, variants, or derivatives thereof.

[0103] Also provided are lipid "variants," such as naturally occurring variants, isobars, and isomers (including stereoisomers) of the lipid biomarkers provided herein. To this end, it is further envisioned that the lipid biomarkers described herein may encompass a collection of one or more isomers thereof. For example, PC(32:1) is a lipid or lipid biomarker that is a collection of one or more phosphatidylcholine isomers with 32 carbons in the acyl chain and one double bond in either of the two acyl chains. Exemplary isomers for the lipid biomarkers described herein are provided in the table below. Preferably, each of the lipid biomarker isomers has the same molecular weight. Although a lipid biomarker can encompass its total number of isomers, a biological sample from a subject may contain only one isomer, two isomers, three isomers, four isomers, five isomers, etc., or any number of isomers less than the total number of all possible isomers of the lipid biomarker. Thus, a lipid biomarker can refer to one or more of the isomers that make up the entire set of possible isomers. [Table 1-1] [Table 1-2] [Table 1-3] [Table 1-4] [Table 1-5] [Table 1-6] [Table 1-7] [Table 1-8] [Table 1-9]

Table 1-10

Table 1-11

Table 1-12

[0104] In certain examples, PE(O-40:6) includes PE(40:5p) and PE(40:6e). Because PE(O-40:6) encompasses both PE(40:5p) and PE(40:6e), for example, one or more lipid biomarkers described herein may include Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:0), LPC(19:1), LPC(20:0), LPC(21:0), LPC(22:0), LPC(23:0), LPC(24:0), LPC(25:0), LPC(26:0), LPC(27:0), LPC(28:0), LPC(29:0), LPC(30:0), LPC(31:0), LPC(32:0), LPC(33:0), LPC(34:0), LPC(35:0), LPC(36:0), LPC(37:0), LPC(38:0), LPC(39:0), LPC(36:0), LPC(39:0), LPC(38:0), LPC(39 ... :3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE( 34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5) (e.g. PE(38:4p) and PE(38:5e)), PE(40:5p), PE(40:6e), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33: 1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4 ), SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0105] In certain examples, PE(O-38:5) includes PE(38:4p) and PE(38:5e). Because PE(O-38:5) encompasses both PE(38:4p) and PE(38:5e), for example, one or more lipid biomarkers described herein may include Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:0e ... :3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE( 34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(38:4p), PE(38:5e)PE(O-40:6) (e.g. PE(40:5p) and PE (40:6e)), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1 ), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4) , SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0106] In another example, PE(34:2p) includes the isomer PE(34:3e), which may be used interchangeably herein.

[0107] It is also contemplated that the lipid biomarkers described herein may encompass or be interchangeable with one or more isobars thereof. The term "isobaric" typically refers to different lipids that have approximately or substantially the same mass (e.g., m / z ratio) and may not be distinguishable from one another on the analytical platform used to detect them (e.g., for mass spectrometry, different lipids in an isobaric may elute at approximately the same time and have similar or identical quantum ions, and therefore cannot be distinguished).

[0108] Lipids can be defined according to the following formula: XXX(YY:ZZ), where XXX is an abbreviation for the lipid class or group (often referring to the lipid head group), YY is the number of carbons in the acyl chain, and ZZ is the number of double bonds in the acyl chain. Similar notation (e.g., XXX(YY1:ZZ1_YY2:ZZ2) or XXX(YY1:ZZ1_YY2:ZZ2_YY3_ZZ3) may be used to define lipid isomers, where the numbers refer to the specific acyl chains of the lipid. However, it is contemplated that the lipids defined herein may be identified by different naming annotations or nomenclature known in the art (see, e.g., Liebisch et al., J Lipid Res, 2013 Jun;54(6):1523-1530; Lipidomics Standards Initiative Consortium, Nat Metab, 2019 Aug;1(8):745-747).

[0109] It is also contemplated that the listed lipid biomarkers may additionally cover one or more additional lipid biomarkers that behave similarly or equivalently to the lipid biomarker in question (e.g., exhibit a similar concentration profile). To this end, a lipid biomarker may exhibit substantial collinearity with one or more additional lipid biomarkers, for example, in that they are diagnostic or indicative of breast cancer in a subject. Collinearity refers to a strong correlation or linear relationship between a pair of predictors (e.g., a pair of lipid biomarkers), and collinearity between multiple predictors is referred to as multicollinearity. Thus, in some examples, the one or more lipid biomarkers include one or more additional lipid biomarkers, such as those outlined in Examples 1-3 below, that exhibit collinearity with one or more of the one or more lipid biomarkers described in the examples provided herein. In other examples, the lipid biomarkers exhibit little or no collinearity with one or more additional lipid biomarkers.

[0110] Also provided are fragments of lipid biomarkers comprising lipid head groups and acyl chains or fragments thereof, which comprise less than 100% of the entire lipid biomarker molecule. In this regard, those skilled in the art will understand that MRM analysis of lipid biomarkers by mass spectrometry can involve fragmenting lipids into their component parts (e.g., lipid head groups and one or more acyl chains) to aid in the identification and quantification of the lipid biomarkers, as described in more detail below.

[0111] High-resolution accurate-mass MS (HRMS) can be used to perform reliable and sensitive quantitative analysis of lipid biomarkers, similar to that of MRM (see Rochat, Trends in Analytical Chemistry, 2016, for a review). Parallel reaction monitoring (PRM) is an ion monitoring technique based on high-resolution and high-accuracy mass spectrometry. PRM can be based, for example, on the Q Exactive Orbitrap™ (Thermo Scientific™) system or the Sciex 7500 (Sciex™) system as representative quadrupole high-resolution mass spectral platforms. Unlike MRM, which monitors specific transitions at a time, the high resolution and mass accuracy of PRM's full scan (MS1) and tandem mass spectrometry (MS / MS) scans can provide sufficient selectivity by monitoring all MS / MS fragment ions for each target precursor lipid.

[0112] Suitably, the levels (e.g., concentrations or expression levels) of two or more of the lipid biomarkers provided herein (e.g., 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, or 55 lipid biomarkers) are determined using the methods described herein. In some examples, the methods described herein include determining the levels or concentrations of three or more lipid biomarkers described herein. In other examples, the methods described herein include determining the levels or concentrations of four or more lipid biomarkers described herein. In certain examples, the methods described herein include determining the levels or concentrations of five or more lipid biomarkers described herein. In some examples, the methods described herein include determining the levels or concentrations of six or more lipid biomarkers described herein. In various examples, the methods described herein include determining the levels or concentrations of seven or more lipid biomarkers described herein. In particular examples, the methods described herein include determining the levels or concentrations of eight or more lipid biomarkers described herein. In some examples, the methods described herein include determining the levels or concentrations of nine or more lipid biomarkers described herein. In other examples, the methods described herein include determining the levels or concentrations of ten or more lipid biomarkers described herein. In certain examples, the methods described herein include determining the levels or concentrations of eleven or more lipid biomarkers described herein. In some examples, the methods described herein include determining the levels or concentrations of twelve or more lipid biomarkers described herein. In various examples, the methods described herein include determining the levels or concentrations of thirteen or more lipid biomarkers described herein.In particular examples, the methods described herein include determining the levels or concentrations of 14 or more lipid biomarkers described herein. In various examples, the methods described herein include determining the levels or concentrations of 15 or more lipid biomarkers described herein. In particular examples, the methods described herein include determining the levels or concentrations of 16 or more lipid biomarkers described herein. In some examples, the methods described herein include determining the levels or concentrations of 17 or more lipid biomarkers described herein. In certain examples, the methods described herein include determining the levels or concentrations of 18 or more lipid biomarkers described herein. In particular examples, the methods described herein include determining the levels or concentrations of 19 or more lipid biomarkers described herein. In various examples, the methods described herein include determining the levels or concentrations of 20 or more lipid biomarkers described herein.

[0113] In one example, the methods of the disclosure include measuring Cer(d36:1) as well as at least one additional lipid biomarker described herein (e.g., Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPC(19:4), LPC(20:4), LPC(21:4), LPC(22:4), LPC(23:4), LPC(24:4), LPC(25:4), LPC(26:4), LPC(27:4), LPC(28:4), LPC(29:4), LPC(30:4), LPC(31:4), LPC(32:4), LPC(33:4), LPC(34:4), LPC(35:4), LPC(36:4), LPC(37:4), LPC(38:4), LPC(39:4), LPC(40:4), LPC(41:4), LPC(42:4), LPC(43:4), LPC(44:4), LPC(45:4), LPC(46:4), LPC(47:4), LPC(48:4), LPC(49 ...9:4), LPC(49:4), LPC(49: E(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE (34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1) , PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36: 1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), T The method includes determining the levels of TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), and fragments, variants, or derivatives thereof.

[0114] In one example, the methods of the disclosure include determining the level of Cer(d42:0) and at least one additional lipid biomarker described herein.

[0115] In one example, the methods of the disclosure include determining the level of Cer(d42:1) and at least one additional lipid biomarker described herein.

[0116] In one example, a method of the disclosure includes determining the level of DG(34:2) and at least one additional lipid biomarker described herein.

[0117] In one example, the methods of the disclosure include determining the levels of Hex2Cer (d34:1) and at least one additional lipid biomarker described herein.

[0118] In one example, the methods of the disclosure include determining the levels of Hex2Cer (d42:2) and at least one additional lipid biomarker described herein.

[0119] In one example, a method of the disclosure includes determining the level of LPC(14:0) and at least one additional lipid biomarker described herein.

[0120] In one example, the methods of the disclosure include determining the levels of LPC(16:0e) and at least one additional lipid biomarker described herein.

[0121] In one example, a method of the disclosure includes determining the level of LPC(17:1) and at least one additional lipid biomarker described herein.

[0122] In one example, a method of the disclosure includes determining the level of LPC(18:3) and at least one additional lipid biomarker described herein.

[0123] In one example, a method of the disclosure includes determining the level of LPE(22:6) and at least one additional lipid biomarker described herein.

[0124] In one example, a method of the disclosure includes determining the level of LPI(20:4) and at least one additional lipid biomarker described herein.

[0125] In one example, a method of the disclosure includes determining the level of PC(32:1) and at least one additional lipid biomarker described herein.

[0126] In one example, a method of the disclosure includes determining the level of PC(34:1) and at least one additional lipid biomarker described herein.

[0127] In one example, a method of the disclosure includes determining the level of PC(35:4) and at least one additional lipid biomarker described herein.

[0128] In one example, a method of the disclosure includes determining the level of PC(36:2) and at least one additional lipid biomarker described herein.

[0129] In one example, a method of the present disclosure includes determining the level of PC(36:3) and at least one additional lipid biomarker described herein.

[0130] In one example, a method of the disclosure includes determining the level of PC(37:4) and at least one additional lipid biomarker described herein.

[0131] In one example, a method of the disclosure includes determining the level of PC(38:5) and at least one additional lipid biomarker described herein.

[0132] In one example, a method of the disclosure includes determining the level of PE(34:1) and at least one additional lipid biomarker described herein.

[0133] In one example, a method of the disclosure includes determining the level of PE(34:2p) and at least one additional lipid biomarker described herein.

[0134] In one example, a method of the disclosure includes determining the level of PE(36:4) and at least one additional lipid biomarker described herein.

[0135] In one example, the methods of the disclosure include determining the level of PE(38:3p) and at least one additional lipid biomarker described herein.

[0136] In one example, a method of the disclosure includes determining the level of PE(38:4) and at least one additional lipid biomarker described herein.

[0137] In one example, the disclosed method includes determining the level of PE(O-38:5) (e.g., PE(38:4p) and / or PE(38:5e)) and at least one additional lipid biomarker described herein. More particularly, the disclosed method can include determining the level of PE(38:4p) and at least one additional lipid biomarker described herein. More particularly, the disclosed method can include determining the level of PE(38:5e) and at least one additional lipid biomarker described herein.

[0138] In one example, the disclosed method includes determining the level of PE(O-40:6) (e.g., PE(40:5p) and / or PE(40:6e)) and at least one additional lipid biomarker described herein. More particularly, the disclosed method can include determining the level of PE(40:5p) and at least one additional lipid biomarker described herein. More particularly, the disclosed method can include determining the level of PE(40:6e) and at least one additional lipid biomarker described herein.

[0139] In one example, the methods of the present disclosure include determining the level of PI(32:1) and at least one additional lipid biomarker described herein.

[0140] In one example, the methods of the present disclosure include determining the level of PI(34:1) and at least one additional lipid biomarker described herein.

[0141] In one example, the methods of the present disclosure include determining the levels of PI(36:1) and at least one additional lipid biomarker described herein.

[0142] In one example, the methods of the disclosure include determining the levels of PI(38:6) and at least one additional lipid biomarker described herein.

[0143] In one example, a method of the disclosure includes determining the level of PS(38:4) and at least one additional lipid biomarker described herein.

[0144] In one example, a method of the disclosure includes determining the level of PS(40:6) and at least one additional lipid biomarker described herein.

[0145] In one example, a method of the disclosure includes determining the level of PS(40:7) and at least one additional lipid biomarker described herein.

[0146] In one example, the methods of the disclosure include determining the level of SM(d32:2) and at least one additional lipid biomarker described herein.

[0147] In one example, the methods of the disclosure include determining the level of SM(d33:1) and at least one additional lipid biomarker described herein.

[0148] In one example, the methods of the disclosure include determining the level of SM(d35:1) and at least one additional lipid biomarker described herein.

[0149] In one example, the methods of the disclosure include determining the level of SM(d36:1) and at least one additional lipid biomarker described herein.

[0150] In one example, the methods of the disclosure include determining the level of SM(d36:2) and at least one additional lipid biomarker described herein.

[0151] In one example, the methods of the disclosure include determining the levels of SM(d38:4) and at least one additional lipid biomarker described herein.

[0152] In one example, the methods of the disclosure include determining the level of SM(d40:3) and at least one additional lipid biomarker described herein.

[0153] In one example, the methods of the disclosure include determining the level of SM(d41:2) and at least one additional lipid biomarker described herein.

[0154] In one example, the methods of the disclosure include determining the level of SM(d41:3) and at least one additional lipid biomarker described herein.

[0155] In one example, the methods of the disclosure include determining the level of SM(d42:1) and at least one additional lipid biomarker described herein.

[0156] In one example, the methods of the disclosure include determining the level of SM(d42:4) and at least one additional lipid biomarker described herein.

[0157] In one example, the methods of the disclosure include determining the level of SM(d44:4) and at least one additional lipid biomarker described herein.

[0158] In one example, the methods of the disclosure include determining the levels of TG(52:3e) and at least one additional lipid biomarker described herein.

[0159] In one example, a method of the disclosure includes determining the level of TG(53:4) and at least one additional lipid biomarker described herein.

[0160] In one example, a method of the disclosure includes determining the level of TG(54:3) and at least one additional lipid biomarker described herein.

[0161] In one example, a method of the disclosure includes determining the level of TG(54:4) and at least one additional lipid biomarker described herein.

[0162] In one example, a method of the disclosure includes determining the level of TG(54:5) and at least one additional lipid biomarker described herein.

[0163] In one example, a method of the disclosure includes determining the level of TG(54:6) and at least one additional lipid biomarker described herein.

[0164] In one example, a method of the disclosure includes determining the level of TG(56:1) and at least one additional lipid biomarker described herein.

[0165] In one example, a method of the disclosure includes determining the level of TG(57:1) and at least one additional lipid biomarker described herein.

[0166] In one example, a method of the disclosure includes determining the level of TG(58:1) and at least one additional lipid biomarker described herein.

[0167] In one example, a method of the disclosure includes determining the level of TG(58:2) and at least one additional lipid biomarker described herein.

[0168] In one example, a method of the disclosure includes determining the level of TG(58:3) and at least one additional lipid biomarker described herein.

[0169] In one example, a method of the disclosure includes determining the level of TG(59:2) and at least one additional lipid biomarker described herein.

[0170] In one example, the methods of the disclosure include determining the levels of TG(60:1) and at least one additional lipid biomarker described herein.

[0171] In one example, a method of the disclosure includes determining the level of TG(62:2) and at least one additional lipid biomarker described herein.

[0172] None of the methods disclosed herein may involve measuring any other biomarkers, and therefore the methods disclosed herein may involve excluding any other biomarkers from the analysis.

[0173] In certain examples, the methods disclosed herein do not include measuring levels of PC(32:1), PC(38:5), and / or TG(54:4). To this end, the one or more lipid biomarkers may not include PC(32:1), PC(38:5), and / or TG(54:4), or fragments, variants, or derivatives thereof.

[0174] In some examples, the one or more lipid biomarkers are Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI( 20:4), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PE(34:1), PE(34:2p), PE(36: 4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38: 6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36 :2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG (52:3e), TG(53:4), TG(54:3), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0175] In some examples, the one or more lipid biomarkers may not include Cer(d36:1), LPC(16:0e), PC(32:1), PC(34:1), PC(36:2), PC(36:3), PC(38:5), PE(38:3p), PE(38:4), PE(O-38:5), SM(d36:1), SM(d42:1), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), and / or TG(54:6), or fragments, variants, or derivatives thereof.

[0176] In some examples, the one or more lipid biomarkers are Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(35:4), PC(37:4), PE(34:1), PE(34:2p), PE(36:4), PE(40:5p), PE(40:6e), PI(32:1), PI(34:1), PI(36:1), PI(38:6). , PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:4), SM(d44:4), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or a fragment, variant, or derivative thereof.

[0177] Determination of lipid biomarker levels It will be understood by those skilled in the art that the expression level, abundance, or concentration of one or more lipid biomarkers can be determined by any means known in the art. The terms "determining," "measuring," "evaluating," "assessing," "quantifying," "calculating," and "assaying" are used interchangeably herein and can include any form of measurement known in the art, such as those described below. Such determination can include detecting the presence or absence and / or determining the concentration level of one or more of the lipid biomarkers in a biological sample obtained from a subject.

[0178] Suitable means for determining the concentration or level of expression of one or more lipid biomarkers include, but are not limited to, nuclear magnetic resonance (NMR) spectroscopy, surface plasmon resonance (SPR), chromatographic techniques, mass spectrometry, biosensors, and any combination of these techniques.

[0179] In certain instances, the concentration or expression level of one or more lipid biomarkers is measured by mass spectrometry. Mass spectrometry (MS) is an analytical technique that measures the mass-to-charge (m / z) ratio of charged particles. It is primarily used to determine the elemental composition of a sample or molecule and to elucidate the chemical structure of molecules such as peptides, lipids, and other chemical compounds. MS works by ionizing chemical compounds to produce charged molecules or molecular fragments and measuring their mass-to-charge ratios. MS instruments typically consist of three modules: (1) an ion source, which can convert gas-phase sample molecules into ions (or, in the case of electrospray ionization, transfer ions present in solution to the gas phase); (2) a mass analyzer, which selects ions by their mass by applying an electromagnetic field; and (3) a detector, which measures the value of an indicator quantity and thus provides data for calculating the abundance of each ion present.

[0180] Suitable mass spectrometry methods to be used with the present disclosure include electrospray ionization mass spectrometry (ESI-MS), ESI-MS / MS, ESI-MS / (MS)n, matrix-assisted laser desorption / ionization time-of-flight mass spectrometry (MALDI-TOF-MS), surface-enhanced laser desorption / ionization time-of-flight mass spectrometry (SELDI-TOF-MS), tandem liquid chromatography-mass spectrometry (LC-MS / MS) mass spectrometry, and on-silicon desorption / ionization (DIOS). ), secondary ion mass spectrometry (SIMS), quadrupole time-of-flight (Q-TOF), atmospheric pressure chemical ionization mass spectrometry (APCI-MS), APCI-MS / MS, APCI-(MS)n, atmospheric pressure photoionization mass spectrometry (APPI-MS), APPI-MS / MS, and APPI-(MS)n, quadrupole mass spectrometry, Fourier transform mass spectrometry (FTMS), and ion trap mass spectrometry, where n is an integer greater than zero. In certain examples, the concentration or expression level of one or more lipid biomarkers is determined, at least in part, by using liquid chromatography-mass spectrometry (LC-MS).

[0181] As mentioned above, MS ionizes lipids and selects ions based on their mass-to-charge ratio. It has been widely used to characterize lipids, especially with the development of soft ionization techniques such as electrospray ionization (ESI) and matrix-assisted laser desorption ionization (MALDI). Lipid extraction is typically the first step for lipid analysis, separating the lipid component (organic phase) from other components such as proteins and nucleic acids (aqueous phase). However, some examples utilize single-phase lipid extraction.

[0182] Extraction methods typically involve the application of a mixture of methanol, chloroform, and water for phase separation. However, shotgun lipidomic methods have also been developed that omit the chromatographic separation and sample processing described above, analyze all lipid classes together, and instead use ionizing additives to help distinguish between specific lipids. This MS method involves lipid digestion, fragmentation, or denaturation, followed by LC-MS or LC-MS / MS (tandem MS), which can derive the mass-to-charge ratios for the specific lipid head groups and / or acyl chains that make up the lipid biomarkers described herein.

[0183] Preferably, the one or more lipid biomarkers or one or more fragments thereof are then subjected to quantitative mass spectrometry, including but not limited to selected reaction monitoring mass spectrometry (SRM), high-resolution data independent analysis (SWATH), multiple reaction monitoring (MRM), and / or MSI-based quantification.

[0184] In one particular example, an MRM assay is used that uses specific lipids and their fragments (transitions) as discriminators for individual lipid biomarkers.

[0185] In a specific example, the MS method is carried out in positive ion mode and / or negative ion mode. Generally, lipids can form small cation adducts in positive ion mode due to the ionization process. The formation of cation adducts of lipid molecular species, resulting from the affinity of cations with the dipoles present in lipid species, depends on the availability of small cations. For example, such adducts can be formed by H + , NH4 + , Li + , Na + , K. + , and (-H20+H) +In negative ion mode, lipid species in deprotonated form or with anionic adducts are displayed depending on whether the lipid molecular species has a net ionic bond. For example, PE, PI, PS, PA, and PG are all of the acidic lipid class (i.e., ionic bonds are present) and therefore can be detected as deprotonated ions. Some lipids are of the polar lipid class with no ionizable bonds, or PC and SM are of the strongly zwitterionic lipid class, all of which may contain small anions (e.g., Cl) depending on the concentrations present and their affinity with these lipid species. - , CH3COO - , and HCOO - ) can be formed as an anionic adduct with

[0186] In some examples, one or more lipid biomarkers described herein include an ion mode as described in Tables 8-26. Preferably, the ion mode is specific to that method of MS described in each example.

[0187] In other examples, one or more lipid biomarkers described herein have retention or elution times at or about those set forth in Tables 8 through 26. Preferably, retention times are determined according to the methods of MS described in the respective Examples.

[0188] In certain examples, one or more lipid biomarkers described herein have the exact mass, neutral mass, or mass-to-charge ratio (m / z) of, or approximately the exact mass, neutral mass, or mass-to-charge ratio (m / z) of, those set forth in the isomer table above or in Tables 8-23. Preferably, the neutral mass is determined according to the method of MS described in each example.

[0189] In some applications, various ionization techniques can be coupled to the mass spectrometers provided herein to generate desired information. Non-limiting exemplary ionization techniques that can be used with the present disclosure include, but are not limited to, matrix-assisted laser desorption ionization (MALDI), desorption electrospray ionization (DESI), direct assisted real-time (DART), surface-assisted laser desorption ionization (SALDI), or electrospray ionization (ESI).

[0190] In some applications, HPLC and UHPLC can be coupled to mass spectrometer, so that some other lipid separation techniques can be carried out before mass spectrometry analysis.Some exemplary separation techniques that can be used to separate desired analytes (e.g., lipids) from matrix background include, but are not limited to, reversed-phase liquid chromatography (RP-LC) of lipids, off-line liquid chromatography (LC) before MALDI, first-dimensional gel separation, second-dimensional gel separation, strong cation exchange (SCX) chromatography, strong anion exchange (SAX) chromatography, weak cation exchange (WCX) and weak anion exchange (WAX).One or more of the above techniques can be used before mass spectrometry analysis.

[0191] In some instances, the expression or concentration of a lipid biomarker is higher in a subject compared to a reference value determined from a control; however, for certain lipid biomarkers, the expression or concentration of that biomarker is decreased relative to the reference value from a control.

[0192] Preferably, an increased level of expression or concentration of the first subset of one or more lipid biomarkers is indicative of or correlates with a subject having breast cancer; and / or a decreased level of expression or concentration of one or more lipid biomarkers of the second subset (e.g., not present in the first subset of one or more lipid biomarkers) is indicative of or correlates with a subject having breast cancer. In other examples, a decreased level of concentration or expression of the first subset of one or more lipid biomarkers is indicative of or correlates with a subject not having breast cancer; and / or an increased level of expression or concentration of the second subset of one or more lipid biomarkers (e.g., not present in the first subset of one or more lipid biomarkers) is indicative of or correlates with a subject not having breast cancer.

[0193] As will be understood by those skilled in the art, the level or expression level of any one of the lipid biomarkers described herein may be (i) higher, increased, or greater than the expression level in a control or reference sample, or (ii) lower, decreased, or reduced relative to the expression level in a control or reference sample, or a threshold expression level. In various examples, an expression level can be classified as higher, increased, or greater if it exceeds the average and / or median expression level of the reference population. In some examples, an expression level can be classified as lower, decreased, or reduced if it is lower than the average and / or median expression level of the reference population. In this regard, the reference population can be a group of subjects with breast cancer. Alternatively, the reference population can be a group of subjects known to be cancer-free.

[0194] As used herein, terms such as "higher," "increased," and "greater than" refer to an elevated amount or level of a lipid biomarker, such as in a biological sample, when compared to a control or reference level or amount. The concentration or expression level of a lipid biomarker may be relative or absolute (i.e., relatively or absolutely higher, increased, or greater). In some examples, a level of a lipid biomarker is higher, increased, or greater if its concentration or expression level is greater than or about 0.5%, 1%, 2%, 3%, 4%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 100%, 150%, 200%, 300%, 400%, or at least about 500% greater than the concentration or expression level of the lipid biomarker in the control or reference level or amount.

[0195] As used herein, the terms "lower," "reduced," and "decreased" refer to a lower amount or level of a lipid biomarker, such as in a biological sample, when compared to a control or reference level or amount. The concentration or expression level of a lipid biomarker may be relative or absolute (i.e., relatively or absolutely lower, reduced, or decreased). In some examples, the concentration or expression of a lipid biomarker is lower, reduced, or decreased if the concentration or expression level is less than about 95%, 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, or 10%, or even less than about 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.01%, 0.001%, or 0.0001% of the concentration or expression level or amount of the lipid biomarker in the control or reference level or amount.

[0196] The term "control sample" typically refers to a biological sample from a (healthy) non-diseased individual who does not have cancer, or more particularly does not have breast cancer. In some examples, the control sample may be from a subject who is known to be cancer-free, or more particularly, to be breast cancer-free. Alternatively, the control sample may be from a subject in remission from cancer. The control sample may be a pooled sample, an average sample, or an individual sample. An internal control is a marker from the same biological sample being tested.

[0197] In some instances, the reference level or amount is determined from measurements of the biomarkers in a corresponding panel of biomarkers from a population of healthy individuals. As used herein, the term "healthy individual" refers to a person or population of people known to be free of breast cancer. In some instances, the control or reference level is determined from measurements of the corresponding biomarkers in a "typical population." Preferably, the "typical population" represents various breast cancers at different stages of disease progression. It is particularly preferred that the "typical population" represents the expression characteristics of a cohort of subjects, such as female subjects, as described herein.

[0198] In another example, the reference level or amount may be derived from an established data set that includes one or more of the following: 1. A dataset comprising measurements of lipid biomarkers for a subject or population of subjects known to have breast cancer; 2. A dataset containing lipid biomarker measurements for the subject being tested, where the measurements were previously generated, such as when the subject was known to be healthy (i.e., free of breast cancer), and / or 3. A dataset comprising lipid biomarker measurements for a healthy individual or population of healthy individuals.

[0199] In certain examples, the dataset including measurements of lipid biomarkers may be obtained from a population of subjects known to have breast cancer, healthy individuals, or a population of healthy individuals in a fasting state, a non-fasting state, or a combination thereof.

[0200] As used herein, concentration or expression level can be the absolute or relative amount of expressed lipid.Therefore, in some examples, the concentration or expression level of any one of one or more lipid biomarkers is compared with the control level of concentration or expression, such as the lipid concentration or expression level of one or more " housekeeping " lipids or molecules in the biological sample of the subject.

[0201] In a further example, the concentration or expression level of any one of one or more lipid biomarkers is compared to a threshold concentration or expression level, such as the lipid biomarker concentration or expression level in a biological sample from a control subject without breast cancer and / or the mean or median level of lipid biomarker concentration or expression in a biological sample from a population of breast cancer patients. The threshold concentration or expression level is generally a quantified level of the lipid biomarker concentration or expression. Typically, a concentration or expression level of a lipid biomarker in a sample that exceeds or falls below the threshold concentration or expression level is predictive of a particular disease state or outcome, such as the presence or absence of breast cancer. The nature and value (if any) of the threshold concentration or expression level typically vary based on, for example, the method selected to determine the concentration or expression of one or more lipid biomarkers used in determining a breast cancer diagnosis in a subject.

[0202] Those skilled in the art can use any method for measuring lipid biomarker concentration, abundance, or expression known in the art, such as those described herein, to determine the threshold level of any one of one or more lipid biomarkers in a sample that can be used, for example, to determine the presence or absence of breast cancer in a relevant subject.In various examples, the threshold level is, for example, the average and / or median concentration or expression level (median or absolute) of the lipid biomarker in a reference population with or without breast cancer.In addition, the concept of a threshold level of concentration or expression should not be limited to a single value or result.In this regard, the threshold level of concentration or expression can include multiple threshold concentration or expression levels, which can represent, for example, a high, intermediate, or low probability of a subject having breast cancer.

[0203] In view of the foregoing, any of the methods disclosed herein may include establishing a reference or threshold level of concentration or expression of one or more lipid biomarkers.

[0204] Preferably, the predictive accuracy of the methods described herein, as determined by ROC AUC value, is at least about 0.65 (e.g., at least about 0.65, 0.66, 0.67, 0.68, 0.69, 0.70, 0.71, 0.72, 0.73, 0.74, 0.75, 0.76, 0.77, 0.78, 0.79, 0.80, 0.81, 0.82, 0.83, 0.84, 0.85, 0.86, 0.87, 0.88, 0.89, 0.90, 0.91, 0.92, 0.93, 0.94, 0.95, 0.96, 0.97, 0.98, 0.99, or any range therein). More particularly, the predictive accuracy of the described methods is preferably at least about 0.70. Even more particularly, the predictive accuracy of the described methods is preferably at least about 0.75. Even more particularly, the predictive accuracy of the described methods is preferably at least about 0.80. Even more particularly, the predictive accuracy of the described methods is preferably at least about 0.85. Even even more particularly, the predictive accuracy of the described methods is preferably at least about 0.90.

[0205] Calculating a risk or diagnostic score For the methods described herein, determining the presence or absence of breast cancer in a subject may include calculating a risk score or a diagnostic score.

[0206] The term "risk score" or "disease risk score" refers to a calculated value of one or more feature values or scores that indicates an undesirable physiological condition in a patient, such as the presence of cancer. In certain instances, the term "risk score" refers to a numerical representation of the current degree of risk or probability that a patient has for having a particular disease or condition.

[0207] A risk score can be calculated using the concentration or expression levels or expression signatures of one or more lipid biomarkers, such as in a panel (e.g., 2, 3, 4, 5, etc., or more) of diagnostic lipid biomarkers, including those described above. To this end, the methods described herein include obtaining a risk score for a combination of lipid biomarkers described above or described in any of Examples 1 and 2 (e.g., Tables 8-23). The lipid concentration or expression signature can be determined using the independent diagnostic value of the lipid biomarker based on the normalized level of the lipid biomarker concentration or expression in the sample and the correlation of the lipid biomarker concentration or expression with the presence or absence of disease. Any method for determining a concentration or expression signature for a lipid biomarker known in the art can be utilized. After determining the concentration or expression levels or expression signatures of individual lipid biomarkers, such as in a panel of two or more of the lipid biomarkers described herein, a risk score can be calculated by combining the concentration or expression levels and / or expression signatures of each lipid biomarker in the panel. Methods for calculating the risk score can be as described in the Examples.

[0208] In certain examples, the risk score is calculated at least in part by logistic regression.For example, the linear combination of the concentration or expression level of one or more lipid biomarkers with various coefficients determined through previous training can be generated, and then used to estimate the log-odds of cancer.The log-odds can then be converted into the probability of a subject having breast cancer through logistic regression.In other examples, the risk score is calculated at least in part by partial least squares discriminant analysis.

[0209] Thus, a risk score for a patient may be calculated according to the following formula: Probability of cancer = e I+Σci*Li / (1+e I+Σci*Li )

[0210] where intercept I and coefficient c i are specific logistic regression model parameters pre-calculated based on the training data, and the values L i represents the normalized lipid abundance measured for each patient sample for each lipid in the panel.

[0211] The calculated risk score of the present disclosure can be used to determine the likelihood of the presence or absence of breast cancer in a subject. Generally, the calculated risk score can be compared to a reference risk score. In certain examples, (i) if the risk score is equal to or higher than the reference risk score, the subject has breast cancer; (ii) if the risk score is lower than the reference risk score, the subject does not have breast cancer. It is envisioned that a subject's diagnosis and / or risk score can be used to determine whether the subject should be treated with an anti-cancer drug. Thus, in other examples, (i) if the risk score is equal to or higher than the reference risk score, the subject should be given anti-cancer treatment; and (ii) if the risk score is lower than the reference risk score, the subject should not be given anti-cancer treatment.

[0212] In some examples, the risk score is compared with a threshold risk score, such as a median or average risk score, to determine the diagnosis of breast cancer in the subject.If the risk score is equal to or higher than the threshold risk score, the subject preferably has breast cancer.Alternatively, if the risk score is lower than the threshold risk score, the subject preferably does not have breast cancer.The threshold risk score can be the median or average of the risk scores calculated for each subject in a population of subjects with breast cancer, and / or the median or average of the risk scores calculated for each subject in a population of subjects without breast cancer, such as the subjects described in Example 1.

[0213] kit The present disclosure also contemplates kits for the detection of lipid biomarkers that may be suitable for use in the methods described herein.

[0214] In one broad aspect, the disclosure provides a kit for determining the presence or absence of breast cancer in a subject, the kit including one or more reagents for determining the level of one or more lipid biomarkers in a biological sample obtained from the subject, the one or more lipid biomarkers being Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer( d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35: 4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38 :5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d 33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42: 4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0215] Any agent or probe capable of specifically binding to a biomarker gene product, such as a set of primers, a labeled nucleic acid probe, an aptamer, an antibody, and / or an antibody fragment, is useful. Other components of the kit typically include labels, secondary antibodies, inhibitors, cofactors, and control lipid product preparations to allow the user to quantify the concentration or expression level and / or evaluate whether the measurement is functioning correctly. Biosensors, including optical (e.g., SPR-based sensors, interferometry-based sensors, waveguide-based sensors), electrochemical, and mechanical biosensors, are particularly suitable assays that can be easily performed by those skilled in the art using the kit components.

[0216] In some examples, the kit may include a substrate, such as a microtiter plate, on which are immobilized capture probes or antibodies corresponding to the lipid biomarkers to be measured.

[0217] In some examples, the kits include beads having immobilized capture probes or antibodies thereon that correspond to the lipid biomarkers being measured.

[0218] Optionally, the kit further comprises means for detecting binding of the probe, such as an antibody, to the lipid biomarker. Such means include, for example, a reporter molecule, such as an enzyme (such as horseradish peroxidase or alkaline phosphatase), a dye, a radionucleotide, a luminescent group, a chemiluminescent group, a fluorescent group, biotin, or a colloidal particle, such as colloidal gold or selenium. Preferably, such a reporter molecule is directly linked to the antibody.

[0219] In one example, the kit can additionally comprise a reference sample.Preferably, the reference sample comprises a reference lipid that can be detected by antibody and / or labeled or modified so as to be distinguished from natural lipid.Preferably, the reference lipid has a known concentration.Such a reference lipid is particularly useful as a standard.Therefore, various known concentrations of such reference lipid can be detected using the diagnostic assay described herein.

[0220] The instructions accompanying the kit of the present disclosure are typically written on a label or package insert (e.g., a paper sheet included in the kit), but machine-readable instructions (e.g., instructions stored on a magnetic or optical storage disk) are also acceptable. Instructions relating to the use of the reagents described herein generally include information for determining the concentration or expression level of one or more lipid biomarkers, as well as guidance on dosage, dosage schedule, and administration route for the indicated treatment. The kit may further include instructions for selecting individuals who have breast cancer and are therefore suitable for treatment.

[0221] In certain examples, the reference data resides on a computer-readable medium (e.g., software embodied in or utilized by any one or more of the methodologies or functions described herein). The computer-readable medium may be included on a storage device such as a computer memory (e.g., a hard disk drive or solid-state drive) and may include computer-readable code components that, when selectively executed by a processor, implement one or more aspects of the present disclosure.

[0222] system The present disclosure also contemplates systems for the detection of lipid biomarkers that may be suitable for use in the methods described herein.

[0223] In one broad aspect, the present disclosure provides a system for determining the presence or absence of breast cancer in a subject, the system comprising: 1. A mass spectrometry unit configured to determine a level of one or more lipid biomarkers in a biological sample obtained from a subject, wherein the one or more lipid biomarkers are Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(1 7:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4) ), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1) , PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM (d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), a mass spectrometry unit selected from the group consisting of TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof; and a processing unit configured to use or analyze the levels of one or more lipid biomarkers to determine the presence or absence of breast cancer in a subject.

[0224] It should be noted that the step of determining the concentration level or expression level may be performed by the mass spectrometry unit and / or may be performed at least in part by a pre-processing unit. The pre-processing unit may be the same as or different from the processing unit that performs the step of analyzing the concentration or expression level. For example, the pre-processing unit may receive data from the mass spectrometry unit indicating several fragments (e.g., lipid head groups and / or acyl chains) of the lipid biomarker for each mass value. This data may also represent the retention time of specific fragments. The pre-processing unit may then process this data to determine the lipid biomarker as a combination of fragments, thereby calculating the corresponding concentration or expression level.

[0225] Preferably, the mass spectrometry unit and the processing unit are as described herein.

[0226] Computer Implementation Method It is envisioned that one or more steps of the methods described herein may be automated or implemented by a computer, in the sense that the disclosed methods are implemented as software code stored on a non-volatile data storage medium, which executes the software code, causing the computer to perform the methods disclosed herein.

[0227] For example, comparing the concentration level or expression level of one or more lipid biomarkers with a reference or threshold level or value can be performed by a computer executing software code describing the comparing step. Thus, the comparison can be performed by a computer or computing device, such as by a processing unit. The determined or detected amount values of one or more lipid biomarkers in a sample from a subject and the reference amount can be compared with each other, for example, and the comparison can be performed automatically by a computer program executing a comparison algorithm. Additionally, the calculation of a risk or diagnostic score and / or its comparison with a reference risk or diagnostic score can be performed automatically by a computer program executing a comparison algorithm. Preferably, such an algorithm can be trained on one or more case and / or control samples. In some examples, a processor can utilize the concentration or expression level data and / or the risk or diagnostic score to calculate the likelihood of the subject in question having breast cancer.

[0228] The computer program that performs the evaluation preferably provides the desired evaluation in a suitable output format. For computer-assisted comparison, the determined quantity values can be compared by the computer program with values corresponding to suitable references stored in a database. The computer program can further evaluate the results of the comparison, i.e., automatically provide the desired evaluation in a suitable output format.

[0229] In some examples, the methods of the present disclosure include one or more of the following broad steps: (i) optionally performing a measurement of the concentration or expression level of one or more lipid biomarkers described herein; (ii) inputting or receiving the value from (i) into a processing system configured to determine the presence or absence of breast cancer in the subject; (iii) optionally, calculating, by a processing system, a risk or diagnostic score from the levels or expression levels of one or more lipid biomarkers; (iv) comparing the concentration or expression level and / or risk or diagnostic score obtained in step (iii) with a threshold or reference value by a processing system; and (v) determining the presence or absence of breast cancer in the subject.

[0230] The methods of the present disclosure preferably allow for integration into existing or newly developed pathology architectures or platform systems. For example, the present disclosure provides a method that allows a user to determine the status of a subject (e.g., the presence or absence of breast cancer), the method comprising: (a) receiving data in the form of concentration or expression levels of one or more lipid biomarkers for a test sample, wherein the one or more lipid biomarkers are Cer(d36:1), Cer(d42:0), Cer(d42:1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e ... 7:1), LPC(18:3), LPE(22:6), LPI(20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4) ), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1) , PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS(40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM (d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4) , TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof; (b) optionally processing the subject data, such as in a processing unit or system, via univariate and / or multivariate analysis and / or machine learning algorithms (e.g., lasso-penalized multivariate Cox regression, logistic regression, partial least squares discriminant analysis, random forests, decision trees, gradient boosting) to provide a risk or diagnostic score; (c) determining the subject's status according to the concentration or expression level results and / or risk or diagnostic score results, as compared with predetermined or reference concentration or expression levels and / or risk or diagnostic score values, such as by a processing unit or system; (d) transmitting an indication of the state of the subject to a user.

[0231] In some examples, the above methods further include generating or producing concentration level or expression level data by determining the concentration level or expression level of one or more lipid biomarkers in a biological sample obtained from the subject, wherein the one or more lipid biomarkers are, for example, Cer(d36:1), Cer(d42:0), Cer(d42: 1), DG(34:2), Hex2Cer(d34:1), Hex2Cer(d42:2), LPC(14:0), LPC(16:0e), LPC(17:1), LPC(18:3), LPE(22:6), LPI( 20:4), PC(32:1), PC(34:1), PC(35:4), PC(36:2), PC(36:3), PC(37:4), PC(38:5), PE(34:1), PE(34:2p), PE(36:4), PE(38:3p), PE(38:4), PE(O-38:5), PE(O-40:6), PI(32:1), PI(34:1), PI(36:1), PI(38:6), PS(38:4), PS(40:6), PS (40:7), SM(d32:2), SM(d33:1), SM(d35:1), SM(d36:1), SM(d36:2), SM(d38:4), SM(d40:3), SM(d41:2), SM(d41:3), SM(d42:1), SM(d42:4), SM(d44:4), TG(52:3e), TG(53:4), TG(54:3), TG(54:4), TG(54:5), TG(54:6), TG(56:1), TG(57:1), TG(58:1), TG(58:2), TG(58:3), TG(59:2), TG(60:1), and TG(62:2), or fragments, variants, or derivatives thereof.

[0232] In one example, the method additionally comprises: (a) having a user determine data using a remote end station; (b) transmitting the data from the end station to the base station over the communications network.

[0233] The base station may include first and second processing systems, in which case the method may include: (a) transmitting data to a first processing system; (b) causing the first processing system to perform a univariate or multivariate analytical function to generate a risk or diagnostic score.

[0234] The method also includes: (a) transmitting the results of the univariate or multivariate analytical function and / or the determined concentrations or expression levels of the lipid biomarkers to a second processing system; (b) causing a second processing system to determine the state of the object.

[0235] The second processing system may be coupled to a database adapted to store predetermined data and / or univariate or multivariate analytical functions, such that the computer-implemented method comprises: (a) querying a database to obtain at least selected predetermined data or accessing univariate or multivariate analytical functions from the database; (b) comparing the selected predetermined data with the target data or generating a predicted probability index.

[0236] The second processing system may be coupled to a database, and the method includes storing the data in the database, such as by a memory unit.

[0237] The reference concentration or expression level data comprises a level or concentration or expression level determined for one or more lipid biomarkers in the biological sample selected from the group consisting of: (i) a biological sample from a normal or healthy subject, e.g., a normal or healthy subject without breast cancer; (ii) a biological sample from a subject previously diagnosed or determined to have breast cancer; (iii) an extract of any one of (i) to (ii); (iv) a dataset comprising concentrations or expression levels for lipid biomarkers in a normal or healthy individual or a population of normal or healthy individuals; (vi) a dataset comprising concentrations or expression levels for lipid biomarkers in an individual or population of individuals with breast cancer; and (vii) A dataset comprising concentrations or expression levels for lipid biomarkers in a subject being tested, wherein the concentrations or expression levels are determined for a sample taken at an earlier time point when the subject was known not to have breast cancer.

[0238] Obtaining samples from subjects The methods disclosed herein may further include an initial or earlier step of providing or collecting a biological sample from a subject that preferably contains lipid microvesicles or extracellular vesicles. Such a sample may be obtained by freshly collecting the sample, or may be obtained from a previously collected and stored sample. For example, the sample may be obtained from a blood or serum sample that has been previously collected and stored (e.g., refrigerated or frozen). Preferably, the sample is obtained by freshly collecting the sample from the subject. Alternatively, the sample may be obtained from a previously collected and stored sample from the subject.

[0239] In certain examples, a subject fasts prior to or is in a fasting state at the time of collection of a biological sample for further testing by the methods provided herein. As used herein, the term "fasted" refers to a state in which no food or drink has been consumed for a period of at least about 3 hours to about 12 hours (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 hours, and any range therein) prior to providing a biological sample for testing.

[0240] In an alternative example, the subject does not fast prior to or is in a non-fasting (or "fed") state at the time of collection of a biological sample for further testing by the methods provided herein. The term "non-fasting," as generally used herein, refers to the ingestion of food and / or beverages within at least about 3 to about 12 hours (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 hours, and any range therein) of providing a biological sample for testing.

[0241] The methods of the present disclosure can be performed on various biological samples. As used herein, the term "biological sample" preferably refers to a sample obtained from a subject. For example, the biological sample may be a bodily fluid of the subject. A suitable biological sample typically contains a concentration of lipid microvesicles or extracellular vesicles. A suitable biological sample may also contain circulating lipid microvesicles or extracellular vesicles. Circulating lipid microvesicles or extracellular vesicles may be found in bodily fluids (e.g., blood, plasma, serum, urine, vomit, tears, sputum, etc.) or other excretions (e.g., feces). In certain examples, the biological sample is or contains blood, plasma, or serum. In certain examples, the biological sample is blood. In other examples, the biological sample is plasma. In various examples, the biological sample is serum.

[0242] Preferably, the biological sample is enriched for extracellular vesicles and / or their lipid content. In certain examples, the biological sample is or contains a population of extracellular vesicles or exosomes. In some examples, the biological sample is or contains a population of extracellular vesicles or exosomes obtained from a patient diagnosed with or suspected of having breast cancer. In certain examples, the biological sample is purified or partially purified to isolate extracellular vesicles, including cancer-derived extracellular vesicles, from the biological sample according to the present disclosure. For example, a serum sample may be purified to remove cells. In certain examples, other original components (e.g., debris, albumin) in the biological sample are removed or partially removed from the biological sample before performing the method of the present disclosure. In some examples, the biological sample is substantially free of cells.

[0243] Samples include extracts, derivatives, fractions, or suspensions of the original biological sample obtained from a subject disclosed herein.

[0244] Preferably, the biological sample described herein contains a concentration of extracellular vesicles, such as cancer-derived extracellular vesicles or exosomes. A suitable biological sample may also contain circulating extracellular vesicles. Circulating extracellular vesicles may be found in bodily fluids (e.g., blood, plasma, serum, urine, vomit, tears, sputum, etc.) or other excretions (e.g., feces). In certain examples, the biological sample is or contains an extracellular vesicle sample. In other examples, the biological sample is or contains an exosome sample or exosomal sample. To this end, embodiments of the present disclosure preferably include determining the expression or concentration level of EVs or exosomes of the above-described lipid biomarkers in a biological sample obtained from a subject. In other words, the method may involve determining or measuring the expression levels of two or more EV- or exosome-associated biomarkers provided herein, to the exclusion of protein biomarkers that are not associated with EVs or exosomes (e.g., biomarkers associated with free protein or free RNA in a biological sample).

[0245] Thus, determining the level of EV expression of a biomarker includes determining the expression level of that biomarker associated with EVs or a population of EVs. Similarly, determining the level of exosomal expression of a biomarker includes determining the expression level of that biomarker associated with exosomes or a population of exosomes. Terms such as "EV-associated," "exosome-associated," and the like refer to biomolecules, such as lipid biomarkers, located within or on the surface of EVs or exosomes, respectively. For this purpose, EVs and exosomes may carry or contain cellular biomolecules within their lipid membranes. Because EVs and exosomes are typically formed from cellular membranes, they may also carry molecules on their surface. Additionally, because some biomolecules bind to those on the surface of EVs and exosomes, these biomolecules may also be indirectly considered "EV-associated" or "exosome-associated." Furthermore, the present disclosure contemplates detecting EV- or exosome-associated biomarkers even after they have been separated from the EVs or exosomes with which they may originally be associated.

[0246] The biological sample may be subjected to any suitable pretreatment step before measuring the level of one or more lipid biomarkers to improve the accuracy and / or efficiency of the measurement. Such pretreatment steps may include extraction, centrifugation (e.g., ultracentrifugation), lyophilization, fractionation, separation (e.g., using column or gel chromatography), concentration, or evaporation. In some cases, this treatment may include one or more extractions with a solution containing any suitable solvent or combination of solvents, such as, but not limited to, acetonitrile, water, chloroform, methanol, butylated hydroxytoluene, trichloroacetic acid, toluene, hexane, benzene, or a combination thereof. In some examples, the biological sample may undergo one or more processing steps to isolate, concentrate, or enrich for extracellular vesicles and / or their lipid content. Preferably, the biological sample is an extracellular vesicle sample or a sample enriched for extracellular vesicles.

[0247] As used herein, the term "extracellular vesicles" or "EVs" refers to cell-derived vesicles that contain a membrane enclosing an internal space. Extracellular vesicles include all membrane-bound vesicles, the diameter of which is typically smaller than the diameter of the cell from which they are derived. More specifically, extracellular vesicles can include small extracellular vesicles (50-200 nm), microvesicles (0.2-1 μm), exomeres and exosomes (30-150 nm), and oncosomes (1 μm to 10 μm). In various examples, extracellular vesicles are or include exosomes. In certain instances, the extracellular vesicles have an average diameter of about 10 nm to about 250 nm (e.g., about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250 nm or any range therein), more particularly about 50 nm to about 200 nm, and even more particularly about 30 nm to about 150 nm.

[0248] In view of the above, a biological sample (e.g., blood, plasma, and / or serum) may be enriched for extracellular vesicles, such as cancer-derived extracellular vesicles or exosomes. In some examples, the biological sample is a purified or partially purified population of extracellular vesicles or exosomes. In other examples, the biological sample is a purified population of exosomes. In further examples, the biological sample is a partially purified population of exosomes. Thus, in some examples, the methods described herein include an initial or earlier step of at least partially enriching, concentrating, isolating, or purifying a population or fraction of extracellular vesicles or exosomes from the biological sample of the patient in question, such as by the methods described herein.

[0249] A population or fraction of extracellular vesicles or exosomes can be enriched, concentrated, isolated, or purified from a biological fluid, such as those provided above, to facilitate the removal of contaminating proteins, lipoproteins, and the like. To this end, extracellular vesicles or exosomes can be isolated by any means known in the art, including, but not limited to, polyethylene glycol precipitation, high-MW centrifugal filtration, ultracentrifugation, size-exclusion chromatography, exosome precipitation (e.g., ExoQuick from System Biosciences), tangential flow filtration, affinity-based capture of exosomes (e.g., affinity purification using binding molecules or antibodies against CD63, CD81, CD82, CD9, Alix, annexin, EpCAM, and Rab5), immunomagnetic bead capture (e.g., EXO-NET® system from INOVIQ), and any combination thereof. In various examples, extracellular vesicles or exosomes are at least partially isolated from a biological sample by ultracentrifugation. In certain examples, extracellular vesicles or exosomes are at least partially isolated from a biological sample by size-exclusion chromatography. In other examples, extracellular vesicles or exosomes are at least partially isolated from biological samples by affinity-based methods, and more particularly by immunomagnetic bead capture.

[0250] As used herein, the terms "isolate," "isolating," or "isolation" refer to a material, such as an extracellular vesicle or exosome, that has been removed from its natural state or otherwise subjected to human manipulation. Isolated material can be substantially or essentially free of components that normally accompany it in its natural state, or can be manipulated to be in an artificial state with components that normally accompany it in its natural state. These terms include the total physical separation of materials from their natural environment, including extracellular vesicles (e.g., removal / purification from a biological sample obtained from a subject suspected of having breast cancer).

[0251] Thus, any of the methods disclosed herein may include a step of obtaining a biological sample from a subject and determining the expression level, concentration, or abundance of one or more lipid biomarkers in the sample. Alternatively, any of the methods disclosed herein may not include a step of obtaining a biological sample from a subject and determining the expression level, concentration, or abundance of one or more lipid biomarkers in the sample. Instead, the expression level, concentration, or abundance of one or more lipid biomarkers in the sample may have been previously determined.

[0252] In order that preferred embodiments of the present disclosure may be fully understood and put to practical use, reference is made to the following non-limiting examples. [Example]

[0253] Example 1. Exploratory evaluation of lipidomic biomarkers in four breast cancer cohorts The aim of this example was to establish a predictive model for breast cancer from lipidomic data. [Table 2]

[0254] research design Four sets of plasma samples obtained with informed consent from adult female patients were utilized for lipidomic analysis in 2018, 2019, 2020, and 2021. As summarized in Table 1, the disease characteristics of the individuals from whom the samples were derived spanned various stages of invasive ductal carcinoma (IDC), ductal carcinoma in situ (DCIS), invasive lobular carcinoma (ILC), and normal controls. All individuals fasted prior to blood collection. It is unclear whether any individual was included in more than one dataset. [Table 3]

[0255] Cases In all four datasets, breast cancer type was confirmed clinically and morphologically, and breast cancer stage was confirmed by biopsy.

[0256] Breast cancer patients were excluded according to the following criteria: During pregnancy or breastfeeding History of HBV, HCV, or HIV Previous breast cancer treatment Previous breast cancer surgery

[0257] In datasets 2–4, cases met additional criteria, namely the absence of: Other chronic illnesses History of cardiovascular disease or diabetes Hyperlipidemia or dyslipidemia Radiation or chemotherapy

[0258] Control In all four datasets, normal controls were individuals without a diagnosis of breast cancer and were categorically matched to cases by age and BMI.

[0259] In datasets 2–4, normal controls met additional criteria, namely the absence of: Chronic illness History of cardiovascular disease or diabetes Hyperlipidemia or dyslipidemia History of any cancer Radiation or chemotherapy

[0260] Sample Size Considerations The current analysis was a post-hoc exploratory analysis of existing data. The four datasets included in the current analysis were pilot datasets derived without power calculations relevant to the current analysis.

[0261] Analysis population The following analysis populations were considered for discovery purposes: 1. Dataset 1 2. Dataset 2 3. Dataset 3 4. Dataset 4 5. Dataset 2 + Dataset 3 + Dataset 4

[0262] Test performance calculations involved multiple populations, as shown in Table 3. [Table 4]

[0263] General Considerations for Data Analysis Multiple Batches The looser inclusion criteria for Dataset 1 meant that individuals with a diagnosis of or undergoing treatment for traits such as dyslipidemia that affect lipid profiles could be included.

[0264] Datasets 2 and 3 were similar in design, but all four datasets were acquired at different time points and subjected to separate laboratory runs. Datasets 2 and 3 were each analyzed in two parts, resulting in a total of six laboratory batches.

[0265] Generally, a variety of breast cancer types and stages are considered among the cases. For these reasons, in the first example, a search for predictors was performed on each dataset, one at a time.

[0266] Multiple comparisons and multiplicity In the interest of statistical power, no adjustment for multiple testing was performed.

[0267] Study population Clinical characteristics Tabular summaries of clinical characteristics were generated one dataset at a time according to data availability (Tables 4-7). All tables were stratified by case / control status and included entries for age, BMI, HDL, and LDL. In addition, total cholesterol and triglycerides were tabulated for datasets 2-4. All participants self-reported as white, non-Hispanic. Categorical variables were summarized as n (%), continuously symmetrically distributed variables were summarized as mean (SD), and continuously distributed variables with skewed distribution were summarized as median (IQR).

[0268] Nonparametric tests were used to compare the distributions of these variables in cases and controls. Pearson's chi-squared test was applied for categorical variables, and the Mann-Whitney U test was applied for continuously distributed variables. Results were expressed to three decimal places or p<0.001. Missing values for clinical data were not imputed. It was noted that for dataset 2, batch was confounded with case-control status (p<0.001). This means that the statistical power of this dataset is reduced after adjusting for batch.

[0269] Differences in lipid levels (cholesterol, HDL, and LDL) were observed between cases and controls in cohorts 3 and / or 4. Due to the complex interrelationships between lipids and disease state, it was decided not to impose statistical adjustments on any of these measures. [Table 5] [Table 6] [Table 7] [Table 8]

[0270] Biomarker Data Analysis introduction Lipidomic data was generated using liquid chromatography-mass spectrometry (LC-MS). Lipids were identified and quantified using LipidSearch™ software and Skyline. This process involved peak detection and database matching, and the generation of quality scores for filtering purposes. Statistically analyzed data included normalized lipid concentrations.

[0271] Exploratory Data Analysis The analyses in this section were performed on combined cases and controls, one dataset at a time. In addition, we considered a pooled dataset including datasets 2 through 4, the latter of which is labeled "234."

[0272] No missing values were present in the data; however, zeros were considered as such. Lipids with a coefficient of variation less than 0.20 in the combined case-control set were excluded from further consideration. The distribution of each lipid was plotted as a histogram, and a global natural logarithmic transformation (ln(1 + x)) was applied to improve symmetry.

[0273] Principal component analysis (PCA) was performed, and the results were plotted in two and three dimensions and colored by case-control status. When detected, extreme outliers were excluded and the plots were redrawn. For analysis populations with multiple laboratory batches, an adjustment for batch was applied, adapting to case-control status, by using the Empirical Bayes methodology (Johnson and Rabinovic, 2007). The PCA plots were redrawn.

[0274] After completing the above steps, five analysis-ready datasets were available containing 394 lipids, 386 lipids, 382 lipids, 380 lipids, and 373 lipids for datasets 1, 2, 3, 4, and 234, respectively.

[0275] Training and validation sets for dataset 234 Dataset "234" was split into training and validation sets in a 2:1 ratio, balanced for cohort, case / control status, and age. The training set was used for signature discovery, and the validation set was used for test performance calculations. The following signature discovery step was performed on each dataset individually and on the 234 training set, for a total of 5 rounds.

[0276] Initial Coarse Filter Each lipid was examined one at a time for association with case / control status in a logistic model, adjusting for any covariates established in the previous step. Any lipid with a p-value >= 0.05 was excluded from further consideration.

[0277] Determination of lipid clusters For the purpose of deriving a knowledge-based signature of breast cancer, the remaining lipids were grouped according to their numerical correlation. Similarity matrices were created for the 30 strongest associations arising from each dataset. Hierarchical cluster analysis was applied and dendrograms were plotted.

[0278] Machine Learning Machine learning was performed using gradient boosting (Friedman 2001; Friedman 2002; Chen et al., 2021). Gradient boosting is an approach for determining a regression function that minimizes the expectation of a loss function. It is an iterative method in which the negative gradient of the loss function is calculated, a regression model is fitted, a gradient descent step size is selected, and the regression function is updated. The gradient is approximated by a regression tree, and at each iteration, the gradient determines the direction the function needs to move to improve its fit to the data.

[0279] The loss function for the current binary endpoint was specified as Bernoulli. A learning rate was introduced to suppress proposed moves and protect against overfitting. The minimum number of observations at each terminal node was 10. Two-way interactions were allowed. Random subsampling of half of the observations, without replacement, was applied to achieve variance reduction in the gradient estimates. There were 100 rounds of gradient boosting per model.

[0280] The outcome for each lipid was an estimate of its relative impact or feature importance. To protect against overfitting, variable selection was achieved by selecting the five lipids with the largest estimated relative impact. One or more of these were excluded if their estimated relative impact was <5%. The selected variables (lipids) and their estimated relative impact / feature importance were tabulated (Tables 8-12). Violin plots show the distribution of lipid concentrations by case / control status (Figures 1-5). Machine learning results for dataset 1 [Table 9] Machine learning results for dataset 2 [Table 10] Machine learning results for dataset 3 [Table 11] Machine learning results for dataset 4 [Table 12] Machine learning results for dataset 234 (training set) [Table 13]

[0281] Candidate Signatures For each lipid set, a logistic model was fitted by regressing case / control status against the selected lipids. Some models were simplified when there was no loss of fit, e.g., when two highly correlated lipids were included. The resulting models are shown in Tables 13-17. [Table 14] [Table 15] [Table 16] [Table 17] [Table 18]

[0282] Inspection performance calculation The test performance of each candidate signature arising from a single dataset was determined by applying it to the remaining three individual datasets as independent patient sets.

[0283] Test performance of candidate signatures derived from Dataset 234 (training set) was determined by application to Dataset 1 and Dataset 234 (validation set). Test performance calculations took the following form: Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for a range of predicted probability thresholds. Tables were generated to summarize performance across a range of classification thresholds, allowing greater or lesser weighting to be applied to true-positive findings relative to false-positives. Receiver operating characteristic (ROC) curves were generated to plot sensitivity as a function of (1 specificity). Additionally, the area under the curve (AUC) was estimated.

[0284] Performance of Candidate Signature 1 This signature was supported by data in cohorts 3 and 4 (AUC=0.76 and 0.74, respectively). Figure 6 shows the overlaid ROC curves.

[0285] Performance of Candidate Signature 2 This signature was supported by the data in cohort 3 (AUC=0.79). Performance in cohorts 1 and 4 was weaker (AUC=0.69 and 0.68, respectively), but still supported. Figure 7 shows the overlaid ROC curves.

[0286] Performance of Candidate Signature 3 This signature was supported by the data in cohorts 1 and 4 (AUC=0.73 and 0.72, respectively). Performance in cohort 2 was slightly weaker (AUC=0.68), but still supported. Figure 8 shows the overlaid ROC curves.

[0287] Performance of Candidate Signature 4 This signature was supported by the data in cohorts 2 and 3 (AUC=0.78 and 0.85, respectively). Figure 9 shows the overlaid ROC curves.

[0288] Performance of candidate signature 234 This signature was very similar in composition to candidate signature 4. It was supported by the data in dataset 234 validation (AUC = 0.83). Performance in cohort 1 was weaker (AUC = 0.69), but still supported. Figure 10 shows the overlaid ROC curves.

[0289] Example 2 In this example, we processed and analyzed four cohorts containing blood plasma sequentially over time, with the first cohort available in 2018 and cohort 4 available in 2021 (following the cohorts described above for Example 1). Over this time period, we performed various statistical analyses with the goal of finding a panel of lipids that distinguishes cancer from control samples in the provided cohorts.

[0290] research design The study design, number and description of cases in each cohort, treatment batches (listed as Set 1 to Set 6), and clinical characteristics are described in Example 1 above.

[0291] General Considerations for Data Analysis All stages of data analysis focused on early stage cancer (stages I and II) diagnoses; earlier phases of analysis also considered specific differences.

[0292] Improvements over time in the datasets and data structures used After cohorts 1–3 became available, a merged, manually curated dataset of common lipids was compiled in 2020. This improved version of the data contained approximately 450 lipid identifiers that could be matched across all subsequent cohorts, and cohort 4 was subsequently added. This is referred to as the BCAL lipid database.

[0293] Data merging and lipid identifiers Data merging was performed at the bioinformatics analysis level by a combination of matching any lipid ion as accurately as possible prior to the manually curated BCAL lipid database, and additionally using m / z value matching combined with retention time ordering. This approach was used for the original three cohorts and was superseded by the use of lipid databases as they became available.

[0294] Data Processing For all analyses, the following steps were performed. Low average value <10 -5 Lipids having Discarded lipids with missing values (this only affects earlier analyses, as the lipid database does not contain missing values) The remaining data were normalized to adjust for possible differences in total sample volume. ●Natural logarithm transformation was applied The distribution of the obtained data was visually confirmed using box plots and density plots. The overall clustering was confirmed by generating a principal component plot. Outliers were not removed

[0295] Batch normalization Batch processing differences are evident whenever different datasets are combined, and therefore batch normalization was evaluated. Starting with cohort 2, set 3 batch sample replicates became available in addition to technical injection replicates, and these were used to evaluate the utility of batch normalization using replicate correlations and coefficients of variation. An adapted version of internal reference scaling (IRS) normalization was found to give good performance (Plubell, 2017) and was subsequently used for batch normalization.

[0296] Variable Selection At each stage of the analysis, lipid biomarkers were selected based on their ability to distinguish between controls and cancer in currently available datasets, and the selection was validated in subsequent sets.

[0297] An initial selection of 18 lipid biomarkers was performed using bioinformatically merged Set 1 and Set 2 as described above (see Table 18 below). A mixed-effects model with a random batch effect was performed to find differentially expressed markers for the available sets using the nlme R package implementation (Pinheiro, 2000). [Table 19]

[0298] An early process filter was performed to retain all markers that showed evidence of differential expression between either control-early or control-stage 1-stage 2 using both ANOVA or mixed effects models with batch effects.

[0299] For any of these models, any lipid with a Benjamini and Hochberg adjusted p-value <0.05 (Benjamini, 1995) was selected, resulting in an initial subset of 110 markers of interest.

[0300] This filtered set of markers was generated to perform differential expression between control and early cancer samples, selecting the top differentially expressed markers based solely on p-value <0.05 and fold change criteria (18 initial panels - "Initial 18 Step 1 Set 1 and Set 2"). The volcano plot below shows the lipid fold change and p-value for which lipid ion identifiers are available at that time (Goedhart, 2020).

[0301] When Set 3 became available, the above steps were repeated using the entirety of Cohort 2. Following this process, the data from Set 2 and Set 3 were used to enrich the initial panel of 18 with 9 additional markers, resulting in the panel "Initial 18 + 9 Additional Cohort 2 NewVS," which represented the intermediate panel of interest (Table 19). [Table 20-1] [Table 20-2]

[0302] After completion of Cohort 3 (Sets 4 and 5), differential expression using ANOVA on the combined and IRS batch-normalized data was used to add the top up- and down-regulated lipids to the previously existing panel (18 400-item reoptimized lipid panels—"Cohort 2-3 Final"; see Table 21 below). Similarly, the distribution of fold changes and p-values is visualized on the volcano plot below. The intermediate addition of the 18 most differentially expressed markers for Cohort 3 (top 9 most up-regulated and top 9 most down-regulated based on fold change cancer / control among differentially expressed lipids based on p-value < 0.05) was retained as panel "Cohort 3 Add Cohort 3 Plsda Add 9 Up 9 Down" (Table 20). [Table 21] [Table 22]

[0303] Subsequent filtering to smaller panels was performed using a combination of stepwise regression and the removal of lipids that did not significantly contribute to the logistic model discriminating between control and early-stage disease, using implementations available in the core statistics R package (Dobson, 1990). This resulted in the selection of smaller subsets of lipids (the initial 18 9 lipid panel—"18-9 Lipids Restrict9Lipids"; a 10-subset of the 18 400 optimized panels, "Top10 Optimized for Patents Top10Coeff5050train"; see Tables 22 and 23 below; and a subset obtained in the same manner from Cohort 2 lipid selection, "Initial 18 + 9 Further Cohort 2 NewVS"—"Optimized Cohort 2"; see Table 24 below). [Table 23] [Table 24] [Table 25]

[0304] Cohort 4 was not used directly for variable selection, but the consistency of fold change control / cancer across all cohorts was used to select a smaller panel of 6 lipid biomarkers from the 400 reoptimized panel that showed consistent performance across all batches ("4 markers in 6 panels"; see Table 25). [Table 26]

[0305] Additionally, the top 10 differentially abundant lipids for Cohort 4 were also identified by a combination of fold change and p-value cutoff, as in previous cohorts, and used to select a panel of 6 lipids common to those 12 lipid biomarkers identified in Example 1 ("Common6-ML_GBM"; see Table 26). [Table 27]

[0306] Classification and Machine Learning Approaches Initial analyses performed on Sets 1 and 2 evaluated random forests (Breiman, 2001), PLSDA using the caret R package implementation (Kuhn, 2020), logistic regression, and support vector machines (implemented in the R package e1071) in the context of Sets 1 and 2, with logistic regression selected as giving the best and more easily interpretable results using a smaller lipid panel on the initial dataset.

[0307] Whenever a lipid panel is evaluated, its classification potential is evaluated using leave-one-out cross-validation, where a model is trained on all but one sample, and model performance is evaluated on the remaining samples. This process is repeated for all samples in turn, ensuring that data used for testing is not also used for training. This approach was particularly relevant in the early stages when only a few datasets were available.

[0308] For each panel obtained through variable selection or set selection in one cohort, its classification success was evaluated in the next available cohort.

[0309] As mentioned above, variable selection was performed using only cohorts 1, 2, and 3, which contained only invasive ductal carcinoma (IDC) samples for cancer subcategory, except for refining the subsequent small panel. This ensures that the results obtained in cohort 4 demonstrate that the panel still has classifiability on a dataset containing different early cancer subcategories (DCIS and ILC).

[0310] This sequential analytical path, combined with the different choices of methods for variable selection, may explain, at least in part, the differences in the lipid biomarkers identified by the statistical data analysis of Example 1 and the approach described in Example 2. Despite this, several lipid biomarkers (see overlapping panel of six biomarkers below) overlap between the diagnostic panels of Examples 1 and 2, thereby demonstrating a high degree of consistency in identifying relevant lipid biomarkers by the two different methodologies.

[0311] Panel Details Limited panel of 6 markers - "Panel 6 4 markers" This panel contains two markers common to Example 1. As an extended example, the box plots in Figures 16-23 detail conserved patterns of expression or concentration across several available sets.

[0312] A panel of six markers common to the validation report, "Common6-ML_GBM" This panel is an important overlap panel, as it contains all lipid markers identified as diagnostically relevant by both Examples 1 and 2, including two of the m / z values in the six previous panels and all lipids from Figure 3 (Cohort 4 volcano plot).

[0313] ROC curve The ROC curves for each of the panels from Example 2 are provided in Figures 14, 15, and 24-33. These curves show the diagnostic potential of each of the panels in patients with breast cancer, including early-stage breast cancer. All curves were generated using leave-one-out cross-validation.

[0314] Example 3 Using Cohorts 3 and 4, we randomly selected m / z values (either 6 or 18 panels) to compare with the diagnostic value of the diagnostic panels identified in Examples 1 and 2. For the 6 random panels, 1000 simulations were performed. For the 18 random panels, 100 simulations were performed. We then tabulated the random panels that achieved high random accuracy (defined as >75%) and compared their composition to previously identified panels (see Tables 27-29 below).

[0315] result Panels of six lipid biomarkers were among the best performers, with ML_GBM from Table 23 being the best performer in that analysis (see Table 24), while four markers in the six panel from Table 22 were ranked third in that analysis (see Table 25). A high proportion of the top-performing panels (random or otherwise) also contained the LPC(14:0) marker. Several of the random panels achieved similar diagnostic performance.

[0316] When summarizing the most commonly occurring lipids from well-performing random panels, three from the ML_GBM panel are in the top five (see Table 27). [Table 28] [Table 29] [Table 30] [Table 31-1] [Table 31-2] [Table 31-3]

[0317] Example 4 In this example, we conducted experiments to determine whether breast cancer can be detected using ingested extracellular vesicles (EV) samples along with fasted EV samples, and whether breast cancer can be detected using fasted or fed plasma samples directly. We also aimed to determine whether any of the existing lipid signatures determined using EV-fasted samples from various historical cohorts could be used to evaluate fed EV samples or fasted plasma samples.

[0318] research design Lipids in fasting and nonfasting EV and plasma samples were extracted from 60 control patients. Matched samples were analyzed in two separate laboratories: internally at BCAL (n=26 of 60 total samples) and externally at Baker Heart and Diabetes Institute (n=60 samples). The BCAL laboratory samples were previously measured in historical cohorts and therefore contained the full complement of lipids in the previous lipid signature of interest. The Baker laboratory samples did not identify the full complement of BCAL lipids; however, a larger number of patient samples were evaluated.

[0319] The resulting lipid quantifications were tested, where possible, against existing BCAL signatures and correlations of the classification scores obtained between matched fasted and fed samples were reported.

[0320] Data analysis Identifying lipids that change in fasting / fed states Matched control samples were used to find which lipids change in concentration between fasted and fed states in EV samples and separately in plasma, as well as between plasma and EV samples.

[0321] In terms of data processing, lipids with missing values were removed. Data were shifted by a small value of eps = 0.001 to allow for logarithmic transformation of zero values. Two-tailed paired t-tests were applied separately to all lipids, and any lipid was considered significantly different if the FDR-adjusted (Benjamini and Hochberg) p-value was less than 0.05.

[0322] Identifying lipid signatures that show robust results in fasted or fed samples This lipid signature was examined to determine whether lipid concentrations change depending on fasting / fed state, and we speculated that lipid signatures that do not change upon feeding could potentially be used to classify fed samples as well. The lipid panels examined were 18 400-item reoptimized lipid panels, 6 panels with 4 markers, and a common 6-ML_GBM panel.

[0323] For each of the lipid signatures considered (typically trained using cohorts 2-3), a model prediction script was run to normalize the new data, apply the signature, and generate a cancer probability for each of the samples.

[0324] Fed and fasted samples were matched side-by-side, the resulting scores were compared, and fit lines and statistics (fit line coefficient, R-squared, correlation, correlation significance) were generated.

[0325] result Differential abundance depending on fasting state No statistically significant differences in the concentrations of lipids in the BCAL panel common 6-ML_GBM were observed between fed and fasted states in either EV or plasma samples (Figure 34 and Table 31). The other two panels contained several lipids that showed significant differences in their concentrations between fed and fasted states. Table 31 details the full panel m / z values of lipids whose concentrations changed with fasting / fed states. [Table 32-1] [Table 32-2]

[0326] Correlation of lipid signature scores for the BCAL dataset (n=26) Table 32 summarizes the score alignment matches for the signatures of interest. The common 6-ML_GBM yields the highest score for correlation between fasted and fed matched EV samples, second only to the baseline example consisting of frozen replicate samples previously analyzed by BCAL. Fasted EV and plasma samples are also highly correlated. [Table 33]

[0327] Similarity of lipid signature scores for external laboratory datasets (n=60) Although not all lipids from the BCAL signature were included in this analysis, all six lipids from the common 6-ML_GBM could be matched between both datasets, and Figure 35 and Table 33 below summarize the score alignment obtained with this signature in the new dataset (n=60 controls). The matched scores obtained were significantly correlated. Since all samples were controls, samples with low cancer probability scores represent correct classifications (Figure 36). For both fasted and non-fasted samples, the median score was below 0.5, while for fasted and non-fasted EV samples, more than 75% of the samples have a model probability <0.6. [Table 34]

[0328] conclusion Experiments performed indicate that the lipid biomarker panel described herein (e.g., the common 6-ML_GBM) may not contain any lipids whose concentrations change significantly between fed and fasting states. Therefore, the common 6-ML_GBM lipid signature can potentially be used to evaluate non-fasting patient EV samples and fasting EV samples. Classification algorithm scores generated with new laboratory data correlate significantly between patient-matched fasting and non-fasting EV samples, and between fasting plasma and fasting EV samples, both in data generated internally by BCAL and external laboratory data.

[0329] References Chen T, He T, Benesty M et al (2021). xgboost: Extreme Gradient Boosting. R package version 1.4.1.1. https: / / CRAN.R-project.org / package=xgboost Friedman JH (2001). Greedy function approximation: a gradient boosting machine. Ann Statist 29(5): 1189-1232. Friedman JH (2002). Stochastic gradient boosting. Comput Stat Data Anal 38(4): 367-378. Johnson WE, Li C, Rabinovic A (2007). Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics 8(1):118-27. Mistry DAH, French PW (2016). Circulating phospholipids as biomarkers of breast cancer: a review. Breast Cancer: Basic and Clinical Research 10: 191-196. Society AC. Cancer Facts and Figures 2016. Atlanta: American Cancer Society, Inc.Available at: http: / / www.cancer.org / cancer / breastcancerinmen / detailedguide / breast-cancer-in-men-key-statistics

Claims

[Claim 1] The invention described in the specification.