Biomarkers for lung cancer detection

Abstract

The present application discloses an in vitro method for diagnosing lung cancer in a subject, comprising detecting at least one biomarker selected from the group consisting of Rho GDP dissociation inhibitor beta (ARHGDIB), alpha-tubulin 4A (TUBA4A), glutathione S-transferase omega 1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13) in a biological sample from the subject, and a kit for measuring the at least one biomarker.

Description

[Technical Field]

[0001] The present invention relates to methods and kits for diagnosing lung cancer in a biological sample from a subject. [Background technology]

[0002] Lung cancer is the most common malignant tumor in terms of incidence and the most deadly cancer worldwide. Smoking, particularly cigarette smoking, is by far the leading cause of lung cancer. High lung cancer mortality rates are primarily based on the level of progression at the time of diagnosis. Five-year survival rates drop significantly from 83% for stage IA tumors to 6% for stage IV tumors. Currently, more than half of lung cancer patients are diagnosed at the metastatic stage.

[0003] Due to the serious nature of lung cancer and its incurability in advanced stages, there is a great need for methods that allow for early diagnosis of lung cancer, which is a prerequisite for improving patient survival and treatment outcomes.

[0004] Currently, only 15% of newly diagnosed lung tumors are diagnosed early. For this reason, lung cancer screening using low-dose computed tomography (LDCT) can reduce lung cancer-specific mortality by 20% compared with chest radiography. However, the high rate of false-positive results (96.4% and 94.5% in the LDCT and radiography groups, respectively) and the risk of malignancy associated with cumulative radiation exposure are significant limitations of LDCT. Summary of the Invention [Problem to be solved by the invention]

[0005] Therefore, there is an urgent need for sensitive, preferably minimally invasive, methods for detecting lung cancer, especially during the early stages of the disease, to improve prognosis and reduce overdiagnosis. [Means for solving the problem]

[0006] The present inventors have addressed the challenges of lung cancer detection and diagnosis by developing biomarkers and biomarker panels, particularly biomarker panels. Indeed, the present inventors have demonstrated that certain biomarkers can detect lung cancer regardless of stage, have particularly strong diagnostic performance in early-stage lung cancer, and may therefore be useful for diagnosis. Furthermore, the present inventors have demonstrated that certain methods can be successfully used to identify and validate suitable biomarker panels.

[0007] Therefore, certain aspects of the present invention relate to biomarkers and biomarker panels suitable for use in diagnosing lung cancer. In certain embodiments, the biomarkers are selected from Rho GDP dissociation inhibitor beta (ARHGDIB), alpha-tubulin 4A (TUBA4A), glutathione S-transferase omega 1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13). Each of these markers has its own predictive value. However, generally, a panel of at least two, preferably at least three, biomarkers is used. Therefore, particularly preferred combinations are envisioned. In certain embodiments, the method comprises detecting ARHGDIB and at least one other lung cancer biomarker, preferably a marker selected from the group consisting of α-tubulin 4A (TUBA4A), glutathione S-transferase ω1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13). In certain embodiments, the method comprises detecting ARHGDIB and at least two other markers selected from the group consisting of α-tubulin 4A (TUBA4A), glutathione S-transferase ω1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13). In a further embodiment, the present invention provides a biomarker panel comprising at least two biomarkers selected from the group consisting of Rho GDP dissociation inhibitor beta (ARHGDIB), alpha-tubulin 4A (TUBA4A), glutathione S-transferase omega 1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13), which can distinguish individuals with lung cancer from healthy individuals. In a particular embodiment, the biomarker panel comprises at least ARHGDIB and one or two other markers selected from the defined group.The lung cancer biomarkers of the present invention have excellent performance in terms of a large area under the receiver operating characteristic curve (AUC), a high positive predictive value (PPV), a high negative predictive value (NPV), high sensitivity and / or high specificity, preferably high sensitivity and high NPV or high specificity, more preferably high sensitivity, high NPV and high specificity. This is supported by an AUC of 0.90 or more, a PPV of 0.90 or more, an NPV of 0.90 or more, a specificity of 0.90 or more, and / or a sensitivity of 0.90 or more, preferably a sensitivity of 0.90 or more and an NPV or specificity of 0.90 or more, more preferably a sensitivity of 0.90 or more, an NPV of 0.90 or more and a specificity of 0.90 or more. Furthermore, the lung cancer biomarkers of the present invention can detect lung cancer in a non-invasive manner, regardless of disease stage. As a result, the biomarker panel taught herein can be used as a routine test for high-risk and average-risk individuals (e.g., smokers or former smokers). The biomarkers and biomarker panels taught herein may also effectively complement currently used methods in lung cancer screening, such as LDCT, thereby reducing the number of false-positive cases that often lead to additional invasive tests and unnecessary costs, and expose patients to physical and mental hardship. To make the lung cancer biomarkers of the present invention easy for physicians to use, the inventors have also adopted a threshold-based approach that ascribes a threshold / biomarker and then a score / sample to classify a subject as having or not having lung cancer.

[0008] A first aspect provides an in vitro method for diagnosing lung cancer in a subject, comprising detecting at least one biomarker selected from the group consisting of Rho GDP dissociation inhibitor beta (ARHGDIB), alpha-tubulin 4A (TUBA4A), glutathione S-transferase omega 1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13) in a biological sample from the subject. In certain embodiments, the method comprises detecting ARHGDIB in a biological sample from the subject.

[0009] In certain embodiments, the method comprises detecting CDH13. In certain embodiments, the method comprises detecting GSTO1. In certain embodiments, the method comprises detecting CDH13 and at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, and PRDX6, preferably at least TUBA4A and / or FLNA and / or PRDX6, most preferably at least TUBA4A. In certain embodiments, the method comprises detecting GSTO1 and at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, FLNA, PRDX6, and CDH13, preferably at least TUBA4A and / or FLNA, most preferably at least TUBA4A. In certain embodiments, the method comprises detecting CDH13 and GSTO1, optionally at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, FLNA, and PRDX6, preferably at least TUBA4A and / or FLNA, most preferably at least TUBA4A. In certain embodiments, the method comprises detecting ARHGDIB and, optionally, at least one biomarker selected from the group consisting of TUBA4A, GSTO1, FLNA, and PRDX6 and CDH13.

[0010] In certain embodiments, the method comprises detecting at least two, at least three, at least four, or at least five biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13. In certain embodiments, the method comprises detecting ARHGDIB and at least two other biomarkers selected from the group consisting of TUBA4A, GSTO1, FLNA, PRDX6, and CDH13. In certain embodiments, the method comprises detecting at least three biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13. In certain embodiments, the method comprises detecting ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13.

[0011] In certain embodiments, the biological sample is a body fluid sample; preferably a body fluid sample selected from the group consisting of plasma, serum, whole blood, urine, tissue lysate, cerebrospinal fluid (CSF), saliva, and sweat; more preferably a plasma sample. In certain embodiments, the at least one biomarker is detected using mass spectrometry, a biochemical or molecular biological assay, an immunoassay, a chromatography method, or a combination thereof.

[0012] In certain embodiments, the method comprises: (a) measuring the amount or expression level of at least three biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample from the subject; (b) calculating a score based on the amounts or expression levels of the at least three biomarkers measured in step (a); (c) comparing the score calculated in step (b) with a threshold score; and (d) diagnosing the subject with lung cancer if the score calculated in step (b) is equal to or greater than the threshold score. Includes:

[0013] A further aspect is (a) a means for measuring the amount or expression level of at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample of a subject; and (b) a threshold value representing a known diagnosis of lung cancer for said at least one biomarker, or a means for establishing said threshold value; and a kit, particularly a kit for diagnosing lung cancer, comprising the above compound.

[0014] In certain embodiments, the means are particularly adapted for measuring the amount and / or expression level of said markers. In certain embodiments, the kit comprises means particularly adapted for measuring the amount or expression level of ARHGDIB and at least one biomarker, preferably at least two biomarkers, selected from the group consisting of TUBA4A, GSTO1, FLNA, PRDX6 and CDH13, or any combination of the markers listed above.

[0015] A further aspect provides the use of the kit for diagnosing lung cancer based on the detection of said at least one biomarker in a sample from a subject. [Brief explanation of the drawings]

[0016] [Figure 1] Plasma levels of six protein biomarkers identified as a lung cancer diagnostic panel. Scatter plots of the concentrations of (A) FLNA, (B) TUBA4A, (C) GSTO1, (D) PRDX6, (E) ARHGDIB, and (F) CDH13 obtained from lung cancer patients (n=128) and healthy volunteers (n=93) using an LC-PRM assay targeting proteotypic peptides. Data points and their medians are shown. ****P<0.0001 adjusted using the nonparametric Kruskal-Wallis test. [Figure 2] Parallel reaction monitoring (PRM) readout of six proteins included in the diagnostic panel. Representative PRM traces recorded in lung cancer patient and healthy donor samples are shown for each protein. Detected product ions of the endogenous target peptide (top) and the internal standard peptide (bottom) are displayed. [Figure 3] Forest plot of negative predictive value (NPV), positive predictive value (PPV), sensitivity ("sens"), specificity ("spec") and area under the receiver operating characteristic curve (AUC) for combinations (including subcombinations) of the 6-protein panel and single biomarkers tested in the validation dataset. DETAILED DESCRIPTION OF THE INVENTION

[0017] As used herein, the singular forms "a," "an," and "the" include both singular and plural referents unless the context clearly dictates otherwise.

[0018] The terms "comprising," "comprises," and "comprised of," as used herein, are synonymous with "including," "includes," or "containing," and are inclusive or open-ended and do not exclude additional, unrecited members, elements, or method steps. These terms also encompass "consisting of" and "consisting essentially of," which have well-established meanings in patent language.

[0019] The recitation of numerical ranges by endpoints includes all numbers and fractions subsumed within the respective ranges, as well as the recited endpoints.

[0020] As used herein, the terms "about" or "approximately," when referring to a measurable value, e.g., a parameter, amount, duration, etc., are meant to encompass variation of the specified value and variation from the specified value, e.g., variation of + / - 10% or less, preferably + / - 5% or less, more preferably + / - 1% or less, and even more preferably + / - 0.1% or less, of the specified value, insofar as such variation is appropriate for practicing the disclosed invention. It should be understood that the value to which the modifier "about" refers is itself also specifically, and preferably disclosed.

[0021] The term "one or more" or "at least one," e.g., one or more elements or at least one element of a group of elements, is self-explanatory, but by way of further example, the term includes reference to, among other things, any one of said elements, or any two or more of said elements, e.g., any three, four, five, six, or seven, etc. of said elements, up to and including all said elements. As another example, "one or more" or "at least one" may refer to 1, 2, 3, 4, 5, 6, 7, or more.

[0022] The discussion of the background of the invention herein is included to explain the context of the invention and should not be construed as an admission that any of the material mentioned was published, publicly known, or part of the common general knowledge in any country as of the priority date of any of the claims.

[0023] Throughout this disclosure, various publications, patents, and published patent specifications are referenced by an identifying citation. All documents cited herein are incorporated by reference in their entirety. In particular, the teachings or passages of such documents specifically mentioned herein are incorporated by reference.

[0024] Unless otherwise specified, all terms (including technical and scientific terms) used in disclosing the present invention have the meanings commonly understood by one of ordinary skill in the art to which this invention belongs. As a further guide, term definitions are included to better appreciate the teachings of the present invention. When a particular term is defined in connection with a particular aspect of the present invention or a particular embodiment of the present invention, such connotation is meant to apply throughout the specification, i.e., in the context of other aspects or embodiments of the present invention, unless otherwise specified.

[0025] In the following text, different aspects or embodiments of the invention are defined in more detail. Each aspect or embodiment so defined may be combined with any other aspect or embodiment, unless expressly indicated otherwise. In particular, any feature indicated as being preferred or advantageous may be combined with any other feature or features indicated as being preferred or advantageous.

[0026] Throughout this specification, references to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrase "in one embodiment" or "in an embodiment" in various places throughout this specification may, but do not necessarily all refer to the same embodiment. Furthermore, particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments, as would be apparent to one of ordinary skill in the art from this disclosure. Furthermore, although some embodiments described herein include some features but not other features included in other embodiments, combinations of features from different embodiments are within the scope of the present invention and are meant to form different embodiments, as would be understood by one of ordinary skill in the art. For example, in the appended claims, any of the claimed embodiments may be used in any combination.

[0027] The present inventors have identified a set of biomarkers that are particularly interesting for diagnosing lung cancer, especially early-stage lung cancer. The expression of the biomarkers Rho GDP dissociation inhibitor beta (ARHGDIB), alpha-tubulin 4A (TUBA4A), glutathione S-transferase omega 1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13) has been found to be particularly correlated with the development of lung cancer and is therefore suitable for both individual diagnosis and as part of a panel. While detecting one of these markers often already provides an important indication, combining two or more markers increases the accuracy and sensitivity of diagnosis. Indeed, the present inventors have found that detection of at least one, preferably at least two or three, and more preferably all five or six biomarkers selected from the group consisting of Rho GDP dissociation inhibitor beta (ARHGDIB), alpha-tubulin 4A (TUBA4A), glutathione S-transferase ω1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13) in a subject's biological sample enables accurate discrimination of lung cancer patients from individuals without lung cancer. At least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 has excellent performance in diagnosing lung cancer in a subject's biological sample. This is supported by excellent performance measures, including AUC, PPV, NPV, sensitivity, and specificity. More specifically, the sensitivity of this method is better than the described sensitivity of the CancerSEEK blood test for diagnosing lung cancer (Cohen JD, Li L, Wang Y, Thoburn C, Afsari B, Danilova L, et al., Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science. 2018;359: 926-30).

[0028] Compared with lung cancer screening using imaging techniques such as low-dose CT (LDCT), the method taught herein, which is highly sensitive and highly specific, reduces the rate of false-positive results (i.e., overdiagnosis rate) and avoids exposing the tested subject to radiation. Reducing false-positive cases also avoids further invasive testing and unnecessary costs, as well as exposing patients to physical and mental hardship. Furthermore, the use of one or more biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 allows for the detection of lung cancer in a non-invasive manner, since the biomarkers can be detected in body fluid samples, such as plasma samples. Furthermore, testing for one or more biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 also allows for the detection of lung cancer regardless of disease stage, including stage I lung tumors. Early diagnosis of cancer during routine screening provides patients with a prompt treatment solution. In addition, to make the biomarkers easier for physicians to use, the inventors also adopted a threshold-based approach resulting from thresholds / biomarkers and, optionally, scores / samples to classify subjects as having or not having lung cancer.

[0029] Thus, in a first aspect, the invention provides the use, preferably in vitro or ex vivo, of at least one, such as at least two, at least three, at least four, at least five or all six biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 as biomarkers for lung cancer in a subject.

[0030] The term "biomarker" is widely known in the art and may broadly refer to a biological molecule and / or detectable portion thereof, the qualitative and / or quantitative assessment of which in a subject is predictive or informative (e.g., predictive, diagnostic, and / or prognostic) with respect to one or more aspects of the subject's phenotype and / or genotype, e.g., with respect to the subject's status with respect to a given disease or pathological condition. Reference herein to a "biomarker panel" is made when two or more biomarkers are detected in the methods or uses taught herein.

[0031] In one embodiment, the biomarkers taught herein, such as at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13, among others, may be peptide-based, polypeptide-based, and / or protein-based.

[0032] Reference to any marker, including any peptide, polypeptide, or protein, corresponds to the marker, peptide, polypeptide, or protein commonly known by that name in the art. These terms encompass such markers, peptides, polypeptides, or proteins in any organism found therein, particularly animals, preferably warm-blooded animals, more preferably vertebrates, even more preferably mammals, including humans and non-human mammals, and even more preferably humans. The terms specifically encompass such markers, peptides, polypeptides, or proteins having native sequences, i.e., those whose primary sequence is identical to the sequence of the marker, peptide, polypeptide, or protein found or derived in nature. Those skilled in the art will understand that native sequences may differ between different species due to genetic divergence between such species. Furthermore, native sequences may differ between or within different individuals of the same species due to normal genetic variation within a given species. Native sequences may also differ between or even within different individuals of the same species due to post-transcriptional or post-translational modifications. Any such variants or isoforms of markers, peptides, polypeptides, or proteins are contemplated herein. Therefore, all sequences of markers, peptides, polypeptides, or proteins that are found or derived from nature are considered to be "native." The term encompasses markers, peptides, polypeptides, or proteins when they form part of a living organism, organ, tissue, or cell, when they form part of a biological sample, and when they are at least partially isolated from such a source. The term also encompasses markers, peptides, polypeptides, or proteins when they are produced by recombinant or synthetic means.

[0033] In certain embodiments, the biomarkers taught herein may be human biomarkers, such as human ARHGDIB (also known as Rho GDP-dissociation inhibitor 2 (RhoGDI2)), human TUBA4A, human GSTO1, human FLNA, human PRDX6, or human CDH13.

[0034] As an example, the protein sequence of human ARHGDIB is annotated under NCBI Genbank (http: / / www.ncbi.nlm.nih.gov / ) accession number NP_001308351.1 and UniProtKB / Swiss-prot number P52566.3; the protein sequence of human TUBA4A is annotated under NCBI Genbank accession number NP_005991.1 and UniProtKB / Swiss-prot number P68366.1; the protein sequence of human GSTO1 is annotated under NCBI Genbank accession number NP_004823.1 and UniProtKB / Swiss-prot number P78417.2; the protein sequence of human FLNA is annotated under NCBI Genbank accession number NP_004823.1 and UniProtKB / Swiss-prot number P78417.2; The protein sequence of human PRDX6 is annotated under NCBI Genbank accession number NP_004896.1 and UniProtKB / Swiss-prot number P30041.3; and the protein sequence of human CDH13 is annotated under NCBI Genbank accession number NP_001248.1 and UniProtKB / Swiss-prot number P55290.1.

[0035] Unless the context makes clear otherwise, reference herein to any marker, peptide, polypeptide, or protein, or fragment thereof, may generally also include modified forms of said marker, peptide, polypeptide, or protein, or fragment thereof, such as those having post-expression modifications including, for example, phosphorylation, glycosylation, lipidation, methylation, cysteinylation, sulfonation, glutathionylation, acetylation, oxidation of methionine to methionine sulfoxide or methionine sulfone, etc.

[0036] Reference herein to any marker, peptide, polypeptide or protein also includes fragments thereof. Thus, reference herein to measuring (or measuring the amount of) any one marker, peptide, polypeptide or protein may include measuring that marker, peptide, polypeptide or protein, for example measuring any mature and / or processed soluble / secreted form thereof (e.g., plasma circulating form), and / or measuring one or more fragments thereof.

[0037] For example, any marker, peptide, polypeptide, or protein, and / or one or more fragments thereof may be measured collectively such that the amount measured corresponds to the total amount of the species measured collectively. In a further example, any marker, peptide, polypeptide, or protein, and / or one or more fragments thereof may each be measured individually.

[0038] The term "fragment," in reference to a peptide, polypeptide, or protein, generally refers to an N-terminally and / or C-terminally truncated form of the peptide, polypeptide, or protein. Preferably, a fragment may comprise at least about 30%, e.g., at least about 50% or at least about 70%, preferably at least about 80%, e.g., at least about 85%, more preferably at least about 90%, even more preferably at least about 95% or even about 99% of the amino acid sequence length of said peptide, polypeptide, or protein. For example, so long as it does not exceed the length of the full-length peptide, polypeptide, or protein, a fragment may comprise a sequence of > 5 contiguous amino acids, or > 10 contiguous amino acids, or > 20 contiguous amino acids, or > 30 contiguous amino acids, such as > 40 contiguous amino acids, for example > 50 contiguous amino acids, such as > 60, > 70, > 80, > 90, > 100, > 200, > 300, or > 400 contiguous amino acids of the corresponding full-length peptide, polypeptide, or protein.

[0039] For example, at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 may be measured by measuring a peptide fragment of a full-length protein. For example, ARHGDIB may be measured by measuring YVQHTYR (SEQ ID NO: 1), TUBA4A may be measured by measuring EIIDPVLDR (SEQ ID NO: 2), GSTO1 may be measured by measuring GSAPPGPVPEGSIR (SEQ ID NO: 3), FLNA may be measured by measuring SPFSVAVSPSLDLSK (SEQ ID NO: 4), PRDX6 may be measured by measuring LSILYPATTGR (SEQ ID NO: 5), and CDH13 may be measured by measuring SIVVSPILIPENQR (SEQ ID NO: 6). The peptide YVQHTYR (SEQ ID NO: 1) typically has a mass / charge of 483.74 m / z, the peptide EIIDPVLDR (SEQ ID NO: 2) typically has a mass / charge of 535.30 m / z, the peptide GSAPPGPVPEGSIR (SEQ ID NO: 3) typically has a mass / charge of 660.85 m / z, the peptide SPFSVAVSPSLDLSK (SEQ ID NO: 4) typically has a mass / charge of 767.41 m / z, the peptide LSILYPATTGR (SEQ ID NO: 5) typically has a mass / charge of 596.34 m / z, and the peptide SIVVSPILIPENQR (SEQ ID NO: 6) typically has a mass / charge of 782.96 m / z.

[0040] Reference herein to any protein, polypeptide, or peptide may also encompass variants thereof. The term "variant" of a protein, polypeptide, or peptide refers to a protein, polypeptide, or peptide whose sequence (i.e., amino acid sequence) is substantially identical (i.e., largely but not completely identical) to the sequence of the recited protein or polypeptide, e.g., at least about 80% identical or at least about 85% identical, e.g., preferably at least about 90% identical, e.g., at least 91% identical, 92% identical, more preferably at least about 93% identical, e.g., at least 94% identical, even more preferably at least about 95% identical, e.g., at least 96% identical, even more preferably at least about 97% identical, e.g., at least 98% identical, and most preferably at least 99% identical. Preferably, the variant may exhibit such a degree of identity (i.e., overall sequence identity) to the recited protein, polypeptide, or peptide when the entire sequence of the recited protein, polypeptide, or peptide is referenced in a sequence alignment.

[0041] Sequence identity may be determined using suitable algorithms for performing sequence alignments and determination of sequence identity, as is known per se. Exemplary, but non-limiting, algorithms include those based on BLAST (Basic Local Alignment Search Tool), first described by Altschul et al., 1990 (J Mol Biol 215: 403-10), such as the "Blast2 alignment" algorithm described by Tatusova and Madden 1999 (FEMS Microbiol Lett 174: 247-250), using, for example, the published default settings or other suitable settings (e.g., for the BLASTN algorithm: cost to open a gap=5, cost to extend a gap=2, penalty for mismatch=-2, reward for match=1, gap x_dropoff=50, expectation=10.0, wordsize=28; or for the BLASTP algorithm: matrix=Blosum62, cost to open a gap=11, cost to extend a gap=1, expectation=10.0, wordsize=3).

[0042] A variant of a protein, polypeptide, or peptide may be a homolog (e.g., an ortholog or paralog) of said protein, polypeptide, or peptide. As used herein, the term "homology" generally refers to the structural similarity between two macromolecules, particularly two proteins or polypeptides, from the same or different taxa, where said similarity is due to a common ancestry.

[0043] When the present specification refers to or includes fragments and / or variants of proteins, polypeptides, or peptides, this preferably refers to variants and / or fragments that are "functional," i.e., that at least partially retain the biological activity or intended functionality of the respective protein, polypeptide, or peptide. Preferably, functional fragments and / or variants may retain at least about 20%, e.g., at least 30%, or at least about 40%, or at least about 50%, e.g., at least 60%, more preferably at least about 70%, e.g., at least 80%, even more preferably at least about 85%, even more preferably at least about 90%, and most preferably at least about 95% or even about 100% or more of the intended biological activity or functionality compared to the corresponding protein, polypeptide, or peptide.

[0044] A further aspect provides an in vitro method for diagnosing lung cancer in a subject, comprising detecting at least one, e.g., at least two, at least three, at least four, at least five, or all six biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample from the subject.

[0045] In a related aspect, provided herein is a method for treating a pulmonary arthritis, comprising: (a) detecting at least one, e.g., at least two, at least three, at least four, at least five, or all six biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample from the subject; (b) diagnosing lung cancer in the subject based on the detection of said at least one biomarker; and (c) administering a treatment for lung cancer to the subject diagnosed with lung cancer, for example, administering to the subject an effective amount of a therapeutic agent for lung cancer. A method for diagnosing and treating lung cancer in a subject, comprising:

[0046] Individual biomarkers from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13, especially ARHGDIB, have excellent predictive ability for distinguishing lung cancer patients from healthy subjects.

[0047] More specifically, the performance metrics of ARHGDIB were excellent: NPV of ≥ 0.90, sensitivity of ≥ 0.90, and AUC of ≥ 0.90 in the test cohort, and NPV of ≥ 0.90 and sensitivity of ≥ 0.90 in the validation cohort. Thus, in certain embodiments, an in vitro method for diagnosing lung cancer in a subject comprises detecting ARHGDIB and, optionally, at least one biomarker selected from the group consisting of TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample from the subject. However, ARHGDIB can be used in a panel with other known lung cancer biomarkers.

[0048] In certain embodiments, an in vitro method for diagnosing lung cancer in a subject comprises detecting in a biological sample from the subject at least two (e.g., two, three, four, five, or six), at least three (e.g., three, four, five, or six), at least four (e.g., four, five, or six), at least five (e.g., five or six), or all six biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13.

[0049] The present inventors have shown that a combination of biomarkers including CDH13 and / or ARHGDIB shows particularly good performance measures for distinguishing lung cancer patients from healthy subjects. Thus, in certain embodiments, an in vitro method for diagnosing lung cancer in a subject comprises detecting, in a biological sample from the subject, a first biomarker selected from the group consisting of CDH13 and ARHGDIB, and optionally, if the first biomarker is CDH13, at least one (e.g., one, two, three, four, or all five) additional biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, and PRDX6, or if the first biomarker is ARHGDIB, at least one (e.g., one, two, three, four, or all five) additional biomarkers selected from the group consisting of CDH13, TUBA4A, GSTO1, FLNA, and PRDX6.

[0050] Among biomarker combinations including CDH13 and / or ARHGDIB, 35 biomarker combinations demonstrate superior performance metrics: sensitivity of ≧0.90 and specificity of ≧0.90 or NPV of ≧0.90.

[0051] Thus, in a further particular embodiment, an in vitro method for diagnosing lung cancer in a subject comprises detecting, in a biological sample from the subject, a biomarker panel comprising the following biomarkers: (i) FLNA, TUBA4A, GSTO1, PRDX6, ARHGDIB, and CDH13; (ii) FLNA, TUBA4A, GSTO1, PRDX6, and ARHGDIB; (iii) FLNA, TUBA4A, GSTO1, ARHGDIB, and CDH13; and (iv) FLNA, TUBA4A, PRDX6, ARHGDIB, and CDH13. (v) FLNA, GSTO1, PRDX6, ARHGDIB, and CDH13; (vi) TUBA4A, GSTO1, PRDX6, ARHGDIB, and CDH13; (vii) FLNA, TUBA4A, GSTO1, and ARHGDIB; (viii) FLNA, TUBA4A, GSTO1, and CDH13; (ix) FLNA, TUBA4A, PRDX6, and ARHGDIB; (x) FLNA, TUBA4A, ARHGDIB, and CDH13; (xi) FLNA, GSTO1, PRDX6, and ARHGDIB; (xii) FLNA, GSTO1, ARHG (xiii) FLNA, PRDX6, ARHGDIB and CDH13; (xiv) TUBA4A, GSTO1, PRDX6 and ARHGDIB; (xv) TUBA4A, GSTO1, ARHGDIB and CDH13; (xvi) TUBA4A, PRDX6, ARHGDIB and CDH13; (xvii) GSTO1, PRDX6, ARHGDIB and CDH13; (xviii) FLNA, TUBA4A and ARHGDIB; (xix) FLNA, GSTO1 and ARHGDIB; (xx) FLNA, PRDX6 and ARHGD IB; (xxi) FLNA, ARHGDIB, and CDH13; (xxii) TUBA4A, GSTO1, and ARHGDIB; (xxiii) TUBA4A, GSTO1, and CDH13; (xxiv) TUBA4A, PRDX6, and ARHGDIB; (xxv) TUBA4A, PRDX6, and CDH13; (xxvi) TUBA4A, ARHGDIB, and CDH13; (xxvii) GSTO1, PRDX6, and ARHGDIB; (xxviii) GSTO1, ARHGDIB, and CDH13; (xxix) PRDX6, ARHGDIB, and CDH13;(xxx) FLNA and ARHGDIB; (xxxi) TUBA4A and ARHGDIB; (xxxii) GSTO1 and ARHGDIB; (xxxiii) GSTO1 and CDH13; (xxxiv) PRDX6 and ARHGDIB; or (xxxv) ARHGDIB and CDH13.

[0052] Of the 35 biomarker combinations that include CDH13 and / or ARHGDIB, 6 biomarker combinations show better performance metrics: sensitivity of ≧0.90, specificity of ≧0.90, and NPV of ≧0.90.

[0053] Therefore, in a further particular embodiment, an in vitro method for diagnosing lung cancer in a subject comprises detecting a biomarker panel comprising the biomarkers FLNA, TUBA4A, GSTO1 and CDH13; FLNA, ARHGDIB and CDH13; TUBA4A, GSTO1 and CDH13; TUBA4A, PRDX6 and CDH13; TUBA4A, ARHGDIB and CDH13; or GSTO1 and CDH13 in a biological sample from the subject.

[0054] In certain embodiments, an in vitro method for diagnosing lung cancer in a subject comprises detecting CDH13 and at least one, e.g., one, two, three, four, or all five, biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, and PRDX6 in a biological sample from the subject.

[0055] In a further particular embodiment, an in vitro method for diagnosing lung cancer in a subject comprises detecting in a biological sample from the subject a biomarker panel comprising the following biomarkers: (i) CDH13, FLNA, TUBA4A, GSTO1, PRDX6, and ARHGDIB; CDH13, FLNA, TUBA4A, GSTO1, and ARHGDIB; (ii) CDH13, FLNA, TUBA4A, PRDX6, and ARHGDIB; (i) CDH13, FLNA, GSTO1, PRDX6 and ARHGDIB; (iv) CDH13, TUBA4A, GSTO1, PRDX6 and ARHGDIB; (v) CDH13, FLNA, TUBA4A and GSTO1; (vi) CDH13, FLNA, TUBA4A and ARHGDIB; (vii) CDH13, FLNA, GSTO1 and ARHGDIB; (viii) CDH13, FLNA, PRDX6 and ARHGDIB; (ix) CDH13, TUBA4A, GSTO1 and ARHGDIB; (x) CDH13, TUBA4A, PRDX6 and ARHGDIB; (xi) CDH13, GSTO1, PRDX6 and ARHGDIB; (xii) CDH13, FLNA and ARHGDIB; (xiii) CDH13, TUBA4A and GSTO1; (xiv) CDH13, TUBA4A and PRDX6; (xv) CDH13, TUBA4A and ARHGDIB; (xvi) CDH13, GSTO1 and and ARHGDIB; (xvii) CDH13, PRDX6 and ARHGDIB; (xviii) CDH13 and GSTO1; (xvix) CDH13 and ARHGDIB, preferably FLNA, TUBA4A, GSTO1 and CDH13; FLNA, ARHGDIB and CDH13; TUBA4A, GSTO1 and CDH13; TUBA4A, PRDX6 and CDH13; TUBA4A, ARHGDIB and CDH13; or GSTO1 and CDH13.

[0056] In certain embodiments, an in vitro method for diagnosing lung cancer in a subject comprises detecting ARHGDIB, optionally in combination with at least one biomarker selected from the group consisting of TUBA4A, GSTO1, FLNA, PRDX6, and CDH13, in a biological sample from the subject. Similarly, the present application also provides an in vitro method for diagnosing lung cancer in a subject, comprising detecting GSTO1, optionally in combination with at least one, preferably two, biomarkers selected from the group consisting of TUBA4A, ARHGDIB, FLNA, and CDH13. The present application also provides an in vitro method for diagnosing lung cancer in a subject, comprising detecting CDH13, optionally in combination with at least one, preferably two, biomarkers selected from the group consisting of TUBA4A, ARHGDIB, FLNA, PRDX6, and GSTO1. A further particularly preferred embodiment is a method comprising detecting CDH13 and GSTO1, optionally in combination with one or more biomarkers selected from the group consisting of TUBA4A, ARHGDIB, FLNA and PRDX6.

[0057] In a further particular embodiment, an in vitro method for diagnosing lung cancer in a subject comprises detecting in a biological sample from the subject a biomarker panel comprising the following biomarkers: (i) ARHGDIB, FLNA, TUBA4A, GSTO1, PRDX6, and CDH13; (ii) ARHGDIB, FLNA, TUBA4A, GSTO1, and PRDX6; (iii) ARHGDIB, FLNA, TUBA4A, GSTO1, and CDH13; (iv) ARHGDIB, FLNA, TUBA4A, PRDX6, and CDH13; (v) A (vi) ARHGDIB, FLNA, GSTO1, PRDX6, and CDH13; (vii) ARHGDIB, FLNA, TUBA4A, and GSTO1; (viii) ARHGDIB, FLNA, TUBA4A, and PRDX6; (ix) ARHGDIB, FLNA, TUBA4A, and CDH13; (x) ARHGDIB, FLNA, GSTO1, and PRDX6; (xi) ARHGDIB, FLNA, GSTO1, and CDH13; (xii) ARHGDIB, FLNA, and PRDX6 and CDH13; (xiii) ARHGDIB, TUBA4A, GSTO1 and PRDX6; (xiv) ARHGDIB, TUBA4A, GSTO1 and CDH13; (xv) ARHGDIB, TUBA4A, PRDX6 and CDH13; (xvi) ARHGDIB, GSTO1, PRDX6 and CDH13; (xvii) ARHGDIB, FLNA and TUBA4A; (xviii) ARHGDIB, FLNA and GSTO1; (xix) ARHGDIB, FLNA and PRDX6; (xx) ARHGDIB, FLNA and CDH13; (xxi ) ARHGDIB, TUBA4A and GSTO1; (xxii) ARHGDIB, TUBA4A and PRDX6; (xxiii) ARHGDIB, TUBA4A and CDH13; (xxiv) ARHGDIB, GSTO1 and PRDX6; (xxv) ARHGDIB, GSTO1 and CDH13; (xxvi) ARHGDIB, PRDX6 and CDH13; (xxvii) ARHGDIB and FLNA; (xxviii) ARHGDIB and TUBA4A; (xxvix) ARHGDIB and GSTO1; (xxx) ARHGDIB and PRDX6;or (xxxi) ARHGDIB and CDH13, preferably ARHGDIB, FLNA and CDH13; or ARHGDIB, TUBA4A and CDH13. Alternatively contemplated panels are those further described for the methods above;

[0058] In addition to powerful individual biomarkers, we identified a six-protein panel that exhibited excellent discriminatory power in logistic regression models, especially when compared with the commercially available Xpresys® Lung (XL) test (Biodesix, Boulder, CO) and univariate models: lowest AIC (30.876), highest AUC (0.999), highest PPV (0.992), highest NPV (0.989), highest specificity (0.989), and highest sensitivity (0.992) in the test cohort. In addition, the six-protein panel enables noninvasive detection of lung cancer regardless of disease stage (including stage I tumors). Therefore, the six-protein panel has particularly high potential as a screening tool.

[0059] Therefore, in certain embodiments, an in vitro method for diagnosing lung cancer in a subject comprises detecting ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample from the subject.

[0060] A molecule or analyte, e.g., a marker, peptide, polypeptide, or protein, is "detected" in a sample when the presence or absence and / or amount of said molecule or analyte is detected or determined in the sample, preferably to the substantial exclusion of other molecules and analytes.

[0061] Depending on factors that can be assessed and determined by one of skill in the art, such as the type of biomarker (e.g., peptide, polypeptide, or protein), the type of sample (e.g., whole blood, plasma, serum, lung tissue biopsy, tissue section), the expected abundance of the biomarker in the sample, the type, robustness, sensitivity, and / or specificity of the detection method used to detect the biomarker, among others, the biomarker may be measured directly in the sample, or the sample may be subjected to one or more processing steps aimed at achieving an adequate measurement of the biomarker.

[0062] By way of example, a sample may be subjected to one or more isolation or separation steps aimed at isolating biomarkers from the sample or preparing a fraction of the sample enriched in biomarkers. For example, if the biomarkers are peptides, polypeptides, or proteins, any known protein purification technique may be applied to the sample to isolate peptides, polypeptides, and proteins therefrom. Non-limiting examples of methods for purifying peptides, polypeptides, or proteins may include chromatography, preparative electrophoresis, centrifugation, precipitation, affinity purification, etc.

[0063] As used herein, the term "purified" with respect to a marker, peptide, polypeptide, or protein does not require absolute purity. Instead, it indicates that such marker, peptide, polypeptide, or protein is in a discrete environment in which its abundance (conveniently expressed in terms of mass or weight or concentration) relative to other analytes is greater than in a biological sample. A discrete environment refers to a single medium, e.g., a single solution, gel, precipitate, lyophilized material, etc. Purified proteins, polypeptides, or peptides may be obtained by known methods, including, for example, laboratory or recombinant synthesis, chromatography, preparative electrophoresis, centrifugation, precipitation, affinity purification, etc.

[0064] The purification marker, peptide, polypeptide, or protein may preferably constitute 10% by weight, more preferably 50% by weight, for example 60% by weight, even more preferably 70% by weight, for example 80% by weight, even more preferably 90% by weight, for example 95% by weight, 96% by weight, 97% by weight, 98% by weight, 99% by weight, or even 100% by weight of the protein content of the discrete environment. Protein content may be determined, for example, by the Lowry method (Lowry et al., 1951. J Biol Chem 193: 265), optionally as described in Hartree 1972 (Anal Biochem 48: 422-427). Peptide, polypeptide, or protein purity may be determined by SDS-PAGE under reducing or non-reducing conditions using Coomassie blue or, preferably, silver staining.

[0065] Any existing, available, or conventional separation, detection, and quantification method may be used herein to measure the presence or absence (e.g., readout present vs. absence; or detectable amount vs. undetectable amount) and / or amount (e.g., readout is absolute amount or relative amount, e.g., absolute concentration or relative concentration) of a marker, peptide, polypeptide, or protein in a sample.

[0066] For example, such methods may include mass spectrometry, biochemical assays, immunoassays, or chromatography, or a combination thereof.

[0067] The term "immunoassay" generally refers to any method known for detecting one or more molecules or analytes of interest in a sample, where the specificity of the immunoassay for the molecules or analytes of interest is conferred by the specific binding between a specific binding agent (typically, but not limited to, an antibody) and the molecule or analyte of interest. Immunoassay techniques include, but are not limited to, immunohistochemistry, direct ELISA (enzyme-linked immunosorbent assay), indirect ELISA, sandwich ELISA, competitive ELISA, multiplex ELISA, radioimmunoassay (RIA), ELISPOT technology, and other similar techniques known in the art.

[0068] In general, any mass spectrometry (MS) technique (e.g., tandem mass spectrometry, MS / MS; or post source decay, TOF MS) that can provide accurate information about the mass of a peptide, and preferably also accurate information about the fragmentation and / or (partial) amino acid sequence of a selected peptide, is useful herein. Suitable peptide MS and MS / MS techniques and systems are known per se (see, for example, Methods in Molecular Biology, vol. 146: "Mass Spectrometry of Proteins and Peptides", by Chapman, ed., Humana Press 2000, ISBN 089603609x; Biemann 1990. Methods Enzymol 193: 455-79; or Methods in Enzymology, vol. 402: "Biological Mass Spectrometry", by Burlingame, ed., Academic Press 2005, ISBN 9780121828073) and may be used herein. MS peptide analysis methods may advantageously be combined with upstream peptide or protein separation or fractionation methods, such as chromatography. Data obtained from MS may be processed using software for proteomics and / or metabolomics data analysis known in the art, for example, Skyline software (v19.1.0.193).

[0069] Chromatography may also be used to measure biomarkers. As used herein, the term "chromatography" encompasses methods for separating chemicals and is referred to as such and widely available in the art. Chromatography, as used herein, may preferably be column chromatography (i.e., where a stationary phase is deposited or packed in a column), preferably liquid chromatography, and even more preferably HPLC. Details of chromatography are well known in the art, but for further guidance see, for example, Meyer M., 1998, ISBN: 047198373X, and "Practical HPLC Methodology and Applications", Bidlingmeyer, BA, John Wiley & Sons Inc., 1993.

[0070] Measurement of biomarkers in the present disclosure may involve the use of additional peptide or polypeptide separation, identification, or quantification methods, optionally in conjunction with any of the analytical methods described above, including, but not limited to, chemical extraction partitioning, isoelectric focusing (IEF), including capillary isoelectric focusing (CIEF), capillary isotachophoresis (CITP), capillary electrochromatography (CEC), one-dimensional polyacrylamide gel electrophoresis (PAGE), two-dimensional polyacrylamide gel electrophoresis (2D-PAGE), capillary gel electrophoresis (CGE), capillary zone electrophoresis (CZE), micellar electrokinetic chromatography (MEKC), free-flow electrophoresis (FFE), and the like.

[0071] The inventors have found that each of the biomarkers ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 can be detected using antibody-independent methods, more particularly parallel reaction monitoring (PRM)-based mass spectrometry as described, for example, in Bourmaud A, Gallien S, Domon B. Parallel reaction monitoring using quadrupole-Orbitrap mass spectrometer: Principle and applications. Proteomics. 2016;16: 2146-59.

[0072] In certain embodiments, at least one biomarker is detected using mass spectrometry, preferably liquid chromatography-mass spectrometry (LC-MS). For example, LC-MS may be performed using an LC-MS setup consisting of a Dionex U3000 RSLC liquid chromatography system operated in column switching mode coupled with a Q Exactive Plus mass spectrometer.

[0073] Those skilled in the art will appreciate that prior to mass spectrometry, the biological sample may be depleted and processed, for example (ultra)filtered, denatured, alkylated, digested and / or deglycosylated.

[0074] In certain embodiments, detecting at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in the subject's biological sample comprises measuring the amount or expression level of the at least one biomarker in the subject's biological sample.

[0075] The terms "quantity," "amount," and "level" are synonymous and generally well understood in the art. These terms, as used herein, may refer, inter alia, to the absolute quantification of a molecule or analyte in a sample, or the relative quantification of a molecule or analyte in a sample, i.e., relative to another value, e.g., relative to a reference value taught herein, or relative to a range of values ​​representing baseline expression of a biomarker. These values ​​or ranges can be obtained from a single patient or a group of patients.

[0076] The absolute amount of a molecule or analyte in a sample may advantageously be expressed as a weight or molar amount, or more commonly as a concentration, for example weight / volume or mole / volume.

[0077] The relative amount of a molecule or analyte in a sample may advantageously be expressed as an increase or decrease, or as a fold increase or decrease, relative to another value, e.g., a reference value as taught herein. Performing a relative comparison between a first parameter and a second parameter (e.g., a first amount and a second amount) may, but need not, require first determining absolute values ​​for the first and second parameters. For example, a measurement method may generate quantifiable readouts (e.g., signal intensities) of the first and second parameters, the readouts being a function of the values ​​of the parameters, which can be directly compared to generate relative values ​​for the first parameter versus the second parameter, without the actual need to first convert the readouts to absolute values ​​for the respective parameters.

[0078] The terms "amount" and "expression level" of said at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 are used interchangeably herein to refer to the absolute and / or relative quantification, concentration level or amount of any such product in a sample. Preferably, the biomarker is a protein, and the term "expression level" refers to the "protein expression level."

[0079] The inventors have found that the biomarkers TUBA4A, GSTO1, FLNA, PRDX6, ARHGDIB and / or CDH13 are significantly differentially expressed in biological samples, preferably plasma samples, of patients suffering from lung cancer compared to healthy subjects.

[0080] In certain embodiments, the methods taught herein may comprise comparing the amount or expression level of the at least one biomarker selected from the group consisting of TUBA4A, GSTO1, FLNA, PRDX6, ARHGDIB, and CDH13 in a biological sample from a subject to a reference or threshold amount or expression level of the at least one biomarker selected from the group consisting of TUBA4A, GSTO1, FLNA, PRDX6, ARHGDIB, and CDH13, wherein the reference or threshold amount or expression level of the at least one biomarker may represent a known diagnosis of lung cancer.

[0081] In certain embodiments, the method comprises: (a) measuring the amount or expression level of at least one, e.g., at least two, at least three, at least four, at least five, or all six, biomarkers selected from the group consisting of TUBA4A, GSTO1, FLNA, PRDX6, ARHGDIB, and CDH13 in a biological sample from the subject; (b) comparing the amount or expression level of said at least one biomarker measured in step (a) with a reference value or threshold value representing a known diagnosis of lung cancer; (c) detecting deviation or non-deviation of the amount or expression level of said at least one biomarker measured in step (a) from said reference value or threshold; and (d) attributing said step of finding deviation or absence of deviation to a specific diagnosis of lung cancer. Includes:

[0082] In certain embodiments, the method for diagnosing and treating lung cancer in a subject comprises: (a) measuring the amount or expression level of at least one, e.g., at least two, at least three, at least four, at least five, or all six, biomarkers selected from the group consisting of TUBA4A, GSTO1, FLNA, PRDX6, ARHGDIB, and CDH13 in a biological sample from the subject; (b) comparing the amount or expression level of the at least one biomarker measured in (a) with a reference value or threshold value representing a known diagnosis of lung cancer; (c) diagnosing lung cancer in the subject or diagnosing that the subject is in need of treatment for lung cancer if the amount or expression level of the at least one biomarker measured in (a) deviates from the reference value or threshold; and (d) administering a treatment or procedure for lung cancer to the subject diagnosed with lung cancer, for example, administering to the subject an effective amount of a therapeutic agent for lung cancer. Includes:

[0083] The inventors have found that the levels of TUBA4A, GSTO1, FLNA, PRDX6 and ARHGDIB are increased and the levels of CDH13 are decreased in biological samples, preferably plasma samples, of lung cancer patients (irrespective of the stage of lung cancer) compared to healthy subjects.

[0084] Thus, in certain embodiments, where a reference value or threshold value represents a subject or group of subjects who are not afflicted with lung cancer: - the amount or expression level of ARHGDIB in the subject's sample is increased compared to a reference value or threshold (e.g., by at least about 10% (about 1.1-fold or more), or at least about 20% (about 1.2-fold or more), or at least about 30% (about 1.3-fold or more), or at least about 40% (about 1.4-fold or more), or at least about 50% (about 1.5-fold or more), or at least about 60% (about 1.6-fold or more), or at least about 70% (about 1.7-fold or more), or at least about 90% (about 1.9-fold or more), or at least about 100% (about 2-fold or more), or at least about 400% (about 5-fold or more), or at least about 900% (about 10-fold or more)); - the amount or expression level of TUBA4A in the subject's sample is increased compared to a reference value or threshold (e.g., by at least about 10% (about 1.1-fold or more), or at least about 20% (about 1.2-fold or more), or at least about 30% (about 1.3-fold or more), or at least about 40% (about 1.4-fold or more), or at least about 50% (about 1.5-fold or more), or at least about 60% (about 1.6-fold or more), or at least about 70% (about 1.7-fold or more), or at least about 90% (about 1.9-fold or more), or at least about 100% (about 2-fold or more), or at least about 200% (about 3-fold or more), or at least about 300% (about 4-fold or more), or at least about 400% (about 5-fold or more)); - the amount or expression level of GSTO1 in the subject's sample is increased compared to a reference value or threshold (e.g., by at least about 10% (about 1.1-fold or more), or at least about 20% (about 1.2-fold or more), or at least about 30% (about 1.3-fold or more), or at least about 40% (about 1.4-fold or more), or at least about 50% (about 1.5-fold or more), or at least about 60% (about 1.6-fold or more), or at least about 70% (about 1.7-fold or more), or at least about 90% (about 1.9-fold or more), or at least about 100% (about 2-fold or more), or at least about 200% (about 3-fold or more), or at least about 300% (about 4-fold or more), or at least about 400% (about 5-fold or more), or at least about 900% (about 10-fold or more)); - the amount or expression level of PRDX6 in the subject's sample is increased compared to a reference value or threshold (e.g., by at least about 10% (about 1.1-fold or more), or at least about 20% (about 1.2-fold or more), or at least about 30% (about 1.3-fold or more), or at least about 40% (about 1.4-fold or more), or at least about 50% (about 1.5-fold or more), or at least about 60% (about 1.6-fold or more), or at least about 70% (about 1.7-fold or more), or at least about 90% (about 1.9-fold or more), or at least about 100% (about 2-fold or more), or at least about 200% (about 3-fold or more), or at least about 300% (about 4-fold or more), or at least about 400% (about 5-fold or more), or at least about 900% (about 10-fold or more)); - the amount or expression level of FLNA in the subject's sample is increased compared to a reference value or threshold (e.g., by at least about 10% (about 1.1-fold or more), or at least about 20% (about 1.2-fold or more), or at least about 30% (about 1.3-fold or more), or at least about 40% (about 1.4-fold or more), or at least about 50% (about 1.5-fold or more), or at least about 60% (about 1.6-fold or more), or at least about 70% (about 1.8-fold or more), an increase of at least about 90% (about 1.9-fold or more), or at least about 100% (about 2-fold or more), or at least about 200% (about 3-fold or more), or at least about 300% (about 4-fold or more), or at least about 400% (about 5-fold or more), or at least about 900% (about 10-fold or more), or at least about 1400% (about 15-fold or more), or at least about 1900% (about 20-fold or more); and / or - A decrease in the amount or expression level of CDH13 in the subject's sample compared to a reference value or threshold (e.g., a decrease of at least about 10% (about 0.9-fold or more), or at least about 20% (about 0.8-fold or more), or at least about 30% (about 0.7-fold or more), or at least about 40% (about 0.6-fold or more), or at least about 50% (about 0.5-fold or more), or at least about 60% (about 0.4-fold or more), or at least about 70% (about 0.3-fold or more), or at least about 80% (about 0.2-fold or less), or at least about 90% (about 0.1-fold or less)). allows for the diagnosis of lung cancer in a subject.

[0085] In certain embodiments, the method comprises measuring the amount or expression level of TUBA4A, GSTO1, FLNA, PRDX6, ARHGDIB, and CDH13 in a biological sample from a subject, and where the reference value or threshold value represents a subject or group of subjects not afflicted with lung cancer: - the amount or expression level of ARHGDIB is elevated in the subject's sample compared to a reference value or threshold; - the amount or expression level of TUBA4A is elevated in the subject's sample compared to a reference value or threshold; - the amount or expression level of GSTO1 is elevated in the subject's sample compared to a reference value or threshold value; - the amount or expression level of PRDX6 is elevated in the subject's sample compared to a reference value or threshold value; - the amount or expression level of FLNA is elevated in the subject's sample compared to a reference value or threshold; and - if the amount or expression level of CDH13 is decreased in the subject's sample compared to a reference value or threshold value; The subject is diagnosed with lung cancer.

[0086] Comparison with a standard or threshold

[0087] The distinct reference values ​​or thresholds may represent a diagnosis of lung cancer versus the absence of lung cancer (e.g., healthy or recovered from lung cancer). In another example, the distinct reference values ​​or thresholds may represent diagnoses of lung cancer of varying severity.

[0088] In yet another example, a separate reference value or threshold may represent a subject's need for therapeutic treatment of lung cancer versus a subject's need for therapeutic treatment of lung cancer.

[0089] Such comparisons may generally involve any means for determining the presence or absence of at least one difference between the compared values ​​or profiles, and optionally the magnitude of such difference. Comparisons may include visual inspection, arithmetic, or statistical comparisons of measurements. Such statistical comparisons include, but are not limited to, applying an algorithm. When values ​​or biomarker profiles include at least one standard, the comparison to determine differences in said values ​​or biomarker profiles may also include measurements of these standards, such that measurements of the biomarkers are correlated with measurements of an internal standard.

[0090] The reference value or threshold value for the amount or expression level of said at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 may be established according to known procedures previously adopted for other biomarkers.

[0091] For example, a reference value for the amount of said at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 for a particular diagnosis of lung cancer as taught herein may be established by determining the amount or expression level of said at least one biomarker in a biological sample of an individual or a population (e.g., group) of individuals characterized by said particular diagnosis of said disease or pathological condition. Such a population may include, but is not limited to, ≧2, ≧10, ≧100, or even several hundred or more individuals.

[0092] Thus, as an illustrative example, a reference value for the amount or expression level of said at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 for diagnosing lung cancer versus not having such a disease or pathological condition may be established by determining the amount or expression level of said at least one biomarker in a sample from an individual or population of individuals diagnosed with lung cancer or not having lung cancer, respectively (e.g., based on other sufficiently conclusive means, such as clinical signs and symptoms, imaging, etc.).

[0093] In such cases, measuring the amount or expression level of at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 for the same patient at different time points allows for continuous monitoring of the patient's condition and may lead to a prediction of the worsening or improvement of the patient's pathological condition for a given disease or pathological condition taught herein. Tools such as the kits described herein below can be developed to ensure this type of monitoring. One or more reference values, thresholds, or ranges for the amount or expression level of the at least one biomarker associated with the development of lung cancer can be determined, for example, in the subject in advance or during a monitoring process over a certain period of time. Alternatively, these reference values ​​or ranges can be established through a dataset of several patients with highly similar disease phenotypes (e.g., subjects who have not developed lung cancer). A sudden deviation of the level of the at least one biomarker from the reference value, threshold, or range can predict (e.g., at home or in a clinic) a worsening of the patient's pathological condition before (often serious) symptoms can actually be felt or observed. Monitoring may be applied during the course of a subject's medical treatment, preferably medical treatment aimed at alleviating the disease or pathological condition so monitored. Such monitoring may, for example, be involved in determining whether a patient can be discharged from the hospital, requires a change in treatment, or requires further hospitalization.

[0094] Therefore, also provided herein is an in vitro method for monitoring lung cancer in a subject, comprising detecting at least one, for example at least two, at least three, at least four, at least five or all six biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 in a biological sample from the subject.

[0095] In one embodiment, a reference value or threshold value as contemplated herein may convey an absolute amount of a biomarker as contemplated herein. In another embodiment, the amount of a biomarker in a tested subject's sample may be determined directly compared to the reference value (e.g., in terms of increase or decrease, or fold increase or decrease). Advantageously, this may allow for comparison of the amount or expression level of a biomarker in a subject's sample to the reference value (in other words, measuring the relative amount of a biomarker in a subject's sample relative to the reference value) without first having to determine the absolute amount of each of the biomarkers.

[0096] As explained, the method, use or product may comprise finding a deviation or no deviation between the amount or expression level of at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 taught herein measured in a sample from a subject and a given reference value or threshold value.

[0097] A "deviation" of a first value from a second value or a "difference" between a first value and a second value may generally encompass any direction (e.g., increase: first value > second value; or decrease: first value < second value) and any degree of change.

[0098] For example, a deviation or difference may include, but is not limited to, a decrease in the first value relative to the second value being compared of at least about 10% (about 0.9-fold or less), or at least about 20% (about 0.8-fold or less), or at least about 30% (about 0.7-fold or less), or at least about 40% (about 0.6-fold or less), or at least about 50% (about 0.5-fold or less), or at least about 60% (about 0.4-fold or less), or at least about 70% (about 0.3-fold or less), or at least about 80% (about 0.2-fold or less), or at least about 90% (about 0.1-fold or less).

[0099] For example, a deviation or difference may include, but is not limited to, an increase in a first value relative to the second value being compared, such as, but not limited to, at least about 10% (about 1.1-fold or more), or at least about 20% (about 1.2-fold or more), or at least about 30% (about 1.3-fold or more), or at least about 40% (about 1.4-fold or more), or at least about 50% (about 1.5-fold or more), or at least about 60% (about 1.6-fold or more), or at least about 70% (about 1.7-fold or more), or at least about 80% (about 1.8-fold or more), or at least about 90% (about 1.9-fold or more), or at least about 100% (about 2-fold or more), or at least about 150% (about 2.5-fold or more), or at least about 200% (about 3-fold or more), or at least about 500% (about 6-fold or more), or at least about 700% (about 8-fold or more).

[0100] Preferably, deviation or difference may refer to a statistically significant observed change. For example, deviation or difference may refer to an observed change that is outside the error range of a reference value in a given population (e.g., expressed by a standard deviation or standard error, or a predetermined multiple thereof, for example, ±1×SD or ±2×SD or ±3×SD, or ±1×SE or ±2×SE or ±3×SE). Deviation or difference may also refer to a value that is outside the reference range defined by values ​​in a given population (e.g., outside a range that includes ≧40%, ≧50%, ≧60%, ≧70%, ≧75%, or ≧80%, or ≧85%, or ≧90%, or ≧95%, or even ≧100% of the value in the population).

[0101] In further embodiments, deviation or divergence may be determined when the observed change exceeds a given threshold or cutoff value. Such threshold or cutoff value may be selected as generally known in the art to provide a selected accuracy, sensitivity and / or specificity of the prediction method, for example, at least 50%, or at least 60%, or at least 70%, or at least 80%, or at least 85%, or at least 90%, or at least 95% accuracy, sensitivity and / or specificity.

[0102] For example, receiver operating characteristic (ROC) curve analysis can be used to select optimal threshold or cut-off values ​​for the amount of a given biomarker for clinical use of the diagnostic test based on acceptable overall accuracy, sensitivity and / or specificity, or related performance measures that are known per se, such as positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), negative likelihood ratio (LR-), Youden index, etc.

[0103] For example, optimal threshold or cutoff values ​​may be selected for each individual biomarker as extrema of a receiver operating characteristic (ROC) curve, i.e., the point of maximum distance to the diagonal, as described in Robin X., PanelomiX: a threshold-based algorithm to create panels of biomarkers, 2013, Translational Proteomics, 1(1):57-64.

[0104] Those skilled in the art will understand that it is not relevant to give an exact threshold or cut-off value: relevant threshold or cut-off values ​​can be obtained by correlating the sensitivity and specificity with the sensitivity / specificity for any threshold or cut-off value.

[0105] It is up to the diagnostic technician to decide what level of positive predictive value / negative predictive value / sensitivity / specificity is desirable and how much loss of positive or negative predictive value is acceptable. The threshold or cutoff level selected may depend on other diagnostic parameters used by the diagnostic technician in combination with the method.

[0106] The method, use or product may further comprise attributing the presence or absence of lung cancer to finding a deviation or no deviation between the amount or expression level of at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 taught herein measured in a biological sample of the subject and a given reference value or threshold.

[0107] In the methods provided herein, the observation of a deviation between the amount or expression level of the at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a subject's biological sample and a reference value or threshold value can lead to a conclusion that the diagnosis of lung cancer in the subject is different from that represented by the reference value or threshold value. Similarly, if no deviation is found between the amount or expression level of the at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a subject's sample and a reference value or threshold value, such absence of deviation can lead to a conclusion that the diagnosis of lung cancer in the subject is substantially the same as that represented by the reference value or threshold value.

[0108] In a particular embodiment, the reference value or threshold used in the method according to the present invention is determined from biological samples of a subject or group of subjects not suffering from lung cancer, such as a healthy subject or group of healthy subjects, which may be at high or average risk of developing lung cancer, for example, the healthy subject or group of healthy subjects may be a smoker or group of smokers. The amount or expression level of the at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample from a subject (preferably, but not limited to, a subject with lung cancer) may be elevated (e.g., when the biomarker is ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13) or decreased (e.g., when the biomarker is CDH13) compared to (i.e., relative to) a reference value or threshold value representing the amount or expression level of the at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample from a subject or group of subjects not afflicted with lung cancer, e.g., a healthy subject or group of healthy subjects. Such elevated or decreased amount or expression level may allow for the diagnosis of lung cancer in the subject.

[0109] To facilitate physician use of the diagnostic methods taught herein, the inventors have adopted a threshold-based approach to assign cutoff values / biomarkers, which are extrema of the receiver operating characteristic (ROC) curve, i.e., the points of maximum distance to the diagonal, as described in Robin X., PanelomiX: a threshold-based algorithm to create panels of biomarkers, 2013, Translational Proteomics, 1(1):57-64.

[0110] In certain embodiments of the methods taught herein, where a particularly high overall diagnostic accuracy of the method is desired (including high sensitivity (e.g., 95% or greater) and / or specificity (e.g., 95% or greater)), and the method includes detecting ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample, the reference value or threshold for TUBA4A is 1.69, the reference value or threshold for GSTO1 is 5.36, the reference value or threshold for FLNA is 0.48, the reference value or threshold for PRDX6 is 6.0, the reference value or threshold for ARHGDIB is 0.51, and the reference value or threshold for CDH13 is 69.83.

[0111] In further particular embodiments of the methods taught herein, where a particularly high sensitivity (e.g., 95% or greater) and / or specificity (e.g., 95% or greater) of the method is desired and the method includes detecting ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample, the reference value or threshold for TUBA4A is 0.19, the reference value or threshold for GSTO1 is 5.36, the reference value or threshold for FLNA is 0.48, the reference value or threshold for PRDX6 is 4.04, the reference value or threshold for ARHGDIB is 0.51, and the reference value or threshold for CDH13 is 148.16.

[0112] To make the diagnostic methods taught herein even more practical for physicians, the inventors also employ a single score per sample of a subject at risk of lung cancer to classify the sample as lung cancer or healthy, wherein the single score represents the number of biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 that are differentially expressed in the subject's sample compared to healthy individuals.

[0113] Thus, in certain embodiments, the methods taught herein comprise: (a) measuring the amount or expression level of at least three, at least four, at least five, or all six biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample of the subject; (b) calculating a score based on the amounts or expression levels of the at least three, at least four, at least five, or all six biomarkers measured in step (a); (c) comparing the score calculated in step (b) with a reference score or threshold score; and (d) diagnosing the subject with lung cancer if the score calculated in step (b) is equal to or greater than the reference score or threshold score. Includes:

[0114] In certain embodiments, the methods taught herein comprise: (a) measuring the amount or expression level of at least three, at least four, at least five, or all six biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample of the subject; (b) comparing the amount or expression level of the biomarker measured in step (a) with a reference value or threshold; (c) detecting deviation or non-deviation of the amount or expression level of said biomarker measured in step (a) from said reference value or threshold; (d) calculating a score representing the number of biomarkers measured in step (a) that are found in step (c) to deviate from said reference value or threshold; (e) comparing the score calculated in step (b) with a reference score or threshold score; and (d) diagnosing the subject with lung cancer if the score calculated in step (b) is equal to or greater than the second standard score or threshold score. Includes:

[0115] In certain embodiments, the score representing the number of biomarkers measured in step (a) that are found to deviate from said reference value or threshold in step (c) can be expressed as follows:

[0116]

number

[0117] (In the formula, S p is the score for patient p, n is the number of biomarkers measured in step (a), and X ip is the amount or expression level of the i-th biomarker in subject p, and T i is a reference value or threshold for the i-th biomarker, and I(x) is a characteristic function that takes the value of 1 if x = true and 0 otherwise, as described, for example, in Robin X., PanelomiX: a threshold-based algorithm to create panels of biomarkers, 2013, Translational Proteomics, 1(1):57-64.

[0118] In certain embodiments, where a particularly high overall diagnostic accuracy of the method is desired (including high sensitivity (e.g., 95% or greater) and / or specificity (e.g., 95% or greater)), the methods taught herein may be: (a) measuring the amount or expression level of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a sample from a subject; (b) comparing the amount or expression level of the biomarker measured in step (a) with a reference value or threshold; wherein the reference value or threshold for TUBA4A is 1.69, the reference value or threshold for GSTO1 is 5.36, the reference value or threshold for FLNA is 0.48, the reference value or threshold for PRDX6 is 6.0, the reference value or threshold for ARHGDIB is 0.51, and the reference value or threshold for CDH13 is 69.83; (c) detecting deviation or non-deviation of the amount or expression level of said biomarker measured in step (a) from said reference value or threshold; (d) diagnosing the subject with lung cancer if the amounts or expression levels of at least three biomarkers measured in step (a) deviate from said reference values ​​or thresholds. Includes:

[0119] In certain embodiments, where particularly high sensitivity (e.g., 95% or greater) and / or specificity (e.g., 95% or greater) of the method is desired, the methods taught herein may be (a) measuring the amount or expression level of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a sample from a subject; (b) comparing the amount or expression level of the biomarker measured in step (a) with a reference value or threshold value; wherein the reference value or threshold for TUBA4A is 0.19, the reference value or threshold for GSTO1 is 5.36, the reference value or threshold for FLNA is 0.48, the reference value or threshold for PRDX6 is 4.04, the reference value or threshold for ARHGDIB is 0.51, and / or the reference value or threshold for CDH13 is 148.16; (c) detecting deviation or non-deviation of the amount or expression level of said biomarker measured in step (a) from said reference value or threshold; (d) diagnosing the subject with lung cancer if the amounts or expression levels of at least five biomarkers measured in step (a) deviate from said reference values ​​or thresholds. Includes:

[0120] The terms "sample" or "biological sample," as used herein, include any biological specimen obtained or isolated from a subject. Samples may include, but are not limited to, organ tissue (i.e., lung tissue), whole blood, plasma, serum, whole blood cells, red blood cells, white blood cells (e.g., peripheral blood mononuclear cells), saliva, urine, stool (i.e., feces), tears, sweat, sebum, nipple aspirate, ductal lavage, tumor exudate, synovial fluid, cerebrospinal fluid, lymphatic fluid, fine needle aspirate, amniotic fluid, any other bodily fluid, cell lysate, cell secretion product, inflammatory fluid, semen, and vaginal secretion. Preferably, the sample may be readily obtained by a minimally invasive method, such as blood sampling, which allows for removal / isolation / provision of the sample from the subject. The term "tissue," as used herein, includes cells of an organ, but also encompasses all types of cells from the human body, including blood and other bodily fluids listed above.

[0121] In certain embodiments, the sample is a bodily fluid sample; preferably a bodily fluid sample selected from the group consisting of plasma, serum, whole blood, urine, tissue lysate, cerebrospinal fluid (CSF), saliva, and sweat.

[0122] Identifying solid tumors by simple blood analysis has been a long-standing goal in cancer research, as detection of cancer during routine screening can provide patients with prompt treatment solutions. When combined with highly accurate measurement methods, blood samples may represent an ideal source of cancer diagnosis, being minimally invasive and easily collected.

[0123] Although blood-based early diagnosis of cancer remains a challenge, several proteins circulating in the blood are useful for monitoring treatment response and / or tumor recurrence. To date, only prostate-specific antigen has been routinely measured in blood for early cancer diagnosis. Recently, Cohen and colleagues published results from CancerSeek, a blood test that assesses the presence of eight protein markers and 1,933 genetic mutations in cell-free DNA to diagnose common solid tumors. While the results were promising, the usefulness of this assay to advance cancer management has yet to achieve widespread adoption. The median sensitivity of CancerSeek in lung cancer was approximately 59%, the second lowest among the eight cancer types examined.

[0124] The present inventors have found that each of the biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 can be detected in plasma, for example, by mass spectrometry, and that detection of at least one, preferably at least two, and more preferably at least six biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a plasma sample makes it possible to distinguish lung cancer patients from subjects not suffering from lung cancer. Therefore, the method of the present invention allows for the diagnosis of lung cancer in a minimally invasive or non-invasive manner, as the method taught herein can be performed on body fluid samples, particularly plasma samples, which can be easily collected.

[0125] In certain embodiments, the sample is a plasma sample.

[0126] The terms "diagnosing" or "diagnosis" generally refer to the process or act of recognizing, determining, or concluding a disease or pathological condition in a subject based on symptoms and signs and / or from the results of various diagnostic procedures (e.g., from knowing the presence, absence, and / or amount of one or more biomarkers characteristic of the diagnosed disease or pathological condition). As used herein, "diagnosis" of a disease or pathological condition taught herein in a subject can particularly mean that the subject has such a disease or pathological condition, and thus is diagnosed with such a disease or pathological condition. "Diagnosis" of the absence of a disease or pathological condition taught herein in a subject can particularly mean that the subject does not have such a condition, and thus is diagnosed as not having such a condition. A subject may also be diagnosed as not having such a condition despite exhibiting one or more conventional symptoms or signs associated with such a condition.

[0127] The term "lung cancer," as used herein, refers to malignant tumors that arise in the lungs and includes both small cell lung cancer and non-small cell lung cancer, including adenocarcinoma, squamous cell carcinoma, and large cell carcinoma.

[0128] The present inventors have identified a biomarker that has strong diagnostic performance in lung cancer, especially in early lung cancer, and allows for the diagnosis of lung cancer regardless of the stage (including stage I tumors). As a result, this biomarker is a very interesting tool for screening patients for lung cancer, even at an early stage of the disease.

[0129] In certain embodiments, the lung cancer is stage I (e.g., stage IA or IB), stage II (e.g., stage IIA or IIB), stage III (e.g., stage IIIA, IIIB, or IIIC), or stage IV (e.g., stage IVA or IVB) lung cancer, preferably stage I or stage II lung cancer, more preferably stage I lung cancer.

[0130] The terms "subject" or "patient," as used herein, typically and preferably refer to humans, but may also include reference to non-human animals, preferably warm-blooded animals, more preferably vertebrates, and even more preferably mammals, such as non-human primates, rodents, dogs, cats, horses, sheep, pigs, and the like. Particularly contemplated are subjects known to or suspected of having lung cancer. Suitable subjects may include subjects who see a physician for screening for lung cancer and / or who have symptoms and signs indicative of lung cancer.

[0131] In certain embodiments, the subject is at risk for lung cancer, e.g., a subject at average or high risk for lung cancer. Non-limiting examples of risk factors for lung cancer include genetic susceptibility, diet, occupational exposure (e.g., asbestos, metals, silica, polycyclic aromatic hydrocarbons, diesel exhaust), air pollution, and tobacco smoking.

[0132] In a more particular embodiment, the subject is a tobacco smoker or former tobacco smoker.

[0133] The diagnostic methods taught herein can be used to effectively complement imaging techniques in lung cancer screening.

[0134] In certain embodiments, the subject is one who has been diagnosed with lung cancer, for example, by imaging techniques.

[0135] As used herein, the phrase "subject in need of treatment" includes subjects who would benefit from treatment of a given pathological condition, particularly lung cancer. Such subjects may include, but are not limited to, subjects diagnosed with said pathological condition.

[0136] The term "treat" or "treatment" encompasses both therapeutic treatment of an already established disease or pathological condition, e.g., treatment of an already established proliferative disease, and prophylactic or preventive measures, where the purpose is to prevent or reduce the likelihood of an undesired affliction occurring, e.g., preventing the progression of lung cancer. Beneficial or desired clinical results can include, but are not limited to, alleviation of one or more symptoms or one or more biological markers, a decrease in the extent of disease, a stable (i.e., not worsening) state of disease, a delay or slowing of disease progression, remission or palliation of the disease state, and the like. "Treatment" can also mean extending survival compared to that expected in the absence of treatment. Non-limiting examples of therapeutic treatments for lung cancer are radiation therapy, chemotherapy, targeted drug therapy, immunotherapy, and surgery.

[0137] In certain embodiments, the treatment is selected from the group consisting of radiation therapy, chemotherapy, targeted therapy, immunotherapy, and surgery.

[0138] In certain embodiments, the treatment includes Abraxane, afatinib dimaleate, Afinitor, Afinitor Disperse Disperz, Alecensa, Alectinib, Alimta, Alunbrig, Atezolizumab, Avastin, Bevacizumab, Brigatinib, Carboplatin, Ceritinib, Crizotinib, Cyramza, Dabrafenib mesylate, Dacomitinib, Docetaxel, Doxorubicin hydrochloride, Durvalumab, Entrectinib, Erlotinib hydrochloride, Everolimus, Etoposide, Etoposide phosphate, Gefitinib, Gilotrif, Gemcitabine, Gemzar, Hycamtin, Imfinzi, Iressa, Keytruda, Lobrena, Lorlatinib, Mechlorethamine hydrochloride, Mekinist, Methotrexate The method includes administering an effective amount of a therapeutic agent selected from the group consisting of: sartan, mustalgen, mubashi, navelbine, necitumumab, nivolumab, Opdivo, osimertinib mesylate, paclitaxel, paclitaxel albumin-stabilized nanoparticle formulation, Paraplat, Paraplatin, pembrolizumab, pemetrexed disodium, portraza, ramucirumab, rozlytrek, tafinlar, tagrisso, tarceva, taxol, taxotere, tecentriq, topotecan hydrochloride, trametinib, trexall, vidinpro, vinorelbine tartrate, xalkori, zykadia, carboplatin-taxol, and gemcitabine-cisplatin.

[0139] The term "effective amount," as used herein, may refer to a prophylactically effective amount, which is the amount of an active compound or pharmaceutical agent, more particularly a prophylactic agent, that inhibits or delays the onset of a disorder in a subject as sought by a researcher, veterinarian, physician, or other clinician, or may refer to a therapeutically effective amount, which is the amount of an active compound or pharmaceutical agent, more particularly a therapeutic agent, that elicits a biological or medical response in a subject that is sought by a researcher, veterinarian, physician, or other clinician, which may include, among other things, the alleviation of symptoms of the disease or pathological condition being treated. Methods for determining therapeutically and prophylactically effective doses for the agents taught herein are known in the art.

[0140] The terms "administration" or "administering," as used herein, refer to administering a particular treatment for lung cancer to a subject in need of such treatment. Such treatments may be therapeutic agents as described elsewhere herein. The route of administration may be essentially any route of administration, including, but not limited to, oral administration (e.g., oral ingestion or inhalation), intranasal administration (e.g., intranasal inhalation or intranasal mucosal application), parenteral administration (e.g., subcutaneous, intravenous, intramuscular, intraperitoneal, or intrasternal injection or infusion), transdermal or transmucosal (e.g., oral, sublingual, intranasal) administration, topical administration, rectal, vaginal, or intratracheal instillation, etc. Thus, the therapeutic effect achievable by the methods of the present invention may be, for example, systemic, local, tissue-specific, etc., depending on the specific needs of a given application of the present invention.

[0141] The present method for diagnosing lung cancer may fully qualify as an in vitro method in that one or more in vitro processing and / or analysis steps are applied to a sample removed from a subject. The term "in vitro" generally refers to the outside or exterior of a body, e.g., an animal or human body. Detecting or measuring at least one biomarker selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a biological sample of a subject may generally mean that the testing step of the method includes measuring the amount of the at least one biomarker in a sample of the subject. It is understood that the present method may generally include a testing step in which data is collected from and / or about the subject.

[0142] A further aspect is (a) a means for measuring the amount or expression level of at least one, at least two, at least three, at least four, at least five, or all six, preferably at least two, biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a sample from a subject; and (b) establishing a reference value or threshold representing a known diagnosis of lung cancer for each of said at least one, at least two, at least three, at least four, at least five, or all six, preferably at least two, biomarkers. and a kit, particularly a kit for diagnosing or monitoring lung cancer, comprising the above compound.

[0143] In certain embodiments, the means for measuring a biomarker is specifically adapted for that biomarker. For example, various techniques for measuring a biomarker may employ binding agents for the respective biomarkers. Thus, a means for measuring the amount or expression level of at least one, at least two, at least three, at least four, at least five, or all six, preferably at least two, biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 in a sample from a subject may include a binding agent, such as an antibody or antibody fragment, an aptamer, a photoaptamer, a protein, a peptide, a peptidomimetic, or a small molecule, capable of specifically binding to ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and / or CDH13, and / or a carrier that allows for visualization and / or qualitative readout of the measurement, for example, by spectrophotometry. In some embodiments, the binding agent taught herein may include a detectable label. In some embodiments, the binding agent may be equipped with a tag that allows for detection using another agent (e.g., a probe binding partner). The biomarker-binding agent conjugate may be associated with or bound to a detection agent to facilitate detection.

[0144] Additionally or alternatively, the binding agent may detect expression of said biomarker at the RNA level. The most commonly used methods for RNA detection include Northern blot, polymerase chain reaction (PCR), RNA in situ hybridization, cDNA microarray, and high-throughput sequencing techniques.

[0145] In certain embodiments, the means for measuring the amount or expression level of at least one, at least two, at least three, at least four, at least five or all six, preferably at least two, biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 in a sample from a subject comprises one or more antibodies that specifically bind to at least one, at least two, at least three, at least four, at least five or all six, preferably at least two, biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13. Numerous antibodies against ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 or CDH13 are commercially available from various suppliers. This information can be obtained from the respective suppliers or can be conveniently cataloged and queried in publicly available databases, such as the GeneCards® database maintained by the Weizmann Institute (www.GeneCards.org), under the field "Antibody Products."

[0146] In certain embodiments, the means for measuring the amount or expression level of at least one, at least two, at least three, at least four, at least five or all six, preferably at least two, biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13 in a sample from a subject comprises one or more probes or primers that specifically bind to RNA encoding each of said biomarkers.

[0147] In certain embodiments, the reference value or threshold is a reference value or threshold for the amount or expression level of at least one, at least two, at least three, at least four, at least five or all six, preferably at least two, biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13, wherein said reference value or threshold corresponds to the amount or expression level of said at least one, at least two, at least three, at least four, at least five or all six, preferably at least two, biomarkers in a sample from a subject not affected by lung cancer, such as a sample from a healthy subject or a group of healthy subjects, or said reference value or threshold corresponds to the amount or expression level of said at least one, at least two, at least three, at least four, at least five or all six, preferably at least two, biomarkers in a sample from a subject or group of subjects affected by lung cancer.

[0148] In certain embodiments, the kit further comprises a reference score or threshold score representing the number of biomarkers selected from the group consisting of ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6, and CDH13 that are found to deviate from their respective reference values ​​or thresholds.

[0149] The kit for diagnosing lung cancer in a subject may further comprise a ready-to-use substrate solution, a washing solution, a dilution buffer and instructions. The diagnostic kit may also comprise positive and / or negative control samples.

[0150] Preferably, the instructions included with the diagnostic kit are unambiguous, concise, and easy to understand for one of skill in the art. The instructions typically provide information regarding the contents of the kit, how to collect tissue samples, methodology, experimental readout and interpretation, and precautions and warnings.

[0151] The terms "kit of parts" and "kit," as used throughout this specification, refer to an article of manufacture containing components necessary to perform a specified method (e.g., a method for diagnosing lung cancer in a subject or a method for determining whether a subject is in need of therapeutic treatment for lung cancer as taught herein) and packaged to allow their transport and storage. Materials suitable for packaging the components included in the kit include crystal, plastic (e.g., polyethylene, polypropylene, polycarbonate), bottles, flasks, vials, ampoules, paper, envelopes, or other types of containers, carriers, or supports. When a kit includes multiple components, at least a subset of the components (e.g., two or more of the multiple components) or all of the components may be physically separated, e.g., contained in or on individual containers, carriers, or supports. The components included in a kit may or may not be sufficient to perform the specified method, such that external reagents or substances may not be necessary or required, respectively, to perform the method. Typically, the kits are employed with standard laboratory equipment, such as liquid handling equipment, environmental (e.g., temperature) control equipment, analytical instruments, etc. In addition to the enumerated binding agents taught herein (e.g., antibodies, hybridization probes, amplification primers, and / or sequencing primers, optionally provided on an array or microarray), the kits may also include some or all of the following: solvents, buffers (e.g., histidine buffer, citrate buffer, succinate buffer, acetate buffer, phosphate buffer, formate buffer, benzoate buffer, TRIS (Tris(hydroxymethyl)-aminomethane) buffer, or maleate buffer, or mixtures thereof), enzymes (e.g., thermostable DNA polymerase), detectable labels, detection reagents, and control formulations (positive and / or negative) useful in the specified method. Typically, the kits also include instructions for their use, e.g., on a printed insert or on computer-readable media.These terms may be used interchangeably with the term "article of manufacture," which, when used in this context, broadly encompasses any man-made tangible product of construction.

[0152] In certain embodiments, the kit further comprises a computer-readable storage medium having recorded thereon one or more programs for performing the methods taught herein.

[0153] A further aspect provides the use of a kit as taught herein for diagnosing lung cancer or for monitoring lung cancer based on detection of said biomarkers in a sample from a subject.

[0154] Furthermore, the inventors have discovered that certain methods can be successfully used to identify and validate suitable biomarker panels.

[0155] More specifically, a further aspect of the present invention is (a) measuring the amount or expression level of a predetermined set of proteins in biological samples from a first group of patients diagnosed with the disease or pathological condition and a first group of healthy individuals; (b) identifying proteins whose abundance or expression levels differ significantly between a patient group diagnosed with said disease or pathological condition and a healthy control group; (c) performing hierarchical clustering on the proteins identified in (b) and selecting one protein per group of correlated proteins; (d) applying the absolute shrinkage and selection operator (LASSO) in combination with bootstrap sampling for the proteins selected in (c) to obtain the best protein biomarker panel for predicting the outcome of a disease or pathological condition; (e) optionally, identifying an optimal threshold value for each biomarker in the protein biomarker panel obtained in (d); and (f) validating each biomarker in the protein biomarker panel obtained in (d) in a second group of patients diagnosed with the disease or pathological condition and a second group of healthy individuals. The present invention provides a method for identifying and validating a biomarker panel for a particular disease or disorder, preferably lung cancer, comprising:

[0156] In certain embodiments, the predetermined set of proteins includes proteins known to be associated with a disease or disorder, and optionally, known to be measurable in a biological sample. For example, if the biological sample is human blood, the predetermined set of proteins includes only proteins detectable in human blood. If the biological sample is human blood, the predetermined set of proteins may further include known blood proteins, e.g., known plasma proteins.

[0157] In certain embodiments, proteins known to be associated with a disease or disorder include proteins identified using genomic analysis of diseased and non-diseased human tissues (e.g., genes differentially expressed by at least a 2-fold difference (p<0.05) between tumor vs. tumor-normal biospecimens), proteins identified based on mouse xenograft studies (e.g., human protein candidates identified in the plasma of mice xenografted with human lung cancer cell proteins), and / or proteins identified based on published literature.

[0158] In certain embodiments, the group of patients diagnosed with the disease or pathological condition and the group of healthy individuals are selected so that they are well representative of the intended population for which the biomarker panel will be used for screening.

[0159] In certain embodiments, the step of identifying proteins whose abundance or expression levels differ significantly between a group of patients diagnosed with the disease or pathological condition and a group of healthy individuals is performed using a nonparametric Kruskal-Wallis test and a Bonferroni-adjusted P value. For example, a protein is considered to be significantly different between a group of patients diagnosed with the disease or pathological condition and a group of healthy individuals if P<0.00014 (=0.05 / 351; Bonferroni correction). In certain embodiments, hierarchical clustering is performed on the proteins identified in (b) using Spearman's correlation coefficient. For example, hierarchical clustering of proteins can be performed using a dissimilarity function (=1-absolute value of correlation) to identify all correlated groups.

[0160] In certain embodiments, when an LC-PRM-MS method is used to measure the amount or expression level of a given set of proteins in a biological sample, one protein per correlated protein group is selected to represent the correlated protein group, for example, based on high intensity in the biological sample, fewer missing values, and / or no interference of the PRM signal.

[0161] In certain embodiments, performing hierarchical clustering on the proteins identified in (b) allows for filtering out highly correlated proteins before applying LASSO in combination with bootstrap sampling, thereby preventing LASSO from randomly selecting one protein from a group of highly correlated proteins.

[0162] In certain embodiments, the best protein biomarker panel for predicting the outcome of a disease or pathological condition is selected based on the frequency of carrying certain combinations of proteins.

[0163] In certain embodiments, the optimal threshold for each biomarker in the protein biomarker panel obtained in (d) is determined using the PanelomiX platform, e.g., as described in Robin X., PanelomiX: a threshold-based algorithm to create panels of biomarkers, 2013, Translational Proteomics, 1(1):57-64.

[0164] general conclusion

[0165] While the present invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications, and variations will be apparent to those skilled in the art in light of the foregoing description. It is therefore intended to embrace all such alternatives, modifications, and variations within the spirit and broad scope of the appended claims as hereinafter set forth.

[0166] The above aspects and embodiments are further supported by the following non-limiting examples. [Example]

[0167] Example 1: Use of the biomarker panel taught herein allows for the diagnosis of lung cancer in plasma samples from patients 1. Materials and Methods 1.1 Study cohort The training cohort consisted of 128 lung cancer patients and 93 healthy donors followed within a Luxembourg hospital. The validation cohort included 48 patients and 49 age-, sex-, and smoking-matched non-cancer subjects (not included in the training cohort). All participants provided blood samples with informed consent in accordance with the Declaration of Helsinki. The study was approved by the National Research Ethics Committee (Comite National d'Ethique de Recherche) and the National Commission for Data Protection (Commission Nationale pour la Protection des Données). Blood samples were collected and processed according to standard operating procedures at the Integrated Biobank of Luxembourg, and plasma samples were prepared. Disease diagnosis, staging, and grading were performed by experienced pathologists according to the IASLC / ATS / ERS Histological Classification of Lung Tumors (2011) and the TNM Classification of Lung Cancer (2009). The clinicopathological characteristics of the subjects are summarized in Tables 1 and 2.

[0168] [Table 1]

[0169] [Table 2]

[0170] 1.2 Plasma depletion and processing Training Cohort High-abundance proteins were removed from 40 μL of plasma using an Agilent 1260 Infinity Bio-inert LC system equipped with a Human 14 Multiple Affinity Removal Column (4.6 x 100 mm) (Agilent Technologies, Diegem, Belgium) according to the manufacturer's protocol. After elution, Buffer A was exchanged into 100 mM NH4HCO3 / 10% ACN (pH 8), and the volume was reduced to 100 μL using a Spin Concentrator 5K (Agilent). Proteins were denatured with 1% sodium deoxycholate (SDC), reduced with 10 mM dithiothreitol for 30 min at 37 °C, alkylated with 25 mM iodoacetamide for 30 min at room temperature, and subsequently quenched with 10 mM n-acetyl-L-cysteine. All reagents were prepared in 50 mM Tris buffer. The processed sample was diluted to reduce the SDC concentration to 0.5% and incubated with 13 μg of sequencing-grade trypsin (Promega, Leiden, The Netherlands) for 16 h at 37°C, then with 10 U of PNGase F for 1 h at 37°C, followed by an additional 2 μg of trypsin for 3 h at 37°C. SDC was removed by precipitation with 1% formic acid and centrifugation. The digested sample was washed with a Sep-Pak C18 cartridge (Waters, Milford, MA, USA) and dried under vacuum. The sample was reconstituted with 200 μL of 0.1% formic acid / 4% acetonitrile.

[0171] Validation cohort Aliquots of 40 μl plasma samples from patients and healthy donors were depleted using an LC1260 chromatography system coupled with a Mars14 column (Agilent) depletion column. The depleted plasma samples were buffer-exchanged into 50 mM ammonium bicarbonate buffer by ultrafiltration onto Agilent spin filters (5 kDa MWCO). A 10% sample aliquot was then denatured with 1% (w / v) sodium deoxycholate (SDC) and 10 mM DTT at 37°C for 1 h. The samples were then alkylated with iodoacetamide at a final concentration of 25 mM for 30 min in the dark at room temperature. The reaction was stopped by adding acetylcysteine ​​to a final concentration of 10 mM. The samples were diluted to 1% (w / v) SDC with ammonium bicarbonate buffer and then digested overnight with sequencing-grade trypsin (Promega). The sample was then deglycosylated by incubation with PNGase F for 1 hour at 37°C, followed by a second trypsin digestion for 3 hours at 37°C. SDC was precipitated by adding formic acid to a final concentration of 1% and removed by centrifugation. The supernatant was then purified by solid-phase extraction (Sep Pak C18, Waters). The eluted sample was dried by vacuum centrifugation and finally suspended in 50 μL of 1% acetonitrile and 0.05% trifluoroacetic acid in water.

[0172] Heavy C-terminal lysine or arginine (C-terminal arginine, 13 C6, 15 N4, Δm = 10 Da, C-terminal lysine 13 C6, 15 A mixture of quantified synthetic peptides (aqua quant pro, Thermo Scientific) labeled with N2, Δm=8 Da) was aliquoted for single use and stored at −80°C. The mixture was spiked into digested plasma samples prior to LC-MS analysis.

[0173] 1.3 LC-PRM analysis Training Cohort Stable isotope labeling (SIL) (for C-terminal arginine 13 C6 15 For N4 and C-terminal lysines, 13 C6 15 N2) synthetic peptides were used as internal standards (AQUA QuantPro grade, Thermo Fisher Scientific, Bremen, Germany). LC-MS attributes (retention time, precursor m / z, and most intense fragment ion) were determined for each peptide to develop an LC-PRM method. Samples were analyzed using the scheduled LC-PRM assay for 351 peptides. An Ultimate 3000 RSLCnano system coupled to a Q-Exactive Plus mass spectrometer (Thermo Fisher Scientific) was used as previously described in Kim YJ et al., Quantification of SAA1 and SAA2 in lung cancer plasma using the isotype-specific PRM assays. Proteomics 2015;15:3116-3125. Accurate relative quantification was obtained from the intensity ratio of the light and SIL peptides.

[0174] Validation cohort The LC-MS setup consisted of a Dionex U3000 RSLC liquid chromatography system operated in column-switching mode coupled to a Q Exactive Plus mass spectrometer. The A and B mobile phases for the liquid chromatography consisted of water with 0.1% formic acid and acetonitrile with 0.1% formic acid, respectively. The loading phase consisted of 1% acetonitrile and 0.05% trifluoroacetic acid in water. The sample was applied to a trap column (75 μm × 20 mm, C) by the loading phase at a flow rate of 5 μl / min. 18The sample was loaded onto a 3 μm C18 column (pepmap100, 3 μm). The sample was then eluted from the trap onto an analytical column (75 μm × 150 mm, C18, 2 μm for pepmap100) using a linear gradient ranging from 2% A to 35% B in 66 min. MS acquisition was performed on a Q Exactive Plus (Thermo Scientific) operated in parallel reaction monitoring mode (PRM). The acquisition loop consisted of a timed, targeted PRM acquisition performed at 200 m / z with a resolution of 70,000. The isolation window for the target peptide ion was set to 1 m / z, the normalized collision energy was set to 25, and the maximum fill time was set to 240 ms. The duration of the timed window for each pair of endogenous and isotope-labeled peptides was set to 5 min and centered on their retention times.

[0175] 1.4 Model development and statistical analysis (training cohort) LC-PRM signals were converted to protein concentrations in fmol / μL based on an internal standard peptide. Values ​​for undetected proteins were replaced by the minimum protein concentration / √2. Protein concentrations in lung cancer and healthy samples were compared using the nonparametric Kruskal-Wallis test and Bonferroni-adjusted P value. Proteins with a P value <0.00014 (=0.05 / 351; Bonferroni-corrected) were further considered for analysis. Correlations between proteins were examined using Spearman's correlation coefficient. Hierarchical clustering of proteins was performed using a dissimilarity function (=1 - absolute value of correlation) to identify all correlated groups. One protein per highly correlated protein group was selected to represent the group based on the criteria of high intensity, fewer missing values ​​in lung cancer samples, and no interference in the PRM signal.

[0176] Bootstrap sampling and least absolute value shrinkage and selection operator (LASSO) penalty were used to find the best combination of proteins for outcome prediction. LASSO with 10-fold cross-validation was performed on 4,500,000 bootstrapped datasets using the "glmnet" package in R. To evaluate the predictive ability of proteins and protein combinations, the negative predictive value (NPV), positive predictive value (PPV), sensitivity, specificity, area under the receiver operating characteristic curve (AUC), and Akaike information criterion (AIC) of the logistic regression model were calculated on the original dataset. The AUC of different models was compared using the bootstrap test. To compare sensitivity and specificity, McNemar's chi-squared was used. 2 Tests were used as recommended. For model validation, sensitivity, specificity, NPV, PPV, AUC and their 95% confidence intervals (CI) were calculated on the validation dataset.

[0177] Multinomial logistic regression was used to predict the probability of each cancer stage (six levels, including four cancer stages, one unknown stage, and one healthy state) using a six-protein panel. The most probable level was selected as the final predicted cancer stage (or healthy state). Cohen's kappa test was used to assess the degree of agreement between the clinically annotated and predicted stage classifications.

[0178] Continuous variables were compared using the Kruskal-Wallis test. Dichotomous or categorical variables were compared using the Pearson chi-square test.

[0179] 1.5 MS data processing of validation cohort Data were processed using Skyline software (v19.1.0.193). The MS signals of the four most intense and least interfered product ions for each pair of endogenous peptide and isotope-labeled peptide were extracted as an ion chromatogram (XIC). The relative dot products between each pair of endogenous peptide fragment ion XIC and isotope-labeled peptide fragment ion XIC were calculated using the Skyline embedded calculator, and peptide pairs with scores below 0.99 were rejected for quantitative analysis. Quantitative values ​​were extracted as the sum of the endogenous fragment ion XIC areas divided by the sum of the XIC areas of the corresponding isotope-labeled peptides. The MS signal was converted to concentration by the following formula:

[0180]

number

[0181] 1.6 Using PanelomiX for Threshold Selection (Training Cohort) Robin X. PanelomiX for the Combination of Biomarkers. Methods Mol Biol 2019;1959:261-273. Using the PanelomiX platform, candidate biomarker thresholds were selected to optimize the classification performance of the combination. First, thresholds were defined for each protein, and then scores were assigned to each subject. A patient's score is the number of biomarkers that meet the disease status (referred to as "positive" biomarkers). If the subject's score is at least equal to the panel threshold score identified by PanelomiX, the subject was classified as a lung cancer patient. The thresholds obtained from the training set were applied to the validation set for cancer prediction, and performance metrics were calculated.

[0182] 1.7 Summary of new findings Genomic analysis of human lung tissue DNA copy number / alteration was determined for 19 tumor samples from 38 lung cancer biopsies using the Agilent SurePrint Human CGH Microarray 244K. Copy number variation (CNV) specific to each tumor sample was determined as the log fold change relative to a human reference (Promega, Madison, WI) using DNA Analytics. For gene expression analysis, the Agilent SurePrint G3 Human Exon 2X400K Microarray was used. Of the 19 matched healthy and tumor biopsies, 16 matched samples had high-quality RNA suitable for analysis (RIN > 7.5). A two-tailed t-test was used with GeneSpring software to identify genes that were differentially expressed between tumor and tumor-normal biopsies with at least a 2-fold difference (p < 0.05). 78 genes were commonly overexpressed in tumor samples compared with tumor-normal lung biopsies, while 81 genes were underexpressed in tumor samples compared with tumor-normal. Candidate gene biomarkers were prioritized as previously described (Salhia B, Kiefer J, Ross JT et al., Integrated genomic and epigenomic analysis of breast cancer brain metastasis. PLoS One 2014;9:e85448).

[0183] Mouse xenograft studies An orthotopic lung cancer model was developed using human lung cancer cell lines (H2009 and H1975) in the lungs of compromised mice, as described in a previous study. Two weeks after implantation, mouse plasma was collected from 21 mice that developed tumors, and circulating human proteins were identified by LC-MS / MS-based proteomics. Shotgun-based proteomic analysis was performed on these plasma samples to identify human proteins secreted from lung tumors into the mouse circulation. Bioinformatics analysis identified 436 human-specific proteins in mouse plasma, including proteins previously implicated in the development and progression of lung cancer.

[0184] Integration of discovery data and candidate prioritization For further discovery, we conducted a thorough search of published literature to identify high-quality lung cancer omics datasets and curated 40 datasets that met stringent quality criteria. These datasets were combined with our own discovery data to develop a comprehensive human tissue candidate database. The top 4,000 candidates from these four datasets (proteomics, aCGH, gene expression, and literature curation) were combined with human protein candidates identified in the plasma of mice xenografted with human lung cancer cells, resulting in a total of 4,254 unique biomarker candidates for consideration in the next prioritization step.

[0185] To prioritize candidates based on their detectability in plasma, we performed a two-pronged analysis. First, we performed an exhaustive shotgun analysis of a pool of human plasma derived from 13 patients with late-stage lung cancer, resulting in the detection of 1,245 unique human proteins, 520 of which overlapped with the prioritized tissue-based candidate list. Next, to help detect candidates that may not have been revealed by shotgun proteomics, we applied Accurate Inclusion Mass Screening (AIMS). Using experimentally observed peptides and publicly available databases, we constructed a list of 29,270 peptides representing 3,573 of our protein biomarker candidates. These peptides were subset into 20 distinct inclusion lists, which were split equally between FHCRC and TGen for analysis in depleted plasma by two-dimensional LC-MS / MS separation. Using biomarker candidates observed by shotgun or AIMS analysis of depleted human plasma, we prioritized biomarker candidates for validation studies, yielding 559 proteins.

[0186] 2.Results 2.1 Composition of patients and healthy donors The cohort consisted of 57.92% males, 42.08% females, 14.93% non-smokers, 57.92% former smokers, and 27.15% current smokers. The mean age was 63.56 (±10.03), and the median age was 63 (Table 3). No significant differences in age, gender, or smoking status were found between healthy and cancer individuals.

[0187] 2.2 Extensive selection of potential tumor predictors in plasma Previous multi-omics discovery efforts conducted in our laboratory suggested that 559 proteins are associated with lung cancer and potentially detectable in human blood (see Materials and Methods, 1.8 Summary of New Findings) (Zhang H, Kennedy J, Lee LW, et al., Integrated Strategy for Lung Cancer Biomarker Candidate Discovery by Quantitative Proteomics Profiling on Tumor and Adjacent Normal Lung Tissue (abstract). 59th ASMS Conference on Mass Spectrometry and Allied Topics. Denver, Colorado: 2011: Abstract nr MP 679 and Zhang H, Whiteaker J, Lin C, et al., Prioritization of Plasma-Based Predictive Markers for Chemotherapy in Lung Cancer Using Fractionation and Targeted Mass Spectrometry (abstract). 61st ASMS Conference on Mass Spectrometry and Allied Topics. Minneapolis, Minnesota: 2013: Abstract nr MP 541). The detectability of each protein in human plasma was previously validated, resulting in a set of 323 proteins to be further validated in a larger cohort. In this study, the plasma levels of the 323 proteins were quantified by LC-PRM in the plasma of lung cancer patients and healthy donors. An additional 28 known plasma proteins were also screened. Differential analysis of the PRM data showed that the plasma levels of 229 proteins were significantly different between the lung cancer and healthy groups (data not shown).

[0188] [Table 3]

[0189] 2.3 Improving biomarker selection Of the 229 differentially abundant proteins in the plasma of lung cancer and healthy subjects, 90 proteins showed a correlation of ≥ 0.9 or ≤ -0.9 with one or more proteins, whereas 139 proteins exhibited weaker correlations. When the threshold for dissimilarity or "distance" between proteins was set to 0.1 (as an absolute value), 19 groups with highly correlated proteins were identified. Therefore, 19 alternative proteins were selected (see Materials and Methods for details), and 71 proteins were excluded from further analysis.

[0190] LASSO variable selection was performed using 158 proteins. The most frequently (23-fold) retained combination was FLNA, TUBA4A, GSTO1, PRDX6, ARHGDIB, and CDH13 (hereafter referred to as the 6-protein combination / panel / classifier) ​​(Table 4). The concentrations of the six proteins were significantly different in the plasma of lung cancer patients and healthy donors (Figure 1). The PRM readouts of proteins measured in samples from one lung cancer patient and one healthy donor compared with the internal standard are shown in Figure 2. These proteins were individually selected as the most predictive, regardless of the combination: 74.51% of cases for FLNA, 76.91% for TUBA4A, 44.42% for GSTO1, 54.74% for PRDX6, 45.11% for ARHGDIB, and 81.43% for CDH13 (Table 4). The proteins selected as being predictive in more than 75% of all combinations were TUBA4A, TFPI, and CDH13 (hereafter referred to as the 3-protein combination).

[0191] [Table 4]

[0192] 2.4 Model performance analysis We compared the performance of our model against the commercially available Xpresys® Lung (XL) assay (Biodesix, Boulder, CO), which consists of five diagnostic proteins. The XL assay was originally designed to distinguish benign from malignant lung nodules among indeterminate lung nodules. Our biomarker panel was compared to the five diagnostic protein panel included in the Xpresys® Lung XL assay. However, some caution should be exercised in interpreting the comparative results. Xpresys® Lung XL was developed to help identify nodules likely to be benign after chest computed tomography (CT) scans. Therefore, we validated it in a cohort of subjects presenting with lung nodules measuring 8–30 mm and either a diagnosis of non-small cell lung cancer (i.e., stage IA) or no evidence of cancer. In this study, Xpresys® Lung XL was studied in a cohort of patients with mixed lung cancer types and stages, in addition to a healthy population. Because the primary objective and target population of this study differed from those of the Xpresys® Lung XL study, the comparison between the 6-biomarker panel and the Xpresys® Lung XL is merely an indication of the good performance of the classifier and should not be interpreted as a comparison of the utility of the biomarker panel. Nevertheless, a direct comparison with the XL test on the same pool of plasma samples, albeit limited, based on a different primary objective and target population, can provide a useful benchmark for the panel. Performance index values ​​were best for the 6-protein combination compared with the 3-protein combination, the XL panel, and the univariate model (Table 5): lowest AIC (30.876), highest AUC (0.999) (shared with the 3-protein combination), highest PPV (0.992), highest NPV (0.989), highest specificity (0.989) (shared with ARHGDIB), and highest sensitivity (0.992). When TUBA4A, TFPI and CDH13 were used as classifiers, the value of AIC (31.402) was slightly higher, and the values ​​of PPV (0.984), NPV (0.968), specificity (0.978) and sensitivity (0.977) were slightly lower.When FLNA, TUBA4A, GSTO1, PRDX6, and ARHGDIB were considered as the only classifiers, the performance indices also showed good predictive power. Only CDH13 and TFPI performed poorer, but still had good predictive power (AUC = 0.845 and 0.851, respectively).

[0193] Compared with the six-protein model, the logistic regression model derived using the proteins in the XL panel had a higher AIC (45.592), suggesting a poorer fit to the data. In addition, the PPV, NPV, specificity, and sensitivity were lower than those of the six-protein and three-protein models (Table 5). Next, we tested the ability of the six-protein panel to predict cancer stage. As shown in Table 6, the six-protein panel distinguished between healthy individuals and lung cancer individuals but was unable to predict cancer stage. An unweighted Cohen's Kappa of 0.59 (95% CI, 0.52-0.66) and a weighted Cohen's Kappa of 0.73 (95% CI, 0.73-0.73) were found, suggesting a weak degree of agreement between the predicted stage and the clinically annotated stage. Importantly, the 6-protein panel classified 22 of 23 stage I patients as having lung cancer, demonstrating its strong diagnostic performance in early stage cases.

[0194] [Table 5]

[0195] [Table 6]

[0196] 2.5 Determining Biomarker Thresholds for Sample Classification The PanelomiX platform was used to select the best thresholds for the six identified biomarkers. Three panel optimization options were used: optimizing sensitivity with ≥95% specificity, optimizing specificity with ≥95% sensitivity, and optimizing overall accuracy. When choosing to optimize accuracy or specificity, only one threshold / biomarker was selected by PanelomiX, resulting in one combination / optimization. When optimizing sensitivity, 19,644 combinations were found, the first of which was the same as the threshold combination selected when optimizing specificity. Therefore, the following two threshold combinations were considered: those obtained when optimizing panel accuracy (T A combination) and a common combination for optimizing sensitivity and specificity (T S combination) (Table 7). T A If any three proteins were positive using the threshold, the subject was classified as having lung cancer. S For , any five of the six proteins must be positive to classify an individual as having lung cancer.

[0197] When thresholds were applied to the initial dataset, the panel's performance metrics were excellent: A The combination had a sensitivity of 0.992 and a specificity of 0.989, and T S The combination had a sensitivity of 0.977 and a specificity of 1.0.

[0198] [Table 7]

[0199] 2.6 Panel performance on the validation dataset. The model was then tested on a validation dataset using plasma from 48 lung cancer patients and 49 healthy donors. The logistic regression and PanelomiX threshold model estimates obtained from the training set were applied to the validation set for cancer prediction. The NPV, PPV, sensitivity, and specificity of the XL and 6-protein panels were calculated for the new dataset (Table 8). Comparing the results obtained from the logistic regression model, the values ​​of all performance metrics for the 6-protein combination were at least as high as those of the XL panel. Interestingly, the highest specificity (0.918) was achieved with the T S As predicted by the threshold, a 6-protein panel was obtained. All possible subcombinations of the 6-protein panel were also tested on the validation dataset. Many of them exhibited excellent performance, as shown by the forest plot of NPV, PPV, sensitivity, specificity, and AUC (Figure 3). The subcombinations of biomarkers were filtered for subcombinations with a sensitivity of ≥ 0.90 and an NPV of ≥ 0.90 or a specificity of ≥ 0.90. The 36 combinations shown in Table 9 were obtained after filtering and had the best performance. Of these 36 combinations, six combinations, namely, FLNA-TUBA4A-GSTO1-CDH13 (i.e., combination 8 in Table 9), FLNA-ARHGDIB-CDH13 (i.e., combination 21 in Table 9), TUBA4A-GSTO1-CDH13 (i.e., combination 23 in Table 9), TUBA4A-PRDX6-CDH13 (i.e., combination 25 in Table 9), TUBA4A-ARHGDIB-CDH13 (i.e., combination 26 in Table 9), and GSTO1-CDH13 (i.e., combination 33 in Table 9), have a sensitivity of ≧0.90, an NPV of ≧0.90, and a specificity of ≧0.90.

[0200] [Table 8]

[0201] [Table 9-1]

[0202] [Table 9-2]

[0203] 3. Discussion The objective of this study was to identify a panel of protein biomarkers for use as a noninvasive diagnostic tool in lung cancer. To this end, we screened 351 potential biomarkers discovered and preliminarily validated in human plasma. Here, based on PRM measurements and subsequent logistic regression analysis, we identified a blood-based 6-protein panel as a potential diagnostic tool in lung cancer. To facilitate physician use of this panel, we also adopted a threshold-based approach, resulting in a cutoff value per biomarker and then a score per sample to classify it as lung cancer or healthy.

[0204] The biomarker panel demonstrated excellent performance in the test cohort, supported by AUC (0.999), PPV (0.992), NPV (0.989), specificity (0.989), and sensitivity (0.992) values. This result was confirmed in the validation dataset, which also showed that other subcombinations of these six proteins exhibited excellent discriminatory power. Importantly, the ability of the six-protein panel to noninvasively detect lung cancer regardless of disease stage (including stage I tumors) suggests its high potential as a screening tool.

[0205] The performance of the six-protein biomarker panel described herein was compared to the Xpresys® Lung (XL) test, a commercially available MS-based lung cancer diagnostic test. Although limited, direct comparison with the XL test on the same pool of plasma samples, based on a different primary objective and target population, can provide a useful benchmark for the six-protein biomarker panel described herein. In the training set, values ​​for all performance metrics tended to be better for the six-protein biomarker panel described herein, indicating that the panel exhibited superior diagnostic accuracy in this cohort.

[0206] The biomarker panel is further validated in an independent cohort including patients with different types of cancer (e.g., colon cancer) and donors with and without underlying non-malignant lung disease (e.g., chronic obstructive pulmonary disease). The analysis confirms the specificity of the biomarker panel for lung cancer.

[0207] In conclusion, we have identified a protein-based diagnostic panel for detecting lung cancer in blood. When used as a routine test for high-risk and average-risk individuals (e.g., smokers and ex-smokers), this could effectively complement LDCT in lung cancer screening. This would reduce the number of false-positive cases, which often lead to additional invasive tests and unnecessary costs, and expose patients to physical and mental hardship.

Claims

1. detecting at least three biomarkers in a biological sample from a subject; The at least three biomarkers are -Rho GDP dissociation inhibitor β (ARHGDIB), and - at least two, at least three, at least four, or at least five selected from the group consisting of α-tubulin 4A (TUBA4A), glutathione S-transferase ω1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13); 1. An in vitro method for diagnosing lung cancer in a subject, comprising:

2. At least (a) ARHGDIB, TUBA4A and GSTO1; (b) ARHGDIB, TUBA4A and FLNA; (c) ARHGDIB, TUBA4A, and PRDX6; or (d) ARHGDIB, TUBA4A, and CDH13 The method of claim 1 , comprising detecting

3. (a) ARHGDIB, TUBA4A, GSTO1 and FLNA; (b) ARHGDIB, TUBA4A, FLNA, and PRDX6; (c) ARHGDIB, TUBA4A, GSTO1 and PRDX6; (d) ARHGDIB, TUBA4A, FLNA, and CDH13; (e) ARHGDIB, TUBA4A, GSTO1, and CDH13; or (f) ARHGDIB, TUBA4A, PRDX6, and CDH13 The method of claim 1 , comprising detecting

4. (a) ARHGDIB, TUBA4A, GSTO1, FLNA, and PRDX6; (b) ARHGDIB, TUBA4A, GSTO1, FLNA, and CDH13; (c) ARHGDIB, TUBA4A, FLNA, PRDX6, and CDH13; or (d) ARHGDIB, TUBA4A, GSTO1, PRDX6, and CDH13 The method of claim 1 , comprising detecting

5. The method of claim 1, comprising detecting ARHGDIB, TUBA4A, GSTO1, FLNA, PRDX6 and CDH13.

6. The method of any one of claims 1 to 5, wherein the biological sample is a body fluid sample selected from the group consisting of plasma, serum, whole blood, urine, tissue lysate, cerebrospinal fluid (CSF), saliva, and sweat.

7. The method described in claim 6, wherein the biological sample is a plasma sample.

8. The method of any one of claims 1 to 7, wherein the at least three biomarkers are detected using mass spectrometry, biochemical assay, immunoassay, chromatography, or a combination thereof.

9. (a) measuring the amount or expression level of said at least three biomarkers in a biological sample of a subject; (b) calculating a score based on the amounts or expression levels of the at least three biomarkers measured in step (a); and (c) comparing the score calculated in step (b) to a threshold score; wherein if the score calculated in step (b) is equal to or greater than the threshold score, the calculated score indicates that the subject has lung cancer. The method according to any one of claims 1 to 8.

10. (a) means for measuring the amount or expression level of at least three biomarkers in a biological sample of a subject; and (b) a threshold value representing a known diagnosis of lung cancer for each of said at least three biomarkers, or a means for establishing said threshold value; wherein the at least three biomarkers are: - Rho GDP dissociation inhibitor β (ARHGDIB), and - at least two selected from the group consisting of α-tubulin 4A (TUBA4A), glutathione S-transferase ω1 (GSTO1), filamin A (FLNA), peroxiredoxin 6 (PRDX6), and cadherin 13 (CDH13); The kit.

11. The kit according to claim 10, wherein the means is a binding substance that specifically binds to the protein or RNA encoding the protein.

12. 12. Use of the kit of claim 10 or 11 for diagnosing lung cancer based on the detection of said at least three biomarkers in a sample from a subject.

Images