Methods and systems for predicting response to PD-1 axis-directed therapeutics in mismatch repair deficient colorectal tumors

Abstract

Disclosed are scoring functions for predicting the response of dMMR and / or MSI-H colorectal tumors to PD-1 axis-directed therapy, as well as methods and systems for evaluating tissue samples for the presence of feature metrics useful for calculating such scoring functions. The scoring function integrates one or more spatial relationships between cell types into a numerical representation of the likelihood that the tumor will respond to PD-1 axis-directed therapy. Then, based on the output of the scoring function, the subject can be selected to receive PD-1 axis-directed therapy (if the scoring function indicates a sufficient likelihood of positive response) or alternative therapy (if the scoring function indicates an insufficient likelihood of positive response).

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of priority to U.S. Provisional Application No. 63 / 362,305, filed March 31, 2022, and U.S. Provisional Application No. 63 / 383,688, filed November 14, 2022, both of which are incorporated by reference in their entireties herein for any purpose.

[0002] INCORPORATION BY REFERENCE OF SEQUENCE LISTING The sequence listing, entitled "P37479WO_SEQ_LIST", created on February 1, 2023, and having a size of 10,677 bytes, is incorporated herein by reference. [Technical field]

[0003] FIELD OF THEINVENTION The present invention relates to the detection, characterization and enumeration of biomarkers in tumor samples that are useful for predicting response to immune checkpoint inhibitor therapy. [Background technology]

[0004] Brief description of the related art Programmed death ligand 1 (PD-L1) is an immune checkpoint protein that regulates the immune system through binding of the programmed cell death protein 1 (PD-1) receptor. PD-L1 is expressed on multiple immune cell types and also on many cancer cell types, including colorectal cancer (CRC) cells. PD-L1 can bind to the PD-1 receptor on activated T cells, leading to the suppression of cytotoxic T cells and allowing immune evasion of cancer. See Zou et al (2016). Cancers can escape immune surveillance and eradication through upregulation of the programmed death 1 (PD-1) pathway and its ligand, programmed death ligand 1 (PD-L1), on tumor cells and in the tumor microenvironment. Blockade of this pathway with antibodies against PD-1 or PD-L1 has led to remarkable clinical responses in some cancer patients. However, identification of predictive biomarkers for patient selection remains a major challenge.

[0005] CRC with DNA mismatch repair deficiency (dMMR) has microsatellite instability (MSI) that leads to hypermutation and expression of mutation-specific neopeptides. See Llosa et al. (2015). Treatment of metastatic CRC (mCRC) with the anti-PD-1 antibody pembrolizumab was approved by the US Food and Drug Administration for this subgroup of tumors that progressed after treatment with fluoropyrimidines, oxaliplatin, and irinotecan, due to frequent and durable responses in these patients. However, more than half of dMMR mCRC patients are resistant to PD-1 blockade due to unknown mechanisms. See Le(I) & Le(II).

[0006] PD-L1 is the most widely used predictive biomarker for selecting patients to receive PD-1 axis-directed therapeutics. However, differences related to tumor type have been observed, and to date, PD-L1 expression is not useful for predicting treatment response in patients with colorectal cancer. See Yi.

[0007] Assessment of intratumoral immune infiltrates has been proposed as a promising area of ​​investigation for potential biomarker signatures related to immunotherapy. The presence and intensity of an inflammatory response is known to be a prognostic factor in several different cancer types, including colorectal cancer. See Jass I; Jass II; Galon I; Galon II; Pages. In addition, spatial metrics between specific immune cell types have been investigated in colorectal tumors for their impact on prognosis, survival and response to treatment. See Barrera; Chakrabarti, Wang I; Wang II; Yoon; Zhang I; WO 2020 / 072348 A1; WO 2020 / 161125 A1.

[0008] However, to date, no biomarkers have yet been validated to accurately predict response to PD-1 blockade in patients with dMMR tumors. There remains a need to identify biomarker signatures indicative of response and benefit of immune checkpoint inhibitors in patients with dMMR CRC. Summary of the Invention

[0009] The present invention generally relates to a scoring function for predicting the response of dMMR and / or MSI-H colorectal tumors (including stage III and stage IV tumors) to PD-1 axis-directed therapy, and a method and system for evaluating tissue samples for the presence of feature metrics useful for calculating such a scoring function. The scoring function integrates one or more spatial relationships between cell types into a numerical representation of the likelihood that the tumor will respond to PD-1 axis-directed therapy. Based on the output of the scoring function, the subject can then be selected to receive PD-1 axis-directed therapy (if the scoring function indicates a sufficient likelihood of positive response) or an alternative therapy (if the scoring function indicates an insufficient likelihood of positive response).

[0010] In an exemplary embodiment, a method of treating a subject having a dMMR or MSI-H stage III colorectal tumor is provided, the method comprising administering a PD-1 axis-directed therapy to the subject, wherein the tumor has previously been determined to have a predicted response score (PRS) indicative of response to the PD-1 axis-directed therapy, the PRS being determined from a continuous scoring function incorporating a feature set selected from the group consisting of Feature Set 1, Feature Set 2, Feature Set 3, Feature Set 4, and Feature Set 5. In another embodiment, the continuous scoring function is a Cox proportional hazards model, such as regularized Cox regression with LASSO.

[0011] In an embodiment, a method of treating a subject having a dMMR or MSI-H stage IV colorectal tumor, the method comprising administering to the subject a PD-1 axis-directed therapy, wherein the tumor has previously been determined to have a predicted response score (PRS) indicative of response to the PD-1 axis-directed therapy, the PRS being determined from a continuous scoring function incorporating a feature set selected from the group consisting of Feature Set 1, Feature Set 2, Feature Set 3, Feature Set 4, and Feature Set 5. In another embodiment, the continuous scoring function is a Cox proportional hazards model, such as a regularized Cox regression with LASSO.

[0012] In an exemplary embodiment, the spatial relationship is PD-L1 + PD-1 within a given distance of cells (e.g., the mean or median number of cells) + PD-L1 is a metric that indicates the number of cells that can be used by itself or can be incorporated into a multivariate scoring function (such as a continuous scoring function) for prediction of response to a PD-1 axis-directed therapeutic. In certain embodiments, a method of treating a subject having a dMMR or MSI-H stage III or stage IV colorectal tumor is provided, the method comprising administering to the subject a PD-1 axis-directed therapy, the tumor having a PD-L1 expression level above a predetermined cutoff. + PD-1 within a given distance (e.g., within a range of 5 μm to 50 μm) of cells + It has previously been determined that the average number of cells

[0013] Also disclosed herein are systems, materials and methods useful for making such predictions, including affinity histochemistry assays and reagents, biomarker-specific reagent panels useful for performing such AHC assays, stained samples and slides, image analysis systems programmed to extract features from stained samples, and the like.

[0014] Other features and embodiments will become apparent from the following detailed description. [Brief description of the drawings]

[0015] [Figure 1A] FIG. 1A is a flow chart illustrating an exemplary method for selecting patients to receive a PD-1 axis-directed therapy or an alternative therapy using the scoring functions disclosed herein.

[0016] [Figure 1B] FIG. 1B is a flow chart showing an exemplary method for selecting patients to receive a PD-1 axis-directed therapy or an alternative therapy using the average number of PD-1+ cells within a predetermined distance of at least one PD-L1 cell.

[0017] [Diagram 2] FIG. 2 illustrates exemplary components and modules of the image analysis system disclosed herein.

[0018] [Figure 3A] 3A shows an exemplary workflow for performing object identification and ROI annotation tasks in an image analysis system disclosed herein, where the object identification task is performed before (or simultaneously with) ROI annotation. A digital image of an AHC-stained tissue section is obtained (top panel), and an object identification task (such as that performed by the FI module disclosed herein) is performed across substantially the entire image (middle panel). ROI annotation (bottom panel) does not affect the area of ​​the image where the object identification task is performed.

[0019] [Figure 3B] 3B shows an exemplary workflow for performing object identification and ROI annotation tasks in the image analysis system disclosed herein, where the object identification task is performed after ROI annotation. A digital image of an AHC stained tissue section is obtained (upper panel) and the ROI annotation task is performed (middle panel). The image identification task is performed only on the annotated (lower panel) and does not affect the region of the area where the object identification task is performed.

[0020] [Figure 4] Figure 4 shows examples of additional systems that may be included in the image analysis system to form a sample analysis system. The dashed arrows indicate sample flow. The solid arrows indicate data flow, including sample images, process instructions, and data outputs (such as feature metrics and PRS).

[0021] [Figure 5A] Figure 5A shows the predicted response score (PRS) distribution of the Cox proportional hazards model trained on feature set 1. The black vertical line indicates the median PRS. The white bars and solid curve indicate individuals falling into the low-risk group (i.e., responders). The grey bars and diagonal curve indicate individuals falling into the high-risk group (i.e., non-responders).

[0022] [Figure 5B] Figure 5B shows Kaplan-Meier curves stratified based on PRS risk groups (low risk in light grey; high risk in dark grey), with associated patient stratification charts shown below. "Time" on the x-axis indicates overall survival in days after treatment.

[0023] [Figure 6A] Figure 5A shows the predicted response score (PRS) distribution of the Cox proportional hazards model trained on feature set 2. The black vertical line indicates the median PRS. The white bars and solid curve indicate individuals falling into the low-risk group (i.e., responders). The grey bars and diagonal curve indicate individuals falling into the high-risk group (i.e., non-responders).

[0024] [Figure 6B] Figure 6B shows Kaplan-Meier curves stratified based on PRS risk groups (low risk in light grey; high risk in dark grey), with associated patient stratification charts shown below. "Time" on the x-axis indicates overall survival in days after treatment.

[0025] [Figure 7A]Figure 7A shows the predicted response score (PRS) distribution of the Cox proportional hazards model trained on feature set 3. The black vertical line indicates the median PRS. The white bars and solid curve indicate individuals falling into the low-risk group (i.e., responders). The grey bars and diagonal curve indicate individuals falling into the high-risk group (i.e., non-responders).

[0026] [Figure 7B] Figure 7B shows Kaplan-Meier curves stratified based on PRS risk group (low risk in light grey; high risk in dark grey), with associated patient stratification charts shown below. "Time" on the x-axis indicates progression-free survival in days after treatment.

[0027] [Figure 8A] Figure 8A shows the predicted response score (PRS) distribution of the Cox proportional hazards model trained on feature set 4. The black vertical line indicates the median PRS. The white bars and solid curve indicate individuals falling into the low-risk group (i.e., responders). The grey bars and diagonal curve indicate individuals falling into the high-risk group (i.e., non-responders).

[0028] [Figure 8B] Figure 8B shows Kaplan-Meier curves stratified based on PRS risk group (low risk in light grey; high risk in dark grey), with associated patient stratification charts shown below. "Time" on the x-axis indicates progression-free survival in days after treatment.

[0029] [Figure 9] Figure 9 shows Kaplan-Meier curves stratified based on the mean number of PD-1 cells within 10 μm of at least one PD-L1+ cell (light grey, low risk; dark grey, high risk), with associated patient stratification charts shown below. "Time" on the x-axis indicates progression-free survival in days after treatment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] I. Abbreviations and Definitions Unless otherwise defined, technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art.See, for example, Lackie, DICTIONARY OF CELL AND MOLECULAR BIOLOGY, Elsevier (4th ed.2007); Sambrook et al., MOLECULAR CLONING, A LABORATORY MANUAL, Cold Springs Harbor Press (Cold Springs Harbor, NY1989). The term "a" or "an" is intended to mean "one or more". The terms "comprise", "comprises" and "comprising" when preceding a list of steps or elements are intended to mean that the addition of further steps or elements is optional and not excluded.

[0031] Antibody: The term "antibody" is used herein in the broadest sense and encompasses a variety of antibody structures, including, but not limited to, monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), and antibody fragments, so long as they exhibit the desired antigen-binding activity.

[0032] Antibody fragment: "Antibody fragment" refers to a molecule other than an intact antibody that contains a portion of an intact antibody that binds the antigen to which the intact antibody binds. Examples of antibody fragments include, but are not limited to, Fv, Fab, Fab', Fab'-SH, F(ab')2, diabodies, linear antibodies, single-chain antibody molecules (e.g., scFv), and multispecific antibodies formed from antibody fragments.

[0033] Biomarker: As used herein, the term "biomarker" refers to any molecule or group of molecules found in a biological sample that can be used to characterize the biological sample or the subject from which the biological sample is obtained. For example, a biomarker can be a molecule or group of molecules whose presence, absence, or relative abundance is characteristic of a particular cell or tissue type or condition, or characteristic of a particular pathological condition or condition, or an indicator of the severity of a pathological condition, an indicator of the likelihood of progression or regression of a pathological condition, and / or an indicator of the likelihood of a pathological condition responding to a particular treatment. As another example, a biomarker can be a cell type or microorganism (such as bacteria, mycobacteria, fungi, viruses, etc.), or a substitute molecule or group of molecules thereof.

[0034] Biomarker-specific reagent: A specific detection reagent, such as a primary antibody, that can bind directly and specifically to one or more biomarkers in a cell sample.

[0035] Cell sample: As used herein, the term "cell sample" refers to any sample containing intact cells, such as a cell culture, a body fluid sample, or a surgical specimen taken for pathological, histological, or cytological interpretation.

[0036] Detection Reagent: A "detection reagent" is any reagent used to deposit a stain adjacent to a biomarker-specific reagent in a cell sample. Non-limiting examples include the biomarker-specific reagent (such as a primary antibody), secondary detection reagents (such as a secondary antibody that can bind to the primary antibody), tertiary detection reagents (such as a tertiary antibody that can bind to the secondary antibody), enzymes directly or indirectly associated with the biomarker-specific reagent, chemicals reactive with such enzymes that result in the deposition of a fluorescent or chromogenic stain, wash reagents used between staining steps, and the like.

[0037] Detectable moiety: A molecule or material capable of producing a detectable signal (visual, electronic, etc., or other) that indicates the presence (i.e., qualitative analysis) and / or concentration (i.e., quantitative analysis) of a detectable moiety deposited on a sample. The term "detectable moiety" includes, but is not limited to, chromogenic, fluorescent, phosphorescent, and luminescent molecules and materials, catalysts (such as enzymes) that convert one substance to another resulting in a detectable difference (e.g., by converting a colorless substance to a colored substance or vice versa, or by producing a precipitate, or by increasing the turbidity of the sample), and labels compatible with mass cytometry imaging (e.g., Multiple Ion Beam Imaging ("MIBI" as described in Baharlou, Bodenmiller, and Ptacek) or Imaging Mass Cytometry ("IMB" as described by Baharlou and Bodenmiller)). In some examples, the detectable moiety is a fluorophore that belongs to several general chemical classes, including coumarins, fluoresceins (or fluorescein derivatives and analogs), rhodamines, resorufins, luminophores, and cyanines. Additional examples of fluorescent molecules can be found in Molecular Probes Handbook-A Guide to Fluorescent Probes and Labeling Technologies, Molecular Probes, Eugene, OR, ThermoFisher Scientific, 11 th Edition. In other embodiments, the detectable moiety is a molecule detectable by bright field microscopy, such as dyes including diaminobenzidine (DAB), 4-(dimethylamino)azobenzene-4'-sulfonamide (DABSYL), tetramethylrhodamine (DISCOVERY Purple), N,N'-biscarboxypentyl-5,5'-disulfonato-indo-dicarbocyanine (Cy5), and rhodamine 110 (Rhodamine). In yet other embodiments, the detectable label is compatible with mass cytometric imaging, such as stable metal isotopes (including but not limited to lanthanide series metals).

[0038] Feature metric: A value indicating the amount of a feature in a sample or the relationship between features in a sample. Examples include the number of cells positive for a biomarker, the density of a particular cell type in a particular region (e.g., the number of biomarker positive cells over the area of ​​the ROI, the number of biomarker positive cells over the linear distance of the edge that defines the ROI, etc.), pixel density (i.e., the number of biomarker positive pixels over the area of ​​the ROI, the number of biomarker positive pixels over the linear distance of the edge that defines the ROI, etc.), the average or median distance between cells expressing the biomarker(s), etc. Feature metrics can be sum or global metrics.

[0039] Histochemical detection: A process involving labeling a biomarker or other structure in a tissue sample with a biomarker-specific reagent and a detection reagent in a manner that allows for microscopic detection of the biomarker or other structure in relation to the cross-sectional relationships between the structures of the tissue sample. Examples include immunohistochemistry (IHC), chromogenic in situ hybridization (CISH), fluorescent in situ hybridization (FISH), silver in situ hybridization (SISH), and hematoxylin and eosin (H&E) staining of formalin-fixed, paraffin-embedded tissue sections.

[0040] Immune checkpoint molecule: a protein expressed by immune cells whose activation downregulates cytotoxic T cell responses. Examples include PD-1, TIM-3, LAG-4, and CTLA-4.

[0041] Immune evasion biomarkers: biomarkers expressed by tumor cells that help tumors evade T cell-mediated immune responses. Examples of immune evasion biomarkers include PD-L1, PD-L2, and IDO.

[0042] Immunological biomarker: a biomarker that is characteristic of or affects an immune response to an abnormal cell, including, but not limited to, a biomarker indicative of a particular class of immune cell (such as CD3), a biomarker that characterizes an immune response (e.g., the presence or absence or amount of a cytokine protein or a particular immune cell subtype(s)), or a biomarker expressed by, presented by, or otherwise located on a non-immune cell structure that affects the type or extent of an immune cell response (e.g., cell surface expressed antigens, MHC-ligand complexes, and immune evasion biomarkers).

[0043] Monoclonal antibody: An antibody obtained from a population of substantially homogeneous antibodies, i.e., the individual antibodies constituting the population are identical and / or bind the same epitope, with the exception of variant antibodies that contain naturally occurring mutations or that may arise during the manufacture of a monoclonal antibody preparation, such variants generally being present in minor amounts. In contrast to polyclonal antibody preparations that typically contain different antibodies directed against different determinants (epitopes), each monoclonal antibody of a monoclonal antibody preparation is directed against a single determinant on an antigen. Thus, the modifier "monoclonal" indicates the nature of the antibody being obtained from a population of substantially homogeneous antibodies and is not to be construed as requiring production of the antibody by any particular method. For example, monoclonal antibodies used in accordance with the present invention may be made by a variety of techniques, including, but not limited to, hybridoma methods, recombinant DNA methods, phage display methods, and methods utilizing transgenic animals containing all or part of the human immunoglobulin loci, or combinations thereof.

[0044] Multiplex histochemical staining: a histochemical staining technique in which multiple biomarker-specific reagents that bind to different biomarkers are applied to a single section and stained with different color stains.

[0045] PD-1 axis-directed therapy: a therapeutic agent that disrupts the ability of PD-1 to downregulate T cell activity. Examples include PD-1 specific antibodies (such as nivolumab, pembrolizumab, cemiplimab, tislelizumab, spartalizumab, MEDI0680 (AstraZeneca), toripalimab, sintilimab, cetolimab, and pidilizumab), PD-L1 specific antibodies (such as atezolizumab, durvalumab, and avelumab), PD-1 directed bispecifics (e.g., tebotelimab (PD-1 / LAG3 bispecific DART® molecule); PD-1 ligand fragments and fusion proteins (such as AMP-224 (a fusion of the extracellular domain of PD-L2 with the Fc region of human IgG1)), PD-L1 directed bispecifics (FS118 (PD-L1 / LAG3 bispecific tetravalent antibody (F-Star Therapeutics), and small molecule inhibitors such as CA-170 (a small molecule with binding specificity for PD-L1, PD-L2 and VISTA), and BMS-1001 & BMS-1166 (small molecules predicted to dimerize PD-L1, see e.g. WO2015034820 and WO2015160641).

[0046] Sample: As used herein, the term "sample" is intended to refer to any material obtained from a subject that can be tested for the presence or absence of a biomarker.

[0047] Secondary detection reagent: A specific detection reagent capable of specifically binding to a biomarker-specific reagent.

[0048] Section: as a noun, a thin section of a tissue sample suitable for microscopic analysis, usually cut with a microtome; as a verb, the process of making sections.

[0049] Serial section: As used herein, the term "serial section" refers to any one of a series of sections cut in sequence from a tissue sample by a microtome. Two sections do not necessarily have to be consecutive sections from a tissue to be considered "serial sections" of one another, but generally must contain sufficiently similar tissue structures in the same spatial relationship so that the structures can be matched to one another after histological staining.

[0050] Single histochemical staining: A histochemical staining method in which a single biomarker-specific reagent is applied to a single section and stained with a single color stain.

[0051] Specific detection reagent: any composition of matter capable of specifically binding to a chemical structure of a target in the context of a cell sample. As used herein, the phrases "specific binding," "specifically binds to," or "specific for" refer to a measurable and reproducible interaction between a target and a specific detection reagent that determines the presence of the target in the presence of a heterogeneous population of molecules, including biological molecules. For example, an antibody that specifically binds to a target is an antibody that binds to this target with higher affinity, avidity, more readily, and / or with a longer duration than it binds to other targets. In one embodiment, the extent to which a specific detection reagent binds to an unrelated target is less than about 10% of the binding of an antibody to the target, as measured, for example, by radioimmunoassay (RIA). In certain embodiments, a biomarker-specific reagent that specifically binds to a target has a dissociation constant (Kd) of 1 μM or less, 100 nM or less, 10 nM or less, 1 nM or less, or 0.1 nM or less. In another embodiment, specific binding can include, but does not require, exclusive binding.Exemplary specific detection reagents include nucleic acid probes specific for particular nucleotide sequences, antibodies and antigen-binding fragments thereof, as well as antibodies based on the Z domain of protein A from S. aureus; Affibody AB, Solna, Sweden, AVIMER (domain A / LDL receptor based scaffold; Amgen, Thousand Oaks, CA), dAb (VH or VL antibody domain based scaffold; GlaxoSmithKline PLC, Cambridge, UK), DARPin (ankyrin repeat protein based scaffold; Molecular Partners AG, Zurich, Switzerland), ANTICALIN (lipocalin based scaffold; Pieris AG, Phillysing, Germany), NANOBODY (VHH (camelid Ig based scaffold); Ablynx, Inc., New York, USA), and other antibody-specific detection reagents. Examples of engineered specific binding compositions include: TRANS-BODY (a transferrin-based scaffold; Pfizer Inc., New York, NY), SMIP (Emergent Biosolutions, Inc., Rockville, MD), and TETRANECTIN (a C-type lectin domain (CTLD)-based scaffold; tetranectin; Borean Pharma A / S, Aarhus, Denmark). Descriptions of such engineered specific binding structures are reviewed in Wurch et al., Development of Novel Protein Scaffolds as Alternatives to Whole Antibodies for Imaging and Therapy: Status on Discovery Research and Clinical Validation, Current Pharmaceutical Biotechnology, Vol. 9, pp. 502-509 (2008), the contents of which are incorporated by reference.

[0052] Stain: When used as a noun, the term "stain" shall refer to any substance that can be used to visualize specific molecules or structures in a cell sample for microscopic analysis, including bright field microscopy, fluorescence microscopy, electron microscopy, etc. When used as a verb, the term "stain" shall refer to any process that results in the deposition of a colorant on a cell sample.

[0053] Subject: As used herein, the term "subject" or "individual" refers to a mammal. Mammals include, but are not limited to, domestic animals (e.g., cows, sheep, cats, dogs and horses), primates (e.g., humans and non-human primates, such as monkeys), rabbits and rodents (e.g., mice and rats). In certain embodiments, the individual or subject is a human.

[0054] Test Sample: A tumor sample obtained from a subject whose outcome was unknown at the time the sample was obtained.

[0055] Tissue sample: As used herein, the term "tissue sample" is intended to refer to a sample of cells that preserves the cross-sectional spatial relationships between cells that were present within the subject from which the sample was obtained.

[0056] Tumor mutation burden: quantification of the total number of nonsynonymous mutations per coding area of ​​the tumor genome.

[0057] Tumor sample: A tissue sample obtained from a tumor.

[0058] II. Biomarker Description CD3: CD3 is a cell surface receptor complex that is frequently used as a biomarker to define cells with T cell lineage. The CD3 complex is composed of four different polypeptide chains: CD3-gamma, CD3-delta, CD3-epsilon, and CD3-zeta. CD3-gamma and CD3-delta form heterodimers with CD3-epsilon (εγ-homodimer and εδ-heterodimer), respectively, while CD3-zeta forms a homodimer (ζζ-homodimer). Functionally, the εγ, εδ, and ζζ homodimers form a signaling complex together with the T cell receptor complex. Exemplary sequences of human CD3 gamma, delta, epsilon, and zeta chains (as well as isoforms and variants thereof) can be found in Uniprot Accession Nos. P09693 (canonical amino acid sequence disclosed herein in SEQ ID NO:1), P04234 (canonical amino acid sequence disclosed herein in SEQ ID NO:2), P07766 (canonical amino acid sequence disclosed herein in SEQ ID NO:3), and P20963 (canonical amino acid sequence disclosed herein in SEQ ID NO:4). As used herein, the term "human CD3 protein biomarker" encompasses any CD3 gamma, delta, epsilon and CD3-zeta chain polypeptides having a reference human sequence and naturally occurring variants thereof that maintain the function of the reference sequence; εγ homodimers, εδ heterodimers and ζζ homodimers that contain one or more of the CD3 gamma, delta, epsilon and CD3-zeta chain polypeptides having a reference human sequence and naturally occurring variants thereof that maintain the function of the reference sequence; and any signaling complex that contains one or more of the following: one or more CD3 homodimers or heterodimers.In some embodiments, a human CD3 protein biomarker-specific agent includes any biomarker-specific agent that specifically binds to a structure (e.g., an epitope) within a CD3 gamma chain polypeptide (e.g., the polypeptide of SEQ ID NO:1), a CD3 delta chain polypeptide (e.g., the polypeptide of SEQ ID NO:2), a CD3 epsilon chain polypeptide (e.g., the polypeptide of SEQ ID NO:3), or a CD3 zeta chain polypeptide (e.g., the polypeptide of SEQ ID NO:4), or that binds to a structure (e.g., an epitope) located within the εγ-homodimer, εδ-heterodimer, or ζζ homodimer.

[0059] CD8: CD8 is a heterodimeric disulfide-linked transmembrane glycoprotein found in cytotoxic suppressor T cell subsets, thymocytes, certain natural killer cells, and a subpopulation of myeloid cells. Exemplary sequences of the human alpha and beta chains of the CD8 receptor (and its isoforms and variants) can be found in Uniprot Accession No. P01732 (reference amino acid sequence disclosed herein in SEQ ID NO: 5) and P10966 (reference amino acid sequence disclosed herein in SEQ ID NO: 6). As used herein, the term "human CD8 protein biomarker" encompasses any CD8 alpha chain polypeptide with a reference human sequence and its natural variants that maintain the function of the reference sequence; any CD8 beta chain polypeptide with a reference human sequence and its natural variants that maintain the function of the reference sequence; any dimer containing a CD8 alpha chain polypeptide and / or a CD8 beta chain polypeptide with a reference human sequence and its natural variants that maintain the function of the reference sequence. In some embodiments, a human CD8 protein biomarker-specific agent includes any biomarker-specific agent that specifically binds to a structure (such as an epitope) within a CD8 alpha chain polypeptide (such as the polypeptide of SEQ ID NO:5), a CD8 beta chain polypeptide (such as the polypeptide of SEQ ID NO:6), or binds to a structure (such as an epitope) located within the CD8 dimer.

[0060] CD68: CD68 is a glycoprotein encoded by the CD68 gene located at position 17p13.1 on chromosome 17. CD68 protein is found in cytoplasmic granules of a variety of different blood and muscle cells and is frequently used as a biomarker for cells of the macrophage lineage, including monocytes, histiocytes, giant cells, Kupffer cells and osteoclasts. An exemplary sequence of human CD68 (and its isoforms and variants) can be found in Uniprot Accession No. P 34810 (whose canonical amino acid sequence is disclosed herein in SEQ ID NO: 7). As used herein, the term "human CD68 protein biomarker" encompasses any CD68 polypeptide having the standard human sequence and naturally occurring variants thereof that maintain the function of the standard sequence. In some embodiments, a human CD20 protein biomarker specific agent encompasses any biomarker specific agent that specifically binds to a structure (such as an epitope) within a human CD68 polypeptide (e.g., the polypeptide of SEQ ID NO: 7).

[0061] Pan-cytokeratin: As used herein, "pan-cytokeratin" and "PanCK" refer to any biomarker-specific reagent or group of biomarker-specific reagents that specifically bind to multiple cytokeratins sufficient to specifically stain epithelial tissue in a tissue sample. Exemplary pan-cytokeratin biomarker-specific reagents typically include either: (a) a single cytokeratin-specific reagent that recognizes an epitope common to multiple cytokeratins, where most epithelial cells in the tissue express at least one of the multiple cytokeratins; or (b) a cocktail of biomarker-specific reagents that are specifically reactive with multiple cytokeratins, where most epithelial cells in the tissue express at least one of the multiple cytokeratins. Reference to a "cocktail" in this definition includes both a single composition that includes each member of the multiple, or providing each member of the multiple as a separate composition but staining them with a single dye, or combinations thereof. The PanCK cocktail has been reviewed by NordiQC. In some embodiments, the PanCK biomarker specific reagent comprises an antibody cocktail containing two or more of the antibody clones selected from the group consisting of 5D3, LP34, AE1, AE2, AE3, MNF116, and PCK-26. In certain embodiments, the PanCK cocktail is selected from the group consisting of a cocktail of AE1 and AE3, a cocktail of AE1, AE3, and 5D3, and a cocktail of AE1, AE3, and PCK26. The cocktail of AE1 and AE3 is commercially available from Agilent Technologies (catalog numbers GA05361-2, IS05330-2, IR05361-2, and M351501-2, M351529-2). The cocktail of AE1, AE3, and 5D3 is commercially available from BioCare (catalog numbers CM162, IP162, OAI162, and PM162) and Abcam (catalog number ab86734). A cocktail of AE1, AE3 and PCK26 is available from Roche (catalogue number 760-2135).

[0062] PD-1: Programmed death ligand 1 (PD-1) is a member of the CD28 family receptors encoded by the PDCD1 gene on chromosome 2. An exemplary sequence of the human PD-1 protein (as well as its isoforms and variants) can be found in Uniprot Accession No. Q15116 (the reference amino acid sequence disclosed herein in SEQ ID NO: 8). In some embodiments, a human PD-1 protein biomarker-specific agent encompasses any biomarker-specific agent that specifically binds to a structure (such as an epitope) within a human PD-1 polypeptide (such as the polypeptide of SEQ ID NO: 8).

[0063] PD-L1: Programmed death ligand 1 (PD-L1) is a type 1 transmembrane protein encoded by the CD274 gene on chromosome 9. PD-L1 acts as a ligand for PD-1 and CD80. An exemplary sequence of human PD-L1 protein (and its isoforms and variants) can be found in Uniprot Accession No. Q9NZQ7 (the reference amino acid sequence disclosed herein in SEQ ID NO: 9). In some embodiments, a human PD-L1 protein biomarker-specific agent includes any biomarker-specific agent that specifically binds to a structure (such as an epitope) within a human PD-L1 polypeptide (such as the polypeptide of SEQ ID NO: 9).

[0064] III. Prediction of response to PD-1 axis-directed therapy Tissue-based biomarker signatures have the potential to stratify dMMR / MSI-H metastatic CRC (mCRC) subjects based on their likelihood of deriving survival benefit from anti-PD-1 therapy. Incorporating such biomarker signatures into clinical practice may not only optimize treatment outcomes but also protect patients from unnecessary adverse events.

[0065] Disclosed herein are methods for predicting response to PD-1 axis-directed therapy in colorectal tumors that are one or more of dMMR and MSI-H for which PD-1 axis-directed therapy is being considered, including stage III and stage IV tumors.

[0066] An overview of one exemplary process is outlined in FIG. 1A. One or more sets of feature metrics are derived from one or more affinity histochemistry (AHC)-labeled samples (hereafter referred to as "feature sets") 101 of the tumor. A scoring function combines the feature metrics (optionally including other variables) into a predicted response score (PRS) that correlates with the likelihood that the subject will respond to a PD-1 axis-directed therapy 102 and is compared to one or more predefined cutoffs. In some embodiments, the PRS indicates the likelihood of overall survival (OS) after receiving the PD-1 axis-directed therapy. In another embodiment, the PRS indicates the likelihood of progression-free survival (PFS) after receiving the PD-1 axis-directed therapy. Depending on the PRS, the subject is administered a PD-1 axis-directed therapy 104 or an alternative therapy 105.

[0067] Another exemplary process is outlined in FIG. 1B. In this example, the feature metric 101 is the average number of PD-1+ cells within a predetermined distance of at least one PD-L1+ cell. The PRS is a univariate analysis that compares the average number of PD-1+ cells within a predetermined distance of at least one PD-L1+ cell to a predetermined cutoff that correlates with the likelihood that the subject will respond to a PD-1 axis-directed therapy 103. If the average number is above the predetermined cutoff, the subject is administered a PD-1 axis-directed therapy 104a. If the average number is below the predetermined cutoff, an alternative therapy 105a is administered.

[0068] III.A. Quantifying Feature Sets

[0069] Feature sets useful in the present method are selected from the group consisting of Feature Set 1, Feature Set 2, Feature Set 3, Feature Set 4, and Feature Set 5 as shown in Table 1 (FS: feature set; SD: standard deviation; MnD: mean distance; MdD: median distance; PTO: peripheral outer; EM: epithelial marker). [Table 1]

[0070] For feature set 5, the predetermined distance is generally in the range of 5 to 50 μm, including, for example, 5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 μm. In a particular embodiment, the predetermined distance for feature set 5 is 10 μm.

[0071] These feature sets are extracted from one or more tissue samples from tumors labeled for biomarker(s) indicated by affinity histochemistry (AHC) assays. Each feature set includes (a) a set of features to be identified in the sample (such as cell types classified by one or more biomarker statuses); (b) a metric that describes the spatial relationship between one or more of the features (such as density of a feature or distance between two features); and (c) a region of interest (ROI) from which the metric is described.

[0072] III.A.1. Sample

[0073] The sample used in AHC assay is typically a tissue sample that is processed in a manner compatible with histochemical labeling, including, for example, fixation, embedding in a wax matrix (such as paraffin), and sectioning (such as microtome).No specific processing step is required by the present disclosure, as long as the obtained sample is compatible with multiple histochemical labeling of the sample for the biomarkers of interest, generating a digital image of the labeled sample, and identifying the region of interest in which the feature is identified.In a specific embodiment, the sample is a microtome section of a formalin-fixed paraffin-embedded (FFPE) sample.

[0074] The sample is derived from a tumor previously determined to have one or more of dMMR or MSI-H. In a specific embodiment, the tumor is a stage III tumor. In another specific embodiment, the tumor is a stage IV tumor.

[0075] Mismatch repair status (also referred to as "MMR") typically involves assessing the expression and / or methylation status of four genes involved in mismatch repair: hPMS2, hMLH1, hMSH2, and hMSH6. Tumors that are deficient in expression of any of these four genes are determined to be mismatch repair deficient (referred to as "dMMR"), whereas tumors that are not deficient in expression of any of these genes are determined to be MMR-proficient (referred to as "pMMR"). MMR status can be determined, for example, by protein-based assays (immunoassays such as enzyme-linked immunosorbent assays (e.g., ELISA) or affinity histochemistry assays (AHC) assays) or polymerase chain reaction (PCR) assays (e.g., real-time reverse transcriptase PCR assays).

[0076] Microsatellite unstable ("MSI") tumors are tumors in which length alterations at microsatellite loci have accumulated in the tumor beyond a predefined threshold. In contrast, microsatellite stable (MSS) tumors have not accumulated length alterations at microsatellite loci beyond a predefined threshold. Assays for assessing MSI / MSS status are well known in the art. See, e.g., Murphy et al., J. Mol. Diagn., Vol. 8, Issue 3, pp. 305-11 (Jul. 2006); Esemuede et al., Ann. Surg. Oncol., vol. 17, Issue 12, pp. 3370-78 (Dec. 2010); Mukherjee et al., Hereditary Cancer in Clinical Practice, Vol. 8, Issue 9 (2010); MSI Analysis System (Promega) (assessment of seven markers for the MSI phenotype, including five nearly monomorphic mononucleotide repeat markers (BAT-25, BAT-26, MONO-27, NR-21 and NR-24) and two highly polymorphic pentanucleotide repeat markers (Penta C and Penta D)).

[0077] III.A.2.AHC Panel

[0078] AHC assays rely on one or more panels of biomarker-specific reagents. In certain embodiments, at least one of the panels is a multiplex panel. As used in the context of this disclosure, a "multiplex panel" refers to a set of biomarker-specific reagents useful in a multiplex AHC ("mAHC") assay to differentially label multiple biomarkers in the same sample. At a minimum, the multiplex panel must be sufficient to label tissue in a manner that allows detection of at least one of the features and measurement of at least one of the feature metrics of feature sets 1-4.

[0079] A single multiplex panel may be used to label all features and detect all feature metrics for each Feature Set. Exemplary multiplex panels useful for such embodiments are shown in Table 2 (* indicates any biomarker-specific reagent). [Table 2]

[0080] Alternatively, separate panels may be used for different feature metrics for each Feature Set. Exemplary panels for Feature Sets 1-3 are shown in Tables 3-6 (* indicates any biomarker-specific reagent). [Table 3] [Table 4] [Table 5] [Table 6] [Table 7]

[0081] These examples are not intended to be an exhaustive list of panels useful for detecting features and feature metrics. In an exemplary embodiment, AHC is a multiplexed immunohistochemistry (mIHC) assay comprising applying at least one of a multiplexed panel selected from the group consisting of multiplexed panels A-F to a tissue sample from a tumor by mIHC methodology, and the biomarker-specific reagent of the panel is an antibody. In another embodiment, a multiplexed panel selected from the group consisting of multiplexed panels A-F comprises an antibody against one or more epithelial markers. In yet another embodiment, the antibody against one or more epithelial markers of panels A-F is a panCK antibody.

[0082] III.A.3. Biomarker labeling, counterstaining, and morphological staining

[0083] A panel of biomarker-specific reagents is used in combination with a set of appropriate detection reagents to generate biomarker-labeled sections. Biomarker labeling is typically accomplished by contacting a section of a sample with a biomarker-specific reagent under conditions that facilitate specific binding between the biomarker and the biomarker-specific reagent. The sample is then contacted with a set of detection reagents that interact with the biomarker-specific reagent to promote deposition of detectable moieties in proximity to the biomarker, thereby generating a detectable signal localized to the biomarker. Typically, a washing step is performed between the application of different reagents to prevent undesired non-specific labeling of tissue. The biomarker-labeled section may optionally be further labeled with a contrast agent (such as hematoxylin stain) to visualize macromolecular structures. Additionally, serial sections of the biomarker-labeled section(s) may be labeled with a morphological stain to facilitate ROI identification.

[0084] The detectable moieties used with the panel should be compatible with multiple affinity histochemical labeling methods, such as multiplexed immunohistochemistry (IHC). In some embodiments, the detectable moiety is a fluorophore. Exemplary fluorophores include several common chemical classes, such as coumarins, fluoresceins (or fluorescein derivatives and analogs), rhodamines, resorufins, luminophores, and cyanines. Additional examples of fluorescent molecules can be found in Molecular Probes Handbook-A Guide to Fluorescent Probes and Labeling Technologies, Molecular Probes, Eugene, OR, ThermoFisher Scientific, 11 th Edition. Exemplary fluorescent dyes compatible with multiplex IHC and methodologies for using same are disclosed, for example, in Gorris, Hofman, and Parra. In other embodiments, the detectable moiety is a molecule detectable via brightfield microscopy. Exemplary brightfield dyes compatible with multiplex IHC and methods for using same are disclosed, for example, in Hofman, Ide, Morrison, Parra, Stack, and U.S. Pat. No. 10,041,950 B2. Specific examples include diaminobenzidine (DAB), 4-(dimethylamino)azobenzene-4'-sulfonamide (DABSYL), tetramethylrhodamine (DISCOVERY Purple), N,N'-biscarboxypentyl-5,5'-disulfonato-indo-dicarbocyanine (Cy5), rhodamine 110 (rhodamine). In yet other embodiments, the detectable moiety is a mass spectrometer-detectable label. A review of mass spectrometry-based multiplexing methods and labels can be found, for example, in Levenson and Parra.

[0085] Non-limiting examples of commercially available detection reagents or kits containing detection reagents suitable for use in the methods of the invention include the following: VENTANA ULTRAVIEW Detection System (secondary antibodies conjugated to enzymes, including HRP and AP), VENTANA IVIEW Detection System (biotinylated anti-species secondary antibodies and streptavidin-conjugated enzymes), VENTANA OPTIVIEW Detection System (OptiView) (anti-species secondary antibodies conjugated to haptens and anti-hapten tertiary antibodies conjugated to enzyme multimers), VENTANA Amplification Kit (unconjugated secondary antibodies that can be used with any of the above VENTANA Detection Systems to amplify the number of deposited enzymes at the sites of primary antibody binding ... OPTIVIEW Detection System (OptiView) (anti-hapten secondary antibodies conjugated to haptens and anti-hapten tertiary antibodies conjugated to enzyme multimers), VENTANA OPTIVIEW Detection System (OptiView) (anti-hapten secondary antibodies conjugated to haptens Amplification systems include an anti-species secondary antibody conjugated to a hapten, an anti-hapten tertiary antibody conjugated to an enzyme multimer, and a tyramide conjugated to the same hapten. In use, the secondary antibody is contacted with the sample, resulting in binding to the primary antibody. The sample is then incubated with the anti-hapten antibody, resulting in the association of an enzyme with the secondary antibody. The sample is then incubated with tyramide, resulting in the attachment of additional hapten molecules. The sample is then incubated again with the anti-hapten antibody, resulting in the attachment of additional enzyme molecules.The sample is then incubated with a detectable moiety to allow dye attachment to occur; VENTANA DISCOVERY, DISCOVERY OMNIMAP, DISCOVERY ULTRAMAP anti-hapten antibodies, secondary antibodies, chromogens, fluorophores, and dye kits (each of which is available from Ventana Medical Systems, Inc., Tucson, Arizona); POWERVISION and POWERVISION+ IHC Detection Systems (secondary antibodies polymerized directly with HRP or AP into compact polymers with a high enzyme to antibody ratio); DAKO ENVISION™+ System (enzyme-labeled polymer conjugated to secondary antibodies); ULTRAPLEX Multiplex Chromogenic IHC Technology from CELL IDx (hapten-labeled primary antibodies combined with enzyme- or fluorine-labeled anti-hapten secondary antibodies).

[0086] If desired, biomarker-labeled slides may be counterstained to aid in the manual or automated identification of morphologically relevant areas for identifying ROIs. Examples of counterstains include chromogenic nuclear counterstains such as hematoxylin (stains blue to purple), methylene blue (stains blue), toluidine blue (stains nuclei dark blue and polysaccharides pink to red), nuclear fast red (also known as Köln-Echtrot, stains red), and methyl green (stains green); eosin (4',6-diamino-2-phenylindole (DAPI, stains blue), propidium iodide (stains red), Hoechst stain (stains blue), Nuclear Green DCS1 (stains green), Nuclear Yellow (Hoechst Chromogenic non-nuclear stains such as S769121 (stains yellow at neutral pH and blue at acidic pH), DRAQ5 (stains red), and DRAQ7 (stains red); and fluorescent non-nuclear stains such as fluorophore-labeled phalloidin (stains filamentous actin, color dependent on conjugated fluorophore).

[0087] An automated AHC labeling system may be used to apply the AHC assay and counterstain to the sample. The automated AHC labeling system typically includes at least reservoirs of various reagents used in the labeling protocol, a reagent dispensing unit in fluid communication with the reservoirs to dispense the reagents to the sample, a waste removal system to remove used reagents and other waste from the sample, and a control system to coordinate the operation of the reagent dispensing unit and the waste removal system. In addition to performing the labeling step, many automated AHC labeling systems can also perform steps that are incidental to labeling (or are compatible with separate systems that perform such auxiliary steps), including slide baking (to attach the sample to the slide), dewaxing (also called deparaffinization), antigen retrieval, counterstaining, dehydration and removal, and coverslipping. Prichard, Overview of Automated Immunohistochemistry, Arch Pathol Lab Med., Vol. 138, pp. 1578-1582 (2014), which is incorporated herein by reference in its entirety, describes several specific examples of automated AHC labeling systems and their various features, including the intelliPATH (Biocare Medical), WAVE (Celerus Diagnostics), DAKO OMNIS and DAKO AUTOSTAINER LINK 48 (Agilent Technologies), BENCHMARK (Ventana Medical Systems, Inc.), Leica BOND and LAB VISION AUTOSTAINER (Thermo Scientific) automated AHC labeling systems.Ventana Medical Systems, Inc. is the assignee of several U.S. patents disclosing systems and methods for performing automated analyses, including U.S. Patent No. 5,650,327, U.S. Patent No. 5,654,200, U.S. Patent No. 6,296,809, U.S. Patent No. 6,352,861, U.S. Patent No. 6,827,901, and U.S. Patent No. 6,943,029, as well as U.S. Patent Application Publication No. 20030211630 and U.S. Patent Application Publication No. 20040052685, each of which is incorporated by reference in its entirety. Commercially available labeling units typically operate on one of the following principles: (1) open individual slide labeling, where the slide is positioned horizontally and reagents are dispensed as paddles onto the surface of the slide containing the tissue sample (implemented, for example, in the DAKO AUTOSTAINER Link 48 (Agilent Technologies) and INTELLIPATH (Biocare Medical) labeling instruments); (2) liquid overlay technique, where reagents are covered by or dispensed through an inert fluid layer deposited on the sample (implemented, for example, in the BENCHMARK and DISCOVERY labeling instruments); (3) capillary gap labeling, where the slide surface is placed close to another surface (such as another slide or a cover plate) to create a narrow gap through which capillary forces draw in and keep the liquid reagents in contact with the sample (such as the labeling principle used in the DAKO TECHMATE, Leica BOND, and DAKO OMNIS labeling instruments). Even if capillary gap labeling is repeated several times, the fluids in the gap do not mix (e.g. in DAKO TECHMATE and Leica BOND). In a variation of capillary gap labeling called dynamic gap labeling, the sample is applied to the slide using capillary forces and then parallel surfaces are translated into each other to agitate and mix the reagents during incubation (such as the labeling principle implemented in the DAKO OMNIS slide labeling instrument (Agilent)). In translation gap labeling, a translatable head is placed over the slide.The underside of the head is spaced from the slide by a first gap small enough to allow a meniscus of liquid to form from the liquid on the slide during translation of the slide. A mixing extension having a lateral dimension smaller than the width of the slide extends from the underside of the translatable head to define a second gap smaller than the first gap between the mixing extension and the slide. During translation of the head, the lateral dimension of the mixing extension is sufficient to generate a lateral movement in the liquid on the slide generally in a direction extending from the second gap to the first gap. See WO 2011 / 139978 A1. It has also been proposed to use inkjet technology to deposit reagents on the slide. See WO 2016-170008 A1. This list of labeling techniques is not intended to be comprehensive, and any fully or semi-automated system or manual method for performing biomarker labeling may be incorporated into the method.

[0088] It may also be desirable to morphologically stain serial sections of the biomarker-labeled sections, which may be used to identify ROIs for scoring. Many morphological stains are known, including, but not limited to, hematoxylin and eosin (H&E) stains and Lee's stains (methylene blue and basic fuchsin). In certain embodiments, at least one serial section of each biomarker-labeled slide is H&E stained. Any method of application of H&E stain may be used, including manual and automated methods. In certain embodiments, at least one section of the sample is an H&E stained sample stained with an automated staining system. Automated systems for performing H&E staining typically operate on one of two staining principles: batch staining (also known as "dip'n dunk") or individual slide staining. Batch stainers generally use a reagent vat or bath in which many slides are immersed simultaneously. Individual slide stainers, on the other hand, apply reagents directly to each slide, with no two slides sharing the same aliquot of reagent. Examples of commercially available H&E stainers include the VENTANA HE 600 series H&E stainer (individual slide stainer) manufactured by Roche; the DAKO COVERSTAINER (batch stainer) manufactured by Agilent Technologies; and the LEICA ST4020 Small Linear Stainer, LEICA ST5020 MULTISTAINER, and LEICA ST5010 AUTOSTAINER XL series H&E stainers (batch stainers) manufactured by Leica Biosystems Nussloch GmbH.

[0089] III.A.4. Areas of interest

[0090] Each feature metric is derived from a specific region of interest (ROI) within the sample. The ROI is a biologically relevant location in the tissue section where the relevant spatial relationships between different cell types are evaluated. Exemplary ROIs useful in the present method are listed in Table 8. [Table 8]

[0091] Exemplary predetermined distances away from the leading edge of invasion for use in the PI, PO, and PR regions include, but are not limited to, distances ranging from about 250 μm to about 750 μm, about 400 μm to about 600 μm, or about 500 μm. As used in this context, the term "about" is intended to encompass any distance within 10% of the recited endpoint.

[0092] III.A.5: Spatial Relationships

[0093] The feature metric quantification 101 is based on the spatial relationship of one or more cell types within the associated ROI. Each of the features from feature sets 1-5 is a cell type classified by the presence and / or absence of one or more biomarkers, and the associated feature metric is the spatial relationship between the various cell types. Any method of measuring the associated feature metric in the respective ROI may be used.

[0094] In one embodiment, the spatial relationships are quantified automatically using an image analysis system, an exemplary image analysis system being described below in Section IV.

[0095] III.B: Scoring Functions

[0096] After the feature metric(s) are calculated for the image(s), they are input into a scoring function to calculate a predicted response score (PRS) for the tumor 102. The calculated PRS is compared to one or more cutoffs 103 to determine whether the subject is likely to respond to the PD-1 axis-directed therapy. In an exemplary embodiment, the scoring function is a continuous scoring function based on a feature set selected from the group consisting of Feature Set 1, Feature Set 2, Feature Set 3, and Feature Set 4. In a further embodiment, the continuous scoring function is a Cox proportional hazards model based on a feature set selected from the group consisting of Feature Set 1, Feature Set 2, Feature Set 3, and Feature Set 4.

[0097] Scoring functions are typically modeled on tissue sections obtained from a cohort of subjects with tumors and known responses to PD-1 axis-directed therapy. Candidate scoring function models are generated by inputting selected feature metrics and outcome data for each member of the cohort into the modeling function. The model with the highest concordance with the response is selected as the scoring function. Exemplary modeling functions include quadratic discriminant analysis (QDA), linear discriminant analysis (LDA), support vector machines (SVM), artificial neural networks (ANN), and Cox proportional hazards modeling (COX). In an embodiment, the candidate function is modeled on feature metrics alone. In other embodiments, the candidate function includes other clinical variables such as age, sex, location of metastases, lymph node involvement, etc. In an embodiment, the model is used to correlate feature metrics to the likelihood of progressive disease after treatment within a given time frame (referred to as "progression-free survival"). In one embodiment, the model is used to correlate the feature metric to the likelihood that a patient will survive a given period of time after treatment versus the likelihood that the patient will die within that time frame of any cause (referred to as "overall survival").

[0098] Additionally, one or more stratification cutoffs may be selected to separate patients into "risk bins" according to relative risk (e.g., "high risk" and "low risk", quartiles, deciles, etc.). In one example, the stratification cutoffs are selected using a receiver operating characteristic (ROC) curve. The ROC curve allows the user to balance the sensitivity of the model (i.e., prioritizing capturing as many "positive" or "likely to respond" candidates as possible) with the specificity of the model (i.e., minimizing false positives for "likely to respond" candidates). In an embodiment, the cutoffs are selected between risk bins that are likely to respond and risk bins that are unlikely to respond, and the selected cutoffs balance sensitivity and specificity. In an embodiment, the stratification cutoffs distinguish between (a) patients who are likely to have progressive disease after treatment and (b) patients who are likely to have stable disease, partial response, or complete response to treatment. In one embodiment, the stratification cutoff distinguishes between (a) patients who are likely to have progressive disease after treatment, (b) patients who are likely to have stable disease after treatment, and (c) patients who are likely to have a partial or complete response to treatment. In one embodiment, the stratification cutoff distinguishes between (a) patients who are likely to have progressive or stable disease after treatment, and (b) patients who are likely to have a partial or complete response to treatment. In yet another embodiment, the cutoff can be the mean or median PRS.

[0099] The models may be implemented using a computational statistical analysis software suite, such as The R Project for Statistical Computing (r-project.org), SAS, MATLAB, among others.

[0100] III.C. Treatment

[0101] The PRS 106 is used to determine whether a subject is likely to respond to treatment with a PD-1 directed therapy 107 or would be more likely to benefit from an alternative course of therapy 108.

[0102] In an embodiment, the PD-1 axis-directed therapy course 107 includes PD-1 specific antibodies (such as nivolumab, pembrolizumab, cemiplimab, tislelizumab, spartalizumab, MEDI0680 (AstraZeneca), toripalimab, sintilimab, cetolimab, and pidilizumab), PD-L1 specific antibodies (such as atezolizumab, durvalumab, and avelumab), PD-1 directed bispecifics (such as teboterimub (PD-1 / LAG3 bispecific DART® molecule); PD-1 ligand fusion proteins (such as AMP-224), PD-L1 directed bispecifics (such as FS118), and Small molecule inhibitors (such as CA-170, BMS-1001, or BMS-1166). In an embodiment, the PD-1 axis-directed therapy comprises a PD-1-directed monoclonal antibody or a PD-L1-directed monoclonal antibody. In an embodiment, the PD-1 axis-directed therapy 107 comprises a therapeutic entity selected from the group consisting of nivolumab, pembrolizumab, cemiplimab, tislelizumab, spartalizumab, MEDI0680, toripalimab, sintilimab, cetolimab, atezolizumab, durvalumab, and avelumab. In an embodiment, the PD-1 axis-directed therapy comprises pembrolizumab.

[0103] In some embodiments, the PD-1 axis-directed therapy 107 further comprises a reduced chemotherapy course. A "reduced" chemotherapy course may include a reduction in the number of different chemotherapeutic agents used, the dose of one or more chemotherapeutic agent(s), and / or the duration of treatment with one or more chemotherapeutic agent(s). A reduced chemotherapy course may also include the selection of a chemotherapeutic agent with a reduced toxicity profile compared to other chemotherapeutic agents for the treatment of CRC.

[0104] In other embodiments, the PD-1 axis-directed therapy 107 further comprises another immune checkpoint-specific therapy, for example, a therapy targeting CTLA-4 (such as ipilimumab or tremilumab), IDO (such as NLG919, epacadostat, BMS-986205, PF-06840003, navoximode, indoximod, NLG802 or LY3381916), TIM-3 (such as MGB453, TSR-022, Sym023, BGBA425 (BeiGene)), LAG3 (such as relatlimab, eftilagimod alpha, ieramilimab, REGN3767, or encelimab).

[0105] In other embodiments, the PD-1 axis-directed therapy 107 further comprises another immune checkpoint-directed therapy treatment and a reduced course of chemotherapy.

[0106] In an embodiment, the alternative therapy 108 is a standard course of treatment for stage III or stage IV colorectal cancer. Current treatment protocols usually include surgical removal of the tumor and nearby lymph nodes with surgical removal of distant metastases if tumor count is low, and adjuvant chemotherapy before and / or after surgical removal. For stage III or IV colorectal cancer not suitable for surgery, chemotherapy is usually administered as a primary treatment, optionally in combination with targeted therapy if necessary. Some of the most commonly used regimens include: FOLFOX: leucovorin, fluorouracil (5-FU) and oxaliplatin (ELOXATIN); FOLFIRI: leucovorin, 5-FU, and irinotecan (CAMPTOSAR); CAPEOX or CAPOX: capecitabine (XELODA) and oxaliplatin; FOLFOXIRI: leucovorin, 5-FU, oxaliplatin and irinotecan; one of the above combinations plus a drug that targets VEGF (bevacizumab [AV ASTIN], dib-aflibercept [ZALTRAP], or ramucirumab [CYRAMZA]), or in combination with either an EGFR-targeted drug (cetuximab [ERBITUX] or panitumumab [VECTIBIX]); 5-FU and leucovorin with or without targeted drugs; capecitabine with or without targeted drugs; irinotecan with or without targeted drugs; cetuximab alone; panitumumab alone; regorafenib (Stivarga) alone; trifluridine and tipiracil (Lonsurf).

[0107] IV. Image Analysis System In one embodiment, the feature metric quantification 101 may be performed by an image analysis system. An exemplary image analysis system is shown in FIG.

[0108] Image analysis system 200 may include one or more computing devices, such as a desktop computer, a laptop computer, a tablet, a smartphone, a server, a special-purpose computing device, or any other type(s) of electronic device(s) capable of performing the techniques and operations described herein. In some embodiments, image analysis system 200 may be implemented as a single device. In other embodiments, image analysis system 200 may be implemented as a combination of two or more devices that together achieve various functionality discussed herein. For example, image analysis system 200 may include one or more server computers and one or more client computers communicatively coupled to each other via one or more local area networks and / or wide area networks (such as the Internet).

[0109] The image analysis system 200 includes a memory 201, a processor 202, and one or more output device(s) 203, such as a display, a printer, a non-transitory storage device (e.g., hard drive, flash drive, cloud drive, etc.). The memory 201 may include any combination of any type of volatile or non-volatile memory, such as random access memory (RAM), read-only memory such as electrically erasable programmable read-only memory (EEPROM), flash memory, hard drives, solid state drives, optical disks, etc. For simplicity, the memory 201 is shown in FIG. 2 as a single device, although it will be understood that the memory 201 may be distributed across two or more devices.

[0110] Processor 202 may include one or more processors of any type, such as, for example, a central processing unit (CPU), a graphics processing unit (GPU), a special purpose signal or image processor, a field programmable gate array (FPGA), a tensor processing unit (TPU), etc. For simplicity, processor 202 is illustrated in FIG. 2 as a single device, although it will be understood that processor 202 may be distributed across any number of devices.

[0111] The processor 202 implements a set of instructions stored in the memory 201, the set of instructions including extracting a feature set from one or more digital images of affinity histochemistry (AHC) labeled tissue samples from stage IV colorectal tumors previously determined to be one or more of dMMR or MSI-H, the feature set selected from the group consisting of feature set 1, feature set 2, feature set 3, and feature set 4. In an embodiment, the feature set is extracted by implementing a feature identification (FI) module 204, a region of interest (ROI) module 205, and a scoring module 206 on the digital image(s).

[0112] IV.A.FI module

[0113] The FI module 204 functions to identify features within an image that correlate to cells and to associate the identified cells with a feature vector that describes the biomarker state of the cell. The output of the FI module 204 is effectively a map of the image that annotates the location of every cell in the image (or a specified region of the image) with a feature vector for each marked cell that has sufficient information to classify the cell as being biomarker positive (+) or biomarker negative (-) for the associated biomarker.

[0114] In an embodiment, the FI module 204 is programmed to identify and classify Feature Set 1 objects in a single digital image. In such an embodiment, the image analysis system is programmed to (a) mark all cells, (b) generate feature vectors indicative of CD8, CD68 and PD-L1 status for cells at least in the stromal region, and (c) generate feature vectors indicative of CD8, PD-1 and PD-L1 status for cells at least in the tumor region. The scoring function 206 generates feature metrics, which are then calculated: (1) the standard deviation of the distance of CD8+ cells from CD68+ / PDL1+ cells within 10 μm in the stromal region; and (2) the average distance of CD8+ / PD-1+ cells from CD8+ / PD-L1+ cells within 30 μm in the tumor region. In such embodiments, mAHC-labeled sections of tumors are labeled with a first detectable moiety via a biomarker-specific reagent for CD8, a second detectable moiety via a biomarker-specific reagent for CD68, a third detectable moiety via a biomarker-specific reagent for PD-L1, a fourth detectable moiety via a biomarker-specific reagent for PD-1, and optionally a fifth detectable moiety via a biomarker-specific reagent for EM, wherein the first, second, third, fourth and fifth detectable moieties are distinguishable from one another when labeling the same cells.

[0115] In another embodiment, the FI module 204 is programmed to identify and classify Feature Set 1 objects in separate digital images from the same tumor. In such an embodiment, the image analysis system may be programmed to perform a first FI module 204 on a first digital image of a first mAHC-labeled section of the tumor and a second FI module 204 on a second digital image of a second mAHC-labeled section of the tumor, where (a) the first FI module 204 marks at least all cells in the stromal region that are CD8+ status and at least all cells in the stromal region that are CD68+ / PD-L1+, and (b) the second FI module 204 marks at least all cells in the tumor region that are CD8+ / PD-1+ and all cells in the tumor region that are CD8+ / PD-L1+. The following feature metric: standard deviation of distance of CD8+ cells from CD68+ / PDL1+ cells within 10 μm in the stromal region is calculated from the first digital image, and the following feature metric: mean distance of CD8+ / PD-1+ cells from CD8+ / PD-L1+ cells within 30 μm in the tumor region is calculated from the second digital image. In such embodiments, a first mAHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8, a second detectable moiety via a biomarker-specific reagent for CD68, and a third detectable moiety via a biomarker-specific reagent for PD-L1, where the first, second, and third detectable moieties are distinguishable from one another when labeling the same cells, and a second mAHC-labeled section of the tumor may be labeled with a fourth detectable moiety via a biomarker-specific reagent for CD8, a fifth detectable moiety via a biomarker-specific reagent for PD-1, and a sixth detectable moiety via a biomarker-specific reagent for PD-L1, where the fourth, fifth, and sixth detectable moieties (which may be the same as or different from the first, second, and third detectable moieties) are distinguishable from one another when labeling the same cells.

[0116] In another embodiment, the FI module 204 is programmed to identify and classify Feature Set 2 objects in a single digital image. In such an embodiment, the image analysis system is programmed to (a) mark all cells that are CD8+ cells and all cells that are CD68+ / PDL1+ cells, at least for cells in the outer peritumoral region, and (b) mark all cells that are CD8+, at least for cells in the tumor region. The following feature metrics are then calculated: (1) standard deviation of distance of CD8+ cells from CD68+ / PDL1+ cells within 10 μm in the outer peritumoral region; (2) density of CD8+ cells in the tumor region. In such embodiments, mAHC labeled sections of tumors may be labeled with a first detectable moiety via a biomarker specific reagent for CD8, a second detectable moiety via a biomarker specific reagent for CD68, a third detectable moiety via a biomarker specific reagent for PD-L1, and optionally a fourth detectable moiety via a biomarker specific reagent for EM, wherein the first, second, third and fourth detectable moieties are distinguishable from one another when labeling the same cells.

[0117] In another embodiment, the FI module 204 is programmed to identify and classify Feature Set 2 objects in separate digital images from the same tumor. In such an embodiment, the image analysis system may be programmed to run a first FI module 204 on a first digital image of a first mAHC-labeled section of the tumor and a second FI module 204 on a second digital image of a second mAHC-labeled section of the tumor, where (a) the first FI module 204 marks all cells in the peritumoral outer region that are CD8+ and all cells that are CD68+ / PD-L1+, at least in the peritumoral outer region, and (b) the second FI module 204 marks at least all cells in the tumor region that are CD8+. The following feature metrics are then calculated from the first digital image: standard deviation of distance of CD8+ cells from CD68+ / PDL1+ cells within 10 μm in the peritumoral outer region, and the following feature metric is calculated from the second digital image: density of CD8+ cells in the tumor region. In such embodiments, a first mAHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8, a second detectable moiety via a biomarker-specific reagent for CD68, and a third detectable moiety via a biomarker-specific reagent for PD-L1, and optionally a fourth detectable moiety via a biomarker-specific reagent for an EM marker, where the first, second, third and fourth detectable moieties are distinguishable from one another when labeling the same cells, and a second mAHC-labeled section of the tumor may be labeled with a fifth detectable moiety via a biomarker-specific reagent for CD8, and optionally a sixth detectable moiety via a biomarker-specific reagent for an EM marker, where the fifth and sixth detectable moieties (which may be the same as or different from any of the first, second, third and fourth detectable moieties) are distinguishable from one another when labeling the same cells.

[0118] In another embodiment, the FI module 204 is programmed to identify and classify Feature Set 3 objects within a single digital image. In such an embodiment, the image analysis system is programmed to: (a) mark, at least for cells in the tumor region, all cells that are CD8+, all cells that are CD8+ / PD-1+, all cells that are CD8+ / PD-L1+, and all cells that are EM+ cells; and (b) mark, at least for cells in the stromal region, all cells that are CD8+ and all cells that are PD-L1+ / EM+. The following feature metrics are then calculated: (1) median distance of CD8+ cells from EM+ cells within 30 μm in the tumor region; (2) median distance of CD8+ cells to PD-L1+ / epithelial marker+ (EM+) cells within 10 μm in the stromal region; (3) average number of CD8+ / PD-1+ cells within 10 μm of CD8+ / PD-L1+ cells; (4) median distance of CD8+ / PD-1+ cells to CD8+ / PD-L1+ cells within 30 μm. In such an embodiment, a mAHC labeled section of a tumor may be labeled with a first detectable moiety via a biomarker specific reagent for CD8, a second detectable moiety via a biomarker specific reagent for PD-L1, a third detectable moiety via a biomarker specific reagent for PD-1, and a fourth detectable moiety via a biomarker specific reagent for EM, wherein the first, second, third and fourth detectable moieties are distinguishable from one another when labeling the same cells.

[0119] In another embodiment, the FI module 204 is programmed to identify and classify Feature Set 3 objects in separate digital images from the same tumor. In one such embodiment, the image analysis system may be programmed to run a first FI module 204 on a first digital image of a first mAHC-labeled section of the tumor and a second FI module 204 on a second digital image of a second mAHC-labeled section of the tumor, where (a) the first FI module 204 marks all cells that are CD8+ and all cells that are EM+ in at least the tumor region, and (b) the second FI module 204 marks at least: (b1) all cells in the stromal region that are CD8+, (b2) all cells in the stromal region that are PD-L1+ / EM+, (b3) all cells in the tumor region that are CD8+ / PD-1+, and (b4) all cells in the tumor region that are CD8+ / PD-L1+. The following feature metrics are then calculated from the first digital image: (1) median distance of CD8+ cells from EM+ cells within 30 μm in the tumor region; and the following feature metrics are calculated from the second digital image: (2) median distance of CD8+ cells to PD-L1+ / epithelial marker+ (EM+) cells within 10 μm in the stromal region, (3) average number of CD8+ / PD-1+ cells within 10 μm of CD8+ / PD-L1+ cells, and (4) median distance of CD8+ / PD-1+ cells to CD8+ / PD-L1+ cells within 30 μm.In such embodiments, a first mAHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8 and a second detectable moiety via a biomarker-specific reagent for EM, where the first and second detectable moieties are distinguishable from one another when labeling the same cells, and a second mAHC-labeled section of the tumor may be labeled with a third detectable moiety via a biomarker-specific reagent for CD8, a fourth detectable moiety via a biomarker-specific reagent for PD-L1, a fifth detectable moiety via a biomarker-specific reagent for PD-1, and a sixth detectable moiety via a biomarker-specific reagent for the EM marker, where the third, fourth, fifth and sixth detectable moieties (which may be the same as or different from either of the first and second detectable moieties) are distinguishable from one another when labeling the same cells.

[0120] In another embodiment, in which the FI module 204 is programmed to identify and classify Feature Set 3 objects in separate digital images from the same tumor, the image analysis system may be programmed to run the first FI module 204 on a first digital image of a first mAHC-labeled section of the tumor and the second FI module 204 on a second digital image of a second mAHC-labeled section of the tumor, where (a) the first FI module 204 marks (a1) at least all cells in the tumor region that are CD8+, (a2) at least all cells in the tumor region that are EM+, (a3) ​​at least all cells in the stromal region that are CD8+, and (a4) at least all cells in the stromal region that are PD-L1+ / EM+, and (b) the second FI module 204 marks (b1) at least all cells in the tumor region that are CD8+ / PD-1+, and (b2) at least all cells in the tumor region that are CD8+ / PD-L1+. The following feature metrics are then calculated from the first digital image: (1) median distance of CD8+ cells from EM+ cells within 30 μm in the tumor region, and (2) median distance of PD-L1+ / epithelial marker+ (EM+) cells from CD8+ cells within 10 μm in the stromal region; and the following feature metrics are calculated from the second digital image: (3) mean CD8+ / PD-1+ cell counts from CD8+ / PD-L1+ cells within 10 μm, and (4) median distance of CD8+ / PD-1+ cells from CD8+ / PD-L1+ cells within 30 μm.In such embodiments, a first mAHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8, a second detectable moiety via a biomarker-specific reagent for EM, and a third detectable moiety via a biomarker-specific reagent for PD-L1, where the first, second, and third detectable moieties are distinguishable from one another when labeling the same cells, and a second mAHC-labeled section of the tumor may be labeled with a fourth detectable moiety via a biomarker-specific reagent for CD8, a fifth detectable moiety via a biomarker-specific reagent for PD-L1, a sixth detectable moiety via a biomarker-specific reagent for PD-1, and optionally a seventh detectable moiety via a biomarker-specific reagent for the EM marker, where the third, fourth, fifth, and sixth detectable moieties (which may be the same as or different from either the first and second detectable moieties) are distinguishable from one another when labeling the same cells.

[0121] In another embodiment in which the FI module 204 is programmed to identify and classify Feature Set 3 objects in separate digital images from the same tumor, the image analysis system may be programmed to perform a first FI module 204 on a first digital image of a first mAHC-labeled section of the tumor, a second FI module 204 on a second digital image of a second mAHC-labeled section of the tumor, and a third FI module 204 on a third digital image of a third mAHC-labeled section of the tumor, wherein (a) the first FI module 204 performs 1) marks at least all cells in the tumor area that are CD8+, and (a2) at least all cells in the tumor area that are EM+; (b) the second FI module 204 marks (b1) at least all cells in the stromal area that are CD8+, and (b2) at least all cells in the stromal area that are PD-L1+ / EM+ cells; and (c) the third FI module 204 marks (c1) at least all cells in the tumor area that are CD8+ / PD-1+, and (c2) at least all cells in the tumor area that are CD8+ / PD-L1+. The following feature metrics are then calculated from the first digital image: (1) median distance of CD8+ cells from EM+ cells within 30 μm in the tumor region; then the following feature metrics are calculated from the second digital image: (2) median distance of CD8+ cells to PD-L1+ / epithelial marker+ (EM+) cells within 10 μm in the stromal region; the following feature metrics are calculated from the third digital image: (3) mean CD8+ / PD-1+ cell count of CD8+ / PD-L1+ cells within 10 μm, and (4) median distance of CD8+ / PD-1+ cells to CD8+ / PD-L1+ cells within 30 μm.In such an embodiment, a first mAHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8 and a second detectable moiety via a biomarker-specific reagent for EM, the first and second detectable moieties being distinguishable from one another when labeling the same cells, and a second mAHC-labeled section of the tumor may be labeled with a third detectable moiety via a biomarker-specific reagent for CD8, a fourth detectable moiety via a biomarker-specific reagent for PD-L1, and a fifth detectable moiety via a biomarker-specific reagent for EM, the third, fourth, and fifth detectable moieties (any of the first and second detectable moieties) being distinguishable from one another when labeling the same cells. a third mAHC labeled section of the tumor may be labeled with a sixth detectable moiety via a biomarker specific reagent for CD8, a seventh detectable moiety via a biomarker specific reagent for PD-1, an eighth detectable moiety via a biomarker specific reagent for PD-1, and optionally a ninth detectable moiety via a biomarker specific reagent for an EM marker, and the sixth, seventh, eighth, and ninth detectable moieties (which may be the same as or different from any of the first, second, third, fourth, and fifth detectable moieties) are distinguishable from each other when labeling the same cells;

[0122] In another embodiment, the FI module 204 is programmed to identify and classify Feature Set 4 objects in a single digital image. In such an embodiment, the image analysis system is programmed to mark all cells that are CD8+ and all cells that are EM+, at least for cells in the tumor region. The density of CD8+ / EM- cells in the tumor region is then calculated. In such an embodiment, a first mAHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8 and a second detectable moiety via a biomarker-specific reagent for EM, the first and second detectable moieties being distinguishable from each other when labeling the same cells.

[0123] As used herein, detectable moieties are "distinguishable from each other when labeling the same cell" so long as the presence or absence of each moiety can be detected in the same cell.

[0124] IV.B.ROI Module

[0125] The ROI module 205 is used to generate an ROI within an image from which feature metrics are calculated.

[0126] In some embodiments, the ROI module 205 generates a graphic user interface (GUI) in which a user manually annotates an ROI in a digital image. A trained expert (such as a pathologist) may use a user interaction device (e.g., a mouse, touchpad, stylus, touch-responsive display, etc.) to delineate one or more morphological region(s) (such as tumor areas and / or leading edges of invasion) on a representation of the digital image of the sample in the GUI. The delineated area(s) on the image may then be used as the ROI.

[0127] In other embodiments, the ROI module 205 assists the user in annotating the ROI (referred to as "semi-automated ROI annotation"). For example, the ROI module 205 may generate a GUI that includes the digital image. The user may use and delineate one or more regions on the digital image, which the ROI module 205 then automatically converts to a complete ROI. For example, if the desired ROI is a PI, PO, and / or PR region, the user may delineate the tumor region and the leading edge of the invasion, and the system will automatically draw the PI, PO, and PR regions as defined by the user. In another embodiment where the ROI is an EA or SA, the user may delineate the tumor region, and optionally the leading edge of the invasion, on the image, which is then registered to the biomarker-labeled image, and the system creates the associated EA and SA ROIs by marking all cells within a predetermined distance of EM+ cells as being in the EA, and all cells beyond a predetermined distance as being in the SA. As another example, the ROI module 205 may apply a pattern recognition function that uses computer vision and machine learning to identify regions with similar morphological characteristics to delineate and / or automatically generate an ROI. Thus, for example, a tumor region may be annotated in a semi-automated manner by a method that includes: (a) a user annotates a tumor region in an H&E image of a sample by outlining the tumor region; (b) the ROI module 205 applies a pattern recognition function to identify additional areas of the sample that have morphological characteristics of the outlined area, such that the entire tumor region includes the areas annotated by the user and the areas automatically identified by the system.In another example, the PR, PI, and / or PO ROIs may be annotated in a semi-automated manner by a method that includes: (a) a user annotates a tumor region in an H&E image of the sample by outlining the tumor region and the leading edge of the invasion; (b) the ROI module 205 automatically defines a PR region, a PI region, and / or a PO region(s) that encompasses all pixels within a defined distance of the annotated invasive front; and (c) the ROI module 205 applies a pattern recognition function to identify additional areas of the sample that have morphological characteristics of the PI, PO, and / or PR region identified by step (b). Many other configurations may also be used. In cases where the ROI generation is semi-automated, the user may be given the option to modify the ROI annotated by the computer system, for example, by enlarging the ROI, annotating areas of the ROI or objects within the ROI to be excluded from analysis, etc.

[0128] In other embodiments, the ROI module 205 may automatically suggest ROIs without direct input from a user (referred to as "automated ROI annotation"). For example, a pre-trained tissue partitioning function or other pattern recognition function may be applied to an unannotated image to identify desired morphological regions for use as ROIs. The user may be given the option to modify the ROI annotated by the computer system, for example, by enlarging the ROI, annotating areas of the ROI or objects within the ROI to be excluded from analysis, etc.

[0129] In yet other embodiments, the ROI may be generated by using a registration function, whereby an ROI annotated on one section of a set of serial sections is automatically transferred to other sections of the set of serial sections. This functionality is particularly useful when H&E stained serial sections are provided along with biomarker-labeled sections. In such embodiments, a user may, for example, delineate a tumor region in a digital image of the H&E stained section. The system then registers the ROI from the H&E image to the image of the biomarker-labeled serial section, matching the tissue structure from the H&E image to the corresponding tissue structure of the serial section. Exemplary registration methods can be found, for example, in WO 2013 / 140070 and U.S. Patent Application Publication No. 2016-0321809.

[0130] The FI module 204 and the ROI module 205 may be implemented in any order. For example, the FI module 204 may be applied to the entire image first. The positions and feature vectors of the identified objects can be stored and recalled later when the ROI module 205 is implemented. In such a configuration, the scores can be generated by the scoring module 206 immediately after the generation of the ROI. Such a workflow is illustrated in FIG. 3A. As can be seen in FIG. 3A, an image is obtained with a mixture of different objects (shown as solid ovals and solid diamonds). After the object identification task is implemented, all diamonds in the image are identified (shown as open diamonds). Once the ROI is added to the image (shown as dashed lines), only diamonds located within the ROI area are included in the ROI's metric calculation. Alternatively, the ROI module 205 can be implemented first. Such a workflow is illustrated in FIG. 3B. As can be seen in FIG. 3B, an image is obtained with a mixture of different objects (shown as solid ovals and solid diamonds). The ROI is added to the image (shown as dashed lines), but no objects are yet marked. The FI module 204 may be implemented only in the ROI (minimizing computation time) or may still be implemented in the entire image (allowing for on-the-fly adjustments without rerunning the FI module 204). It is also possible to implement the FI module 204 and the ROI module 205 simultaneously.

[0131] Although these modules are shown in FIG. 2 as stand-alone modules, it will be apparent to one of ordinary skill in the art that each module may instead be implemented as multiple sub-modules, and in some embodiments, any two or more modules may be combined into a single module. Additionally, in some embodiments, image analysis system 200 may include additional modules (e.g., input devices, networking and communication modules, etc.) that are not shown in FIG. 2 for the sake of brevity. Additionally, in some embodiments, some of the blocks shown in FIG. 2 may be disabled or omitted. As described in more detail below, the functionality of some or all of the modules in image analysis system 200 may be implemented as hardware, software, firmware, or any combination thereof. Exemplary commercially available software packages useful for implementing the modules disclosed herein include VENTANA VIRTUOSO; DEFINIENS TISSUE STUDIO, DEVELOPER XD, and IMAGE MINER; and VISOPHARM BIOTOPIX, ONCOTOPIX, and STEREOTOPIX software packages.

[0132] IV.C. Scoring Module

[0133] After the FI module 204 and the ROI module 205 are implemented, the scoring module 206 extracts relevant feature metrics from the relevant ROI(s) and optionally calculates the PRS by applying the feature metrics to a scoring function as described herein. The scoring module 206 is adapted to extract relevant feature metrics from the annotated ROIs based on feature vectors associated with the marked cells. This may be done after the final ROIs are selected, or it may be done continuously as the ROIs are adjusted. For example, the FI module 204 is applied to the entire image as described in FIG. 3A to generate a high-level feature vector for each cell that includes the biomarker state for all relevant cell types (regardless of ROI). The ROI module 205 and the scoring module 206 may then be applied simultaneously, in which case, when the ROI is adjusted by the ROI module 205, the scoring module adjusts the calculated feature metrics (and optionally the calculated PRS) accordingly.

[0134] After the scoring module 206 finishes calculating the feature metrics (and optionally the PRS), it may communicate the final calculated feature metrics and / or PRS to the output device 203. If the scoring module 206 does not apply a scoring function to the feature metrics, the feature metrics are sent to the output device 203, which may apply a scoring function to the feature metrics or send the feature metrics to an end user in a format that can be applied to the scoring function to generate the PRS (e.g., by displaying or printing the raw feature metric data, or by exporting the raw feature metric data to a spreadsheet or other data analysis software program). If the scoring module 206 applies a scoring function to the extracted feature metrics to calculate the PRS, the PRS and, optionally, the raw feature metric data are exported to the output device 203. The PRS may be output as a numerical value, or as a text or graphical representation of the patient's relative risk. For example, the output may be a numerical output of the scoring function when applied to the feature metrics (optionally with appropriate cutoff values ​​between risk buckets). As another example, the output may be a textual representation of the patient's relative risk (i.e., by assigning the patient to a response bucket such as "likely to respond" or "unlikely to respond"). As another example, the output may be a graphical representation of the patient's relative risk, such as a graph showing where the patient's score ranks in a population distribution. Many other outputs may be contemplated.

[0135] IV.D. Exemplary Contemplated Embodiments

[0136] IV.D.1. Feature Set 1

[0137] In an embodiment, the image analysis system 200 functions to calculate feature metrics of Feature Set 1 from a digital image of a single mAHC-labeled tissue sample. In such an embodiment, (a) the ROI module 205 annotates stromal and tumor regions on the digital image; (b) the FI module 204 (b1) marks objects corresponding to cells in the digital image; (b2) generates feature vectors indicative of CD8, CD68, PD-L1, and optionally EM status for cells at least in the stromal region; (b3) generates feature vectors indicative of CD8, PD-1, PD-L1, and optionally EM status for cells at least in the tumor region; (c) the scoring module 206 extracts feature metrics of Feature Set 1 from their respective ROIs and, optionally, calculates a PRS from the extracted feature metrics. In such embodiments, mAHC-labeled sections of tumors are labeled with a first detectable moiety via a biomarker-specific reagent for CD8, a second detectable moiety via a biomarker-specific reagent for CD68, a third detectable moiety via a biomarker-specific reagent for PD-L1, a fourth detectable moiety via a biomarker-specific reagent for PD-1, and optionally a fifth detectable moiety via a biomarker-specific reagent for EM, wherein the first, second, third, fourth and fifth detectable moieties are distinguishable from one another when labeling the same cells.

[0138] In another embodiment, image analysis system 200 is operable to calculate feature metrics of feature set 1 on digital images of separate AHC-labeled tissue samples from the same tumor. In such an embodiment, image analysis system 200 may be programmed to perform a set of functions on a first digital image of a first AHC-labeled section of the tumor and a second digital image of a second AHC-labeled section of the tumor, where (a) first ROI module 205 annotates stromal regions in the first digital image; (b) first FI module 204 marks objects corresponding to cells in at least the stromal regions and generates feature vectors for each marked cell indicative of CD8, CD68, PD-L1, and optionally EM status; (c) first scoring module 206 calculates CD8, CD68, PD-L1, and EM status from CD68+ / PDL1+ cells within 10 μm in the stromal regions of the first digital image; (d) the second ROI module 205 annotates the tumor region on the second digital image; (e) the second FI module 204 marks objects corresponding to cells in at least the tumor region of the second digital image and generates feature vectors indicative of CD8, PD-1, PD-L1, and optionally EM status for the cells in at least the tumor region; (f) the second scoring module 206 calculates the average distance of CD8+ / PD-1+ cells from CD8+ / PD-L1+ cells within 30 μm in the tumor region of the second digital image; (g) the optional third scoring module 206 calculates a PRS from the extracted feature metrics.In such embodiments, a first digital image is obtained from a first mAHC labeled section of a tumor labeled with a first detectable moiety via a biomarker specific reagent for CD8, a second detectable moiety via a biomarker specific reagent for CD68, and a third detectable moiety via a biomarker specific reagent for PD-L1, and optionally a fourth detectable moiety via a biomarker specific reagent for EM (such as pan-cytokeratin), where the first, second, third and optional fourth detectable moieties are distinguishable from one another when labeling the same cells, and a second digital image is obtained from a first mAHC labeled section of a tumor labeled with a first detectable moiety via a biomarker specific reagent for CD8, a second detectable moiety via a biomarker specific reagent for CD68, and a third detectable moiety via a biomarker specific reagent for PD-L1, and optionally a fourth detectable moiety via a biomarker specific reagent for EM (such as pan-cytokeratin), where the first, second, third and optional fourth detectable moieties are distinguishable from one another when labeling the same cells, from a second mAHC labeled section of the tumor labeled with a fifth detectable moiety via a biomarker specific reagent for PD-1, a sixth detectable moiety via a biomarker specific reagent for PD-L1, and optionally an eighth detectable moiety via a biomarker specific reagent for EM (such as pan-cytokeratin), wherein the fifth, sixth, seventh, and any eighth detectable moieties (which may be the same as or different from the first, second, third, and any fourth detectable moieties) are distinguishable from one another when labeling the same cells.

[0139] IV.D.2. Feature Set 2

[0140] In an embodiment, the image analysis system 200 functions to calculate feature metrics of Feature Set 2 from a digital image of a single mAHC-labeled tissue sample. In such an embodiment, (a) the ROI module 205 annotates the outer tumor perimeter (PO region and tumor region) on the digital image; (b) the FI module 204: (b1) marks objects corresponding to cells in the digital image; (b2) generates feature vectors indicative of CD8, CD68, PD-L1, and optionally EM status for cells at least in the PO region; (b3) generates feature vectors indicative of CD8 and optionally EM status for cells at least in the tumor region; (c) the scoring module 206 extracts feature metrics of Feature Set 2 from their respective ROIs and, optionally, calculates a PRS from the extracted feature metrics. In such embodiments, mAHC labeled sections of tumors may be labeled with a first detectable moiety via a biomarker specific reagent for CD8, a second detectable moiety via a biomarker specific reagent for CD68, a third detectable moiety via a biomarker specific reagent for PD-L1, and optionally a fourth detectable moiety via a biomarker specific reagent for EM, wherein the first, second, third and fourth detectable moieties are distinguishable from one another when labeling the same cells.

[0141] In another embodiment, image analysis system 200 is operable to calculate feature metrics of feature set 2 on digital images of separate AHC-labeled tissue samples from the same tumor. In such an embodiment, image analysis system 200 may be programmed to perform a set of functions on a first digital image of a first AHC-labeled section of the tumor and a second digital image of a second AHC-labeled section of the tumor, where (a) first ROI module 205 annotates a peritumoral outer (PO) region in the first digital image; (b) first FI module 204 marks objects corresponding to cells in at least the PO region and generates a feature vector for each marked cell indicative of CD8, CD68, PD-L1, and optionally EM status; (c) first scoring module 206 annotates the PO region of the first digital image; (d) the second ROI module 205 annotates the tumor region on the second digital image; (e) the second FI module 204 marks objects corresponding to cells within at least the tumor region of the second digital image and generates a feature vector indicative of the CD8 and optionally EM status of the cells within at least the tumor region; (f) the second scoring module 206 calculates the density of CD8+ cells within the tumor region of the second digital image; (g) the optional third scoring module 206 calculates the PRS from the extracted feature metrics.In such embodiments, a first digital image is obtained from a first mAHC-labeled section of the tumor labeled with a first detectable moiety via a biomarker specific reagent for CD8, a second detectable moiety via a biomarker specific reagent for CD68, and a third detectable moiety via a biomarker specific reagent for PD-L1, and optionally a fourth detectable moiety via a biomarker specific reagent for EM (such as pan-cytokeratin), where the first, second, third and any fourth detectable moieties are distinguishable from one another if they label the same cells, and a second digital image is obtained from a second mAHC-labeled section of the tumor labeled with a fifth detectable moiety via a biomarker specific reagent for CD8, and any sixth detectable moiety via a biomarker specific reagent for EM (e.g. pan-cytokeratin), where the fifth and any sixth detectable moieties (which may be the same as or different from the first, second, third and any fourth detectable moieties) are distinguishable from one another if they label the same cells.

[0142] IV.D.3. Feature Set 3

[0143] In an embodiment, the image analysis system 200 functions to calculate feature metrics of Feature Set 3 from a digital image of a single mAHC-labeled tissue sample. In such an embodiment, (a) the ROI module 205 annotates tumor and stromal regions on the digital image; (b) the FI module 204 (b1) marks objects corresponding to cells in the digital image and generates feature vectors indicative of CD8, PD-1, PD-L1, and EM status for at least cells in the tumor region and CD8, PD-L1, and EM status for at least cells in the stromal region; and (c) the scoring module 206 extracts feature metrics of Feature Set 3 from their respective ROIs and, optionally, calculates a PRS from the extracted feature metrics. In such embodiments, mAHC labeled sections of tumors may be labeled with a first detectable moiety via a biomarker specific reagent for CD8, a second detectable moiety via a biomarker specific reagent for PD-1, a third detectable moiety via a biomarker specific reagent for PD-L1, and a fourth detectable moiety via a biomarker specific reagent for EM (such as pan-cytokeratin), where the first, second, third and fourth detectable moieties are distinguishable from one another when labeling the same cells.

[0144] In another embodiment, image analysis system 200 is operable to calculate feature metrics of feature set 3 on digital images of separate AHC-labeled tissue samples from the same tumor. In such an embodiment, image analysis system 200 may be programmed to perform a set of functions on a first digital image of a first AHC-labeled section of the tumor and a second digital image of a second AHC-labeled section of the tumor, where (a) first ROI module 205 annotates tumor regions in the first digital image; (b) first FI module 204 marks objects corresponding to cells in at least the tumor region and generates feature vectors for each marked cell indicative of CD8, PD-1, PD-L1 and EM status; (c) first scoring module 206 calculates (c1) median distance of CD8+ cells to EM+ cells within 30 μm, (c2) mean CD8+ / PD-1+ cell count of CD8+ / PD-L1+ cells within 10 μm, and (c3) median distance of CD8+ cells to EM+ cells within 10 μm. ) calculating the average distance from CD8+ / PD-1+ to CD8+ / PD-L1+ cells within 30 μm in the tumor region of the first digital image; (d) the second ROI module 205 annotates the stromal region on the second digital image; (e) the second FI module 204 marks objects corresponding to at least cells in the stromal region of the second digital image and generates a feature vector indicative of the CD8, PD-L1 and EM status of at least cells in the stromal region; (f) the second scoring module 206 calculates the average distance from CD8+ cells to PD-L1+ / EM+ cells within 10 μm in the stromal region of the tumor region of the second digital image; (g) the optional third scoring module 206 calculates a PRS from the extracted feature metrics.In such embodiments, a first digital image is obtained from a first mAHC-labeled section of the tumor labeled with a first detectable moiety via a biomarker specific reagent for CD8, a second detectable moiety via a biomarker specific reagent for PD-1, a third detectable moiety via a biomarker specific reagent for PD-L1, and a fourth detectable moiety via a biomarker specific reagent for EM (such as pan-cytokeratin), where the first, second, third, and fourth detectable moieties are distinguishable from one another when labeling the same cells, and a second digital image is obtained from a second mAHC-labeled section of the tumor labeled with a fifth detectable moiety via a biomarker specific reagent for CD8, a sixth detectable moiety via a biomarker specific reagent for PD-L1, and a seventh detectable moiety (which may be the same as or different from the first, second, third, and fourth detectable moieties) via a biomarker specific reagent for EM, where the fifth, sixth, and seventh detectable moieties are distinguishable from one another when labeling the same cells.

[0145] In another embodiment, the image analysis system 200 is operable to calculate feature metrics of Feature Set 3 on digital images of separate AHC-labeled tissue samples from the same tumor. In such an embodiment, the image analysis system 200 may be programmed to perform a set of functions on a first digital image of a first AHC-labeled section of the tumor and a second digital image of a second AHC-labeled section of the tumor, where (a) the first ROI module 205 annotates tumor and stromal regions in the first digital image; (b) the first FI module 204 (b1) marks objects corresponding to cells in the digital image, (b2) generates a feature vector for at least each marked cell in the tumor region that indicates CD8 and EM status, and (b3) generates a feature vector for at least each marked cell in the stromal region that indicates CD8, PD-L1 and EM status; (c) the first scoring module 206 (c1) determines the median distance of CD8+ cells from EM+ cells within 30 μm in the tumor region of the first digital image, and (c2) determines the median distance of CD8+ cells from EM+ cells within 30 μm in the tumor region of the first digital image. (d) the second ROI module 205 annotates the tumor region on the second digital image; (e) the second FI module 204 marks objects corresponding to cells in at least the tumor region of the second digital image and generates feature vectors indicative of CD8, PD-1, PD-L1, and optionally EM status for cells in at least the tumor region; (f) the second scoring module 206 calculates (f1) the mean CD8+ / PD-1+ cell count of CD8+ / PD-L1+ cells within 10 μm in the tumor region of the second digital image, and (f2) the distance of CD8+ / PD-1+ cells to CD8+ / PD-L1+ cells within 30 μm in the tumor region of the second digital image; (g) the optional third scoring module 206 calculates a PRS from the extracted feature metrics.In such embodiments, a first mAHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8, a second detectable moiety via a biomarker-specific reagent for EM, and a third detectable moiety via a biomarker-specific reagent for PD-L1, where the first, second, and third detectable moieties are distinguishable from one another when labeling the same cells, and a second mAHC-labeled section of the tumor may be labeled with a fourth detectable moiety via a biomarker-specific reagent for CD8, a fifth detectable moiety via a biomarker-specific reagent for PD-L1, a sixth detectable moiety via a biomarker-specific reagent for PD-1, and optionally a seventh detectable moiety via a biomarker-specific reagent for the EM marker, where the third, fourth, fifth, and sixth detectable moieties (which may be the same as or different from either the first and second detectable moieties) are distinguishable from one another when labeling the same cells. In another embodiment, the image analysis system 200 is operable to calculate feature metrics of Feature Set 3 on digital images of separate AHC-labeled tissue samples from the same tumor. In such an embodiment, the image analysis system 200 may be programmed to perform a set of functions on a first digital image of a first AHC labeled section of the tumor, a second digital image of a second AHC labeled section of the tumor, and a third digital image of a third AHC labeled section of the tumor, where (a) the first ROI module 205 annotates tumor regions in the first digital image; (b) the first FI module 204 marks objects corresponding to cells in at least the tumor region of the first digital image and generates a feature vector for at least each marked cell in the tumor region indicative of CD8 and EM status; (c) the first scoring module 206 calculates a median distance of CD8+ cells from EM+ cells within 30 μm in the tumor region of the first digital image; (d) the second ROI module 205 annotates stromal regions on the second digital image; (e) the second FI module 204 marks objects corresponding to at least cells in the stromal region of the second digital image and generates a feature vector for at least each marked cell in the tumor region indicative of CD8 and EM status. (f) the second scoring module 206 calculates the median distance from CD8+ cells to PD-L1+ / EM+ cells within 10 μm in the stromal region of the second digital image; (g) the third ROI module 205 annotates tumor regions on the third digital image; (h) the third FI module 204 marks objects corresponding to cells in at least the tumor region of the third digital image; generate feature vectors indicative of CD8, PD-1, PD-L1, and optionally EM status for cells in at least the tumor region; (i) a third scoring module 206 calculates (i1) the mean CD8+ / PD-1+ cell count of CD8+ / PD-L1+ cells within 10 μm and (i2) the median distance from CD8+ / PD-1+ cells to CD8+ / PD-L1+ cells within 30 μm; (j) an optional fourth scoring module 206 calculates a PRS from the extracted feature metrics.In such an embodiment, a first mAHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8 and a second detectable moiety via a biomarker-specific reagent for EM, the first and second detectable moieties being distinguishable from one another when labeling the same cells, and a second mAHC-labeled section of the tumor may be labeled with a third detectable moiety via a biomarker-specific reagent for CD8, a fourth detectable moiety via a biomarker-specific reagent for PD-L1, and a fifth detectable moiety via a biomarker-specific reagent for EM, the third, fourth, and fifth detectable moieties (any of the first and second detectable moieties) being distinguishable from one another when labeling the same cells. a third mAHC labeled section of the tumor may be labeled with a sixth detectable moiety via a biomarker specific reagent for CD8, a seventh detectable moiety via a biomarker specific reagent for PD-1, an eighth detectable moiety via a biomarker specific reagent for PD-1, and optionally a ninth detectable moiety via a biomarker specific reagent for an EM marker, and the sixth, seventh, eighth, and ninth detectable moieties (which may be the same as or different from any of the first, second, third, fourth, and fifth detectable moieties) are distinguishable from each other when labeling the same cells;

[0146] IV.D.4. Feature Set 4

[0147] In an embodiment, the image analysis system 200 functions to calculate feature metrics of Feature Set 4 from a digital image of a single AHC-labeled tissue sample. In such an embodiment, (a) the ROI module 205 annotates tumor regions on the digital image; (b) the FI module 204 (b1) marks objects corresponding to cells in the digital image and generates a feature vector indicative of the CD8 and optionally EM status of cells in at least the tumor region; (c) the scoring module 206 extracts feature metrics of Feature Set 4 from their respective ROIs and optionally calculates a PRS from the extracted feature metrics. In such an embodiment, the AHC-labeled sections of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for CD8 and, optionally, a second detectable moiety via a biomarker-specific reagent for EM (such as pan-cytokeratin), the first and second detectable moieties being distinguishable from each other when labeling the same cells.

[0148] IV.D.5. Feature Set 5

[0149] In an embodiment, the image analysis system 200 functions to calculate feature metrics of Feature Set 5 from a digital image of a single AHC-labeled tissue sample. In such an embodiment, (a) the ROI module 205 annotates tumor regions on the digital image; (b) the FI module 204 (b1) marks objects corresponding to cells in the digital image and generates a feature vector indicative of the PD-1 and PD-L1 status of the cells at least in the tumor region; (c) the scoring module 206 extracts feature metrics of Feature Set 5 from their respective ROIs and, optionally, calculates a PRS from the extracted feature metrics. In such an embodiment, a first AHC-labeled section of the tumor may be labeled with a first detectable moiety via a biomarker-specific reagent for PD-1 and a second detectable moiety via a biomarker-specific reagent for PD-L1, the first and second detectable moieties being distinguishable from each other when labeling the same cells. In some embodiments, the detectable moieties are bright-field dyes or fluorescent dyes.

[0150] IV.E. Sample Analysis System

[0151] In some embodiments, the image analysis system 200 may be implemented in combination with one or more additional systems to form a sample analysis system. An exemplary sample analysis system is shown in FIG.

[0152] For example, image analysis system 200 may operate in combination with image acquisition system 400. Image acquisition system 400 generates digital images of AHC-stained samples and provides those images to image analysis system 200 for analysis and presentation to a user. Image acquisition system 400 may include a scanning platform, such as a slide scanner, capable of scanning slides containing AHC-labeled samples at 20x, 40x, or other magnifications to generate high-resolution whole-slide digital images. At a basic level, a typical slide scanner includes at least the following: (1) a microscope with a lens objective, (2) a light source (such as halogen, light-emitting diode, white light, and / or multispectral light source, depending on the dye), (3) robotics to move the glass slide (or move an optical element around the slide), (4) one or more digital cameras for image capture, and (5) a computer and associated software to control the robotics and to manipulate, manage, and display the digital slides. Digital data for many different XY locations on the slide (possibly in multiple Z planes) is captured by a camera's charge-coupled device (CCD), and the images are combined together to form a merged image of the entire scanned surface. Common methods for achieving this include: (1) Tile-based scanning, in which the slide stage or optics are moved in very small increments to capture square image frames that slightly overlap adjacent squares. The captured squares are then automatically matched together to create a composite image; and (2) line-based scanning, in which the slide stage moves in a single axis during acquisition to capture multiple merged image “strips.” The image strips can then be aligned with each other to form a larger merged image. A detailed overview of the various scanners (both fluorescent and bright field) can be found in Farahani et al., Whole slide imaging in pathology: advantages, limitations, and emerging perspectives, Pathology and Laboratory Medicine Int'l, Vol. 7, p. 23-33 (June 2015), the entire contents of which are incorporated by reference. Examples of commercially available slide scanners include: 3DHistech PANNORAMIC SCAN II; DigiPath PATHSCOPE; Hamamatsu NANOZOOMER RS, HT, and XR; Huron TISSUESCOPE 4000, 4000XT, and HS; Leica SCANSCOPE AT, AT2, CS, FL, and SCN400; Mikroscan D2; Olympus VS120-SL; Omnyx VL4 and VL120; PerkinElmer LAMINA; Philips ULTRA-FAST SCANNER; Sakura Finetek VISIONTEK; Unic PRECICE 500, and PRECICE 600x; VENTANA ISCAN COREO and ISCAN HT; and Zeiss AXIO SCAN.Z1. Other exemplary systems and features can be found, for example, in International Publication No. 2011-049608, or U.S. patent application Ser. No. 61 / 533,114, filed Sep. 9, 2011, entitled IMAGING SYSTEMS, CASSETTES, AND METHODS OF USING THE SAME, the contents of which are incorporated by reference in their entireties.

[0153] In some embodiments, images generated by image acquisition system 400 may be communicatively coupled to image analysis system 200, e.g., the images may be transferred directly 401 over one or more local and / or wide area networks or via shared memory. In some embodiments, image acquisition system 400 may not be communicatively coupled to image analysis system 200, in which case the images may be stored in storage medium 410 (e.g., any type of non-volatile storage medium (e.g., a flash drive) or a non-volatile storage medium on a server or database accessible by the image analysis system). In such a case, image analysis system 200 may download images 403.

[0154] The sample analysis system may also include one or more sample labeling platforms 420, such as an automated AHC platform and / or an automated H&E staining platform. Stained samples produced by the sample labeling platforms are transferred for imaging 421 to the image acquisition system 400. The resulting digital images are then transferred 401-403 to the image analysis system 200 for evaluation.

[0155] An automated AHC platform typically includes at least the following: reservoirs for the various reagents used in the labeling protocol, a reagent dispensing unit in fluid communication with the reservoir(s) for dispensing the reagents onto the slides, a waste removal system for removing used reagents and other waste from the slides, and a control system for coordinating the operation of the reagent dispensing unit and the waste removal system. In addition to performing the labeling step, many automated slide stainers can also perform steps incidental to labeling (or are compatible with another system that performs such incidental steps), such as slide baking (to adhere the sample to the slide), degreasing (also known as deparaffinization), antigen retrieval, counterstaining, dehydration and clearing, coverslipping, etc. Prichard, Overview of Automated Immunohistochemistry, Arch Pathol Lab Med., Vol. 138, pp. 1578-1582 (2014), which is incorporated herein in its entirety, describes several specific examples of automated IHC / ISH slide stainers and their various features, including intelliPATH (Biocare Medical), WAVE (Celerus Diagnostics), DAKO OMNIS and DAKO AUTOSTAINER LINK 48 (Agilent Technologies), BENCHMARK (Ventana Medical Systems, Inc.), Leica BOND, and Lab Vision Autostainer (Thermo Scientific) automated slide stainers. Ventana Medical Systems, Inc. is the assignee of several U.S. patents disclosing systems and methods for performing automated analyses, including U.S. Patent No. 5,650,327, U.S. Patent No. 5,654,200, U.S. Patent No. 6,296,809, U.S. Patent No. 6,352,861, U.S. Patent No. 6,827,901, and U.S. Patent No. 6,943,029, as well as U.S. Patent Application Publication No. 20030211630 and U.S. Patent Application Publication No. 20040052685, each of which is incorporated by reference in its entirety.Commercially available staining units typically operate on one of the following principles: (1) open individual slide staining, where the slide is positioned horizontally and reagents are dispensed as paddles onto the surface of the slide containing the tissue sample (implemented e.g. in DAKO AUTOSTAINER Link 48 (Agilent Technologies) and intelliPATH (Biocare Medical) stainers); (2) liquid overlay technique, where reagents are covered by or dispensed through an inert fluid layer deposited on the sample (implemented e.g. in BENCHMARK and DISCOVERY stainers); (3) capillary gap staining, where the slide surface is placed close to another surface (e.g. another slide or cover plate) creating a narrow gap through which capillary forces are drawn in and liquid reagents remain in contact with the sample (e.g. the staining principle used in DAKO TECHMATE, Leica BOND, and DAKO OMNIS stainers). Even if capillary gap staining is repeated several times, the fluids in the gap do not mix (e.g. in DAKO TECHMATE and Leica BOND). In a variation of capillary gap staining called dynamic gap staining, the sample is applied to the slide using capillary forces and then parallel surfaces are translated into each other to agitate and mix the reagents during incubation (such as the staining principle implemented in the DAKO OMNIS slide stainer (Agilent)). In translation gap staining, a translatable head is placed on the slide. The lower surface of the head is separated from the slide by a first gap that is small enough for a meniscus of liquid to form from the liquid on the slide during translation of the slide. A mixing extension, having a lateral dimension smaller than the width of the slide, extends from the lower surface of the translatable head to define a second gap smaller than the first gap between the mixing extension and the slide. During translation of the head, the lateral dimension of the mixing extension is sufficient to create a lateral movement in the liquid on the slide generally in a direction extending from the second gap to the first gap.See WO 2011 / 139978 A1. It has recently been proposed to use inkjet technology to deposit reagents onto slides. See WO 2016-170008 A1. This list of staining techniques is not intended to be comprehensive and any fully or semi-automated system for performing biomarker staining may be useful.

[0156] H&E staining platforms typically operate on one of two staining principles: batch staining (also known as "dip'n dunk") or individual slide staining. Batch stainers generally use a reagent vat or bath into which many slides are immersed simultaneously. Individual slide stainers, on the other hand, apply reagent directly to each slide, with no two slides sharing the same aliquot of reagent. Examples of commercially available H&E stainers include the VENTANA SYMPHONY (individual slide stainer) and VENTANA HE 600 (individual slide stainer) series H&E stainers from Roche; the DAKO COVERSTAINER (batch stainer) from Agilent Technologies; and the LEICA ST4020 SMALL LINEAR STAINER (batch stainer), LEICA ST5020 MULTISTAINER (batch stainer), and LEICA ST5010 AUTOSTAINER XL series (batch stainer) H&E stainers from Leica Biosystems Nussloch GmbH. The H&E staining platform is typically used in workflows where morphologically labeled serial sections of the biomarker labeled section(s) are desired.

[0157] The sample analysis system may further include a laboratory information system (LIS) 420. The LIS 430 typically performs one or more functions selected from: recording and tracking operations performed on samples and on images obtained from the samples; instructing different components of the sample analysis system to perform specific operations on samples, slides and / or images and to track information about specific reagents applied to the samples and / or slides (lot numbers, expiration dates, dispensed amounts, etc.). The LIS 430 typically comprises at least a database containing information about the samples; labels associated with the samples, slides and / or image files (e.g., barcodes (including one-dimensional and two-dimensional barcodes), radio frequency identification (RFID) tags, alphanumeric codes attached to the samples, etc.); and a communication device for reading the labels on the samples or slides and / or communicating information about the slides between the LIS 430 and other components of the sample analysis system. Thus, for example, communication devices can be located in each of the sample processing station (not shown), the sample labeling platform(s) 420, and the image acquisition system 400. When the sample is first processed into sections, information about the sample (such as patient ID, sample type, processing performed on the section(s)) is entered into a communication device and a label is created for each section made from the sample. At each subsequent station, this label is entered into the communication device (e.g., by scanning a barcode or RFID tag or by manually entering an alphanumeric code) and the station communicates electronically with the LIS 430, for example to instruct the station or station operator to perform a particular processing on the section and / or record the processing performed on the section 431. The image acquisition system 400 may also encode each image with a computer readable label or code correlating to the section or sample from which the image is derived, such that when an image is transmitted to the image analysis system 200, the image processing steps performed may be transmitted from the LIS to the image analysis system and / or the image processing steps performed on the image by the image analysis system are recorded by the database of the LIS 432.Additionally, the LIS 430 may function as an output device for the image analysis system 200, with the extracted feature metrics and / or the calculated PRS being transmitted to and stored in the LIS 432. Commercially available LIS systems useful in the present methods and systems include, for example, the VENTANA VANTAGE WORKFLOW system (Roche).

[0158] V. Working Examples Example 1: Multiplex fluorescent immunohistochemical assay

[0159] Three multiplex fluorescent IHC (mfIHC) panels were designed as described in Table 9. [Table 9]

[0160] The following antibody clones were used: CD3 (SP162), CD8 (SP239), CD68 (SP251), PD-L1 (SP263), pan-cytokeratin (panCK) (AE1 / AE3 / PCK26 cocktail), PD1 (NAT105), LAG3 (E17B4), MHC-I (EP1395Y), β2-microglobulin (B2M) (ERP21752-214), CD14 (EPR3653), and TGF-β receptor 2 (TGFBR2) (MBS2400063).

[0161] The mfIHC panel was applied to 4 μM tissue sections of formalin-fixed paraffin-embedded (FFPE) tissue blocks from a cohort of dMMR / MSI-H mCRC patients treated with pembrolizumab monotherapy. A detailed description of epitope retrieval from FFPE tissue sections, antibody titration, incubation and image acquisition was previously described (Zhang II). Briefly, for each target, the corresponding 1° antibody (1° Ab) was incubated on the slides, followed by horseradish peroxidase (HRP)-conjugated 2° Ab goat anti-mouse HRP (#760-7060) for PD1, LAG3 and PanCK, and goat anti-rabbit HRP (#760-7058) for CD8, CD68, CD14, MHCI, TGFBR2, B2M, CD3 and PDL1. Targets were then detected with tyramide-conjugated fluorophores (TSA-FL): DISCOVERY Red 610 Kit (#760-245, Roche) (a fluorophore with an excitation wavelength of 580 nm and an emission wavelength of 625 nm ("R610")), DISCOVERY Rhodamine 6G kit (No. 760-244, Roche) (fluorophore with excitation wavelength 546 nm, emission wavelength 572 nm ("R6G")), DISCOVERY FAM kit (No. 760-243, Roche) (fluorophore with excitation wavelength 490 nm, emission wavelength 520 nm ("FAM")), DISCOVERY Cy5 kit (No. 760-238, Roche) (fluorophore with excitation wavelength 650 nm, emission wavelength 670 nm ("Cy5")), DISCOVERY DCC kit (No. 760-240, Roche) (fluorophore with excitation wavelength 436 nm, emission wavelength 480 nm ("DCC")). The next target detection followed the same scheme. To prevent potential cross-reaction of 1 antibodies of the same species, a heating step was introduced to inactivate 1Ab and 2Ab complexes before detecting the next target. Slides were then counterstained with 4',6-diamidino-2-phenylindole (DAPI) (catalog number 760-4196, Roche).The slides were coverslipped using microcover glass, 24 x 50 mm No. 1.5 (VWR, Catalog Number: 48393241) and PROLONG DIAMOND Antifade Mountant with DAPI (ThermoFisher Scientific, Catalog Number: P36962). Once the slides are stained, they are coverslipped. The coverslips are mounted with Diamond Long Mounting Medium and reach their optimal optical properties after 24 hours. The tissue is then outlined using a Sharpie on the slides by hand under a dissecting scope and cleaned to remove dust and debris.

[0162] In addition, serial sections of each sample labeled with the mfIHC assay were stained with hematoxylin and eosin (H&E) on a VENTANA HE 600 slide stainer.

[0163] Example 2: Image scanning and image analysis

[0164] Fluorescence image acquisition was performed on a ZEISS AXIO SCAN.Z1 slide scanner (Oberkochen, Germany). Slides were placed in the slide scanner and imaged based on the panel of markers. Filters were custom narrowband filters designed for our six markers per channel. Exposure times for each marker were maintained throughout the experiment to ensure intensity and population compared to the DAB under validation. Images were assessed for quality and rescanned (if necessary) until a clear, clean image was obtained.

[0165] Image analysis was performed with the HALO image analysis platform (Indica Labs). ROIs were annotated by a pathologist using stained H&E slides from the same case. Within the HALO software, trained classifiers were applied to distinguish epithelial tumor from tumor stroma, remove artifacts, detect individual nuclei, and extend the cytoplasmic radius range from the detected nuclear border. Each cell was defined as positive or negative for each marker based on intensity thresholding. Analysis settings were run on the annotated regions and quality checked by a second certified user.

[0166] From the HALO platform, a csv file containing the raw data for each cell was downloaded. A Python script was developed and validated to read the raw data and input features including cell count, density, intensity, ratio / fraction, percentage, and spatial relationship specific to each panel. The script was automated to report a batch of readouts from the batch of raw data output for regions of interest from each case.

[0167] Example 3: Model Generation

[0168] The features generated using the panel-specific readout scripts were used together to develop a model to stratify subjects into low- and high-risk groups. The model development procedure can be divided into three steps: feature preprocessing, feature selection, model fitting, and case stratification.

[0169] Example 3A: Pretreatment

[0170] Feature preprocessing was performed to remove data artifacts that may adversely affect the quality of feature selection and model fitting results in subsequent steps.

[0171] First, features with missing read data for >50% of cases were removed to reduce model bias where the available data may not accurately describe the underlying feature distribution. Imputation was performed for features with 50% or less of the data missing across cases.

[0172] Second, features providing redundant information were removed based on an assessment of pairwise correlation. Features that calculated the mean and median cell intensity for the same phenotype were evaluated, and if the correlation was greater than 99%, the mean feature was removed from consideration. This step removed features depicting identical information in order to reduce the risk of overfitting in downstream feature selection and model fitting.

[0173] Third, features whose values ​​depend on tissue size were removed or reformatted to normalize by tissue size. In the feature set under consideration, this includes all features that are cell number counts. Since density features are a normalized representation of the number of cells, features consisting of several cells were excluded from consideration in downstream feature selection and model fitting.

[0174] Example 3B: Feature Selection

[0175] Feature selection was performed using a Cox regression model with LASSO regularization. The selection of regularization weights was determined by varying the regularization parameters and performing cross-validation with each weight. The weights that minimized the partial likelihood deviance were selected as candidates. Due to the small size of the training dataset, defined as the number of observations (cases) per feature, and the randomness of the data split during cross-validation, several iterations (e.g., 1000) of cross-validation were performed for a given regularization weight, and the average candidate weight across iterations was selected as the final weight. Taking the final weights into account, the final regularized Cox regression was trained. The features with nonzero coefficients from the final model were then selected as features.

[0176] Separate models were trained using both progression-free survival and overall survival as response variables. Furthermore, training was performed with variables from each panel and all panels considered for feature selection.

[0177] Example 3C: Model fitting

[0178] The set of features selected during feature selection was used to fit a Cox proportional hazards model. The coefficients of the fitted hazards model were used to calculate a risk score for each subject. The median risk score was used as a cutoff to divide the case population into low-risk and high-risk groups, with predicted responders in the low-risk group and predicted non-responders in the high-risk group. Four separate models were developed.

[0179] Example 3D: Model 1

[0180] Model 1 was fitted using one feature metric from mfIHC panel A and one feature from mfIHC panel B in Table 9, using overall survival as the relevant outcome. Details of Model 1 are summarized below in Table 10. Risk score distributions are shown in Figure 5A (vertical black lines indicate stratification cutoffs). Kaplan-Meier curves and associated patient stratification charts are shown in Figure 5B. [Table 10]

[0181] Example 3E: Model 2

[0182] Model 2 was fitted using one feature metric from mfIHC panel A and one feature from mfIHC panel B in Table 9, using overall survival as the relevant outcome. Details of Model 2 are summarized below in Table 11. Risk score distributions are shown in Figure 6A (vertical black lines indicate stratification cutoffs). Kaplan-Meier curves and associated patient stratification charts are shown in Figure 6B. [Table 11]

[0183] Example 3F: Model 3

[0184] Model 3 was fitted using two feature metrics from mfIHC panel A and two feature metrics from mfIHC panel B in Table 9, with progression-free survival as the relevant outcome. Details of Model 3 are summarized below in Table 12. Risk score distributions are shown in Figure 7A (vertical black lines indicate stratification cutoffs). Kaplan-Meier curves and associated patient stratification charts are shown in Figure 7B. [Table 12]

[0185] Example 3G: Model 4

[0186] Model 4 was fitted using the feature metrics from Panel A of Table 9, with progression-free survival as the relevant outcome. Details of Model 4 are summarized below in Table 13. Risk score distributions are shown in Figure 8A (vertical black lines indicate stratification cutoffs). Kaplan-Meier curves and associated patient stratification charts are shown in Figure 8B.

[0187] [Table 13]

[0188] Example 3H: Model 5

[0189] Model 5 was fitted using the mean number of PD-1+ cells within a predefined distance of at least one PD-L1+ cell from the samples stained in Panel B of Table 9, with progression-free survival as the relevant outcome. Several predefined distances were tested, including 5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 μm. A radius of 10 μm was most significant. Details of Model 5 are summarized in Table 14 below. Kaplan-Meier curves of feature values ​​for the 5.01 chart and associated patient stratification are shown in Figure 9. [Table 14]

[0190] Example 4A: Validation Study (Prophetic Example)

[0191] To validate the predictive features discovered in Example 3, patient cases from an independent clinical series are studied. FFPE tumor tissue specimens from patients with MSI-H / dMMR mCRC treated with pembrolizumab are stained with the multiplexed fluorescent marker panel used in Example 3. Feature metrics for each of Models 1-5 are then extracted from digitized slide images, and a risk score is calculated for each patient case by combining these features.

[0192] Multiple cutoffs are evaluated for their ability to balance the sensitivity and specificity of the model. The cutoff values ​​are used to assign each patient case to either a low-risk or high-risk group. Survival outcomes in these subgroups are compared to see if the model is indeed predictive for survival endpoints.

[0193] Example 4B: Additional Validation Studies (Prophetic Examples)

[0194] To validate the predictive features of Model 5, patient cases from an independent clinical series are studied. FFPE tumor tissue specimens from patients with MSI-H / dMMR mCRC treated with PD-1 axis-directed therapy are stained with a chromogenic dual IHC assay for PD-1 and PD-L1. Model 5 feature metrics are then extracted from the digitized slide images, and the average number of PD-1+ cells within a predefined distance of at least one PD-L1+ cell is calculated for each patient case. Multiple predefined distances are tested for each patient sample.

[0195] Multiple cutoffs are evaluated for their ability to balance the sensitivity and specificity of the model. The cutoff values ​​are used to assign each patient case to either a low-risk or high-risk group. Survival outcomes in these subgroups are compared to see if the model is indeed predictive for survival endpoints.

[0196] References

[0197] Aeffner et al., Introduction to Digital Image Analysis in Whole-slide Imaging: A White Paper from the Digital Pathology Association, Journal of Pathology Informatics, 2019, Vol.10, No.9, doi:10.4103 / jpi.jpi_82_18

[0198] Baharlou et al., Mass Cytometry Imaging for the Study of Human Diseases - Applications and Data Analysis Strategies, Frontiers in Immunology, 2019, Vol.10, Art.2657.

[0199] Barrera et al., Computer - extracted features relating to spatial arrangement of tumor infiltrating lymphocytes to predict response to nivolumab in non - small cell lung cancer (NSCLC). ASCO Annual Meeting 2018: Abstract#:12115.

[0200] Bodenmiller, Multiplexed Epitope - Based Tissue Imaging for Discovery and Healthcare Applications, Cell Systems, 2016, Vol.2, Issue 4, pp.225 - 38.

[0201] Chakrabarti et al.,Intratumoral CD3+and CD8+T-Cell Densities in Patients With DNA Mismatch Repair-Deficient Metastatic Colorectal Cancer Receiving Programmed Cell Death-1 Blockade,JCO Precision Oncology,2019,Vol.3,1-7.

[0202] Galon et al.,Validation of the Immunoscore(IM)as a prognostic marker in stage I / II / III colon cancer:Results of a worldwide consortium-based analysis of 1,336 patients.,J.Clin.Oncol.,Vol.34,suppl.Abstract No.3500(2016)(“Galon I”).

[0203] Galon et al.,Validation of the Immunoscore(IM)as a prognostic marker in stage I / II / III colon cancer:Results of a worldwide consortium-based analysis of 1,336 patients,2016 ASCO Annual Meeting,Oral Abstract Session,Abstract#3500,(available at meetinglibrary.asco.org)(“Galon II”).

[0204] Gorris et al.,Eight-Color Multiplex Immunohistochemistry for Simultaneous Detection of Multiple Immune Checkpoint Molecules within the Tumor Microenvironment,2018,Journal of Immunology,Vol.200,Issue 1,pp.347-54.

[0205] Hofman et al.,Multiplexed Immunohistochemistry for Molecular and Immune Profiling in Lung Cancer-Just About Ready for Prime-Time?,Cancers,2019,Vol.11,No.283.

[0206] Ide et al.,Chromogenic Multiplex Immunohistochemistry Reveals Modulation of the Immune Microenvironment Associated with Survival in Elderly Patients with Lung Adenocarcinoma,Cancers(Basel),2018,Vol.10,Issue 9,No.326.

[0207] Jass,Lymphocytic infiltration and survival in rectal cancer,J.Clin.Pathol.,1986,Vol.39,Issue 6,pp.585-589(Jass I).

[0208] Jass et al.,A new prognostic classification of rectal cancer,The Lancet,1987,Vol.329,Issue 8545,pp.1303-1306(Jass II).

[0209] Le et al.,PD-1 Blockade in Tumors with Mismatch-Repair Deficiency,New England Journal of Medicine,2015,Vol.372,Issue 26,pp.2509-20(Le I).

[0210] Le et al.,Mismatch-repair deficiency predicts response of solid tumors to PD-1 blockade,Science,10.1126 / science.aan6733(2017)(Le II).

[0211] Levenson,Immunohistochemistry and mass spectrometry for highly multiplexed cellular molecular imaging,Laboratory Investigation,2015,Vol.95,pp.397-405.

[0212] Llosa et al.,The vigorous immune microenvironment of microsatellite instable colon cancer is balanced by multiple counter-inhibitory checkpoints.Cancer discovery,2015,Vol.5,Issue 1,pp.43-51.

[0213] Morrison et al.,Brightfield multiplex immunohistochemistry with multispectral imaging,Laboratory Investigation,2020,Vol.100,pp.1124-36.

[0214] Pages et al.,Immune infiltration in human tumors:a prognostic factor that should not be ignored,Oncogene,Vol.29,pp.1093-1102(2010).

[0215] Parra et al.,State-of-the-Art of Profiling Immune Contexture in the Era of Multiplexed Staining and Digital Analysis to Study Paraffin Tumor Tissues,Cancers,2019,Vol.11,Issue 2,No.247.

[0216] Ptacek et al.,Multiplexed ion beam imaging(MIBI)for characterization of the tumor microenvironment across tumor types,Laboratory Investigation,2020,Vol.100,pp.1111-1123.

[0217] Stack et al.,Multiplexed immunohistochemistry,imaging,and quantitation:A review,with an assessment of Tyramide signal amplification,multispectral imaging and multiplex analysis,2014,Vol.70,Issue 1,pp.46-58.

[0218] Wang et al.,Case classification with tumor antigen presenting and TGF-β signaling biomarkers to predict anti-PD-1 outcome in GI tract tumors using automated quantitative fluorescence multiplex IHC[abstract],Cancer Research,2019,Vol.79,Issue 13(Suppl),Abstract 4030(Wang I).

[0219] Wang et al.,Exploration of PD-1 / PD-L1 Spatial Interaction and T Cells Functionality to Predict Anti-PD1 Treatment Outcome in GI Tract Tumors using Automated Quantitative Fluorescence Multiplexed IHC,Society for Immunotherapy of Cancer Annual Meeting 2018,Poster#P703(Wang II).

[0220] Wang,et al.,Prediction of recurrence in early stage non-small cell lung cancer using computer extracted nuclear features from digital H&E images.,Scientific Reports 7.1(2017):13543

[0221] WO 2020 / 072348 A1,Methods and systems for predicting response to pd-1 axis directed therapeutics,published 09-APR-2020.

[0222] WO 2020 / 161125 A1,Methods and systems for evaluation of immune cell infiltrate in stage iv colorectal cancer,published 13-AUG-2020.

[0223] Yoon et al.,Inter-tumoral heterogeneity of CD3(+)and CD8(+)T-cell densities in the microenvironment of DNA mismatch repair-deficient colon cancers:implications for prognosis.Clin Cancer Res.,2019,Vol.25,Issue 1,pp.125-133.

[0224] Zhang et al.,Characterization of PD-L1,CD8,CD3,CD68 and PanCK in Tumor Microenvironment of Gl Tract Tumors with respect to Patients’ Mismatch Repair Status and Anti-PD-1 Treatment Outcome using 5Plex IHC and Whole Slide Image Analysis,Annals of Oncology,2018,Vol.29,Suppl.8),VIII36-VIII37(Zhang I).

[0225] Zhang et al.,Fully automated 5-plex fluorescent immunohistochemistry with tyramide signal amplification and same species antibodies,Laboratory Investigation,2017,Vol.97,Issue 7,pp.873-85(Zhang II).

[0226] Zou et al.,PD-L1(B7-H1)and PD-1 pathway blockade for cancer therapy:Mechanisms,response biomarkers,and combinations.Science Translational Medicine,2016,Vol.8,Issue 328,328rv324.

Claims

1. An ex vivo method for selecting subjects having stage III or stage IV dMMR and / or MSI-H colorectal tumors to receive PD-1 axis-directed therapy, - In a portion of the aforementioned tumor, PD-L1 ＋ PD-1 within a predetermined distance from the cell ＋ Determining the average number of cells, -PD-L1 ＋ PD-1 within the predetermined distance of the cell ＋ Comparing the average number of cells with a predetermined cutoff, -PD-L1 ＋ PD-1 within a predetermined distance from the cell ＋ If the average number of cells exceeds the predetermined cutoff, the subject is selected to receive the PD-1 axis-directed therapeutic agent. Methods that include...

2. A pharmaceutical agent for treating a subject having a stage III or stage IV dMMR and / or MSI-H colorectal tumor, wherein the pharmaceutical agent comprises PD-1 axis-directed therapy, wherein the tumor has been previously determined to have an average number of PD-L1+ cells within a predetermined distance of PD-L1+ cells above a predetermined cutoff.

3. The method according to claim 1, wherein the predetermined distance is within a range selected from the group consisting of 5 μm to 50 μm, 5 μm to 40 μm, 5 μm to 30 μm, 5 μm to 25 μm, 5 μm to 20 μm, 5 μm to 15 μm, 5 μm to 10 μm, 10 μm to 50 μm, 10 μm to 40 μm, 10 μm to 30 μm, 10 μm to 25 μm, 10 μm to 20 μm, and 10 μm to 15 μm.

4. The drug according to claim 1, wherein the predetermined distance is 10 μm.

5. PD-L1 ＋ PD-1 within the predetermined distance of the cell ＋ The method according to claim 1, wherein the average number of cells is determined in a tissue section of the tumor, and the tumor is affinity histochemically stained for human PD-1 and human PD-L1, respectively.

6. The method according to claim 5, wherein human PD-1 is affinity histochemically stained with a first bright-field dye, and human PD-L1 is affinity histochemically stained with a second bright-field dye.

7. The method according to claim 5, wherein the tissue section is a formalin-fixed, paraffin-embedded tissue section.

8. the aforementioned PD-L1 ＋ PD-1 within the predetermined distance of the cell ＋ The method according to claim 5, wherein the average number of cells is extracted from a digital image of a tissue section histochemically stained with the affinity by an image analysis system.

9. The method according to claim 5, wherein the tumor is affinity histochemically stained with an anti-human PD-1 monoclonal antibody and an anti-human PD-L1 monoclonal antibody.

10. The method according to claim 1, wherein the colorectal tumor is stage III.

11. The method according to claim 1, wherein the colorectal tumor is stage IV.

12. The method according to claim 1, wherein the PD-1 axis-targeted therapy comprises an anti-PD1 antibody.

13. The method according to claim 12, wherein the anti-PD1 antibody is selected from the group consisting of nivolumab, pembrolizumab, semiprimab, tislerizumab, spartalizumab, MEDI0680, tripalimab, cintilimab, cetrimab, and pizilizumab.

14. The method according to claim 13, wherein the anti-PD1 antibody is pembrolizumab.

15. The method according to claim 1, wherein the PD-1 axis-targeted therapy comprises an anti-PD-L1 antibody.

16. The method according to claim 15, wherein the anti-PD-L1 antibody is selected from the group consisting of atezolizumab, durvalumab, and avelumab.

17. An image analysis system comprising memory and a processor, wherein the processor implements an instruction set stored in the memory, and the instruction set includes extracting a feature set from one or more digital images of affinity histochemistry (AHC) labeled tissue samples from colorectal tumors previously determined to be one or more dMMR or MSI-H, The aforementioned feature set is PD-L1 ＋ PD-1 within 10 μm of a cell ＋ Including the average number of cells, Image analysis system.

18. The image analysis system according to claim 17, wherein the feature set is extracted by implementing a region of interest (ROI) module, a feature identification (FI) module, and a scoring module on the digital image(s), the ROI module annotates one or more ROIs on the digital image(s), the FI module marks cells in at least the relevant ROIs and generates a feature vector for each cell having the state of the relevant biomarker, and the scoring module extracts relevant feature metrics from the relevant ROIs.

19. The image analysis system extracts the feature metric of the feature set from the digital image of the AHC-labeled tissue sample. (d1) The ROI module annotates the tumor region on the digital image, (d2) The FI module marks objects corresponding to cells in the digital image and generates feature vectors indicating PD-1 and PD-L1 states for at least the cells in the tumor region. (d3) The image analysis system according to claim 18, wherein the scoring module extracts feature metrics of the feature set from each of their ROIs.

20. The image analysis system according to claim 17, wherein the AHC-labeled section of the tumor is labeled with a first detectable portion via a PD-1 biomarker-specific reagent and a second detectable portion via a PD-L1 biomarker-specific reagent, and the first and second detectable portions are distinguishable from each other.

21. The image analysis system according to any one of claims 17 to 20, wherein the scoring module further calculates a predictive response score (PRS) by applying the extracted feature metrics to a continuous scoring function that correlates the feature metrics with the likelihood of a response to PD-1 axis-directed therapy.

22. The image analysis system according to claim 21, wherein the PD-1 axis-directed therapy includes an anti-PD1 antibody.

23. The image analysis system according to claim 22, wherein the anti-PD1 antibody is selected from the group consisting of nivolumab, pembrolizumab, semiprimab, tislerizumab, spartalizumab, MEDI0680, tripalimab, cintilimab, cetrimab, and pizilizumab.

24. The image analysis system according to claim 23, wherein the anti-PD1 antibody is pembrolizumab.

25. The image analysis system according to claim 21, wherein the PD-1 axis-directed therapy includes an anti-PD-L1 antibody.

26. The image analysis system according to claim 25, wherein the anti-PD-L1 antibody is selected from the group consisting of atezolizumab, durvalumab, and avelumab.

27. The image analysis system according to claim 21, wherein the continuous scoring function is a Cox proportional hazards model.

28. The image analysis system according to claim 27, wherein the Cox proportional hazards model is a regularized Cox regression using LASSO.

29. A sample analysis system, - An image acquisition system optionally connected to a sample labeling platform. - A storage medium on which one or more images of AHC-labeled tissue samples from the colorectal tumor are stored, and • Laboratory Information System (LIS) A sample analysis system comprising the image analysis system according to claim 17, which is communicably connected to one or more systems selected from the group consisting of the following.

30. A multiaffinity histochemistry (mAHC) assay-stained tissue section of a colorectal tumor previously determined to be one or more dMMR or MSI-H, wherein the tissue section is differentially stained for a panel of biomarkers comprising biomarker-specific reagents specific to PD-1 and PD-L1 proteins.

31. The mAHC-stained tissue section according to claim 30, wherein the mAHC assay is a multiple immunohistochemical assay (mIHC).

Images