Quantifying and assessing biomarker topology based on patient response across multiple tissue types

Abstract

A method and computer system for classification of CD8 T cell topology using a linear cutoff model based on patient response is described herein. A computer system receives a plurality of histology images of tumor samples from a plurality of patients. Image analysis of the plurality of histology images is performed to obtain the abundance of CD8+ T cells in tumor parenchyma and stroma in each of the plurality of histology images. Real inflammation score and tumor infiltration score are determined based on polar coordinate transformation of the abundance of CD8+ T cells in tumor parenchyma and stroma. A feature space is generated based on the real inflammation score and tumor infiltration score, and a linear boundary or linear cutoff between a plurality of classifications in the feature space is identified based on the real inflammation score, tumor infiltration score, and patient response data.

Description

[Technical field]

[0001] CROSS-REFERENCE TO PREVIOUSLY FILED APPLICATIONS This PCT application claims the benefit of priority to U.S. Provisional Application No. 63 / 153,723, filed February 25, 2021, which is incorporated by reference in its entirety.

[0002] Embodiments of the present disclosure relate to methods and devices for quantification, classification, and evaluation of tissue-based biomarker topologies using linear cutoff and patient response based models. [Background technology]

[0003] Biomarkers can be used to identify and evaluate biological processes in the body. Biomarkers are increasingly being used to evaluate the likelihood of a particular patient's outcome to various types of treatment so that the appropriate treatment (e.g., medical, pharmaceutical, etc.) can be provided to a given patient. Some biomarkers are generated, for example, as an immune system response to the presence of cancerous cells or tumors, fibrosis, gastrointestinal disorders, cardiac diseases, etc.

[0004] One example of such a biomarker is CD8.CD8 is a transmembrane glycoprotein that can be expressed in cytotoxic T lymphocytes.Measure the number of CD8+ tumor infiltrating lymphocytes (TIL) can be a reliable marker for evaluating the immune response to cancer and determining whether a given patient is or will be responsive to various cancer immunotherapies. Summary of the Invention [Problem to be solved by the invention]

[0005] In the embodiments presented herein, methods and systems are described for utilizing inflammation and tumor infiltration scores for classification of biomarker topologies and patient responses in numerous clinical and commercial samples for various cancers. [Means for solving the problem]

[0006] In one embodiment, a computer-implemented method for classifying CD8 tumor topology is described. The method includes: receiving a plurality of histology images of tumor samples from a plurality of patients; performing image analysis of the plurality of histology images to obtain the abundance of CD8+ T cells in tumor parenchyma and stroma in each of the plurality of histology images; determining, by at least one processor, a real inflammation score and a tumor infiltration score based on a polar transformation of the abundance of CD8+ T cells in tumor parenchyma and stroma; generating a feature space based on the real inflammation score and the tumor infiltration score; and identifying a linear boundary or linear cutoff between a plurality of classifications in the feature space based on the real inflammation score, the tumor infiltration score, and patient response data.

[0007] Another embodiment includes a system for classification of CD8 tumor topology. The system may include a memory and a processor coupled to the memory. In one embodiment, the processor is configured to receive a plurality of histology images of tumor samples in a plurality of patients, perform image analysis of the plurality of histology images to obtain the abundance of CD8+ T cells in tumor parenchyma and stroma in each of the plurality of histology images, determine a real inflammation score and a tumor infiltration score based on a polar transformation of the abundance of CD8+ T cells in tumor parenchyma and stroma, generate a feature space based on the real inflammation score and the tumor infiltration score, identify a linear boundary or linear cutoff between a plurality of classifications in the feature space based on the real inflammation score, the tumor infiltration score, and patient response data, and store data related to the feature space and the linear boundary or linear cutoff in the memory.

[0008] A further embodiment includes a non-transitory computer readable medium having instructions stored thereon, the execution of which by one or more processors of a device causes the one or more processors to perform operations. In one embodiment, the operations include receiving a plurality of histology images of tumor samples in a plurality of patients, performing image analysis of the plurality of histology images to obtain an abundance of CD8+ T cells in the tumor parenchyma and stroma in each of the plurality of histology images, determining, by the at least one processor, an actual inflammation score and a tumor infiltration score based on a polar transformation of the abundance of CD8+ T cells in the tumor parenchyma and stroma, generating a feature space based on the actual inflammation score and the tumor infiltration score, and identifying a linear boundary or linear cutoff between a plurality of classifications in the feature space based on the actual inflammation score, the tumor infiltration score, and patient response data.

[0009] A further embodiment includes and describes a computer-implemented method for classification of biomarker topologies, comprising receiving a plurality of histology images of tumor samples in a plurality of patients, performing image analysis of the plurality of histology images to obtain information regarding biomarkers in each of the plurality of histology images, determining a real inflammation score and a tumor infiltration score based on a polar transformation of the abundance of CD8+ T cells in the tumor parenchyma and stroma, generating a feature space based on the real inflammation score and the tumor infiltration score, and identifying a linear boundary or linear cutoff between a plurality of classifications in the feature space based on the real inflammation score, the tumor infiltration score, and patient response data.

[0010] Further embodiments include a method including receiving, by at least one processor of a computing device, information including stromal CD8+ T cell parameters and parenchymal CD8+ T cell parameters for each of a plurality of classified histology images; obtaining, by the at least one processor, an abundance of CD8+ T cells in the tumor parenchyma and stroma for each of the plurality of histology images; determining, by the at least one processor, a real inflammation score and a tumor infiltration score based on a polar transform of the abundance of CD8+ T cells in the tumor parenchyma and stroma; generating a feature space based on the real inflammation score and the tumor infiltration score; and identifying, by the at least one processor, a linear boundary or linear cutoff between the plurality of classifications in the feature space based on the real inflammation score, the tumor infiltration score, and patient response data.

[0011] Further features and advantages of various embodiments, as well as the structure and operation, are described in detail below with reference to the accompanying drawings. It should be noted that the specific embodiments described herein are not intended to be limiting. Such embodiments are presented herein for illustrative purposes only. Other embodiments will be apparent to one of ordinary skill in the relevant art based on the teachings contained herein.

[0012] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments of the present disclosure and, together with the description, further serve to explain the principles of the present disclosure and to enable a person skilled in the relevant art to make and use the present disclosure. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 illustrates exemplary images of tumor tissue samples with various classifications using CD8+ histological images obtained by immunostaining, according to an exemplary embodiment. [Diagram 2]FIG. 2 is an exemplary diagram illustrating a method for image analysis using a patient response model for tumor topology classification, according to an exemplary embodiment. [Diagram 3] FIG. 3 is another exemplary diagram illustrating a method for tumor topology classification using a linear cutoff approach based on image analysis and patient response, according to an exemplary embodiment. [Figure 4] FIG. 4 is a flow chart illustrating a process for CD8 tumor topology classification, according to an exemplary embodiment. [Diagram 5] FIG. 5 is a flowchart illustrating a process for classifying CD8 tumor topology of histology images using a feature space obtained by applying a patient response model and a linear cutoff approach, according to an exemplary embodiment. [Figure 6] FIG. 6 is a block diagram of example components of a device in accordance with an example embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0015] The following detailed description describes exemplary embodiments consistent with the present disclosure with reference to the accompanying drawings. This detailed description sufficiently reveals the general nature of the present disclosure so that others can easily modify and / or adapt such exemplary embodiments to various applications by applying the knowledge of those skilled in the relevant art without undue experimentation and without departing from the spirit and scope of the present disclosure. Thus, such adaptations and modifications are intended to be within the meaning and multiple equivalents of the exemplary embodiments based on the teaching and guidance presented herein. It should be understood that the terms or terminology herein are for purposes of description and not of limitation, and thus the terms or terminology herein should be interpreted by those skilled in the relevant art in light of the teachings herein. definition

[0016] In order that this disclosure may be more readily understood, some specific terms are first defined. As used in this application, unless expressly stated otherwise herein, each of the following terms shall have the meaning set forth below. Additional definitions are set forth throughout this application.

[0017] Whenever an embodiment is described herein using the word "comprising," it is understood that otherwise similar embodiments described using the terms "consisting of" and / or "consisting essentially of" are also provided.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this disclosure pertains.For example, Concise Dictionary of Biomedicine and Molecular Biology, Juo, Pei-Show, 2nd ed., 2002, CRC Press, The Dictionary of Cell and Molecular Biology, 3rd ed., 1999, Academic Press, and the Oxford Dictionary Of Biochemistry And Molecular Biology, Revised, 2000, Oxford University Press provide those skilled in the art with a general dictionary for many of the terms used in this disclosure.

[0019] Units, prefixes, and symbols are expressed in the format accepted by the International System of Units (SI). Numeric ranges are inclusive of the numerical values ​​that define the range. When a range of values ​​is listed, each intervening integer value between the listed upper and lower limits of the range, and each fraction thereof, is to be understood as specifically disclosed, along with each subrange between such values. The upper and lower limits of any range can be independently included or excluded from the range, and each range that includes either, neither, or both limits is also encompassed within the scope of the disclosure. Thus, ranges listed herein are understood to be shorthand for all values ​​within the range, including the listed endpoints. For example, a range of 1 to 10 is understood to include any number, combination of numbers, or subrange from the group consisting of 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10.

[0020] When a value is explicitly recited, it should be understood that values ​​that are approximately the same quantity or amount as the recited value are also within the scope of this disclosure. When a combination is disclosed, each subcombination of the elements of the combination is also specifically disclosed and is within the scope of this disclosure. Conversely, when different elements or groups of elements are individually disclosed, their combinations are also disclosed. When any element of the disclosure is disclosed as having multiple options, examples of the disclosure in which each option is excluded, alone or in any combination with other options, are also disclosed hereby. Multiple elements of the disclosure can have such exclusions, and all combinations of elements with such exclusions are disclosed hereby.

[0021] "Cancer" refers to a broad group of different diseases characterized by the uncontrolled growth of abnormal cells in the body. Uncontrolled cell division and proliferation leads to the formation of malignant tumors that can invade nearby tissues and also metastasize to distant sites in the body via the lymphatic system or bloodstream.

[0022] The term "immunotherapy" refers to the treatment of a subject suffering from or at risk of suffering from a disease or recurrence of a disease by methods that involve inducing, enhancing, suppressing, or otherwise modifying the immune response. "Treatment" or "therapy" of a subject refers to any type of intervention or process performed on a subject, or the administration of an active agent to a subject, with the purpose of reversing, alleviating, ameliorating, inhibiting, delaying, or preventing the onset, progression, onset, severity of symptoms, or recurrence of a symptom, complication, or condition, or biochemical manifestation associated with a disease.

[0023] A "subject" includes any human or non-human animal. The term "non-human animal" includes, but is not limited to, vertebrates, such as non-human primates, sheep, dogs, and rodents, such as mice, rats, and guinea pigs. In a preferred embodiment, the subject is a human. The terms "subject" and "patient" are used interchangeably herein.

[0024] The term "biological sample" as used herein refers to biological material isolated from a subject. A biological sample may contain any biological material suitable for determining target gene expression, for example, by sequencing nucleic acid in a tumor (or circulating tumor cells) and identifying genomic alterations in the sequenced nucleic acid. A biological sample may be any suitable biological tissue or fluid, for example, tumor tissue, blood, plasma, and serum. In one embodiment, the sample is a tumor sample. In some embodiments, a tumor sample can be obtained from a tumor tissue biopsy sample, such as a formalin-fixed paraffin-embedded (FFPE) tumor tissue or a fresh frozen tumor tissue. In another embodiment, a biological sample is a liquid biopsy sample, which in some embodiments includes one or more of blood, serum, plasma, circulating tumor cells, exoRNA, ctDNA, and cfDNA.

[0025] As used herein, a "tumor sample" refers to a biological sample that includes tumor tissue. In some embodiments, the tumor sample is a tumor biopsy sample. In some embodiments, the tumor sample includes tumor cells and one or more non-tumor cells present in the tumor microenvironment (TME). For the purposes of this disclosure, the TME is composed of at least two regions. The tumor "parenchyma" is the region of the TME that contains primarily tumor cells, e.g., the portion of the TME that contains the majority of tumor cells. The tumor parenchyma does not necessarily consist of only tumor cells, but rather other cells, e.g., stromal cells and / or lymphocytes, may also be present in the parenchyma. The "stromal" region of the TME includes adjacent non-tumor cells. In some embodiments, the tumor sample includes all or a portion of the tumor parenchyma and one or more cells of the stroma. In some embodiments, the tumor sample is obtained from the parenchyma. In some embodiments, the tumor sample is obtained from the stroma. In other embodiments, the tumor sample is obtained from the parenchyma and the stroma.

[0026] In some aspects, the TME can be classified as immune desert, immune excluded, immune inflamed, or immune balanced. The term "immune desert" indicates that very few or no T cells are present in the TME. In some embodiments, the immune desert classification can be referred to herein as "desert" or "cold". The term "immune excluded" indicates that T cells are accumulated in the tumor stroma without efficient infiltration of the tumor parenchyma. In some embodiments, the immune excluded classification can be referred to herein as "stromal". The term "immune inflammatory" indicates that T cells are infiltrating the tumor parenchyma. In some embodiments, the immune inflammatory classification can be referred to herein as "parenchymal". The term "immune balanced" indicates an intermediate classification level between excluded and inflammatory, in which the number of T cells accumulated in the tumor stroma and the number of T cells accumulated in the tumor parenchyma may be comparable.

[0027] The use of the alternative (e.g., "or") should be understood to mean either one, both, or any combination thereof of the alternatives. As used herein, the indefinite article "a" or "an" should be understood to refer to "one or more" of any listed or enumerated components.

[0028] The term "about" or "essentially includes" refers to a value or composition that is within an acceptable error range of a particular value or composition as determined by one of ordinary skill in the art, which will depend in part on how the value or composition is measured or determined, i.e., the limitations of the measurement system. For example, "about" or "essentially includes" can also mean within 1 or more than 1 standard deviation according to the practice in the art. Alternatively, "about" or "essentially includes" can mean a range of up to 10%. Furthermore, particularly with respect to biological systems or processes, the term can mean up to an order of magnitude or up to 5 times the value. When a particular value or composition is provided in this application and claims, unless otherwise stated, the meaning of "about" or "essentially includes" should be assumed to be within an acceptable error range of that particular value or composition.

[0029] As described herein, any concentration range, percentage range, ratio range, or integer range, unless otherwise indicated, should be understood to include any integer value within the recited range, and, where appropriate, fractions thereof (such as tenths and hundredths of integers).

[0030] Various aspects of the disclosure are described in further detail in the following subsections. An exemplary study of patient response-based classification of CD8 topology

[0031] Here, a method is described for patient response-based classification of CD8 topology in multiple tumor biopsy and resection samples using linear cutoffs and image analysis of CD8 slides and histology images obtained by immunostaining. Identification of CD8+ T cell abundance and CD8 topology may be particularly useful for stratifying patient outcomes in solid tumors based on spatial CD8+ cell patterns. Understanding the role of CD8 topology in various clinical settings may enable more personalized treatment options for patients. In some embodiments, such studies may be difficult to conduct in a reproducible manner, as manual interpretation of these complex patterns is subject to considerable inter-reviewer variability. However, optimization of linear cutoffs using retrospective clinical data and patient response-based approaches may be useful for quantifying CD8 topology in a biologically meaningful, reproducible, and scalable manner. In particular, linear cutoff optimization for feature space and patient response-based methods can be utilized to evaluate CD8 topology in numerous clinical and commercially available CD8 histology slides for various cancers.

[0032] Although the examples herein are described in relation to CD8, the methods described herein can also be used to identify and classify other antigens and biomarker topologies for the purpose of detecting additional tumors or cancerous cells.Biomarkers include, but are not limited to, PD-L1, PD-1, LAG3, CLTA-4, TIGIT, TIM3, NKG2a, CSF1R, OX40, ICOS, MICA, MICB, CD137, KIR, TGFβ, IL-10, IL-8, B7-H4, Fas ligand, CXCR4, mesothelin, CD27, GITR, and any combination thereof.Markers can also include those that are morphologically identified without the use of staining antibodies, such as lymphocytes, fibroblasts, macrophages, neutrophils, eosinophils, or any combination thereof. Similarly, although the examples described herein are described in the context of tumors, the patient response based methods described herein can also be applied to other tissue types in a variety of therapeutic applications, e.g., fibrosis, cardiac, gastrointestinal, and other oncological and non-oncological therapeutic areas. Exemplary embodiments of cutoff optimization and assessment of tumor topology based on patient response

[0033] Tumor microenvironment (TME) inflammation, characterized by infiltration of CD8+ T cells, is associated with improved clinical outcomes across multiple tumor types. Parenchymal infiltration of CD8+ T cells is associated with improved survival with immuno-oncology (IO) treatments, and intratumoral localization also impacts outcomes, highlighting the importance of spatial analysis of CD8+ T cells within the TME. Intratumoral CD8+ T cell patterns, assessed by immunostaining of histological images, are diverse and can be classified as: (i) immune desert (negligible T cell infiltration), (ii) immune exclusion (T cells are confined to the tumor stroma or infiltrative margin), or (iii) immune inflammation (T cells infiltrate the tumor parenchyma and are located in close proximity to tumor cells). Artificial intelligence (AI)-based image analysis can be used to characterize tumor parenchymal and stromal compartments in the TME.

[0034] FIG. 1 illustrates exemplary images of tumor tissue samples with various classifications using CD8+ histology images obtained by immunostaining, according to an exemplary embodiment. The tumor images show various classifications of CD8+ T cell patterns within the TME. The top row of images in FIG. 1 shows immune desert and immune exclusion classifications, and the bottom row of images in FIG. 1 shows immune inflammation classification.

[0035] The immune desert classification indicates that T cells are present in the TME in very few or no cases. In some embodiments, the immune desert classification may be referred to herein as "desert" or "cold". The immune exclusion classification indicates that T cells are accumulated in the tumor stroma without efficient infiltration of the tumor parenchyma. In some embodiments, the immune exclusion classification may be referred to herein as "stromal". The immune inflammation classification indicates that T cells are infiltrating the tumor parenchyma. In some embodiments, the immune inflammation classification may be referred to herein as "parenchymal".

[0036] In some embodiments, there may be different levels (e.g., first and second elimination levels, first, second and third inflammatory levels, etc.) within the classification of immune exclusion and immune inflammation, depending on the progression of T cells moving in the TME. In some embodiments, the third inflammatory level may indicate that the number of T cells infiltrating the parenchyma is greater than the number of T cells infiltrating the parenchyma at the first inflammatory level. Although not shown in FIG. 1, there may be an intermediate classification between the exclusion type and the inflammatory type, referred to herein as "balanced type". The term "balanced type" indicates an intermediate classification level between the exclusion type and the inflammatory type, in which the number of T cells accumulated in the tumor stroma and the number of T cells accumulated in the tumor parenchyma may be comparable.

[0037] In some embodiments, the tumor sample in the histological image obtained by immunostaining can be obtained by tissue biopsy and / or by resection of tumor tissue. In some embodiments, the tumor sample is a tumor tissue biopsy sample. In some embodiments, the tumor sample is a formalin-fixed paraffin-embedded tumor tissue or a fresh frozen tumor tissue. In some embodiments, the tumor sample is obtained from the stroma of the tumor. In some embodiments, the histological image obtained by immunostaining can be referred to as a histological image herein.

[0038] In some embodiments, CD8 topology method is not standardized, which may result in inter-reviewer variability between different pathologists reviewing histology images.The interpretation of CD8 topology from histology images may be confounded by various factors, such as different tumor types, limited tumor structure due to biopsy or sampling, and heterogeneity of inflammation within tumor samples.

[0039] To address these problems in the art, the embodiments described herein present a solution that provides a standardized, scalable approach using image analysis and patient response-based techniques to facilitate the review and assessment of CD8 topology of tumor tissue in patients.

[0040] FIG. 2 is an exemplary diagram illustrating a method of an image analysis and patient response based approach for tumor topology classification, according to an exemplary embodiment. In particular, FIG. 2 shows three different stages of the method, including image analysis, polar coordinate transformation, and application of a patient response based model. The images used for image analysis can include histology images obtained by immunostaining, which show CD8+ T cell patterns within the TME for multiple patients. These images can be labeled by trained topologists to be classified into various categories. In some embodiments, the classification categories are "desert type", "exclusion type", and "stromal type". In some embodiments, the classification categories include "balanced type". In some embodiments, the classification category "desert type" can be referred to herein as "cold" and the classification category "stromal type" can be referred to herein as "inflammatory type".

[0041] In the first stage, the histology images are processed to extract information from each histology image. In some embodiments, image parameters are known, and the image analysis process selects a subset of parameters for further analysis. Such parameters may include, for example, the number of stromal CD8+ T cells, the number of parenchymal CD8+ T cells, and the number of total CD8+ T cells in each image. Other parameters may include the density of stromal CD8+ T cells and the density of parenchymal CD8+ T cells in each image, which may be particularly useful when the total number of total CD8+ T cells is unknown or cannot be determined.

[0042] In some embodiments, image analysis can provide the abundance of CD8+ T cells in the tumor parenchyma and stroma in each histology image. In some embodiments, the abundance of CD8+ T cells can be displayed via a graphical display of the relationship between the percentage of stromal CD8+ T cells and the percentage of parenchymal CD8+ T cells relative to the total number of T cells present in each of the multiple histology images, as shown by the "Image Analysis Readout" plot in FIG. 2. In some embodiments, the graphical display can show the density, percentage, and / or quantity of stromal and parenchymal CD8+ T cells in each image. In some embodiments, image analysis can include any image recognition, processing, and / or analysis algorithm. In some embodiments, image analysis can be performed by applying an artificial neural network (e.g., a convolutional neural network) to the multiple histology images.

[0043] In a second stage, a polar transformation can be performed on the results from the image analysis to convert the image analysis readout graph into a polar plot having polar coordinates. In some embodiments, the polar transformation can include a mathematical transformation of features derived during image analysis into a polar feature space.

[0044] In the third stage, a patient response model can be applied to the transformation results of the image analysis and the abundance of CD8+ T cells in the tumor parenchyma and stroma. In some embodiments, the patient response model can include calculations for determining the value of actual inflammation (r) and the value of tumor infiltration (t) based on the polar coordinate transformation, and for identifying the linear boundary or linear cutoff between multiple classifications in the feature space based on the actual inflammation score, the tumor infiltration score, and the patient response data.

[0045] In some embodiments, the real inflammation score (r) and the tumor infiltration score (t) are defined by a polar coordinate transformation of the x-axis showing the percentage of stromal CD8+ T cells and the y-axis showing the percentage of parenchymal CD8+ T cells. In some embodiments, the percentage of stromal CD8+ T cells represents the number of CD8+ T cells in the stroma divided by the total number of T cells in the stroma, and the percentage of parenchymal CD8+ T cells represents the number of CD8+ T cells in the parenchyma divided by the total number of T cells in the parenchyma. In some embodiments, the multiple classifications for classifying the CD8 topology can include inflammatory, desert / cold, or exclusionary. In some embodiments, a feature space can be generated by plotting the values ​​of the real inflammation score and the tumor infiltration score. In some embodiments, the multiple classifications in the feature space can be determined by identifying a linear boundary or linear cutoff across the derived real inflammation and tumor infiltration scores. In some embodiments, the classification in the feature space can be used to classify the histology images as cold, exclusionary, or inflammatory cases. In some embodiments, if the value of the actual inflammation score is less than a predetermined value on the X-axis of the plot of the feature space, the histology image can be classified as a cold case. In some embodiments, if the value of the actual inflammation score is equal to or greater than a predetermined value on the X-axis of the plot of the feature space and the value of the tumor infiltration score is less than a predetermined value on the Y-axis of the plot of the feature space, the histology image can be classified as an exclusion type case. In some embodiments, if the value of the actual inflammation score is equal to or greater than a predetermined value on the X-axis of the plot of the feature space and the value of the additional tumor infiltration score is greater than a predetermined value on the Y-axis of the plot of the feature space, the histology image can be classified as an inflammation type case.

[0046] In addition to the actual inflammation and tumor infiltration scores, patient response data can also be used to identify linear boundaries or linear cutoffs between the multiple classifications in the feature space. In some embodiments, the patient response data can include clinical response data, such as retrospective clinical response data that indicates whether each patient in the multiple patients is or was responding to a particular treatment or therapy. In some embodiments, the boundary definition values ​​for the actual inflammation and tumor infiltration can be iteratively adapted in the feature space as guided by the patient response data until the linear boundaries or linear cutoffs between the multiple classifications are optimized.

[0047] In some embodiments, immunotherapy or treatment recommendations for a patient's tumor can be generated based on determining a classification for at least one histology image of the patient's tumor using the feature space obtained from the patient response model.

[0048] Figure 3 is another exemplary diagram illustrating a method for tumor topology classification using an image analysis and patient response based approach, according to an exemplary embodiment. In some embodiments, Figure 3 illustrates further details for the method embodiment shown in Figure 2. Figure 3 illustrates four stages for classifying new images using a patient response model for tumor topology classification, where the stages include image analysis, feature extraction, patient response model, and prediction.

[0049] First, as shown in FIG. 3-(1), image analysis can be performed to identify CD8 positive cells and the division of parenchymal and stromal compartments in a histology image of a tumor. In some embodiments, image analysis can include applying a neural network (e.g., a convolutional neural network) to a plurality of histology images to evaluate CD8+ T cells in different parts of the tumor (e.g., tumor epithelium, stroma, and parenchyma) in each image. The image analysis tool can result in the identification of values ​​for a plurality of different parameters for each of the images in the plurality of histology images. In some embodiments, two parameters (e.g., the number of stromal CD8+ T cells and the number of parenchymal CD8+ T cells) can be selected for further analysis. In some embodiments, the abundance of CD8+ T cells in the tumor parenchyma and stroma for a plurality of histology images can be obtained from the image analysis.

[0050] Feature extraction can then be performed by applying a mathematical transformation of the features derived from image analysis to transform the data into a polar feature space, as shown in Figure 3-(2). In some embodiments, feature extraction can be part of the image analysis process to identify relationships between stromal and parenchymal CD8+ T cells.

[0051] After mathematical transformation, a patient response model can be applied to determine linear cutoffs of feature maps for defining CD8 topology based on real inflammation score, tumor infiltration score, and retrospective clinical response data, as shown in FIG. 3-(3). In some embodiments, applying the patient response model can include generating a feature space including multiple classifications (e.g., inflammatory, desert / cold, or exclusionary). In some embodiments, determining linear cutoffs or linear boundaries for multiple classifications can enable adaptive identification of relevant patient populations for stratification. In some embodiments, the methods described herein can enable optimization for stratifying treatment response.

[0052] After generating the feature space, the patient response model can classify the CD8 topology in new histology images as inflammatory, desert, or exclusionary, as shown in Figure 3-(4). Such classification for a given patient's image can then be used to diagnose the patient's condition, determine the patient's immune response, and / or recommend or exclude treatment options for the patient.

[0053] 4 is a flow chart illustrating a process for classification of CD8 tumor topology, according to an exemplary embodiment. Method 400 may be performed by processing logic, which may include hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, etc.), software (e.g., instructions executed on a processing device), or a combination thereof. It should be understood that not all operations are required to accomplish the disclosure provided herein. Furthermore, as will be understood by those skilled in the art, some of the operations may be performed simultaneously or in a different order than shown in FIG. 4.

[0054] In operation 402, a plurality of histology images of tumor samples in a plurality of patients may be received by at least one processor of a computing device. In some embodiments, the histology images may include tumor tissue samples showing CD8+ T cell patterns within the TME for a plurality of patients obtained using CD8+ immunostaining techniques.

[0055] In operation 404, image analysis of the plurality of histology images may be performed to obtain an abundance of CD8+ T cells in the tumor parenchyma and stroma in each of the plurality of histology images. In some embodiments, performing image analysis of the plurality of histology images may include applying an artificial neural network (e.g., a convolutional neural network) to the plurality of histology images. In some embodiments, performing image analysis of the plurality of histology images may include applying an artificial neural network (e.g., a convolutional neural network) to the plurality of histology images. In some embodiments, the abundance of CD8+ T cells in the tumor parenchyma and stroma may be displayed via a graphical representation of a relationship between a percentage of stromal CD8+ T cells and a percentage of parenchymal CD8+ T cells relative to a total number of T cells present in each of the plurality of histology images.

[0056] In operation 406, the actual inflammation score and the tumor infiltration score can be determined based on a polar transformation of the abundance of CD8+ T cells in the tumor parenchyma and stroma. In some embodiments, a polar transformation can be applied to the graphical representation of the relationship between stromal and parenchymal CD8+ T cells, and the resulting polar plot can be used to determine the actual inflammation score and the tumor infiltration score.

[0057] In operation 408, a feature space can be generated based on the actual inflammation score and the tumor infiltration score.

[0058] In operation 410, a linear boundary or linear cutoff between the multiple classifications in the feature space can be identified based on the actual inflammation score, the tumor infiltration score, and the patient response data. In some embodiments, the multiple classifications include inflammatory, desert, or exclusionary. In some embodiments, the data regarding the feature space and the linear boundary or linear cutoff between the multiple classifications in the feature space can be stored in a memory of a computing device or computer system. In some embodiments, the patient response data includes clinical response data, such as retrospective clinical response data indicating whether each patient in the multiple patients is or was responding to a treatment or therapy.

[0059] 5 is a flow chart illustrating a process for classifying CD8 tumor topology of histology images using a feature space obtained by applying a patient response model and a linear cutoff approach according to an exemplary embodiment. Method 500 may be performed by processing logic that may include hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, etc.), software (e.g., instructions executed on a processing device), or a combination thereof. It should be understood that not all operations are required to accomplish the disclosure provided herein. Furthermore, as will be understood by those skilled in the art, some of the operations may be performed simultaneously or in a different order than shown in FIG. 5.

[0060] In operation 502, a new histology image of a tumor sample from a patient may be received by at least one processor of a computing device. In some embodiments, the new histology image may include a tumor tissue sample obtained using CD8+ immunostaining techniques showing a CD8+ T cell pattern within the TME.

[0061] In operation 504, image analysis of the new histology image can be performed to obtain CD8+ T cell abundance in the tumor parenchyma and stroma in the new histology image, which can be performed, for example, by the same image analysis algorithms as in operation 404 of FIG.

[0062] In operation 506, a real inflammation score and a tumor infiltration score can be determined based on a polar transform of CD8+ T cell abundance in the tumor parenchyma and stroma in the new histology image.

[0063] In operation 508, a classification of the new histology image can be determined by mapping values ​​of real inflammation score and tumor infiltration score with the multiple classifications in the feature space. In some embodiments, the feature space and linear boundaries or linear cutoffs between the multiple classifications in the feature space can be generated by method 400 of FIG. 4. In some embodiments, the patient response model can determine where the pattern of stromal and parenchymal CD8+ T cells in the new histology image falls within the boundaries of the multiple classifications in the feature space. Based on this mapping, the patient response model can output a classification of the new histology image.

[0064] 6 is a block diagram of exemplary components of a computer system 600. One or more computer systems 600 can be used to implement, for example, any of the embodiments discussed herein, as well as combinations and subcombinations thereof. In some embodiments, one or more computer systems 600 can be used to implement methods 400 and 500 shown in FIGS. 4 and 5, respectively. Computer system 600 can include one or more processors (also referred to as central processing devices or CPUs), such as processor 604. Processor 604 can be connected to a communication infrastructure or bus 606.

[0065] The computer system 600 may also include a user input / output interface 602 , such as a monitor, keyboard, pointing device, etc., which may communicate with a communications infrastructure 606 via a user input / output interface 603 .

[0066] One or more of the processors 604 may be graphics processing units (GPUs). In one embodiment, a GPU may be a processor that is a specialized electronic circuit designed to handle mathematically intensive applications. A GPU may have a parallel structure that is efficient for parallel processing of large blocks of data, e.g., mathematically intensive data common in computer graphics applications, images, videos, etc.

[0067] The computer system 600 may also include a main or primary memory 608, such as a random access memory (RAM). The main memory 608 may include one or more levels of cache. The main memory 608 may store therein control logic (i.e., computer software) and / or data.

[0068] The computer system 600 may also include one or more secondary storage devices or memory 610. The secondary memory 610 may include, for example, a hard disk drive 612 and / or a removable storage drive 614.

[0069] The removable storage drive 614 can interact with a removable storage unit 618. The removable storage unit 618 can include a computer usable or readable storage device that stores computer software (control logic) and / or data. The removable storage unit 618 can be a program cartridge and cartridge interface (e.g., as found in a video game device), a removable memory chip (e.g., EPROM or PROM) and associated socket, a memory stick and USB port, a memory card and associated memory card slot, and / or any other removable storage unit and associated interface. The removable storage drive 614 can read from and / or write to the removable storage unit 618.

[0070] The secondary memory 610 may include other means, devices, components, means, or other approaches for allowing computer programs and / or other instructions and / or data to be accessed by the computer system 600. Such means, devices, components, means, or other approaches may include, for example, a removable storage unit 622 and an interface 620. Examples of removable storage units 622 and interfaces 620 may include a program cartridge and a cartridge interface (e.g., as found in a video game device), a removable memory chip (e.g., an EPROM or PROM) and associated sockets, a memory stick and a USB port, a memory card and associated memory card slot, and / or any other removable storage unit and associated interface.

[0071] Computer system 600 may further include a communications interface or network interface 624. Communications interface 624 may enable computer system 600 to communicate and interact with any combination of external devices, external networks, external entities, etc. (individually or collectively referred to by reference number 628). For example, communications interface 624 may enable computer system 600 to communicate with external or remote devices 628 via communications path 626, which may be wired and / or wireless (or a combination thereof) and may include any combination of a LAN, a WAN, the Internet, etc. Control logic and / or data may be communicated to and from computer system 600 via communications path 626.

[0072] Computer system 600 may also be any or any combination of a personal digital assistant (PDA), a desktop workstation, a laptop or notebook computer, a netbook, a tablet, a smartphone, a smart watch or other wearable, an appliance, part of the Internet of Things, and / or an embedded system, to name a few non-limiting examples.

[0073] The computer system 600 may be a client or server that accesses or hosts any application and / or data via any delivery paradigm, including, but not limited to, remote or distributed cloud computing solutions, local or on-premise software ("on-premise" cloud-based solutions), an "as a service" model (e.g., Content as a Service (CaaS), Digital Content as a Service (DCaaS), Software as a Service (SaaS), Management Software as a Service (MSaaS), Platform as a Service (PaaS), Desktop as a Service (DaaS), Framework as a Service (FaaS), Backend as a Service (BaaS), Mobile Backend as a Service (MBaaS), Infrastructure as a Service (IaaS), etc.), and / or a hybrid model including any combination of the foregoing examples or other service or delivery paradigms.

[0074] Any applicable data structures, file formats, and schemas in computer system 600 may be derived from standards including, but not limited to, JavaScript Object Notation (JSON), Extensible Markup Language (XML), Yet Another Markup Language (YAML), Extensible Hypertext Markup Language (XHTML), Wireless Markup Language (WML), MessagePack, XML User Interface Language (XUL), or any other functionally similar representations, alone or in combination. Alternatively, proprietary data structures, formats, or schemas may be used exclusively or in combination with known or open standards.

[0075] In some embodiments, a tangible, non-transitory device or article of manufacture that includes a tangible, non-transitory computer usable or readable medium having control logic (software) stored thereon may also be referred to herein as a computer program product or program storage device, including, but not limited to, a tangible article of manufacture that embodies computer system 600, main memory 608, secondary memory 610, and removable storage units 618 and 622, and any combination thereof. Such control logic, when executed by one or more data processing devices (e.g., computer system 600), can cause such data processing devices to operate as described herein.

[0076] References to "one exemplary embodiment," "an exemplary embodiment," "an exemple exemplary embodiment," and the like in the detailed description indicate that the exemplary embodiment being described may include a particular feature, structure, or characteristic, but not all exemplary embodiments necessarily include the particular feature, structure, or characteristic. Moreover, such phrases do not necessarily refer to the same exemplary embodiment. Moreover, when a particular feature, structure, or characteristic is described in relation to an exemplary embodiment, it is within the knowledge of a person skilled in the art to affect such feature, structure, or characteristic in relation to other exemplary embodiments, whether or not explicitly described.

[0077] The exemplary embodiments described herein are provided for illustrative purposes and are not limiting. Other exemplary embodiments are possible, and modifications can be made to the exemplary embodiments within the spirit and scope of the present disclosure. Therefore, the detailed description is not intended to limit the present disclosure. Instead, the scope of the present disclosure is defined only by the claims and their equivalents.

[0078] The embodiments may be implemented in hardware (e.g., circuits), firmware, software, or any combination thereof. The embodiments may also be implemented as instructions stored on a machine-readable medium that may be read and executed by one or more processors. A machine-readable medium may include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computing device). For example, a machine-readable medium may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media, optical storage media, flash memory devices, electrical, optical, acoustic, or other forms of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.), and others. Furthermore, firmware, software, routines, instructions may be described herein as performing certain operations. However, it should be understood that such description is merely for convenience and that such operations actually result from a computing device, processor, controller, or other device executing the firmware, software, routines, instructions, etc. Furthermore, any of the variations of implementation may be performed by a general-purpose computer, as described above.

[0079] Embodiments have been described herein with the aid of functional building blocks that illustrate the implementation of certain functions and relationships thereof. The boundaries of these functional building blocks have been arbitrarily defined herein for convenience of description. Alternative boundaries may be defined so long as the specified functions and relationships (or their equivalents) are appropriately performed. Alternative embodiments may also execute functional blocks, steps, operations, methods, etc. using an order different from that described herein.

[0080] The foregoing description of specific embodiments sufficiently reveals the general nature of the present disclosure, so that others, by applying the knowledge of those skilled in the art, can easily modify and / or adapt such specific embodiments to various applications without undue experimentation and without departing from the general concept of the present disclosure. Such adaptations and modifications are therefore intended to be within the meaning and range of equivalents of the disclosed embodiments based on the teaching and guidance presented herein. It should be understood that the terms or terminology herein are for purposes of description and not of limitation, and thus should be interpreted by those skilled in the art in light of the teaching and guidance.

[0081] The breadth and scope of the present disclosure should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with the following claims and their equivalents.

Claims

1. receiving, by at least one processor of a computing device, a plurality of histology images of tumor samples in a plurality of patients; performing, by at least one processor, image analysis of the plurality of histology images to obtain an abundance of CD8+ T cells in the tumor parenchyma and stroma in each of the plurality of histology images; determining, by at least one processor, a real inflammation score and a tumor infiltration score based on a polar transform of the abundance of CD8+ T cells in the tumor parenchyma and stroma; generating, by at least one processor, a feature space based on the actual inflammation score and the tumor infiltration score; and Identifying, by the at least one processor, a linear boundary or linear cutoff between the plurality of classifications in the feature space based on the actual inflammation score, the tumor infiltration score, and the patient response data. The method includes:

2. The method of claim 1 , wherein performing image analysis of the plurality of histological images comprises applying an artificial neural network to the plurality of histological images.

3. 2. The method of claim 1, wherein the abundance of CD8+ T cells is displayed via a graphical display of the relationship between the percentage of stromal CD8+ T cells and the percentage of parenchymal CD8+ T cells relative to the total number of T cells present in each of a plurality of histology images.

4. 4. The method of claim 3, wherein the percentage of stromal CD8+ T cells comprises the number of CD8+ T cells in the stroma divided by the total number of T cells in the stroma, and the percentage of parenchymal CD8+ T cells comprises the number of CD8+ T cells in the parenchyma divided by the total number of T cells in the parenchyma.

5. applying, by at least one processor of the computing device, a polar transformation of the graphical representation to produce a polar plot; and Using polar plots to generate feature space The method of claim 3 further comprising:

6. The method of claim 1 , further comprising determining a classification for each of the plurality of histology images based on the feature space, the plurality of classifications comprising inflammatory, cold, and exclusionary.

7. 10. The method of claim 1, wherein the patient response data comprises retrospective clinical response data for a plurality of patients indicating whether each patient responded to a treatment or therapy.

8. Identifying linear boundaries or linear cutoffs between multiple classifications in the feature space Iteratively fitting the actual inflammation score and the tumor infiltration score in the feature space and using the patient response data to identify a linear boundary or linear cutoff between the multiple classifications. The method of claim 7 further comprising:

9. receiving, by at least one processor of the computing device, an additional histology image, the additional histology image corresponding to a particular patient; performing additional image analysis of the additional histology images to obtain additional CD8+ T cell abundance in the tumor parenchyma and stroma in the additional histology images; determining an additional actual inflammation score and an additional tumor infiltration score based on a polar transformation of the abundance of additional CD8+ T cells in the tumor parenchyma and stroma in the additional histology images; and Determining a classification of the additional histology image by mapping additional real inflammation score values ​​and additional tumor infiltration score values ​​using multiple classifications in the feature space. The method of claim 1 further comprising:

10. The method of claim 9 , wherein the additional histology image classification comprises inflammatory, cold, or exclusionary.

11. The method of claim 10, wherein the additional histology image is classified as cold if the value of the additional actual inflammation score is less than a predetermined value on a first axis of the plot of the feature space.

12. The method of claim 10, wherein the additional histology image is classified as rejected if the value of the additional actual inflammation score is greater than or equal to a predetermined value on a first axis of the plot of the feature space and the value of the additional tumor infiltration score is less than a predetermined value on a second axis of the plot of the feature space.

13. The method of claim 10, wherein the additional histology image is classified as inflammatory type if the value of the additional actual inflammation score is greater than or equal to a predetermined value on a first axis of the plot of the feature space and the value of the additional tumor infiltration score is greater than a predetermined value on a second axis of the plot of the feature space.

14. memory; and Processor connected to memory A system comprising: The processor: receiving a plurality of histology images of tumor samples in a plurality of patients; performing image analysis of the plurality of histology images to obtain an abundance of CD8+ T cells in the tumor parenchyma and stroma in each of the plurality of histology images; determining a real inflammation score and a tumor infiltration score based on a polar transformation of the abundance of CD8+ T cells in the tumor parenchyma and stroma; Generate a feature space based on the actual inflammation score and the tumor infiltration score; Identifying linear boundaries or linear cutoffs between multiple classifications in the feature space based on the actual inflammation score, the tumor infiltration score, and the patient response data; Store data in memory regarding the feature space and linear boundaries or linear cutoffs The system is configured as follows.

15. 15. The system of claim 14, wherein performing image analysis of the plurality of histology images comprises applying an artificial neural network to the plurality of histology images.

16. The system of claim 14, wherein the abundance of CD8+ T cells is displayed via a graphical display of the relationship between the percentage of stromal CD8+ T cells and the percentage of parenchymal CD8+ T cells relative to the total number of T cells present in each of a plurality of histology images.

17. The processor: receiving an additional histology image, the additional histology image corresponding to a particular patient; performing additional image analysis of the additional histology images to obtain additional CD8+ T cell abundance in the tumor parenchyma and stroma in the additional histology images; determining an additional actual inflammation score and an additional tumor infiltration score based on a polar transformation of the abundance of additional CD8+ T cells in the tumor parenchyma and stroma in the additional histology images; Determining a classification for the additional histology images by mapping additional real inflammation score values ​​and additional tumor infiltration score values ​​using multiple classifications in the feature space. The system of claim 14 further configured to:

18. 20. The system of claim 17, wherein the additional histology image classification comprises inflammatory, cold, or exclusionary.

19. 20. The system of claim 18, wherein the additional histology image is classified as cold if the value of the additional actual inflammation score is less than a predetermined value on a first axis of the plot of the feature space.

20. 20. The system of claim 18, wherein the additional histology image is classified as rejected if the value of the additional actual inflammation score is greater than or equal to a predetermined value on a first axis of the plot of the feature space and the value of the additional tumor infiltration score is less than a predetermined value on a second axis of the plot of the feature space.

21. 20. The system of claim 18, wherein the additional histology image is classified as inflammatory type if the value of the additional actual inflammation score is greater than or equal to a predetermined value on a first axis of the plot of the feature space and the value of the additional tumor infiltration score is greater than a predetermined value on a second axis of the plot of the feature space.

22. 15. The system of claim 14, wherein the patient response data comprises retrospective clinical response data for a plurality of patients indicating whether each patient responded to a treatment or therapy.

23. Identifying linear boundaries or linear cutoffs between multiple classifications in the feature space Iteratively fitting the actual inflammation score and the tumor infiltration score in the feature space and using the patient response data to identify a linear boundary or linear cutoff between the multiple classifications.

23. The system of claim 22, further comprising:

24. A non-transitory computer-readable medium having instructions stored thereon, the execution of the instructions by one or more processors of a device causing the one or more processors to: receiving a plurality of histology images of tumor samples in a plurality of patients; performing image analysis of the plurality of histology images to obtain abundance of CD8+ T cells in the tumor parenchyma and stroma in each of the plurality of histology images; determining, by at least one processor, a real inflammation score and a tumor infiltration score based on a polar transform of the abundance of CD8+ T cells in the tumor parenchyma and stroma; Generating a feature space based on the actual inflammation score and the tumor infiltration score; and Identifying linear boundaries or linear cutoffs between multiple classifications in feature space based on real inflammation scores, tumor infiltration scores, and patient response data. A non-transitory computer-readable medium for causing a computer to perform operations including:

25. 25. The non-transitory computer readable medium of claim 24, wherein performing image analysis of the plurality of histology images comprises applying an artificial neural network to the plurality of histology images.

26. 25. The non-transitory computer readable medium of claim 24, wherein the patient response data comprises retrospective clinical response data for a plurality of patients indicating whether each patient responded to a treatment or therapy.

27. Identifying linear boundaries or linear cutoffs between multiple classifications in the feature space Iteratively fitting the actual inflammation score and the tumor infiltration score in the feature space and using the patient response data to identify a linear boundary or linear cutoff between the multiple classifications.

27. The non-transitory computer readable medium of claim 26, further comprising:

28. 25. The non-transitory computer readable medium of claim 24, wherein the abundance of CD8+ T cells is displayed via a graphical display of the relationship between the percentage of stromal CD8+ T cells and the percentage of parenchymal CD8+ T cells relative to the total number of T cells present in each of a plurality of histology images.

29. The operation is receiving additional histology images, the additional histology images corresponding to a particular patient; performing additional image analysis of the additional histology images to obtain additional CD8+ T cell abundance in the tumor parenchyma and stroma in the additional histology images; determining an additional actual inflammation score and an additional tumor infiltration score based on a polar transformation of the abundance of additional CD8+ T cells in the tumor parenchyma and stroma in the additional histology image; and Determining a classification of the additional histology image by mapping additional real inflammation score values ​​and additional tumor infiltration score values ​​using multiple classifications in the feature space.

25. The non-transitory computer readable medium of claim 24, further comprising:

30. 30. The non-transitory computer readable medium of claim 29, wherein the classification of the additional histology images comprises inflammatory, cold, or exclusionary.

31. 31. The non-transitory computer readable medium of claim 30, wherein if the value of the additional actual inflammation score is less than a predetermined value on a first axis of the plot of the feature space, the additional histology image is classified as cold.

32. 31. The non-transitory computer readable medium of claim 30, wherein the additional histology image is classified as rejected if the value of the additional actual inflammation score is greater than or equal to a predetermined value on a first axis of the plot of the feature space and the value of the additional tumor infiltration score is less than a predetermined value on a second axis of the plot of the feature space.

33. 31. The non-transitory computer readable medium of claim 30, wherein the additional histology image is classified as inflammatory type if the value of the additional actual inflammation score is greater than or equal to a predetermined value on a first axis of the plot of the feature space and the value of the additional tumor infiltration score is greater than a predetermined value on a second axis of the plot of the feature space.

34. receiving, by at least one processor of a computing device, a plurality of histology images of tumor samples in a plurality of patients; performing, by at least one processor, image analysis of the plurality of histology images to obtain information regarding the biomarkers in each of the plurality of histology images; determining, by at least one processor, a real inflammation score and a tumor infiltration score based on a polar transform of the abundance of CD8+ T cells in the tumor parenchyma and stroma; generating, by at least one processor, a feature space based on the actual inflammation score and the tumor infiltration score; and Identifying, by the at least one processor, a linear boundary or linear cutoff between the plurality of classifications in the feature space based on the actual inflammation score, the tumor infiltration score, and the patient response data. The method includes:

35. receiving, by at least one processor of a computing device, information including stromal CD8+ T cell parameters and parenchymal CD8+ T cell parameters for each of the plurality of classified histology images; obtaining, by at least one processor, an abundance of CD8+ T cells in the tumor parenchyma and stroma for each of the plurality of histology images; determining, by at least one processor, a real inflammation score and a tumor infiltration score based on a polar transform of the abundance of CD8+ T cells in the tumor parenchyma and stroma; generating, by at least one processor, a feature space based on the actual inflammation score and the tumor infiltration score; and Identifying, by the at least one processor, a linear boundary or linear cutoff between the plurality of classifications in the feature space based on the actual inflammation score, the tumor infiltration score, and the patient response data. The method includes:

36. receiving, by at least one processor, information including stromal CD8+ T cell parameters and parenchymal CD8+ T cell parameters for an additional histology image, the additional histology image corresponding to a particular patient; obtaining, by at least one processor, an abundance of CD8+ T cells in the tumor parenchyma and stroma for the new histology images; and identifying, by the at least one processor, a classification of the new histology image by comparing the abundance of CD8+ T cells for the new histology image to boundaries in the feature space; 36. The method of claim 35, further comprising:

37. Determining, by at least one processor, a disease state of a particular patient based on the classification of the new histology image.

37. The method of any one of claims 9 and 36, further comprising:

38. Determining, by at least one processor, the likelihood of a particular patient's response to a given treatment based on the classification of the new histology image.

37. The method of any one of claims 9 and 36, further comprising:

39. Determining, by at least one processor, a particular patient's response to a given treatment based on the classification of the new histology image.

37. The method of any one of claims 9 and 36, further comprising:

40. Recommending, by at least one processor, a particular procedure for a particular patient based on the classification of the new histology image.

37. The method of any one of claims 9 and 36, further comprising:

41. Determining, by at least one processor, a disease state of a particular patient based on the classification of the new histology image. The method of any one of claims 1 to 8, 34 and 35, further comprising:

42. Determining, by at least one processor, the likelihood of a particular patient's response to a given treatment based on the classification of the new histology image. The method of any one of claims 1 to 8, 34 and 35, further comprising:

43. Determining, by at least one processor, a particular patient's response to a given treatment based on the classification of the new histology image. The method of any one of claims 1 to 8, 34 and 35, further comprising:

44. Recommending, by at least one processor, a particular procedure for a particular patient based on the classification of the new histology image. The method of any one of claims 1 to 8, 34 and 35, further comprising:

45. The processor: Determine the disease status of a particular patient based on classification of new histology images The system of claim 17 further configured to:

46. The processor: Determine the disease status of a particular patient based on classification of new histology images The system according to any one of claims 14 to 16, further configured to:

47. The processor: Determine the likelihood of a particular patient's response to a given treatment based on classification of new histology images The system of claim 17 further configured to:

48. The processor: Determine the likelihood of a particular patient's response to a given treatment based on classification of new histology images The system according to any one of claims 14 to 16, further configured to:

49. The processor: Determine a particular patient's response to a given treatment based on classification of new histology images The system of claim 17 further configured to:

50. The processor: Determine a particular patient's response to a given treatment based on classification of new histology images The system according to any one of claims 14 to 16, further configured to:

51. The processor: Recommend specific procedures for specific patients based on new histology image classification The system of claim 17 further configured to:

52. The processor: Recommend specific procedures for specific patients based on new histology image classification The system according to any one of claims 14 to 16, further configured to:

53. The operation is Determining the disease status of a particular patient based on classification of additional histology images.

30. The non-transitory computer readable medium of claim 29, further comprising:

54. The operation is Determining the disease status of a particular patient based on classification of new histology images 29. The non-transitory computer readable medium of any one of claims 25 to 28, further comprising:

55. The operation is Determining the likelihood of a particular patient's response to a given treatment based on classification of additional histology images.

30. The non-transitory computer readable medium of claim 29, further comprising:

56. The operation is Determining the likelihood of a particular patient's response to a given treatment based on classification of new histology images.

29. The non-transitory computer readable medium of any one of claims 25 to 28, further comprising:

57. The operation is Determining a particular patient's response to a given treatment based on classification of additional histology images.

30. The non-transitory computer readable medium of claim 29, further comprising:

58. The operation is Determining a particular patient's response to a given treatment based on classification of new histology images.

29. The non-transitory computer readable medium of any one of claims 25 to 28, further comprising:

59. The operation is Recommending specific procedures for specific patients based on classification of additional histological images 30. The non-transitory computer readable medium of claim 29, further comprising:

60. The operation is Recommending specific procedures for specific patients based on new histology image classifications 29. The non-transitory computer readable medium of any one of claims 25 to 28, further comprising:

Images