Methods for assessing molecular changes associated with molecular effects in biological samples

Abstract

The present invention relates to a method for evaluating ex vivo or in vitro the effect of a molecule of interest on at least one molecular marker in a administered biological sample. The method of the present invention is based on segmenting the biological sample on the basis of the concentration of the molecule of interest and comparing molecular changes associated with the presence of the molecule of interest in different parts / segments of the same biological sample or between different biological samples that may contain different amounts of the molecule of interest and / or different molecules of interest.

Description

Technical Field

[0001] The present invention relates to a method for manually or automatically evaluating molecular changes following the concentration or intensity of at least one molecule of interest in a biological sample. More specifically, the present invention relates to a method in which imaging techniques are used to detect changes relative to molecular biomarkers associated with the presence of a molecule of interest, such as an active compound, in a biological sample. The present invention allows defining areas related to the location of the active compound in order to calculate a score, ratio or any value of the molecular changes following the concentration of the molecule of interest.

[0002] The method of the present invention can be applied to all fields involving the study of the behavior of molecules of interest in biological samples. The method of the present invention can be advantageously used in proteomics, polypeptideomics, lipidomics, metabolomics, glycomics or pharmaceutical studies in order to screen candidate molecules and assess their therapeutic or diagnostic potential. Background Art

[0003] In the pharmaceutical, nutraceutical, agrochemical or cosmetic industries, the development of active compounds requires the identification of the mechanism of action (MoA), efficacy and toxicity in in vitro cell models at the early preclinical stage or later in in vivo animal models. At a later stage, these assays are developed and transferred to the clinical stage to validate the efficacy and toxicity of the drug in humans. Mass spectrometry (MS), immunohistochemistry (IHC), enzyme-linked immunosorbent assay (ELISA), in situ hybridization (ISH), polymerase chain reaction (PCR) are the gold standard methods used for the evaluation and quantification of molecular changes at different stages. All of these methods have been developed to quantify molecules in a sample and identify their role in drug efficacy or toxicity.

[0004] All these methods are commonly used to study tissue responses by evaluating and quantifying biomarkers for diagnosis, prognosis, efficacy or response to treatment. When used for control and drug-treated (using active compounds) samples, mass spectrometry is one of the most advanced methods because it does not require labeling of the molecules of interest to be quantified. Mass spectrometry provides molecular information about the biological impact of the active compound as an increase or decrease in the concentration of lipids, metabolites, peptides or proteins in living systems. Other techniques (such as IF, IHC, ELISA, ISH, spectrophotometry or UV fluorescence) allow the evaluation and quantification of peptides, proteins, RNA (ribonucleic acid) and DNA (deoxyribonucleic acid) targeted using labeled antibodies or nucleic acid sequences that bind to the target.

[0005] In the pharmaceutical industry, pharmacodynamics is a field in which scientists study the effects of a drug at a certain time after administration and assess changes in target molecules (biomarkers) or changes in non-target molecules (to study drug responses on a larger scale and identify potential unknown compounds).

[0006] At the tissue level, all the methods used that have been described so far compare the tissue as a whole, which means that a lot of information is lost. These practical methods use relative or absolute quantification of molecular changes in whole tissues or tissue sections and do not take into account where the concentration of the active compound of interest is located in the tissue, cell or other specific area of ​​interest. The methods do not take into account the impact of the localization of the drug on its neighborhood.

[0007] At the cellular level, current cell viability-based measurements often result in biased response estimates due to the analysis of all cells without correlation to the relative or absolute concentration of the drug or active compound that has penetrated or entered the cells.

[0008] Therefore, there is a need for methods that allow the concentration of active compounds to be correlated with biomarkers of efficacy with greater accuracy and precision in order to study the response of cells to active compounds. Summary of the invention

[0009] In this context, the inventors have now proposed a method that combines the identification and localization of a compound of interest (e.g., a drug candidate) in a target tissue with image analysis of molecular changes in the region where the compound of interest is located. Observing the location of the compound of interest and selecting the region where it is specifically located and / or the region where it presents different concentrations allows the response to the compound and changes in related biomarkers to be studied at a very fine scale. According to the present invention, comparisons between different samples or between different regions of the same sample lead to an understanding of the molecular environment that may or may not be related to the presence and / or concentration of the compound of interest. The method of the present invention also allows the molecular impact of the compound of interest to be studied and scored by comparing different calculated ratios between the compound of interest and related biomarkers (e.g., markers of efficacy or toxicity), leading to an understanding of the impact of the compound of interest at the neighborhood level (e.g., efficacy or toxicity).

[0010] Therefore, an object of the present invention is to provide a method for evaluating the effect of at least one molecule of interest on at least one molecular marker in a administered biological sample in vitro or in vitro, the method comprising:

[0011] - selecting a biological sample for administration, which has previously been exposed to said at least one molecule of interest;

[0012] - detecting the presence of said at least one molecule of interest in said administered biological sample using a molecular imaging method, and obtaining a molecular profile of said administered biological sample for said at least one molecule of interest;

[0013] - spatially segmenting the molecular map of the biological sample for medication according to the spectral information to obtain a segmented map of the at least one molecule of interest in the biological sample for medication;

[0014] - selecting a first region of interest (ROI) from the segmented map, the first ROI having a first intensity of the at least one molecule of interest;

[0015] - measuring the intensity or amount of at least one molecular marker in the first ROI;

[0016] - comparing the intensity or amount of at least one molecular marker in said first ROI with the intensity or amount of at least one molecular marker in a second ROI, wherein said second ROI is selected from said administered biological sample or another biological sample.

[0017] The methods can be performed using any kind of biological sample that can be analyzed using molecular imaging methods, including tissue samples, organoids, biological fluid samples such as urine samples, plasma samples, cerebrospinal fluid, and cell suspensions.

[0018] In a specific embodiment, the administered biological sample is a tissue section obtained by sampling an animal previously administered with the molecule of interest. Alternatively or in addition, the administered biological sample has been contacted with the molecule of interest in vitro.

[0019] In certain embodiments, the molecule of interest is a candidate molecule, and the administered biological sample is previously obtained by sampling an animal previously administered with the candidate molecule.

[0020] In a specific embodiment, the second ROI is selected from a segmentation map of the medicated biological sample, the second ROI being physically different from the first ROI and having a second intensity of the molecule of interest. In another embodiment, the second ROI is selected from a second biological sample, the second biological sample has been previously exposed to the molecule of interest at a dose concentration different from the dose concentration used for the medicated biological sample, the second biological sample being from the same biological origin as the medicated biological sample. Optionally, the second ROI is selected from a second biological sample that has not been previously exposed to the molecule of interest, the second biological sample being from the same biological origin as the medicated biological sample. In another embodiment, the second ROI is selected from a second biological sample that has been previously exposed to a second molecule of interest, the second biological sample being from the same biological origin as the medicated biological sample.

[0021] In certain embodiments, the comparison between the first and second ROIs allows for the identification of at least one biomarker specific for the biological effect and response of the molecule of interest.

[0022] In a particular embodiment, the first and second ROIs have been contacted with two molecules of interest corresponding to two different candidate molecules, and comparison between the first ROI and the second ROI allows differentiation of the candidate molecules. In such an embodiment, the biomarkers compared in the first and second ROIs may be the same or different.

[0023] In another embodiment, the first and second ROIs have been contacted with several (ie more than one) molecules of interest, which may be the same or different between the ROIs. In a particular embodiment, both ROIs have been contacted with the same two or more molecules of interest.

[0024] In certain embodiments, the biomarkers compared in the first and second ROIs are the same, even if the molecule of interest of the second ROI is different from the molecule of interest of the first ROI. Thus, the methods of the present invention can be used to assess and compare the effects of at least two different molecules of interest on at least one molecular marker.

[0025] The method of the present invention can be carried out using a molecular imaging method selected from MRI imaging, PET imaging, CT imaging, IF, ISH, IHC and mass spectrometry or cytometer imaging. The method of the present invention is particularly suitable for mass spectrometry imaging, such as MALDI, DESI, LESA, LA-ICP-MS and SIMS, for detecting the presence of the molecule of interest in the biological sample, and thereby mapping the biological sample based on the detection.

[0026] Measurement of the amount or intensity of the biomarker in the selected ROI can be performed using molecular imaging methods or by other bioanalytical techniques (HPLC, LC-MS / MS, GC / MS, magnetic bead multiplex immunoassay, ELISA).

[0027] In a specific embodiment, the method includes a step consisting of establishing a dose-effect curve between one or different molecules of interest and molecular markers, wherein the intensity or amount of each of the one or more molecules of interest in the ROI is expressed according to the intensity or amount of the molecular marker in the same ROI.

[0028] In certain embodiments, the ROI is extracted from the biological sample by laser capture microdissection. The ROI can then be cultured in culture medium prior to performing the bioanalysis to assess the impact of the molecule of interest, for example, by comparing changes in gene or transcript expression, lipidomics, peptideomics, proteomics, and / or metabolomics between the first and second ROIs.

[0029] In certain embodiments, the molecule of interest is a therapeutic antibody, and wherein the presence of the therapeutic antibody is detected by contacting the biological sample with a labeled anti-therapeutic antibody antibody. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 : Molecular profiles of drug-exposed biological samples ( Figure 1 A) and segmenting the molecular profile into regions where the drug has a low intensity ( Figure 1 B, left side) and wherein the drug has a high strength ( Figure 1 The segmentation allows to distinguish the regions that are differently affected by the drug. The two ROIs show different concentrations of biomarkers associated with the drug ( Figure 1 C), which may be associated with differences in the concentration of the drug, in contrast to global methods such as LCMS ( Figure 1 D).

[0031] Figure 2 : Molecular profile of tissue samples not exposed to the drug ( Figure 2 Molecular profiles of tissue samples (A–CTRL: control) and those exposed to the drug Figure 2 B – TRT: treated). The entire first molecular profile was used as ROI, while the second molecular profile was segmented to select the ROI with high concentration of the drug. Both tissue samples have the same origin (tumor tissue). The ROI from the control shows a high concentration of the biomarker associated with the absence of the drug, while the ROI from the treated sample shows a low concentration of the biomarker and a high concentration of the drug ( Figure 2 C).

[0032] Figure 3 : Molecular profiles of tissue samples not exposed to the drug and of tissue samples from three biological replicates ( Figure 3 A – CTRL, TRT Tumor 1, TRT Tumor 2 and TRT Tumor 3). The entire molecular profile was used as ROI for the corresponding samples. All tissue samples had the same origin (tumor tissue). ROIs from controls showed high concentrations of biomarkers associated with the absence of the drug, while ROIs from treated samples showed lower concentrations of the biomarkers, which may be associated with their corresponding drug concentrations ( Figure 3 B).

[0033] Figure 4 : Molecular profile of a tissue sample that has been exposed to the first drug ( Figure 4 A – drug A), molecular profile of tissue samples that have been exposed to the second drug ( Figure 4B – Drug B) and the molecular profile of the same biomarker (Biomarker, Figure 4 A and 4B). The efficacy of drug A and drug B is evaluated by reference to the concentration of the biomarker in two or three ROIs previously identified relative to the concentration of the drug ( Figure 4 C and 4D).

[0034] Figure 5 :High-level image segmentation allows the acquisition of multi-segment maps (pixel maps). Each pixel contains drug (drug A or B) and biomarker-related information from C1 to C11 concentrations in μg / g tissue ( Figure 5 A). Thus, the biological effect of the drug is obtained for each segment (pixel), which corresponds to the different concentration or intensity levels of the biomarker based on the different concentration or intensity levels of the drug ( Figure 5 B) Different curves can then be plotted which give the dose effect of the drug in the individual tissues. ED50: median effective dose, i.e. the dose required to achieve 50% of the desired response.

[0035] Figure 6 : Molecular profiles of tissue samples that have been exposed to drugs ( Figure 6 A), automatic segmentation of the molecular map and identification of ROI ( Figure 6 B) Schematic diagram of ROI extraction by laser capture microdissection ( Figure 6 C).

[0036] Figure 7 : Epacadostat (EPA) calibration curve / QMSI (A), EPA calibration curve / LC-MSMS (B), EPA quantification by QMSI and LC-MS / MS (C).

[0037] Figure 8 : Epacadostat drug detection, quantification and histological localization. After 1,5-DAN matrix deposition, 10 μm thick CT26 control and treated tumor sections were analyzed in duplicate by MALDI MSI. Epacadostat drug histological localization showed presence in the treated tumors, but not in the control tumors. Two parallel samples (1 and 2) were subjected to absolute quantification of EPA in μg / g and shown in the table at the bottom (A). Comparison of LC-MS / MS and QMSI EPA quantification in treated CT26 tumors showed a 17% variation between the two techniques (B). Scale color and scale bars for all sections are shown.

[0038] Fig. 9: Target exposure analysis. Three regions of interest (1, 2, and 3) showing different amounts of EPA were selected. ROIs 1 and 2 accounted for 38% and 62% of the entire tumor surface. Specific EPA localization was found, showing 61% concentrated in 38% of the surface and 39% present on 62% of the remaining tumor surface (A). Serial tissue sections were semi-quantitatively immunostained for IDO1 and inset. Figure 1 and 2 Different expression levels of the IDO1 enzyme on the two regions are shown (B). All these overlays were performed using ImaBiotech's multi-mode platform Multimaging software.

[0039] Fig.10 : Histological localization and absolute quantification. EPA, Trp and Kyn histological localization on both control and treated CT26 tumor sections are shown (A). QMSI and LC-MS / MS analysis of Trp and Kyn was then performed, and the Kyn / Trp ratio was then obtained from control and treated CT26 tumors. Both QMSI and LC-MS / MS analysis noted a 6-fold reduction in Kyn / Trp ratio levels (B).

[0040] Fig.11 : Normalized Trp and Kyn calibration curve: x = concentration in mg / g; y = intensity ( Fig.11 A). Kyn / Trp ratios of plasma and blood samples ( Fig.11 B).

[0041] Fig.12 : Exposure to response and response efficacy analysis. Based on the amount of Kyn in three segmented ROIs, the regional pharmacological effects and response efficacy of EPA are highlighted; 15% of Kyn is concentrated in 38% of the entire tumor (ROI 1), and 85% is concentrated in ROI 2 (A). The relative intensity of EPA, Kyn and lactate was also extracted from the molecular image (B) to show the response efficacy when EPA is highly or poorly present (C). Regional segmentation, relative and absolute quantification were performed using ImaBiotech's multi-mode platform Multimaging software. DETAILED DESCRIPTION OF THE INVENTION

[0043] The method of the present invention is based on segmenting a biological sample on the basis of the concentration of a molecule of interest and comparing molecular changes associated with the presence of the molecule of interest in different parts / segments of the same biological sample or between different biological samples that may contain different amounts of the molecule of interest and / or different molecules of interest. More specifically, according to the present invention, imaging techniques are used to detect and locate the compound in a biological sample previously exposed to a compound of interest. The biological sample is then segmented on the basis of the intensity of the compound in the sample, and the molecular changes are analyzed by comparing the sample segments. The method allows identification of molecules (i.e., biomarkers) that increase or decrease due to the presence of the target molecule and / or due to different doses of the target molecule, even at the cellular level. The method of the present invention can also allow comparison of the molecular effects of the compound of interest between two or more regions of the same biological sample, so as to accurately determine the therapeutic and / or preventive and / or toxic effects of the compound in the target biological sample. New potential drugs can then be identified as well as their dose effects and ultimately associated biomarkers.

[0044] According to the present invention, imaging techniques are used to locate the molecule of interest, and then image segmentation is used to analyze the molecular image of the biological sample based on the relative or absolute amount of the molecule of interest in the entire biological sample. Analysis of selected segments allows the determination of molecular changes that may be attributed to the presence or specific dose of the molecule of interest, and thereby selects valuable biomarkers of the molecule of interest and / or screens valuable drugs, etc.

[0045] In contrast to prior art methods, the method of the present invention can provide information on the molecular efficacy associated with a molecule of interest in different parts of a biological sample, including between different cells of the same biological tissue. Thus, the real-time impact of a molecule in a target biological sample can be assessed. Assuming that a molecular profile of the biological sample represents a certain concentration range of the compound, the active compound / biomarker ratio can be calculated for each image pixel, which will give the drug dose effect in a single tissue.

[0046] Preparation of biological samples

[0047] According to the method of the present invention, all kinds of biological samples previously exposed to molecules of interest can be analyzed. More generally, the biological sample encompasses all systems that exhibit cell activity. This includes, for example, cell cultures, 3D in vitro models such as spheroids or micro-organs, in vivo systems such as animal models, plants, xenograft tissues, tumors and animal biopsies, biological fluids, etc. In favorable cases, the biological sample is selected from tissue samples (e.g., biopsy samples), biological fluid samples (e.g., as urine samples, plasma samples, cerebrospinal fluid) and cell suspensions.

[0048] In the context of the present invention, the term "tissue" refers to a group of functional grouped cells. The target tissue can be a group of similar or different cells, an organ, a part of an organ, a specific area of ​​an organ optionally having a multicellular aggregate, of the same origin. For example, the target tissue can be a tumor located in an organ. In another embodiment, the tissue sample is a plant tissue sample. More generally, a tissue sample refers to any type of tissue of biological origin in a form that can be analyzed by an imaging method.

[0049] Biological fluids encompass all fluids of biological origin, which contain cells and / or are obtained from animals or plants.

[0050] In the context of the present invention, "exposure to" a molecule of interest means that the biological sample has been in contact with the molecule. In the context of the present invention, a biological sample that has been exposed to the target molecule is referred to as a "medicated or treated biological sample". In contrast, a "control sample" refers to a biological sample that has not been exposed to the target molecule. According to the present invention, a control sample has the same biological origin as the medicated sample to which it is compared. For example, if the medicated tissue sample consists of a liver tissue slice sampled from a mouse that has been exposed to the target molecule, the control sample will consist of a liver tissue slice sampled from a mouse that has not been exposed to the target molecule.

[0051] In a specific embodiment, the molecule of interest has been previously administered to an animal model, preferably a mammal, including a human and a non-human mammal, and the animal has been sampled prior to the implementation of the method. In a specific embodiment, the animal model is a non-human mammal. In another specific embodiment, the animal model is a human.

[0052] Depends on required research, described animal model can change.Professional and technical personnel understand which kind of animal model is well adapted to the target tissue, interested molecule, biological property etc. being assessed.For example, in the case of preclinical test of candidate molecule needing to put to death animal, preferably use non-human mammals such as rodents (mice, rats, rabbits, hamsters etc.).Other non-human mammals, particularly monkeys, dogs etc. can be used.In some cases where biological sample can be sampled in the case of no harmful side effects to animals, mankind can be used as animal model (for example, for 1-4 phase clinical trial).Other animal models such as fish, insects can also be used, for example, to study the impact of molecules on environment or specific ecological medium.

[0053] According to the present invention and in a broad sense, all routes of administration of the target molecule can be used, such as enteral routes (i.e., the drug is administered during the digestion process of the gastrointestinal tract) or parenteral routes (i.e., other routes of administration outside the gastrointestinal tract). For example, the molecule can be administered by different routes, such as epidermal, epidural, intraarterial, intravenous, subcutaneous (with specific positioning), intracardiac, intracavernous injection, intracerebral, intradermal, intramuscular, intraosseous infusion, intraperitoneal, intrathecal, intravesical, intravitreal, nasal, oral, rectal, intravaginal or topical administration.

[0054] The administration route can be selected according to the molecule of interest, the tissue targeted by the method, etc. The method of the present invention can also allow the selection of the most suitable administration route. In fact, the method of the present invention allows the evaluation of the ability of the molecule of interest to cross the biological barrier to reach the target tissue.

[0055] According to the invention, the method is preferably performed ex vivo and / or in vitro. In some cases, in vivo analysis can also be performed on living intact animals.

[0056] In a specific embodiment, the method of the present invention is an in vitro analysis, such as an in vitro analysis of a tissue section. In this case, the tissue sample is sampled at a given time (t1) after administration. The sampling can be a biopsy, particularly when the animal is a human. According to one embodiment, after the molecule of interest is administered to a non-human animal, the non-human animal is killed and the tissue sample of interest is sampled.

[0057] In a particular embodiment, the tissue sample is obtained from fresh tissue, frozen tissue or tissue fixed / embedded, for example, in paraffin.All means suitable for obtaining thin tissue sections, for example a few micrometers thick, can be used.

[0058] If necessary, the tissue section can be subjected to pretreatment, which depends particularly on the molecule to be detected, analytical techniques, etc. Therefore, chemical or biochemical reagents can be used to optimize the detection of the molecule of interest and biomarker. For example, solvents and / or detergents can be used to allow the detection of molecules of limited categories or to improve the direct extraction of molecules from tissues. In addition, specific enzymes that can cut peptides or proteins can be used so that targeting, for example, has the same location and / or amount of the parent molecule on tissues. Certain analytes can also be added on or in tissues for the chemical derivatization of the molecule of interest, because it solves certain sensitivity problems, allowing to improve the distribution and quantitative research on the tissue. Antibody labeling (coupling or without label) or fluorescently labeled molecules or radioactivity can also be carried out on tissue sections to allow the detection of the molecule of interest and biomarker.

[0059] The animal model used and / or the target tissue and / or tissue section can also be changed to modify their ability to combine or be incorporated into the molecule of interest. Therefore, this treatment can include the chemical or biological modification of the animal model and / or target tissue and / or tissue section, which allows to improve or suppress the penetration or targeting ability of the molecule of interest to a given target tissue. This treatment can be carried out before, after or simultaneously with the administration of the molecule of interest. For example, in the case where a molecule must cross the blood-brain barrier (BBB), there are some efflux transporters in the barrier that can discharge the molecule of interest across the BBB. The effect of these transporters can be regulated (reduced or suppressed) using inhibitors or genetic modification such as "knockout" to the gene of the transporter or gene expression.

[0060] In the case of a liquid sample, it can be dried on a surface in order to produce an MS image of the dried sample and then characterize the sample using MSI.

[0061] If a biological sample is studied using mass spectrometry imaging, in particular MALDI or ME-SIMS (matrix enhanced secondary ion mass spectrometry), which requires a matrix, the matrix is ​​adapted to the molecule of interest in an advantageous case. For example, the selection can take into account the mass range covered. The skilled person knows which of the existing liquid or solid matrices can be used depending on the molecule and / or target tissue being studied. Similarly, all deposition methods of the matrix can be used, in particular manual spraying, automatic spraying, sublimation, sieving and micro-spotting.

[0062] In another embodiment, the molecule of interest is contacted with a biological sample such as a tissue sample, a biological fluid, a cell suspension, etc. in vitro or in vitro. For example, the biological sample is a cell suspension, and the molecule of interest is added to the suspension and then analyzed. Alternatively, the biological sample can be deposited on a support (and in the case of a fluid sample, it can be optionally dried), and the molecule of interest is sprayed on the sample and then analyzed.

[0063] Detection of molecules of interest

[0064] The biological sample is analyzed to detect the presence and optionally the concentration of the molecule of interest in the biological sample.

[0065] This step can be performed using any technique that allows accurate identification and visualization of molecules in a biological sample in vivo, in vitro or ex vivo.

[0066] Particularly, in the case of in vivo analysis, tomography techniques such as magnetic resonance imaging (MRI), autoradiography, positron emission tomography (PET), single photon emission tomography, CT imaging, etc. can be used.

[0067] In the case of in vitro / ex vivo analysis, IF, IHC, ISH and mass spectrometry imaging (MSI) can be used. Such as MALDI imaging (matrix-assisted laser desorption / ionization), LDI (laser desorption / ionization), DESI (by electrospray desorption), LESA (liquid extraction surface analysis), LAESI (laser ablation electrospray ionization), DART (real-time direct analysis), SIMS (secondary ion mass spectrometry), JEDI (jet desorption electrospray ionization) technology can be used, and optionally combined with different types of mass analyzers such as TOF (time of flight), Orbitrap, FTICR (Fourier transform ion cyclotron resonance), quadrupole (single or triple) etc. Preferably, the mass spectrometry imaging is selected from MALDI, DESI, ICP-MS and SIMS.

[0068] Experimental parameters such as mass range, laser energy density, laser focusing, etc. are fixed to optimize the intensity, sensitivity and resolution of target detection. Therefore, the acquisition of mass spectrogram is performed to obtain a signal. From the mass spectrogram, data useful for target molecule research can be obtained. For data processing, different spectral features can be used, such as peak intensity, signal-to-noise ratio (S / N), peak area, etc. on the mass spectrogram.

[0069] As a general rule, all techniques that allow visualization of molecules within a biological sample may be used.

[0070] Region of Interest (ROI) Selection

[0071] The analysis of the biological sample provides at least a molecular atlas of the biological sample for the molecule of interest. That is, an image of the molecule of interest throughout the biological sample is obtained. The image allows for direct visualization of the distribution and concentration of the molecule of interest in the biological sample. Thus, the presence and / or absence of the molecule in different regions of the biological sample can be visualized, and, assuming that a molecular atlas of the biological sample represents a concentration range, the concentration differences between the different regions can also be visualized. The molecular atlas of the biological sample is then segmented into different regions.

[0072] The segmentation of the biological sample can generate regions of various sizes from pixels to segments (comprising a plurality of pixels). The segmentation can be performed manually or automatically so as to obtain different regions characterized by the presence / absence / intensity / amount of the molecule of interest (i.e., the spectral information of the molecule of interest). For example, the segmentation can be performed automatically using an image segmentation technique selected from region-based segmentation, edge detection segmentation, cluster-based segmentation, segmentation based on weakly supervised learning in CNN, etc. A region of interest (ROI) can then be selected.

[0073] The size of the ROI can be different from each other. The ROI can be the average value of many image pixels or segments. Each image pixel or segment can also represent a ROI with a different intensity or concentration of the molecule of interest.

[0074] In a specific embodiment, the biological sample is a tissue section that has been analyzed by MSI. The image obtained shows the superimposed distribution of the molecules of interest. Therefore, the segmentation is performed on the basis of the peak intensity, peak area or signal-to-noise ratio of the molecules of interest.

[0075] For example, the molecule of interest is a candidate molecule, and the administered biological sample is previously obtained by sampling an animal that has been previously administered with the candidate molecule.

[0076] In one embodiment, the second ROI is selected from the segmentation atlas of the administered biological sample, the second ROI being physically different from the first ROI and having a second intensity of the molecule of interest. In an advantageous case, the selected first ROI exhibits the most significant signal (and therefore the most significant concentration) and the selected second ROI does not exhibit any signal of the molecule of interest (i.e. is deprived of the molecule of interest). Alternatively, the second ROI of the administered biological sample exhibits a low signal (and therefore a low concentration) of the molecule of interest.

[0077] Figure 1 The molecular profile of a tissue section obtained for a drug using MSI is shown. The tissue section was sampled from an animal that had previously been administered the drug. The molecular profile ( Figure 1 A) is based on the intensity associated with the molecule. From the spectral data visualized on the tissue section, two ROIs can be drawn within the same tissue section ( Figure 1 B) The first ROI corresponds to a region of the tissue sample that exhibits a high signal of the molecule, and the second ROI corresponds to a region of the tissue sample that exhibits a low signal of the molecule.

[0078] In another embodiment, the second ROI is selected from a second biological sample that has not been previously exposed to the molecule of interest, the second biological sample being from the same origin as the administered biological sample (ie, control sample).

[0079] In the context of the present invention, "same origin" means that the biological samples are sampled on the same animal model (e.g., two mice) and that the samples are consistent (e.g., two biopsy tissues from the same tumor cell lineage tissue, two liquid samples of urine, etc.).

[0080] Figure 2The molecular profiles of two tissue sections of the same origin (tumor tissue) obtained using MSI for drug are shown. The first tissue section ( Figure 2 A) Samples were collected from animals not exposed to the drug (control, CTRL). The second tissue section ( Figure 2 B) Sampling in animals previously administered the drug (treated, TRT). The first tissue section as a whole is used as the first ROI. The second ROI is selected in the second tissue section and corresponds to the region of the second tissue section where the molecule is highly detected.

[0081] In another embodiment, the second ROI is selected from a second biological sample previously exposed to the molecule of interest at a different dose concentration than that used for the administered biological sample, the second biological sample being from the same biological origin as the administered biological sample.

[0082] Figure 3 The molecular profiles of 4 tissue sections of the same origin (tumor tissue) obtained for the drug using MSI are shown. The first tissue section is sampled in an animal that has not been in contact with the drug (control, CTRL). All other tissue sections are sampled in animals that have been previously administered the drug of the same concentration (TRT1, TRT2, TRT3), but the concentrations in the samples are different. The ROI of each biological sample is the entire corresponding tissue section. Of course, the ROI can be limited to a specific area of ​​each tissue section, such as a ROI that exhibits the most important concentration of the drug.

[0083] In another embodiment, the second ROI is selected from a second biological sample that has been previously exposed to the same or a second molecule of interest, the second biological sample being from the same origin as the administered biological sample.

[0084] Figure 4 Molecular profiles of two tissue sections of the same origin (tumor tissue) obtained for two different drugs using MSI are shown ( Figure 4 A: Drug A; Figure 4 B: Drug B). The tissue sections were sampled from animals that had been previously administered either Drug A or Drug B. The ROI1 and ROI2 were manually segmented on each biological sample. In order to understand the possible drug effects of Drug A and B on the biomarker, its intensity / concentration in each ROI (1 and 2) was then calculated and plotted for each corresponding ROI ( Figure 4 C and 4D).

[0085] In a specific embodiment, the method of the present invention can be used to establish a dose-effect curve between one or more molecules of interest and a molecular marker, wherein each intensity or amount of a molecule of interest in a ROI (a single pixel or several pixels) is expressed according to the intensity or amount of the molecular marker in the same ROI.

[0086] Figure 5 A molecular map of a biological sample representing a range of concentrations of the compound of interest is shown. A higher level of image segmentation allows the acquisition of multi-segment maps (pixel maps). Each pixel contains drug and biomarker related information (in μg / g tissue from C1 to C11 concentrations). Figure 5 A). Thus, the biological effect of the drug is obtained for each segment (pixel), which corresponds to the different concentration or intensity levels of the biomarker based on the different concentration or intensity levels of the drug ( Figure 5 B) Different curves can then be obtained which give the dose effect of the drug in the individual tissues. ED50: median effective dose, ie the dose required to achieve 50% of the desired response.

[0087] Biomarker analysis

[0088] Molecular analysis is performed on each selected ROI to measure the amount or intensity of one or more biomarkers in the ROI of the corresponding biological sample.

[0089] The molecular analysis results in an understanding of the composition of the ROI at a molecular level, and / or the detection and optionally quantification of one or more biomarkers. Advantageously, all ROIs are analyzed using the same method.

[0090] According to one embodiment, the measurement of the amount or intensity of the biomarker in the ROI is performed using a molecular imaging method. Preferably, the molecular imaging method is selected from MSI, ISH, IF, IHC, MRI, imaging mass spectrometry, CT and PET imaging.

[0091] In a specific embodiment, both ROIs are analyzed by MSI. MSI refers to ICP-MS, LA-ICPMS, LAESI, MALDI, DESI, SIMS, LESA and similar surface extraction strategies, SIMS, etc. Basically, mass spectrometry imaging (MSI) can record the distribution of hundreds of biomolecules simultaneously directly from tissue or fluid samples without labeling and without prior knowledge. MSI using different tissue / fluid preparation procedures based on matrix-assisted laser desorption / ionization (MALDI) can be used to analyze proteins, peptides, polysaccharides, lipids, metabolites and drugs.

[0092] According to another embodiment, the measurement of the amount or intensity of the biomarker in the ROI is performed by a bioanalytical method. Preferably, the bioanalytical method is selected from mass spectrometry, electrophoresis, ligand binding assay, nuclear magnetic resonance, microdialysis and chromatography.

[0093] In a specific embodiment, the ROI is extracted from the biological sample before analyzing the biomarkers. In the context of the present invention, "extraction of the ROI" means physically separating the ROI from the rest of the sample. Physical extraction can be performed manually or by microdissection using a laser. Figure 6 C shows laser capture microdissection (LCM) performed on a tissue section. The ROI of interest is thus sampled and separated from the rest of the tissue section. This extraction allows the bioanalysis to be performed only on the portion of the tissue section that must be considered and to distinguish different parts of the same tissue section with greater accuracy.

[0094] Figure 6 shows the molecular profile of tissue samples that have been exposed to the drug ( Figure 6 A), then the molecular map is manually or automatically segmented and the ROI is identified ( Figure 6 B), followed by ROI extraction by laser capture microdissection ( Figure 6 C).

[0095] According to a specific embodiment, the extracted ROI is cultured in culture medium prior to performing said bioassay.

[0096] In certain embodiments, biological analysis is performed on two ROIs and the effect of the molecule of interest is assessed by comparing changes in gene and transcript expression, lipidomics, peptidomics, proteomics and / or metabolomics between the first ROI and the second ROI (or other ROIs).

[0097] Molecular analysis of the ROIs and comparative studies of the results may yield different interpretations.

[0098] According to the present invention, the method for implementing the molecular analysis can be a direct method or an indirect method (e.g., using a labeling molecule such as an antibody or an RNA sequence). For example, the molecular analysis can be performed using IHC, ISH or FISH imaging of the region of interest, LA-ICP imaging using a labeled antibody or separate LA-ICP imaging for detecting metals on tissues, genomic analysis, SNIPS, LC-MS analysis.

[0099] In a specific embodiment, the biomarker associated with the molecule of interest is known, and the analysis step focuses on the detection and / or quantification of the biomarker. The method of the present invention can be used to identify the target tissue at a very fine level. The method of the present invention can also be used to determine the optimal dose of the expected effect of the drug in the target tissue and / or for screening drugs.

[0100] Figure 1 An embodiment is shown in which a biomarker (BM) associated with a drug is known. The molecular profile of the tumor tissue shows that the drug is located in a specific part of the tissue ( Figure 1 B). Analysis of the concentration of the biomarker in the corresponding ROIs showed that the concentration of the BM was reduced in the ROI where the drug was concentrated compared to the ROI where the drug was almost absent ( Figure 1 C) The method of the invention allows to distinguish different parts of the same tissue section, whereas global analysis using standard methods (LCMS) only shows the global impact of the drug in the whole tissue section.

[0101] Figure 4 An embodiment is shown in which the method of the invention is used to screen drugs. The expected effects (ie, reduction in biomarker concentration) associated with two different candidates (Drug A compared to Drug B) are analyzed based on the drug levels in ROI1 and ROI2.

[0102] Figure 5 An embodiment is shown in which a biomarker (BM) associated with a drug is known. The method of the invention is performed to determine the dose effect of the drug from a sample containing analytical replicates and avoiding biological bias. Analysis of the concentration of the biomarker in different pixels shows the effect of the concentration of the drug in the tissue on the concentration of the biomarker ( Figure 5 B). The method of the invention allows determination of the dose of a drug required to obtain a desired effect or even the median effective dose (the dose required to achieve 50% of the desired response). Figure 5 As shown in B, for an efficacy of 50%, drug B is more effective than drug A. Conversely, for an efficacy of 25%, drug A is more effective than drug B.

[0103] According to another embodiment, analysis of the ROI is performed to identify biomarkers associated with the molecule of interest.

[0104] Mass spectrometry provides molecular information about the biological impact of the molecule of interest as an increase or decrease in the concentration of lipids, metabolites, peptides or proteins in a living system. Other techniques (IF, IHC, ELISA, ISH, spectrophotometry or UV fluorescence) allow the evaluation and quantification of peptides, proteins, RNA (ribonucleic acid) and DNA (deoxyribonucleic acid) targeted using labeled antibodies or nucleic acid sequences that bind to the target. All or part of these embodiments can be combined to show the effects of different molecules of interest on the same biological sample and / or the effects of molecules of interest at different concentrations and / or the effects of molecules on different biological samples of interest, etc.

[0105] According to the present invention, multiplex imaging techniques (e.g., mass spectrometry) can be used to normalize the molecular information using endogenous compounds to segment images associated with the molecules of interest. This may help minimize variability. In this case, an external or internal standard compound is used as a reference compound.

[0106] application

[0107] The following examples briefly disclose different applications of the method of the present invention. After administering a drug to at least one biological sample (e.g., in vitro test, in vivo animal), imaging techniques are used to detect and locate the drug. A molecular profile of the sample is obtained for the drug. The molecular profile is segmented based on intensity and two regions of interest (ROIs) are selected.

[0108] 1- Use different analytical methods to identify molecular changes (e.g., use mass spectrometry or mass spectrometry imaging methods to target metabolites, DNA, RNA, proteins, lipids, peptides, metals, or use other imaging techniques ISH, IHC, MRI, PET imaging to compare the two regions of interest). Then, identify the molecules that are increased or decreased where the drug is located. The results related to the drug can be scored and the effect of the drug on certain cells or organ substructures can be confirmed.

[0109] 2- Extract the ROI manually or automatically (using a laser microdissection instrument) and use analytical methods to identify molecular changes. Use bioanalytical methods such as mass spectrometry or other techniques such as PCR to compare the contents of the two regions of interest. Results related to the drug can be scored and the effect of the drug on certain cells or organ substructures can be confirmed.

[0110] 3- Use different analytical methods to identify molecular changes, such as using mass spectrometry or mass spectrometry imaging to target metabolites, DNA, RNA, proteins, lipids, peptides, metals, or other imaging techniques ISH, IHC, MRI, PET imaging to compare the two regions of interest. Identify molecules that are increased or decreased where the drug is located. Results for non-targeted compounds can be scored and confirm the effect of the drug on certain cells or organ substructures.

[0111] 4-Use two different drugs in two biological samples. Select two ROIs, which correspond to the location of the corresponding compounds in the corresponding biological samples. Use different analytical methods to identify molecular changes, such as using proteomics, metallomics, transcriptomics, genomics and metabolomics (including lipidomics) methods to compare the two regions of interest. Then, identify the molecules that increase or decrease where the drug is located. The results of targeted or non-targeted compounds can be scored and the dose response (efficacy or toxicity) in the specific area where the drug is located can be evaluated.

[0112] 5- Use two different drugs in the same biological sample. Using the same process of ROI selection and comparison using analysis tools, select two ROIs. Use different analysis methods to study the differences in molecular expression. It is also possible to select ROIs where two drugs are located and study their effects on molecular changes and possible synergistic effects.

[0113] 6- More ROIs can be generated as image pixels in order to determine the dose effect of a drug based on a range of drug concentrations from one sample containing analysis replicates and avoiding biological bias. The analysis of the concentration of the biomarker within the different pixels shows the effect of the concentration of the drug in the tissue on the concentration of the biomarker. The method of the present invention allows the determination of the dose of the drug required to obtain the desired effect or even the median effective dose (the dose required to achieve 50% of the desired response).

[0114] 7- Expose the biological sample in vitro with a drug that will penetrate into the cells and induce molecular interactions and physiological changes. Using the same process of ROI selection and comparison using analysis tools, select potentially different ROIs corresponding to different cell phenotypes and analyze the corresponding response to the drug (mutagenicity, toxicity, etc.). Example

[0115] Indoleamine-2,3-dioxygenase (IDO1) is an enzyme that converts tryptophan (Trp) to kynurenine (Kyn). IDO1 is one of the first targets for small molecule drug discovery in the field of immuno-oncology (IO) due to its key role in tumor immune escape by reducing Trp and increasing Kyn levels in the microenvironment. Potent and selective IDO1 inhibitors such as Epacadostat (EPA) have been shown to enhance anti-tumor activity by restoring immune system health.

[0116] Quantitative mass spectrometry imaging (QMSI) was used in this study, with the goal not only to quantify metabolites and drugs, but also to reach a higher level of information through their histological localization and quantification. Quantitative ex vivo studies are reported here, which allow highlighting the exposure and pharmacological effect efficacy of Epacadostat (EPA) targets in specific tumor areas. Since exposure at the site of action and exposure to specific targets are identified as the most important factors for successful drug development, the purpose of this invention is to explore the target exposure of IDO1 inhibitors and their intratumoral pharmacodynamic effects on tumor metabolism.

[0117] 1. Materials and Methods

[0118] 1.1 Chemicals and reagents

[0119] All chemicals including 1,5-diaminonaphthalene (1,5-DAN), 2,5-dihydroxybenzoic acid (2,5-DHB), kynurenine-d4, tryptophan e-d5, formic acid, and trifluoroacetic acid were purchased from Sigma-Aldrich (St. Louis, MO). 13 C was purchased from Alsachim. Methanol, acetonitrile, and Optima LC-MS water were purchased from Fisher Scientific (Waltham, MA). Indium tin oxide (ITO)-coated glass slides were purchased from Bruker Daltonics (Bremen, Germany).

[0120] Rabbit anti-mouse IgG IDO primary antibody [#106134] and goat anti-rabbit IgG H&L (HRP) secondary antibody [#205718] and detection system by horseradish peroxidase followed by Steady DAB / plus (brown chromogen) were purchased from AbCam (Cambridge, UK).

[0121] 1.2 Sample collection and tissue preparation

[0122] Colon cancer CT26 cell line was subcutaneously transplanted into BALB / c mice (Charles River Laboratories, France). Mice were randomly divided into groups and the tumor size was 70 to 120 mm. 3 Treatment was started when the average size of the tumor was . The mice were treated with 100 mg / kg of the IDO1 inhibitor Epacadostat (Syngene, India) by oral gavage and then killed 2 hours later. The tumors were sampled and flash frozen in liquid nitrogen for 15 s. The samples were kept at -80 ° C until use. A constant temperature box slicer (CM-3050S, Leica, Germany) was used to freeze the slicer chamber and sample clamp at -23 ° C to obtain 10 micron thick tissue sections. The tissue sections were melted and mounted on ITO-coated slides for downstream MALDI imaging, and the continuous sections on the SuperFrost slides were used for histological and immunohistochemical (IHC) analysis. Two biological parallel samples (long-distance continuous sections) were analyzed to ensure the repeatability of the analysis. For LC-MS / MS analysis, 5 10 μm thick tissue sections were harvested for the preparation of calibration curves and the quantification of EPA, Kyn and Trp. All animal experiments complied with the European Directive 2010 / 63 / UE on the welfare of experimental animals and were approved by the ethics committee.

[0123] 1.3 Qualitative and quantitative MALDI analysis of epacadostat

[0124] The EPA calibration curve was measured in two steps (large range, then limited range). Then, the lower limit of detection (LOD), lower limit of quantification (LLOQ) and linear parameter limits were determined for the calibration curve made on 10 μm thick control tumor tissue sections using at least 10 points of concentration. The concentration range of EPA was 125 to 1 pmol, which was validated as a linear regression model. The calibration curve and tissue sections of interest (1 control and 3 treated in duplicate) were deposited on the same ITO slide and analyzed. Then, a uniform layer of 10 mg / mL of filtered 1,5-diaminonaphthalene (1,5-DAN) prepared using 50 / 50 ACN / H2O matrix was deposited on the tumor tissue sections using an HTX-TM spray device (HTX Technologies, LLC, Carrboro, NC). MALDI MSI analysis was performed using a 7TMALDI-FTICR (SolariX XR, Bruker Daltonics, Bremen, Germany) with a SmartBeam II laser. MSI data of Epacadostat were recorded in CASI negative ion mode (m / z range 436.0 + / - 30) with a spatial resolution of 120 μm using online calibration. Data acquisition was performed using Flex software (FtmsControl 2.1.0) from BrukerDaltonics. The calculated equations were then obtained using Multimaging software (ImaBiotech SAS) and extracted in pmol / mm 2 The conversion of the quantities to μg / g tissue was then obtained in the generated region of interest (ROI).

[0125] 1.4 Qualitative and quantitative MALDI analysis of Trp and Kyn

[0126] Since Kyn was undetectable and Trp was marginally visible in the CT26 tumor model using classical methods, a derivatization strategy was necessary. The derivatization reaction is equivalent to adding C to the neutral analyte. 11 H 16N+ (as X) units. The reagent was applied using an automated sprayer (Suncollect, Sunchrom) and left to incubate at room temperature for 30 min. A uniform layer of a mixture of 40 mg / mL 2,5-dihydroxybenzoic acid (2,5-DHB) and 0.5 μM Kyn-d4-derived internal standards prepared using a 50 / 50 MeOH / H2O+1% TFA matrix was then sprayed on the tumor tissue sections and calibration curve using an HTX-TM spray device (HTX Technologies, LLC, Carrboro, NC). Using isotopically modified compounds, the tissue inhibitory effect should allow the same inhibitory effect to be obtained if present in the tissue. Note that the use of deuterated standards (Trp-d5 and Kyn- 13 C) perform the calibration curve to avoid the internal interference of endogenous compounds. Then, use the 7T MALDI-FTICR (SolariX XR, Bruker Daltonics, Bremen, Germany) with SmartBeam II laser to carry out MALDIMSI analysis. The MSI data of kynurenine and tryptophan are recorded with a spatial resolution of 120 μm using online calibration in positive ion mode (CASI, m / z range is 350.0+ / -150). Data acquisition is carried out using the Flex software (FtmsControl 2.1.0) from Bruker Daltonics.

[0127] 1.5 MALDI-FTICR imaging of other metabolites

[0128] For MALDI MSI of other metabolites, a uniform layer of 10 mg / mL filtered 1,5-diaminonaphthalene (1,5-DAN) prepared with 50 / 50 ACN / H2O matrix was deposited on the tumor tissue sections using an HTX-TM spray device (HTX Technologies, LLC, Carrboro, NC). Then, MALDI MSI analysis was performed using a 7TMALDI-FTICR (SolariX XR, Bruker Daltonics, Bremen, Germany) with a SmartBeam II laser. MSI data for metabolites were recorded with a spatial resolution of 120 μm using online calibration in full scan negative ion mode (m / z range 100–1000). Data acquisition was performed using Flex software (FtmsControl 2.1.0) from Bruker Daltonics.

[0129] 1.6 LC-MS / MS analysis

[0130] The calibration curve (0 to 500nM) was prepared in water containing 5nM internal standards (Kyn-d4 and Trp-d5). Then, continuous tumor sections between 0.5 and 2mg were collected in a methanol / water extraction solution containing 10nM internal standards. Overnight stirring extraction was performed at 4°C, followed by centrifugation at 3000×g for 15min at 4°C, allowing supernatants to be recovered from both Kyn and Trp calibration curves and all sections for LC-MS / MS analysis. For tumors, dilution was performed in water by 1 / 2 before analysis. A total of 5 μL was injected into the LC-MS / MS system. No additional filtration step was required. The LC-MS / MS system consisted of an ultra-high performance liquid chromatography-focusing + LC system, and the LC system consisted of an RS column oven, an RS pump, and an RS autosampler (Dionex Ultimate 3000, Thermo Fisher Scientific), containing a Waters column (Cortecs C18 75×3 mm, particle size=2.7 μ) coupled to a TSQ Quantiva ThermoScientific (Thermo Fisher Scientific). Data acquisition and processing were performed using TSQ Quantiva software 2.0 version 1292 and Xcalibur 3.0 software.

[0131] 1.7 MSI data processing and analysis

[0132] 1.2 software (ImaBiotech SAS). This multimodal imaging platform combines quantitative mass spectrometry imaging (QMSI) and microscope platform with statistical analysis for understanding omics information at the cellular level. MSI data are obtained from each tissue section and the matrix control area adjacent to the tissue section to check the dispersion / displacement of analytes during sample preparation. Therefore, the ROI related to the presence of EPA and / or Kyn is given by an image segmentation algorithm. First, the algorithm divides the sample into different categories (in the case of this study, 2 categories or ROIs for both EPA and Kyn) based on the molecular signal threshold. Then, the algorithm smoothes the two ROIs to convert them into connected spaces. Finally, using Multimaging software, the following formula is used to calculate the exposure score:

[0133]

[0134] Where N ROI is the number of pixels in the ROI, N 总 is the number of pixels in the entire sample.

[0135] 1.8 Immunohistochemical analysis, digital scanned images and high-resolution overlays

[0136] Serial sections were stained for IDO1 histological localization using a rabbit polyclonal antibody purchased from Abcam (Cambridge, UK) and adapted for fresh frozen tissue sections. Sections were first exposed to 0.5% Triton-X for 15 min at room temperature and washed with phosphate-buffered saline, then anti-IDO1 primary antibody (1:50 dilution) was added and treated with goat anti-rabbit IgG H&L (HRP) secondary antibody (1:2000). The detection system was by horseradish peroxidase followed by steady-state DAB / plus (brown chromogen).

[0137] After MSI data acquisition, any residual matrix was removed using 100% methanol washing, and the tissue samples were then stained with hematoxylin and eosin (H&E) solution. High-resolution histological images of H&E staining or IHC were then recorded using a digital slide scanner (3D Histech Pannoramic) and then loaded into Multimaging technology for high-definition overlays using complex molecular images.

[0138] 2. Results

[0139] 2.1 Epacadostat drug detection, quantification and histological localization

[0140] QMSI and LC-MS / MS analysis were performed to quantify the absolute levels of Epacadostat in CT26 tissue, plasma, and blood samples. First, sample preparation and method development were performed for EPA QMSI analysis. A calibration curve was spotted on control CT26 tissue sections, allowing the acquisition of a calibration curve showing a linear range of 12 to 870 μg / g (R 2 =0.996), the limit of detection (LOD) was 12 μg / g, and the lower limit of quantification (LLOQ) was 15 μg / g ( Figure 7 A). Then, molecular images are shown in duplicate in one control and three treated sections each, which show the histological localization of the first EPA isotope (m / z 435.9844). QMSI was then performed using the resulting calibration curve, giving amounts between 38 and 53 μg / g for the three biological replicates ( Figure 8 A). LC-MS / MS quantification was also performed on CT26 serial tissue sections. Figure 7 B shows the calibration curve of EPA obtained in nM. When using LC-MS / MS versus QMSI, the average EPA amount was 38.5 μg / g versus 46 μg / g, respectively. Figure 8 B), showing a 17% variation between the two techniques. EPA was also quantified in plasma and blood ( Figure 8 C).

[0141] 2.2 Target exposure analysis

[0142] For target exposure analysis, first, automatic segmentation of the EPA signal highlighted two ROIs (1 and 2). Then the absolute EPA quantification was evaluated (using QMSI) and automatically extracted from three regions (ROI 1 for high EPA, ROI 2 for low EPA, and ROI 3 for the entire treated tumor). The results showed an amount of 68 μg / g in ROI 1, 25 μg / g in ROI 2, and 42 μg / g in the entire section (ROI 3); this means that almost 61% of the EPA signal (26 μg / g) was concentrated in 38% of the entire tissue section (ROI 1), and 39% (15 μg / g) of the signal was located in the remaining 62% of the tumor area (ROI2) ( Fig. 9 A). Then, the intensity of IDO1 enzyme expression was highlighted by immunostaining. Thus, the semi-quantitative analysis allowed the distinction of two areas showing different IDO1 expression levels, from the highest expression (1) to the lowest expression (2) ( Fig. 9 B). Interestingly, a positive correlation was also noted between IDO1 expression levels (semi-quantitatively assessed by histology) and EPA levels.

[0143] 2.3 Pharmacodynamic studies of Kyn and Trp metabolites

[0144] For endogenous metabolites, the optimized derivatization step allowed the detection and quantification of both Trp and Kyn while improving their detection sensitivity. MSI analysis was used to show the histological localization of Epacadostat, Trp, and Kyn on CT26 tumors ( Fig.10 A). Molecular images showed higher Kyn levels and lower Trp levels on control CT26 tissues compared to treated tissues. Subsequently, for QMSI, calibration curves for both Trp and Kyn were normalized using the same derivatization strategy ( Fig.11 A). Absolute quantification of both Kyn and Trp using QMSI and LC-MS / MS analysis on control and treated CT26 tissue sections, and the Kyn / Trp ratio was then calculated and plotted on Fig. 9 B, which showed a good correlation between the two techniques. It was noted that the Kyn / Trp ratio decreased after EPA treatment, and a 6-fold decrease was found regardless of the technique used. Plasma and blood samples were also analyzed, and the results showed a 3-fold decrease in the Kyn / Trp ratio during treatment ( Fig.11 B).

[0145] 2.4 From target exposure to response efficacy analysis: regional / pharmacological effects of EPA in the CT26 tumor mouse model

[0146] The pharmacological efficacy of EPA drugs was followed by absolute quantification of Kyn metabolites, which are direct enzymatic products of IDO1 enzyme. The absolute Kyn amounts in the three ROIs showed that 15% of Kyn was present in 38% of the whole tumor (ROI 1), and 85% of Kyn was present in ROI 2 ( Fig.12 A). Finally, the histological distribution of lactate metabolites together with EPA and Kyn is shown in Fig.12 In B, the relative intensities of all molecules extracted from each x and y position of the high and low EPA regions (1 and 2, respectively) are then plotted. A negative correlation between EPA, Kyn, and lactate metabolites is found, as high EPA levels on ROI 1 correspond to low expression levels of both Kyn and lactate ( Fig.12 C).

[0147] Results Analysis

[0148] The results showed that the amounts in plasma and blood were between 6.6+ / -1μg / g and 7.3+ / -2μg / g, respectively. Figure 7 C). Then, using LC-MS / MS and QMSI on treated CT26 tumors, the available amount of EPA on tissue was 3 or 4 times higher than plasma (between 35 and 48 μg / g), with less than 20% variation between the two techniques. Correlating with the EPA inhibitory effect, a dose-dependent decrease in kynurenine levels was noted in plasma. In this experiment, >50% inhibition was observed for approximately 24 hours, with a dose of 100 mg / Kg over the course of the entire day.

[0149] Our final study showed that Kyn levels decreased from 25-34 μg / g (P815 high IDO1) to <60 ng / g (P815 low IDO1) in the P815 mouse model.

[0150] In this study, it is known that CT26 murine colon cancer cells express IDO1, so it is used to determine the effect of IDO1 inhibition on tumor growth. In fact, the CT26 model has been well used for the pharmacological evaluation of IDO1 inhibitors. Historically, the abundance and distribution of drugs have been evaluated by established techniques such as quantitative whole body autoradiography (QWBA) or tissue homogenate and LC-MS / MS analysis. However, QWBA cannot distinguish active drugs from their metabolites, and LC-MS / MS, although highly sensitive, cannot report spatial distribution. Mass spectrometry imaging (MSI) can distinguish drugs and their metabolites from endogenous compounds, and simultaneously report their distribution. MSI data is affecting drug development, and is currently used for investigative studies such as DMPK, PD and toxicity fields.

[0151] This study shows that the use of QMSI technology has a significant impact on target exposure studies. When comparing control and treated CT26 tumors, specific regions were segmented that correlated with the amount of EPA contained within. Tumor exposure to EPA was thus confirmed and two distinguishable regions were then extracted (1 for high EPA and 2 for low EPA). Almost 61% of the EPA drug, corresponding to 68 μg / g, was located in 38% of the entire tumor. Semi-quantitative analysis of the IDO1 enzyme showed high expression of IDO1 in ROI 1 compared to ROI 2, supporting exposure of EPA to its IDO1 target.

[0152] Since the basal levels of Kyn and Trp are quite low, a derivatization step was initiated to allow tissue localization of endogenous metabolites. Since MALDI MSI is used for multiplex detection of diverse analytes in a wide mass range, analyte coverage is highly limited to more abundant compounds. Chemical derivatization of analytes on or in tissues solves these problems by selectively labeling functional groups specific to the analyte class, while simultaneously changing their molecular mass and improving their ionization efficiency and therefore their sensitivity. Therefore, the sensitivity of Trp and Kyn is improved, allowing them to be distributed and quantitatively studied on tissues. Several studies have shown that serum from cancer patients has a higher Kyn / Trp ratio than serum from normal volunteers, consistent with increased IDO1 activity. Using both QMSI and LC-MS / MS, our data showed that the Kyn / Trp ratio was reduced by 6 times and 3 times, respectively, when the control was compared with the plasma and blood of the treated CT26 tissue ( Fig.11 B).

[0153] Likewise, the Kyn / Trp ratio has recently been validated as a prognostic tool in a number of cancers: cervical cancer, glioblastoma, and lung cancer, where LC-MS / MS was used for metabolite quantification. Kyn expression was then extracted in comparison to areas where EPA was absent (control tumors), showing a regional correlation between the presence of EPA and its pharmacological effect on Kyn.

[0154] Finally, IDO1 enzyme immunostaining was achieved, allowing the observation of the histological localization of EPA targets. Beyond the substrate and product changes associated with IDO1 inhibition, cancer cells exhibit metabolic changes that distinguish them from healthy tissues and sensitize their metabolic processes to pharmacological targeting. Cellular metabolites have a wide variety of physicochemical properties, so their detection and quantification require a combination of various analytical methods. The presence of specific metabolites may be indicative of the metabolic state of the tumor; however, it is not always possible to directly infer the activity of specific metabolic pathways from measurements of metabolite abundance alone.

[0155] The results of this study show histological counter-localization between lactate and EPA molecules. Since lactate is the end product of the glycolytic pathway, its reduction in the same histological region where EPA is highly concentrated corresponds to a significant reduction in glycolysis. Assessment of regional levels of metabolites such as lactate and glucose was performed using quantitative bioluminescence imaging of snap-frozen biopsy samples for cerebral ischemia.

[0156] In a variety of different tumors and cancer types, lactate has been shown to be a prognostic indicator, as its elevated levels have been associated with worse patient outcomes, poor disease-free or metastasis-free survival, and poor overall survival in human cervical cancer, head and neck cancer, high-grade glioma, and non-small cell lung cancer. This feature makes lactate metabolism of interest for further investigation, not only as a biomarker, but also as a potential therapeutic endpoint or target.

[0157] Subsequently, by inhibiting IDO1 and reducing Kyn in tumor cells, EPA restored the proliferation, activation and regulation of various different cells. In fact, this will support the pharmacological effect of EPA. In addition, MALDI MSI was used to highlight the TCA cycle and energy pathway metabolism, showing different histological localization and relative intensity (data not shown). The presence of a group of specific metabolites can indicate the metabolic state of the tumor. Extracellular tumor microenvironment characterization was achieved, showing the characteristics of glycolysis such as lactic acid, glucose, malic acid, citric acid, glutamine and proline. It is common that in tumors, a higher proportion of glucose carbon is transferred from mitochondria and converted into lactic acid. In addition, glutamine metabolism and functions related to proline synthesis are also shown. Finally, the degradation of purine nucleotides is also shown.

[0158] 3. Conclusion

[0159] In conclusion, these results report a quantitative ultra-high mass resolution study that allowed to highlight in situ PK / PD studies of EPA. In fact, different studies were achieved:

[0160] ■Biodistribution and bioavailability studies by quantifying drug in tissues.

[0161] ■ Targeted tissue exposure studies after tissue segmentation based on biomarker and / or drug histological localization.

[0162] ■ Pharmacodynamic studies were also performed which showed regional exposure of the drug to the response effect, which enabled efficacy analysis.

[0163] Indeed, since exposure at the site of action and to its specific target has been identified as the most important factor for success in drug discovery and chemical probe design, these results demonstrate and confirm the tremendous contribution of QMSI, and more generally MSI, to studying the relationship between target occupancy and drug efficacy.

Claims

1. A method for evaluating ex vivo or in vitro the effect of at least one molecule of interest on at least one molecular marker in a biological sample for administration, the method comprising: - selecting a biological sample for administration, which has previously been exposed to said at least one molecule of interest; - using a molecular imaging method to detect the presence of the at least one molecule of interest in the biological sample for medication, and obtaining a molecular profile of the at least one molecule of interest in the biological sample for medication; - spatially segmenting the molecular map of the biological sample for medication according to the spectral information to obtain a segmented map of the at least one molecule of interest in the biological sample for medication; - selecting a first region of interest ROI from the segmented map, the first ROI having a first intensity of the at least one molecule of interest; - measuring the intensity or amount of at least one molecular marker in the first ROI; - comparing the intensity or amount of at least one molecular marker in the first ROI with the intensity or amount of at least one molecular marker in the second ROI, wherein selecting a second ROI from the segmented map of the medicated biological sample, the second ROI being physically different from the first ROI and having a second intensity of the molecule of interest, Wherein the molecule of interest is a drug candidate molecule.

2. The method of claim 1, wherein the biological sample is selected from the group consisting of a tissue sample and a biological fluid sample. The method according to claim 1 , wherein the biological sample is a cell suspension.

4. The method of claim 2, wherein the biological fluid sample is selected from the group consisting of a urine sample, a plasma sample, and cerebrospinal fluid.

5. The method according to any one of claims 1 to 4, wherein the administered biological sample is a tissue section obtained by sampling an animal previously administered with the molecule of interest, or wherein the administered biological sample has been contacted with the molecule of interest in vitro.

6. The method according to any one of claims 1 to 4, wherein the molecular imaging method is selected from MRI imaging, PET imaging, CT imaging, IF, ISH, IHC, mass spectrometry and cytometry imaging.

7. The method according to claim 6, wherein the molecular imaging method is mass spectrometry imaging selected from MALDI, DESI, LESA, LA-ICP-MS and SIMS.

8. The method according to any one of claims 1 to 4, wherein the administered biological sample is previously obtained by sampling an animal previously administered with the candidate molecule.

9. The method according to any one of claims 1 to 4, wherein the measurement of the amount or intensity of the molecular marker in the first ROI and / or the second ROI is performed using a molecular imaging method, and / or wherein the measurement of the amount or intensity of the molecular marker in the first ROI and / or the second ROI is performed by a bioanalysis method.

10. The method of claim 9, wherein the molecular imaging method is selected from the group consisting of MSI, ISH, IF, IHC, MRI, imaging mass cytometry, CT, and PET imaging.

11. The method of claim 9, wherein the bioanalytical method is selected from the group consisting of mass spectrometry, electrophoresis, ligand binding assay, nuclear magnetic resonance, microdialysis, chromatography, magnetic bead multiplex immunoassay, and ELISA.

12. The method according to any one of claims 1 to 4, wherein the ROI is extracted from the biological sample by laser capture microdissection.

13. The method of claim 9, wherein the ROI is cultured in culture medium prior to performing the bioassay.

14. The method of claim 12, wherein the impact of the molecule of interest is assessed by comparing changes in gene or transcript expression, lipidomics, peptidomics, proteomics and / or metabolomics between the first ROI and the second ROI.

15. The method of any one of claims 1 to 4, wherein the molecule of interest is a therapeutic antibody, and wherein the presence of the therapeutic antibody is detected by contacting the biological sample with a labeled anti-therapeutic antibody antibody.

Images