Method and devices for immunoconjugate screening

Abstract

The present invention relates to the field of preclinical testing in drug development and research. In particular, the invention relates to the field of immunoconjugates design, including antibody drug conjugates (ADC) design, and to methods for identifying high efficacy immunoconjugates.

Description

METHOD AND DEVICES FOR IMMUNOCONJUGATE SCREENING TECHNICAL FIELD

[0001] The present disclosure relates to the field of preclinical testing in drug development and research. In particular, the invention relates to the field of immunoconjugates design, including antibody drug conjugates (ADC) design, and to methods for identifying high efficacy immunoconjugates.

[0002] Yet more specifically, the present disclosure relates to a method, device and program for the screening of immunoconjugates, and / or for determining an efficacy of input immunoconjugates, and / or for predicting a response of a patient or a population of patients to such immunoconjugates.BACKGROUND

[0003] Drug’s efficacy assessment has developed significantly in the context of personalized medicine, for example precision oncology. There are some existing solutions to predict individual patient responses to a given drug, or alternatively to predict the efficacy of a drug toward novel indications.

[0004] These examples of the omics-based approach have been directed to targeted therapies, for example for predicting the efficacy of immunotherapies, and for the efficacy of chemotherapies (e.g., transcriptomic markers of gemcitabine sensitivity in pancreatic ductal adenocarcinoma (PDAC)).

[0005] Clinical trials are one of the major barriers in contemporary drug development. Functional assays based on patient-derived organoids (PDO) are a promising new tool for derisking clinical development of new therapies, but their use has been limited due to small cohort sizes and the absence of systematic validation studies.

[0006] Despite the current developments, it is thus still a challenge to predict the response rate of a patient or a group of patients. This is particularly the case for multivariate response predictions, i.e., when the prediction is realized using a plurality of variables to forecast possible outcomes. Current methods are each optimised for one or a set of drugs and are generally tumour-type specific. Furthermore, current approaches do not generalise to drugs in development, i.e., to predict responses to drugs that have been never given to patients in the clinic. In particular, ADC designers must worry about the multiple potential points of failure in the design of a new drug. The efficacy of the ADC may depend on the payload efficacy, the linker strength (when present) and the target antigen abundance.

[0007] Fang et al. (“The target atlas for antibody-drug conjugates across solid cancers”. Cancer Gene Therapy 31, 273-284. 2024) provides an algorithm to identify ADC targets, based on the selection of surface membrane proteins as target antigens, and through the combination of datasets of solid cancers and normal tissues.

[0008] Kathad et al. (“Expanding the repertoire of Antibody Drug Conjugate (ADC) targets with improved tumor selectivity and range of potent payloads through in-silico analysis”. PLoS ONE 19(8). 2024) proposes to combine transcriptomics, proteomics, immunohistochemistry and cell surface membrane dataset for evidence-based filtering and identification of ADC targets.

[0009] The use of organoids for functional precision medicine is not unheard of, especially in the field of oncology, as illustrated in Boileve et al. (“Organoids for Functional Precision Medicine in Advanced Pancreatic Cancer”. 2024. Gastroenterology; 167:961-976). Also, methods to generate lists of target antigens and payload inputs already exist. Yet, none of the methods can match the expression levels of antigens with the sensitivity to payloads at the individual payload level; let alone for determining the efficacy of immunoconjugates.

[0010] Therefore, the available methods and databases do not necessarily take into account, alone, the specificities of each sub-population of patients, nor the synergy which may occur for certain combinations of payload and antigen-binding sites; let alone in a complex environment, such as a solid tissue and / or an organ.

[0011] Also, the available methods for screening of immunoconjugates are not entirely agnostic. They generally tend to limit the number of tested payloads or tested antigens to the narrow selection of pay load inputs for which proof of efficacy is already demonstrated individually; for example, through clinical trials.

[0012] Within this context, there is a need for improved methods for predicting the efficacy of immunoconjugates. In particular, there remains a need for unbiased and / or agnostic methods for predicting the efficacy of immunoconjugates in vivo in particular with respect to solid cancers. In particular, there is still a need for improved methods for optimizing immunoconjugates, and methods which remain applicable to high-throughput analysis.

[0013] There is also a need for methods for predicting the efficacy of input immunoconjugates, which are capable of identifying immunoconjugates with an IC50 that is compatible with drug development.

[0014] There is also a need for identifying new sub-populations of patients which may benefit from previously known immunoconjugates.

[0015] The invention has for purpose to meet the above-mentioned needs.SUMMARY

[0016] The invention first relates on a computer-implemented method for determining the efficacy of an input immunoconjugate (10) with respect to PDOs comprising at least an input antigen-binding site (11) and an input payload (12); the method comprising the following steps:- providing organoid response data (21) indicative of efficacies of the input payload (12) on a respective plurality of patient-derived organoids, PDOs, of a patient population, and derived by bringing the input payload (12) into contact with the respective PDOs;- providing expression data (20) indicative of levels of expression of antigens expressed by organoid-derived cells of the respective PDOs; and wherein the antigens are capable of binding to the input antigen-binding site (11); determining the efficacy of the input immunoconjugate (10) based on the organoid response data (21) and the expression data (20).

[0017] The computer-implemented method as defined herein above, wherein the efficacy of the input immunoconjugate (10) is compared with the efficacy of a reference immunoconjugate (14).

[0018] The computer-implemented method as defined herein above, wherein the immunoconjugates (10) also comprises at least an input linker (13), and the method comprises providing linker data (23), indicative of efficacies of the input linker (13).

[0019] The computer-implemented method as defined herein above, further comprising providing internalization rate data and / or recycling rate data (22) indicative of the efficacies of the input antigen-binding site (11).

[0020] The computer-implemented method as defined herein above, wherein the step of determining the efficacy further comprises determining the input immunoconjugate (10) as efficient when at least a predetermined portion (34) of the patient population shows respective levels of expression above a minimum reference level of expression (35) and shows an efficacy of the input pay load (12) above a minimum efficacy threshold (36).

[0021] The computer-implemented method as defined herein above, wherein the efficacy of the input immunoconjugate (10) is measured by bringing the input immunoconjugate (10) into contact with the respective PDOs and testing the level of expression of the input antigen and the level of viability of the input pay load (12).

[0022] The computer-implemented method as defined herein above further comprising:providing a data store comprising expression data (20), organoid response data (21), and subject data associated with PDOs from subjects;- selecting the patient population as a subset of the PDOs that are affected by the same or similar tumour.

[0023] The computer-implemented method as defined herein above, wherein the patient-derived organoids (PDO) are selected from a group consisting of cerebral organoids, gastrointestinal organoids, lingual organoids, tooth organoids, thyroid organoids, thymic organoids, testicular organoids, prostate organoids, hepatic organoids, pancreatic organoids, epithelial organoid, lung organoids, kidney organoids, gastruloid organoids, blastoid organoids, endometrial organoids, cardiac organoids, retinal organoids, breast cancer organoids, colorectal cancer organoids, glioblastoma organoids, neuroendocrine tumor organoids, myelin organoids, blood-brain barrier (BBB) organoids, and ovarian organoids.

[0024] The computer-implemented method as defined herein above, wherein the levels of expression is measured on antigens expressed at a surface of the organoid-derived cells; and, for example, wherein the level of expression of antigens is determined by a method selected from the group consisting of immunohistochemistry, mass cytometry, and combinations thereof.

[0025] The computer-implemented method as defined herein above, wherein the expression data (20) comprises at least one of the following data selected from the group consisting of aproteomic data, atranscriptomic data, a genomic data, an epigenomic data, a metabolic data, a microbiome data, an imaging data, a histological data, a cytotoxic data, a cytostatis data, and a combination thereof.

[0026] The computer-implemented method as defined herein above, further comprises providing reference expression data (24) indicative of levels of expression of the antigens expressed by reference non tumour cells.

[0027] The computer-implemented method as defined herein above, wherein the organoid response data (21) comprises cytotoxicity data.

[0028] The computer-implemented method as defined herein above, wherein the input immunoconjugate (10) is recovered or manufactured completely or in part.

[0029] The invention further relates to a computer-implemented method for selecting an optimal immunoconjugate from a set of input immunoconjugates, comprisingdetermining the efficacy of the set of input immunoconjugates according to the method described herein above,selecting the optimal immunoconjugate based on the efficacy of the set of input immunoconjugates.

[0030] The invention also relates to a computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the methods according to the present invention.

[0031] The invention also relates to a computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the methods according to the invention.

[0032] The invention also relates to a device for determining the efficacy of an input immunoconjugate (10) against PDOs comprising at least an input antigen-binding site (11) and an input payload (12); the device comprising:at least one input configured to receive:• an organoid response data (21) indicative of efficacies of the input payload (12) on a respective plurality of patient-derived organoids, PDOs, of a patient population, and derived by bringing the input payload (12) into contact with the respective PDOs;• an expression data (20) indicative of levels of expression of antigens expressed by organoid-derived cells of the respective PDOs; and wherein the antigens are capable of binding to the input antigen-binding site (11); at least one processor configured to:• determine the efficacy of the input immunoconjugate (10) against PDOs based on the organoid response data (20) and the expression data.At least one output configured to provide the efficacy of the input immunoconjugate (10) against PDOs.

[0033] According to a first main embodiment, the invention relates to a method for determining the efficacy of immunoconjugate(s), the immunoconjugate(s) comprising at least one antigen-binding site and at least one payload, the method comprising the steps ofa) providing biological data retrieved from a plurality of patient-derived organoids (PDO) brought into contact with the at least one payload, thereby defining an organoid response data indicative of the efficacy of the isolated payload; b) determining the efficacy of the immunoconjugate(s) based on (i) the organoid response data and (ii) a level of expression of one or more antigen(s) expressed by corresponding organoid-derived cells;wherein:- the antigen(s) expressed by the cells is / are capable of binding to the at least one antigen -binding site, and- the plurality of patient-derived organoids (PDO) is associated to a plurality of avatars, each avatar comprising at least one set of digital data corresponding to patient(s) from a population.

[0034] According to a second main embodiment, the invention relates to a device for predicting an efficacy of an immunoconjugate, or for comparing the efficacy of two immunoconjugate(s), the device comprising:at least one input configured to receive at least:o an avatar of a patient and respective data of organoid response for said patient, said avatar comprising biological data and digital data of said patient, said avatar and said respective data of organoid response defining at least one patient feature and at least one organoid response feature respectively;o a first ensemble of one or more trained learning models each configured to receive as input, with respect to said patient, one or moreof said at least one patient feature and one or more of said at least one organoid response feature, and to provide a first output vector; at least one processor configured to:o calculate the first output vector by providing said at least one patient feature and at least one organoid response feature as input to said first ensemble of one or more trained learning models; ando obtain a prediction of the efficacy of the immunoconjugate using the first output vector.

[0035] According to a third main embodiment, the invention relates to a computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method for determining the efficacy of immunoconjugate(s), and / or or for comparing the efficacy of two immunoconjugate(s).

[0036] According to a fourth main embodiment, the invention relates to a computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the method for determining the efficacy of immunoconjugate(s), and / or for comparing the efficacy of two immunoconjugate(s).DEFINITIONS

[0037] In the present invention, the following terms have the following meanings:

[0038] As used herein, the terms “two or more” and “plurality” are considered as synonymous, unless stated otherwise.

[0039] The term “about”, preceding a figure encompasses plus or minus 10%, or less, of the value of said figure. It is to be understood that the value to which the term “about” refers is itself also specifically, and preferably, disclosed.

[0040] As used herein, an “agent” may refer to any agent / compound / nucleic acid / polypeptide which is susceptible to be administered to, or brought into contact with, a patient / subject, or an organoid. Thus, the term may comprise or consist of a physical treatment (for example radiation therapy), a small molecule, a therapeutic compound, atherapeutic candidate, a polypeptide, an antibody or antigen-binding fragment thereof, an aptamer, a nucleic acid (for example a guide RNA, a silencing RNA, a micro-RNA, a coding RNA, a non-coding RNA), a nanoparticle, an expression vector, a virus, a cell, a prokaryotic cell (for example a bacteria), an eukaiy otic cell (for example an immune cell), combinations thereof, and / or compositions thereof Hence, the agent may be a therapeutic agent or non-therapeutic agent, i.e., it does not have the finality of treating a patient. Examples of therapeutic agents which are thus considered under this definition include, for example, any agent which is susceptible to treat or prevent or reduce the likelihood of occurrence or re-occurrence of a given condition. Hence, the term may, in particular, refer to any therapeutic agent, such as those selected from: an antibody or antigen-binding fragment thereof, an immunoconjugate, a cytotoxin, a chemotherapeutic agent, a cytokine, an immunosuppressant, an immune stimulator, a lytic peptide, a radioisotope, and antiviral agent, an antiparasitic agent, an antimicrobial agent, a chimeric antigen receptor, a nucleic acid, an expression vector and the like. Examples of cells may, in particular, include live cells and / or immune cells, and engineered forms thereof such as CAR T-cells or CAR-NK cells.

[0041] The terms “antibody” and “immunoglobulin” may be used interchangeably and refer to a protein having a combination of two heavy and two light chains whether or not it possesses any relevant specific immunoreactivity. “Antibodies” refers to such assemblies which have significant known specific immunoreactive activity to an antigen of interest. Antibodies and immunoglobulins comprise light and heavy chains, with or without an interchain covalent linkage between them. Basic immunoglobulin structures in vertebrate systems are relatively well understood. The generic term “immunoglobulin” comprises five distinct classes of antibody that can be distinguished biochemically. Although the following discussion will generally be directed to the IgG class of immunoglobulin molecules, all five classes of antibodies are within the scope of the present disclosure. With regard to IgG, immunoglobulins comprise two identical light polypeptide chains of molecular weight of about 23 kDa, and two identical heavy chains of molecular weight of about 53-70 kDa. The four chains are joined by disulfide bonds in a “Y” configuration wherein the light chains bracket the heavy chains starting at the mouth of the ‘Y” and continuing through the variable region. The light chains of anantibody are classified as either kappa (K) or lambda (X). Each heavy chain class may be bonded with either a K or X light chain. In general, the light and heavy chains are covalently bonded to each other, and the “tail” regions of the two heavy chains are bonded to each other by covalent disulfide linkages or non-covalent linkages when the immunoglobulins are generated either by hybridomas, B cells or genetically engineered host cells. In the heavy chain, the amino acid sequences run from an N -terminus at the forked ends of the Y configuration to the C -terminus at the bottom of each chain. Those skilled in the art will appreciate that heavy chains are classified as gamma (y), mu (p), alpha (a), delta (5) or epsilon (s) with some subclasses among them (e.g., yl-y4). It is the nature of this chain that determines the “class” of the antibody as IgG, IgM, IgA IgD or IgE, respectively. The immunoglobulin subclasses or “isotypes” (e.g., IgGl, IgG2, IgG3, IgG4, IgAl, etc.) are well characterized and are known to confer functional specialization. Modified versions of each of these classes and isotypes are readily discernable to the skilled artisan in view of the instant disclosure and, accordingly, are within the scope of the present invention. As indicated above, the variable region of an antibody allows the antibody to selectively recognize and specifically bind epitopes on antigens. That is, the light chain variable domain (VL domain) and heavy chain variable domain (VH domain) of an antibody combine to form the variable region that defines a three-dimensional antigen binding site. This quaternary antibody structure forms the antigen binding site presents at the end of each arm of the “Y”.

[0042] The term “antigen-binding fragment”, as used herein, refers to a part or region of the antibody according to the present disclosure, which comprises fewer amino acid residues than the whole antibody. An “antigen-binding fragment” binds antigen and / or competes with the whole antibody from which it was derived for antigen binding. Antibody antigen-binding fragments encompasses, without any limitation, single chain antibodies, Fv, Fab, Fab', Fab'-SH, F(ab)’2, Fd, defucosylated antibodies, diabodies, triabodies and tetrabodies.

[0043] The term “epitope” refers to a specific arrangement of amino acids located on a protein or proteins to which an antibody or antigen-binding fragment thereof or an antibody mimetic bind. Epitopes often consist of a chemically active surface grouping ofmolecules such as amino acids or sugar side chains, and have specific three-dimensional structural characteristics as well as specific charge characteristics. Epitopes can be linear (or sequential) or conformational, i.e., involving two or more sequences of amino acids in various regions of the antigen that may not necessarily be contiguous.

[0044] An antibody or antigen-binding fragment thereof is said to be “specific for”, “immunospecific” or to “specifically bind” an antigen if it reacts at a detectable level with said antigen, preferably with an affinity constant (KA) of greater than or equal to about 106M'1, preferably greater than or equal to about 107M'1, 108M'1, 5*108M'1, 109M'1, 5*109M'1or more. Affinity of an antibody or antigen-binding fragment thereof for its cognate antigen is also commonly expressed as an equilibrium dissociation constant (KD). An antibody or antigen-binding fragment thereof is said to be “immunospecific”, “specific for” or to “specifically bind” an antigen if it reacts at a detectable level with said antigen, preferably with a KD of less than or equal to IO'6M, preferably less than or equal to IO'7M, 5* IO'8M, 10'8M, 5* IO'9M, IO'9M or less. Affinities of antibodies or antigenbinding fragment thereof can be readily determined using conventional techniques, for example, those described by Scatchard, 1949. Ann NY Acad Sci. 51:660-672. Binding properties of an antibody or antigen-binding fragment thereof to antigens, cells or tissues may generally be determined and assessed using immunodetection methods including, for example, ELISA, immunofluorescence-based assays, such as immuno-histochemistry (IHC) and / or fluorescence-activated cell sorting (FACS) or by surface plasmon resonance (SPR, e.g., using BIAcore®).

[0045] A “chimeric antibody”, as used herein, refers to an antibody or antigen-binding fragment thereof comprising a first amino acid sequence linked to a second amino acid sequence with which it is not naturally linked in nature. The amino acid sequences may normally exist in separate proteins that are brought together in the fusion protein or they may normally exist in the same protein but are placed in a new arrangement in the fusion protein. A chimeric protein may be created, for example, by chemical synthesis, or by creating and translating a polynucleotide in which the peptide regions are encoded in the desired relationship. The term “chimeric antibody” encompasses herein antibodies and antigen-binding fragment thereof in which the constant region, or a portion thereof, isaltered, replaced or exchanged so that the variable region is linked to a constant region of a different or altered class, effector function and / or species, or an entirely different molecule which confers new properties to the chimeric antibody, e.g., an enzyme, toxin, hormone, growth factor, drug, etc.; or the variable region, or a portion thereof, is altered, replaced or exchanged with a variable region, or portion thereof, having a different or altered antigen specificity; or with corresponding sequences from another species or from another antibody class or subclass.

[0046] A “therapeutic antibody”, as used herein, refers to any antibody which is adapted to clinical care.

[0047] Accordingly, an “antigen-binding fragment from a therapeutic antibody” may comprise or consist of one or more antigen-binding sites from said therapeutic antibody.

[0048] The term “immunoconjugate” refers herein to an antibody or antigen-binding fragment, which is conjugated to a payload (e.g. a therapeutic moiety or a detectable moiety), e.g. a radioactive moiety and / or radiodetectable moiety and / or a drug / active agent. Such immunoconjugates, in the sense of the invention may thus comprise at least one antigen-binding site, at least one payload, and optionally at least one linker region (e.g. a linker region between the one or more antigen-binding site(s) and the one or more payload(s)).

[0049] The payload moiety may, in particular be a drug, such as those selected from a group consisting of a cytotoxin, a chemotherapeutic agent, a cytokine, an immunosuppressant, an immune stimulator, a lytic peptide, a nucleic acid, a radioisotope and / or a radiosensitizing agent. Such conjugates are referred to herein as an "antibodydrag conjugates" or "ADCs".

[0050] Accordingly, the term “ADC” may, in particular, refer to such antibodies or antigen-binding fragments conjugated to (either directly or indirectly through a linker) a payload, in particular a cytotoxic moiety.

[0051] The term “linker” typically refers to a linker unit comprised between the conjugate / drug unit and the antibody / antigen-binding fragment unit. Unless stated otherwise, the term shall encompass cleavable and non-cleavable linkers, and peptidic or non-peptidic linkers; for example linkers comprising from 1 to 100 amino acids, e.g. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, and 100 amino acids, or more than 100 amino acids.

[0052] The terms “adapted” and “configured” are used in the present disclosure as broadly encompassing initial configuration, later adaptation or complementation of the present device, or any combination thereof alike, whether effected through material or software means (including firmware).

[0053] The term “processor” should not be construed to be restricted to hardware capable of executing software, and refers in a general way to a processing device, which can for example include a computer, a microprocessor, an integrated circuit, or a programmable logic device (PLD). The processor may also encompass one or more Graphics Processing Units (GPU), whether exploited for computer graphics and image processing or other functions. Additionally, the instructions and / or data enabling to perform associated and / or resulting functionalities may be stored on any processor-readable medium such as, e.g., an integrated circuit, a hard disk, a CD (Compact Disc), an optical disc such as a DVD (Digital Versatile Disc), a RAM (Random- Access Memory) or a ROM (Read-Only Memory). Instructions may be notably stored in hardware, software, firmware or in any combination thereof.

[0054] “Avatar” designates a representation of a subj ect / patient. An avatar comprises at least one set of biological data corresponding to said patient and at least one set of digital data corresponding to said patient. As reported herein, an avatar of a patient is a representation of said patient. Said avatar and said respective data of organoid response define at least one patient feature and at least one organoid response feature, respectively. By a feature, it is meant a characterising parameter. Thereby, a patient feature may be oneor more parameters which defme / describe a health status of the patient. Similarly, an organoid response feature may be one or more parameters which defme / describe a response of an organoid. An organoid response feature may be equivalently called a “functional feature”.

[0055] The “biological data”, as in “biological data from the subject / patient”, may comprise or consist of data retrieved from all or part of a biological sample, or fraction thereof, for example from a patient or a group of patients. Accordingly, a biological sample may be any sample that may be taken from the subject, such as a serum sample, a plasma sample, a urine sample, a blood sample, a lymph sample, or a biopsy, and transformed samples such as patient-derived cells and / or patient-derived biological fluids, and fractions thereof. The biological data may have a time stamp which defines the time of the data collection. New biological data may be created over time with each additional experiment conducted on one or more organoids (i.e. patient-derived organoid(s) or patient-derived organoid cell line(s)). The new data may be added using the distinguishing time stamp or replace the existing biological data by updating the existing time stamp. In that context, the biological data may comprise or consist of one or more set(s) of data(s) obtained from PDOs or PDO cell lines, which can be associated to the avatar. The biological data may thus comprise at least one first group of biological data, obtained from the PDO or PDO cell lines. Said first group of biological data may include one of more organoid response feature; in particular selected from the group consisting of a proteomic feature, a genomic feature, an epigenomic feature, a transcriptomic feature, a metabolic feature, a microbiome feature, an imaging feature, a histological feature, a cytotoxic feature, a cytostatis feature. Optionally, said biological data may comprise at least one further group of data, in particular a one or more further organoid response data, such as a second group of data selected from the group consisting of proteomic data, genomic data, epigenomic data, transcriptomic data, metabolic data, microbiome data, imaging data, a histological data, a cytotoxic data, cytostatis data; with the second (or further) group(s) of data being different from the first group.

[0056] The “digital data”, as in “digital data from the subject / patient”, may comprise at least one first group of digital data, e.g. from a patient or a group of patients, such aselectronic health records (EHRs). Said first group of digital data may include one of more selected from the group consisting of a proteomic feature, a genomic feature, an epigenomic feature, a transcriptomic feature, a metabolic feature, a microbiome feature, an imaging feature, a histological feature. Optionally, said digital data may comprise at least one further group of data, such as a second group of data selected from the group consisting of proteomic data, genomic data, epigenomic data, transcriptomic data, metabolic data, microbiome data, imaging data, histological data; with the second (or further) group(s) of data being different from the first group. Yet optionally, said digital data may comprise patient longitudinal history, including past treatments and their respective responses. The digital data may have a time stamp which defines the time of the data collection. New digital data may be created over time. The new data may be added using the distinguishing time stamp or replace the existing digital data by updating the existing time stamp.

[0057] As used herein, an “organoid” refers to an artificial three-dimensional tissue construct comprising a plurality of cells; which is functionally capable of retaining all or part of an organ tissue, or a portion thereof. Such organoids may comprise at least one, in particular more than one, differentiated cell type(s); and optionally one or more than one undifferentiated cell type(s). The organ tissue which capability is retained by such organoids may be a healthy tissue or a tumoral (i.e., cancerous) counterpart. In a non-exhaustive manner, organoid may thus be selected from the group consisting of cerebral organoids, gastrointestinal organoids (for example foregut organoids, midgut organoids, hindgut organoids, intestinal organoids, or gastric organoids), ovarian organoids, lingual organoids, tooth organoids, thyroid organoids, thymic organoids, testicular organoids, prostate organoids, hepatic organoids, pancreatic organoids, epithelial organoid, lung organoids, kidney organoids, gastruloid organoids, blastoid organoids, endometrial organoids, cardiac organoids, retinal organoids, breast cancer organoids, colorectal cancer organoids, gliobastoma organoids, neuroendocrine tumor organoids, myelin organoids, blood-brain barrier (BBB) organoids; for example those selected from the group consisting of; colorectal cancer organoids, pancreatic cancer organoids, lung cancer organoids.

[0058] Accordingly, an organoid which is submitted to a stimulus / agent (e.g. an isolated payload), for example which is brought in contact with one or more physical, chemical and / or biological agent such as a therapeutic agent, will generate an “organoid response”, which is constitutive of what is referred herein as an “organoid response data”, thus defining an “organoid feature” with respect to the tested agent. By an “organoid response of said patient” it is meant a response of an organoid which is predicted to match with said patient. Accordingly, an organoid response to a matching organoid relates to a response from an organoid, for which all or part of the organoid is predicted to match the patient (or a population of patients). Such an organoid response may be, for example, a response to all or part of an organoid matched with said patient. Advantageously, and as the organoid response is valid even for patients from whom the cell or cell line of the organoid is not directly obtained, the devices and methods of the disclosure may thus also be applicable to patients for which there is no personalized organoid(s) available, thus overcoming the need for conceiving, or maintaining in culture, unnecessary biological material.

[0059] As used herein, the term “patient-derived organoid” or PDO refers to organoids which are derived from all or part of a culture of cells (in particular tumor cells) from individual patients. The PDOs which are suitable for the present invention, and methods, include those which can be stored, cultivated and grown in a controlled environment. As used herein, the term “organoid-derived cells” thus refers to cells which are isolated from a corresponding organoid. Accordingly, for the purposes of the presentation invention, the term may encompass cells isolated from PDOs.

[0060] As used herein the term "biological sample" encompasses a variety of sample types, such as those obtained from a subject and which may be used in a diagnostic or monitoring assay. Biological samples include but are not limited to blood and other liquid samples of biological origin, solid tissue samples such as a biopsy specimen or tissue cultures or cells derived therefrom, and the progeny thereof. For example, biological samples include cells obtained from a tissue sample collected from an individual suspected of having a disease. Such biological samples may encompass clinical samples,cells in culture, cell supernatants, cell lysates, serum, plasma, biological fluid, and tissue samples.

[0061] As used herein, a “proteomic data” or corresponding “proteomic feature” refers to any data which is indicative of the occurrence of a protein, or group of proteins which is expressed or susceptible to be expressed by / in a given cell or group of cells (e.g., an organoid or a tissue), under a given set of conditions. For example, proteomic data can be retrieved from any known method for determining the occurrence of, or amount of, a given protein / polypeptide or fragment thereof, such as (in a non-exhaustive manner).

[0062] As used herein, a “genomic data” or corresponding “genomic feature” refers to any data which is indicative of the structure (e.g. sequence and / or accessibility) of a corresponding genome, or part thereof, in a given cell or group of cells (e.g., an organoid or a tissue). Hence, the term may refer, in particular, to the detection or quantification of any nucleotide sequence corresponding to all or part of the genome of the cell or group cell.

[0063] As used herein, an “epigenomic data” or corresponding “epigenomic feature” refers to any data which is indicative of one or more epigenetic modification(s) of the said genome or part thereof, in a given cell or group of cells (e.g., an organoid or a tissue); this may also refer, in particular, to the detection or quantification of any epigenetic modification of the said genome or part thereof, such as any covalent modification of a deoxyribonucleic acid (e.g. DNA methylation) or post-translation modification of the histones.

[0064] As used herein, a “transcriptomic data” or corresponding “transcriptomic feature” refers to any data which is indicative of the occurrence of a nucleic acid (e.g. any ribonucleic acid, such as siRNAs, miRNA, coding and non-coding RNAs, pre-mature messenger RNAs, and mature messenger RNAs) which is transcribed or susceptible to be transcribed in a given cell or group of cells thereof. Accordingly, this term may refer to the detection or quantification of such nucleic acid(s).

[0065] As used herein, a “metabolic data” or corresponding “metabolic feature” refers to any data which is indicative of the occurrence of a small-molecule or compound whichis present or produced or susceptible to be present or susceptible to be produced in - or by- a given cell or group of cells (e.g., an organoid or a tissue), under a given set of conditions. Such small-molecules or compounds may, in particular, be selected from the group consisting of sugars, nucleotides, amino acids and lipids.

[0066] As used herein, a “microbiome data” or corresponding “microbiome feature” refers to any data which is indicative of the occurrence of one or more micro-organisms (e.g. bacteria, archaea, fungi, algae, and the like) which are present or susceptible to be present in, or associated to, a given cell or group of cells (e.g., an organoid or a tissue), under a given set of conditions.

[0067] As used herein, an “imaging data” or corresponding “imaging feature” refers to any data which is indicative of the morphology and / or anatomy and / or structure of a given cell or group of cells (e.g., an organoid or a tissue) or part thereof, under a given set of conditions. In a non-exhaustive manner, an imaging data or imaging feature may be retrieved from medical imaging, microscopy (e.g. optic microscopy, electron microscopy, X-ray microscopy) and live-cell imaging methods of the cell or group of cells, such as those selected from the group consisting of phase-contrast microscopy, fluorescent microscopy, quantitative phase contrast microscopy and holotomography.

[0068] As used herein, a “histological data” or corresponding “histological feature” refers to a subset of imaging data or imaging features, respectively, which is indicative of the microscopy morphology and / or anatomy and / or structure of a given cell or group of cells (e.g., an organoid or a tissue) or part thereof.

[0069] As used herein, a “cytotoxic data” or corresponding “cytotoxic feature, refers to any data which is indicative of the occurrence of cytotoxicity in a cell or group of cells (e.g. an organoid or a tissue) or part thereof, under a given set of conditions. Accordingly, a cytotoxic data may correspond to a subset of other above-mentioned data, such as a subset of metabolic data, or imaging data and / or of a histological data. For example data which is indicative of a higher cytotoxicity, under a given condition (e.g. in the presence of a given payload) may also be indicative of a lower viability for the same condition.

[0070] As used herein, a “cytostatis data” or corresponding“cytostatis feature” refers to any data which is indicative of the occurrence of cytostasis in a cell or group of cells (e.g. an organoid or a tissue) or part thereof, under a given set of conditions. Accordingly, a cytostasis data may correspond to a subset of other above-mentioned data, such as a subset of metabolic data, or imaging data and / or of a histological data. For example data which is indicative of a higher cytostasis, under a given condition (e.g. in the presence of a given payload) is indicative of the inhibition of cell growth and multiplication, which may also be indicative of cell death induction, and hence of a lower viability for a given set of conditions.

[0071] For a clinical trial, there are known parameters to assess how well a new treatment works. “Progression-free survival (PFS)” is the length of time during and after the treatment of a disease, such as cancer, that a patient lives with the disease, but it does not get worse. “Overall Survival (OS)” is the length of time from either the date of diagnosis or the start of treatment for a disease, such as cancer, that patients diagnosed with the disease are still alive. “Overall Response (OR)” or “Overall Response Rate (ORR) is the percentage of people in a study or treatment group who have a partial response or complete response to the treatment within a certain period of time. A partial response is a decrease in the size of a tumour or in the amount of cancer in the body, and a complete response is the disappearance of all signs of cancer in the body.

[0072] “Electronic Health Record (EHR)” designates an electronic version of patients’ medical history, which is maintained by the provider over time.

[0073] “Omics data” refers to data generated from high-throughput technologies used to study the various "omes" of an organism, such as the genome (all the genetic material), transcriptome (all the RNA molecules), proteome (all the proteins), metabolome (all the small molecules), and interactome (all the interactions between biomolecules). Omics data is often used in systems biology and functional genomics to study the relationships between different molecules and how they interact to affect the overall function of cells, tissues, and organisms. Omic data can be complex, high-dimensional, and noisy and requires specialized computational methods and tools for analysis and interpretation.

[0074] As known per se a “cohort” is a group of people with a shared characteristic. A “cohort study” designates a type of epidemiological study in which a group of people with a common characteristic is followed over time to find how many reach a certain health outcome of interest (disease, condition, event, death, or a change in health status or behavior).

[0075] “Machine learning (ML)” designates in a traditional way computer algorithms improving automatically through experience, on the ground of training data enabling to adjust parameters of computer models through gap reductions between expected outputs extracted from the training data and evaluated outputs computed by the computer models.

[0076] A “hyper-parameter” presently means a parameter used to carry out an upstream control of a model construction, such as a remembering-forgetting balance in sample selection or a width of a time window, by contrast with a parameter of a model itself, which depends on specific situations. In ML applications, hyper-parameters are used to control the learning process.

[0077] “Datasets” are collections of data used to build an ML mathematical model, so as to make data-driven predictions or decisions. In “supervised learning” (i.e. inferring functions from known input-output examples in the form of labelled training data), three types of ML datasets (also designated as ML sets) are typically dedicated to three respective kinds of tasks: “training”, i.e. fitting the parameters, “validation”, i.e. tuning ML hyperparameters (which are parameters used to control the learning process), and “testing”, i.e. checking independently of a training dataset exploited for building a mathematical model that the latter model provides satisfying results.

[0078] A “neural network (NN)” designates a category of ML comprising nodes (called “neurons”), and connections between neurons modelled by “weights”. For each neuron, an output is given in function of an input or a set of inputs by an “activation function”. Neurons are generally organized into multiple “layers”, so that neurons of one layer connect only to neurons of the immediately preceding and immediately following layers.

[0079] The above ML definitions are compliant with their usual meaning, and can be completed with numerous associated features and properties, and definitions of related numerical objects, well known to a person skilled in the ML field. Additional terms willbe defined, specified or commented wherever useful throughout the following description.

[0080] By “ determining the efficacy of an input immunoconjugate”, it is meant in particular:determining the efficacy of immunoconjugates with respect to PDOs or a population of PDOs;determining the efficacy of the combination of one or more antigen-binding sites / antigen-binding fragments and one or more payloads with respect to PDOs or a population of PDOs, the said combination being thus an input immunoconjugate; and / orcomparing the efficacy of two or more immunoconjugates with respect PDOs or a population of PDOs.

[0081] By “efficacy”, it is meant any property of said immunoconjugates which is beneficial, for a given patient or population of patients, for therapeutic efficacy or for diagnosis purpose; e.g. for treating or preventing or reducing the likelihood of occurrence of a given condition. Hence, the assessment of efficacy may, according to particular embodiments, be achieved with any feature selected from the group consisting of a proteomic feature, a genomic feature, an epigenomic feature, a metabolic feature, a microbiome feature, an imaging feature, a histological feature, a cytotoxic feature, a cytostatis feature, or a combination thereof.

[0082] In the context of therapeutic efficacy, the term encompasses in particular the property of an immunoconjugate, or its corresponding payload, to modulate cell viability, for example to selectively modulate cell viability, for example to decrease cell viability in tumor cells and / or to increase cell viability in non-tumor (i.e. “normal” cells). Assessment of cell viability may be achieved by any cell viability assay known to the skilled in the Art, for example an ATP assay. The therapeutic efficacy may be evaluated by immunoconjugate sensitivity values such as area under the curve (AUC) and half-maximal inhibitory concentration (IC50), calculated based on a single-drug, doseresponse curve.BRIEF DESCRIPTION OF THE DRAWINGS

[0083] The present disclosure will be better understood, and other specific features and advantages will emerge upon reading the following description of particular and non-restrictive illustrative embodiments, the description making reference to the annexed drawings.

[0084] Figure 1 presents a process for determining the efficacy of a given immunoconjugate, in this particular case an antibody-drug conjugate susceptible to be active against tumor cells. The process is visually represented for a given antigen, recognized by the antigen-binding sites of the corresponding antibody (x-axis) and the efficacy of the payload alone (y-axis). For each Patient-Derived Organoid, an avatar of a patient is generated, thus providing one dot in the corresponding representation. In this example, the representation thus defines one threshold for each value (i.e. one threshold value of payload efficacy in the y-axis and one threshold value of antigen abundance in the x-axis); thus defining four reference area, with one reference area marked as the immunoconjugate (e.g. ADC) efficacy area in the upper-right comer of the representation (in this case the efficacy area is defined when both corresponding values are at or above their respective thresholds). The lower-left comer of the representation defined an area wherein neither the payload alone is potent nor the antigen sufficiently abundant. The two remaining area define an area wherein the corresponding avatars are associated with high efficacy of either the payload alone or high antigen abundance alone. For one given immunoconjugate, or combination of payload-antigen binding site, a dot is generated. An increased number of dots in the reference efficacy area is indicative of a higher efficacy of said candidate with respect to the patient population defined by the selection of avatars. Alternatively, a larger number of dots in the reference efficacy area for a first immunoconjugate, with respect to the number of dots in the reference efficacy area for a second immunoconjugate, is indicative of an increased efficacy of said first immunoconjugate over the second one. In this example, no avatars are identified within the (reference) ADC efficacy area.

[0085] Figure 2 presents for two individual avatars (i.e. avatar 1 and avatar2) the results of an in vitro preliminary result. For each avatar, the left column displays the effect of an antibody-drug conjugate (ADC), specifically fam-trastuzumab deruxtecan, on the corresponding Patient-Derived Organoid (PDO), as assessed using an in-house cell viability assay. The right column displays the effect of the SN-38 payload alone on the same PDOs using the same viability assay: avatarl is characterized by the absence of HER2 expression and resistance to the SN-38 pay load, avatar2 also lacks HER2 expression but is sensitive to the SN-38 payload. These results indicate that in the absence of HER2 expression, the ADC shows limited efficacy, whereas the SN-38 payload alone can retain its activity when the target cells are intrinsically sensitive. This suggests that the efficacy of the SN-38 payload is independent of HER2 expression levels, in contrast to the ADC whose activity is HER2-dependent.

[0086] Figure 3 presents a method for quantifying antigen abundance in tumor cells vs adjacent and non-adjacent normal (non-tumor) cells. In this example, according to a first step, tumor samples, adjacent normal samples, and non-adjacent normal samples are provided, which can be associated to a same PDO or to a same population of PDOs. In a second step, the abundance of overexpressed genes is assessed and the corresponding relevant antigen receptors are determined. In a third step, antigen receptors highly expressed by tumor cells are identified and further screened in PDO-derived cells, thus corresponding to the data of the x-axis in figure 1.

[0087] Figure 4 presents a method for evaluating the efficacy of payload in tumor cells. In this example, each PDO is associated to a set of data. Organoid response data is retrieved from either in vitro testing or from previously determined PDO-response data, and drug sensitivity is computed for each PDO, this corresponding to the data of the y-axis in figure 1.

[0088] Figures 5A and 5B presents a dose-response curve empirically obtained from two distinct PDO lines in the presence of an antibody-drug conjugate directed toward the TROP2 antigen. The tested ADC is a Sacituzumab-govitecan conjugate and is predicted as efficient prior to experience. Cell viability is defined in the y-axis.

[0089] Figure 6 presents a dose-response curve empirically obtained from a PDO in the presence of an antibody-drug conjugate directed toward the HER2 antigen. The tested ADC is a famtrastuzumab-deruxtecan conjugate and is predicted as inefficient prior to experience. Cell viability is defined in the y-axis.

[0090] Figure 7 presents a graph showing mRNA expression levels of TROP2 gene and the housekeeping gene ACTB, in two distinct PDO lines (PDO-A and PDO-B).

[0091] Figures 8A and 8B presents a dose-response curve empirically obtained from the two distinct PDO lines PDO-A (Figure 8A) and PDO-B (Figure 8B) in the presence of an antibody-drug conjugate directed toward the TROP2 antigen. The tested ADC is a Datopotamab-deruxtecan conjugate and is predicted as efficient prior to experience. Cell viability is defined in the y-axis.

[0092] Figures 9 presents a schematic showing the mechanism of internalization, drug release and recycling of an ADC binding to an antigen expressed at the surface of PDOs.DETAILED DESCRIPTION

[0093] Without wishing to be bound by the theory, the inventors are of the opinion that Patient-Derived Organoids (PDO) are particularly adapted to the high-throughput screening of immunoconjugates, and / or payload-antigen-binding site couples; especially in the context of drug development.

[0094] According to a first embodiment, the invention relates to a computer-implemented method for determining the efficacy of an input immunoconjugate (10) with respect to PDOs comprising at least an input antigen-binding site (11) and an input payload (12); the method comprising the following steps:providing organoid response data (21) indicative of efficacies of the input payload (12) on a respective plurality of patient-derived organoids, PDOs, of a patient population, and derived by bringing the input payload (12) into contact with the respective PDOs;providing expression data (20) indicative of levels of expression of antigens expressed by organoid-derived cells of the respective PDOs; and wherein the antigens are capable of binding to the input antigen-binding site (11); and determining the efficacy of the input immunoconjugate (10) against PDOs based on the organoid response data (21) and the expression data (20).

[0095] In some embodiments, the efficacy of the input immunoconjugate (10) is compared with the efficacy of an input reference immunoconjugate (14). The input reference immunoconjugate (14) can be any input immunoconjugate (10) already known as efficient with respect to PDOs. The efficiency of the reference input immunoconjugate (14) can be determined with a reference organoid response data (21) and a reference expression data (20), thereby defining a reference threshold.

[0096] In some embodiments, the organoid response data (21) of the input immunoconjugate (10) and the expression data of the input antigen-binding site are respectively compared with the organoid response data (21) of the input reference immunoconjugate (14) and the expression data of the input reference antigen-binding site, in order to determine the efficiency of the input immunoconjugate (10).

[0097] In some embodiments, the efficacy of the input immunoconjugate (10) is based on the cytotoxic data of the input payload (12) and the expression data (20) of the antigens expressed by organoid-derived cells of the PDOs.

[0098] In some embodiments, the step of determining the efficacy further comprises determining the input immunoconjugate (10) as efficient when at least a predetermined portion (34) of the patient population shows respective levels of expression above a minimum reference level of expression (35) and shows an efficacy of the input payload above a minimum efficacy threshold (36).

[0099] In some embodiments, the efficacy of the input immunoconjugate (10) can be measured by bringing the input immunoconjugate (10) into contact with the respective PDOs and testing the level of expression of the input antigen and the level of viability of the input payload.

[0100] In some embodiments, the computer-implemented method further comprises providing a data store comprising expression data (20), organoid response data (21), and subject data associated with PDOs from subjects; and selecting the patient population as a subset of the PDOs that are affected by the same or similar tumour.

[0101] According to some embodiments, the plurality of Patient-Derived Organoids (PDO) is brought into contact with one isolated input pay load (12). According to some embodiments, the plurality of Patient-Derived Organoids (PDO) is brought into contact with a plurality of isolated input payloads (12).

[0102] In some embodiments, the input immunoconjugate (10) also comprises at least an input linker (13), and the method comprises providing linker data (23), indicative of efficacies of the input linker (13). The input linker (13) can be selected from the list consisting of cleavable and non-cleavable linkers, and peptidic or non-peptidic linkers; for example linkers comprising from 1 to 100 amino acids, e.g. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82,T183, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, and 100 amino acids, or more than 100 amino acids.

[0103] According to some embodiments, the computer-implemented method further comprises providing internalization rate data (22) and / or recycling rate data (22) indicative of the efficacies of the input candidate antigen-binding site (11), preferably on a respective plurality of patient-derived organoids, of a patient population.

[0104] In some embodiment, the internalization rate data and / or recycling rate data (22) may be obtained from bibliographic data.

[0105] The internalization rate is the kinetic rate at which a cell-surface-bound immunoconjugate (usually bound to its target antigen / receptor) is taken up into the cell via endocytosis. It quantifies how fast the immunoconjugate-antigen complex is removed from the plasma membrane and enters intracellular vesicles (early endosomes). The unity can be in min-1.

[0106] The recycling rate is the kinetic rate at which internalized immunoconjugateantigen complexes (or free receptors) are returned from intracellular compartments back to the cell surface instead of being trafficked to lysosomes for degradation. The unity can be in min-1.

[0107] The person skilled in the art knows how to measure internalization rate data and / or recycling rate and how to obtain such data. Internalization rate can be measured by flow cytometry with acid quenching, use of pH-sensitive probes, and / or quantitative confocal microscopy. Recycling rate can be measured by pulse-chase with re-stripping, antibody-feeding assay, and / or real-time imaging with dual labeling.

[0108] In some embodiment, the internalization rate data and / or recycling rate data (22) may be derived by bringing the candidate antigen-binding site into contact with the respective PDOs. Figure 9 illustrates an example of process of internalization and recycling of an immunoconjugate.

[0109] According to some embodiments, the Patient-Derived Organoids (PDO) are selected from a group consisting of cerebral organoids, gastrointestinal organoids, lingual organoids, tooth organoids, thyroid organoids, thymic organoids, testicular organoids, prostate organoids, hepatic organoids, pancreatic organoids, epithelialorganoid, lung organoids, kidney organoids, gastruloid organoids, blastoid organoids, endometrial organoids, cardiac organoids, retinal organoids, breast cancer organoids, colorectal cancer organoids, glioblastoma organoids, neuroendocrine tumor organoids, myelin organoids, blood-brain barrier (BBB) organoids, ovarian organoids.

[0110] According to some embodiments, the Patient-Derived Organoids (PDO) are selected from a group consisting of: cerebral cancer organoids, gastrointestinal cancer organoids, lingual cancer organoids, tooth cancer organoids, thyroid cancer organoids, thymic cancer organoids, testicular cancer organoids, prostate cancer organoids, hepatic cancer organoids, pancreatic cancer organoids, epithelial cancer organoid, lung cancer organoids, kidney cancer organoids, gastruloid cancer organoids, blastoid cancer organoids, endometrial cancer organoids, cardiac cancer organoids, retinal cancer organoids, breast cancer organoids, colorectal cancer organoids, glioblastoma organoids, neuroendocrine tumor organoids, myelin organoids, blood-brain barrier (BBB) cancer organoids, ovarian cancer organoids.

[0111] According to some embodiments, the plurality of said Patient-Derived Organoids (PDO) comprises at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, or more than 100 PDOs of the same type. According to some embodiments, the plurality of said Patient-Derived Organoids (PDO) comprises at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, or more than 100 PDOs of a distinct type; for example at least one PDO which is a cancer organoid from one type, and at least one PDO which is a cancer organoid from a different type.

[0112] According to some non-mutually exclusive embodiments, the Patient-Derived Organoids may comprise or consist of tumor cells. For example, the PDOs may comprise tumor cells and non-tumor cells. Alternatively, the PDOs may substantially comprise tumor cells. Alternatively, the PDOs may substantially comprise non-tumor cells. According to preferred embodiments, the PDOs may substantially comprise tumor cells.

[0113] It will be understood herein that the plurality of PDOs is brought in contact with the input payload(s) (12) in an isolated form; that is, the payload(s) when not associated to the immunoconjugate.

[0114] Examples of input payloads (12) may be those selected from a group consisting of taxol; cytochalasin B; gramicidin D; ethidium bromide; emetine; mitomycin; etoposide; tenoposide; vincristine; vinblastine; colchicin; doxorubicin; daunorubicin; dihydroxy anthracin dione; a tubulin- inhibitor such as maytansine or an analog or derivative thereof; an antimitotic agent such as monomethyl auristatin E or F or an analog or derivative thereof; dolastatin 10 or 15 or an analogue thereof; irinotecan or an analogue thereof; mitoxantrone; mithramycin; actinomycin D; 1 -dehydrotestosterone; a glucocorticoid; procaine; tetracaine; lidocaine; propranolol; puromycin; calicheamicin or an analog or derivative thereof; an antimetabolite such as methotrexate, 6 mercaptopurine, 6 thioguanine, cytarabine, fludarabin, 5 fluorouracil, decarbazine, hydroxyurea, asparaginase, gemcitabine, or cladribine; an alkylating agent such as mechlorethamine, thioepa, chlorambucil, melphalan, carmustine (BSNU), lomustine (CCNU), cyclophosphamide, busulfan, dibromomannitol, streptozotocin, dacarbazine (DTIC), procarbazine, mitomycin C; a platinum derivative such as cisplatin or carboplatin; duocarmycin A, duocarmycin SA, rachelmycin (CC-1065), or an analog or derivative thereof; an antibiotic such as dactinomycin, bleomycin, daunorubicin, doxorubicin, idarubicin, mithramycin, mitomycin, mitoxantrone, plicamycin, anthramycin (AMC)); pyrrolo[2,l-c] [1,4] -benzodiazepines (PDB); diphtheria toxin and related molecules such as diphtheria A chain and active fragments thereof and hybrid molecules, ricin toxin such as ricin A or a deglycosylated ricin A chain toxin, cholera toxin, a Shiga-like toxin such as SLT I, SLT II, SLT IIV, LT toxin, C3 toxin, Shiga toxin, pertussis toxin, tetanus toxin, soybean Bowman-Birk protease inhibitor, Pseudomonas exotoxin, alorin, saporin, modeccin, gelanin, abrin A chain, modeccin A chain, alpha-sarcin, Aleurites fordii proteins, dianthin proteins, Phytolacca americana proteins such as PAPI, PAPII, and PAP-S, momordica charantia inhibitor, curcin, crotin, sapaonaria officinalis inhibitor, gelonin, mitogellin, restrictocin, phenomycin, and enomycin toxins; ribonuclease (RNase); DNase I, Staphylococcal enterotoxin A; pokeweed antiviral protein; diphtherin toxin; and Pseudomonas endotoxin. The term may also encompass an antibody or antigen-binding fragment conjugated to a nucleic acid or nucleic acid-associated molecule, such as those selected from the group consisting of a cytotoxic ribonuclease (RNase) or deoxy-ribonuclease (e.g., DNase I), an antisense nucleic acid, aninhibitory RNA molecule (e.g., a siRNA molecule) or an immunostimulatory nucleic acid (e.g., an immunostimulatory CpG motif-containing DNA molecule).

[0115] According to some non-mutually exclusive embodiments, the input pay load can be selected from the group consisting of 5FU (capecitabine), Azacitidine, Bortesomib, Carboplatin, Cobimetinib, Docetaxel, Doxorubicin, Etoposide, Everolimus, Fludarabine, Gefitinib, Gemcitabine, Lapatinib, Mitomycin C, Olaparib, Oxaliplatin (platinium analog), Paclitaxel, Pemetrexed, Raltitrexed, Regorafenib, SN-38 (irinotecan), Sunitinib, Trifluridin-tipiracil, Vinorelbine, Vorinostat.

[0116] In some embodiments, the input antigen-binding site (11) is an input antigenbinding fragment.

[0117] According to some non-mutually exclusive embodiments, the input antigenbinding fragments and / or related antigen-binding sites may comprise or consist of antigen-binding fragments from a therapeutic antibody, or antigen-binding fragment thereof.

[0118] According to some non-mutually exclusive embodiments, the antigen-binding fragments and / or related antigen-binding sites may comprise or consist of antigen-binding fragments from a therapeutic antibody, or antigen-binding fragment thereof, selected from the group consisting of Abagovomab, Abatacept, Abciximab, Abituzumab, Abrilumab, Actoxumab, Adalimumab, Adecatumab, Aducanumab, Aflibercept, Afutuzymab, Alacizumab, Alefacept, Alemtuzumab, Alirocumab, Altumomab, Amatixumab, Anatumomab, Anetumab, Anifromumab, Anrukinzumab, Apolizumab, Arcitumomab, Ascrinvacumab, Aselizumab, Atezolizumab, Atinumab, Altizumab, Atorolimumab, Bapineuzumab, Basiliximab, Bavituximab, Bectumomab, Begelomab, Belatacept, Belimumab, Benralizumab, Bertilimumab, Besilesomab, Bevacizumab, Bezlotoxumab, Biciromab, Bimagrumab, Bimekizumab, Bivatuzumab, Blinatumomab, Blosozumab, Bococizumab, Brentuximab, Briakimumab, Brodalumab, Brolucizumab, Bronticizumab, Canakinumab, Cantuzumab, Caplacizumab, Capromab, Carlumab, Catumaxomab, Cedelizumab, Certolizumab, Cetixumab, Citatuzumab, Cixutumumab, Clazakizumab, Clenoliximab, Clivatuzumab, Codrituzumab, Coltuximab, Conatumumab, Concizumab, Crenezumab, Dacetuzumab, Daclizumab, Dalotuzumab,Dapirolizumab, Daratumumab, Dectrekumab, Demcizumab, Denintuzumab, Denosumab, Derlotixumab, Detumomab, Dinutuximab, Diridavumab, Dorlinomab, Drozitumab, Dupilumab, Durvalumab, Dusigitumab, Ecromeximab, Eculizumab, Edobacomab, Edrecolomab, Efalizumab, Efungumab, Eldelumab, Elgemtumab, Elotuzumab, Elsilimomab, Emactuzumab, Emibetuzumab, Enavatuzumab, Enfortumab, Enlimomab, Enoblituzumab, Enokizumab, Enoticumab, Ensituximab, Epitumomab, Epratuzomab, Erlizumab, Ertumaxomab, Etaracizumab, Etrolizumab, Evinacumab, Evolocumab, Exbivirumab, Fanolesomab, Faralimomab, Farletuzomab, Fasimumab, Felvizumab, Fezkimumab, Ficlatuzumab, Figitumumab, Firivumab, Flanvotumab, Fletikumab,Fontolizumab, Foralumab, Foravirumab, Fresolimumab, Fulramumab, Futuximab, Galiximab, Ganitumab, Gantenerumab, Gavilimomab, Gemtuzumab, Gevokizumab, Girentuximab, Glembatumumab, Golimumab, Gomiliximab, Guselkumab, Ibalizumab, Ibritumomab, Icrucumab, Idarucizumab, Igovomab, Imalumab, Imciromab, Imgatuzumab, Inclacumab, Indatuximab, Indusatumab, Infliximab, Intetumumab, Inolimomab, Inotuzumab, Ipilimumab, Iratumumab, Isatuximab, Itolizumab, Ixekizumab, Keliximab, Labetuzumab, Lambrolizumab, Lampalizumab, Lebrikizumab, Lemalesomab, Lenzilumab, Lerdelimumab, Lexatumumab, Libivirumab, Lifastuzumab, Ligelizumab, Lilotomab, Lintuzumab, Lirilumab, Lodelcizumab, Lokivetmab, Lorvotuzumab, Lucatumumab, Lulizumab, Lumiliximab, Lumretuzumab, Mapatumumab, Margetuximab, Maslimomab, Mavrilimumab, Matuzumab, Mepolizumab, Metelimumab, Milatuzumab, Minetumomab, Mirvetuximab, Mitumomab, Mogamulizumab, Morolimumab, Motavizumab, Moxetumomab, Muromonab-CD3, Nacolomab, Namilumab, Naptumomab, Namatumab, Natalizumab, Nebacumab, Necitumumab, Nemolizumab, Nerelimomab, Nesvacumab, Nimotuzumab, Nivolumab, Nofetumomab, Obiltoxaximab, Obinutuzumab, Ocaratuzumab, Ocrelizumab, Odulimomab, Ofatumumab, Olaratumab, Olokizumab, Omalizumab, Onartuzumab, Ontuxizumab, Opicinumab, Oportuzumab, Oregovomab, Orticumab, Otelixizumab, Oltertuzumab, Oxelumab, Ozanezumab, Ozoralizumab, Pagibaximab, Palivizumab, Panitumumab, Pankomab, Panobacumab, Parsatuzumab, Pascolizumab, Pasotuxizumab, Pateclizumab, Patritumab, Pembrolizumab, Pemtumomab, Perakizumab, Pertuzumab, Pexelizumab, Pidilizumab, Pinatuzumab, Pintumomab, Polatuzumab, Ponezumab, Priliximab, Pritumumab,Quilizumab, Racotumomab, Radretumab, Rafivirumab, Ralpancizumab, Ramucirumab, Ranibizumab, Raxibacumab, Refanezumab, Regavirumab, Reslizumab, Rilonacept, Rilotumumab, Rinucumab, Rituximab, Robatumumab, Roledumab, Romosozumab, Rontalizumab, Rovelizumab, Ruplizumab, Sacituzumab, Samalizumab, Sarilumab, Satumomab, Secukimumab, Seribantumab, Setoxaximab, Sevirumab, Sibrotuzumab, Sifalimumab, Siltuximab, Siplizumab, Sirukumab, Sofituzumab, Solanezumab, Solitomab, Sonepcizumab, Sontuzumab, Stamulumab, Sulesomab, Suvizumab, Tabalumab, Tacatuzumab, Tadocizumab, Talizumab, Tanezumab, Taplitumomab, Tarextumab, Tefibazumab, Telimomab aritox, Tenatumomab, Teneliximab, Teplizumab, Tesidolumab, TGN 1412, Ticlimumab, Tildrakizumab, Tigatuzumab, TNX-650, Tocilizumab, Toralizumab, Tosatoxumab, Tositumomab, Tovetumab, Tralokimumab, Trastuzumab, TRBS07, Tregalizumab, Tremelimumab, Trevogrumab, Tucotuzumab, Tuvirumab, Ublituximab, Ulocuplumab, Urelumab, Urtoxazumab, Ustekimumab, Vandortuzumab, Vantictumab, Vanucizumab, Vapaliximab, Varlimumab, Vatelizumab, Vedolizumab, Veltuzumab, Vepalimomab, Vesencumab, Visilizumab, Volocixumab, Vorsetuzumab, Votumumab, Zalutumimab, Zanolimumab, Zatuximab, Ziralimumab, Ziv-Aflibercept, and Zolimomab.

[0119] In some embodiments, the levels of expression are measured on antigens expressed at a surface of the organoid-derived cells of the respective PDOs. For example, the level of expression of antigens can be determined by a method selected from the group consisting of immunohistochemistry, mass cytometry, and combinations thereof

[0120] According to some non-mutually exclusive embodiments, the antigen(s) for which expression is determined, may be selected from the group comprising or consisting of ADAM9, ALCAM, AXL, CA6, CD276, CD46, CDX2, CLDN18, CEACAM5, DLL3, EGFR, ERBB2, ERBB3, F3, FOLR1, GPNMB, IGF1R, IL20RB, KRT19, KRT20, LY75, MET, MKI67, MMP14, MUC5AC, MSLN, NCAM1, PTK7, ROR1, ROR2, SEZ6L2, SLC34A2, SLC39A6, SSTR2, TACSTD2, TFRC, TPBG.

[0121] According to some non-mutually exclusive embodiments, the level of expression of one or more antigen(s) expressed by / in corresponding organoid-derived cells may comprise or consist of determining the level of expression of one or more antigen(s) expressed at the surface of the cells. Examples of antigen(s) which are susceptible to beexpressed at the surface of the cells may include antigen(s) which constitute all or part of a cell receptor and / or of a transmembrane protein, including those which constitute all or part of an extracellular domain, and antigen(s) which constitute all or part of a pore channel, and those antigen(s) which are associated to another antigen susceptible to be present at the surface of the cells, for example susceptible to be associated to a cell receptor or a transmembrane protein.

[0122] Alternatively, the level of expression of one or more antigen(s) expressed by corresponding organoid-derived cells may comprise or consist of determining the level of expression of the one or more nucleic acids (e.g. genes) susceptible to encode for antigen(s) expressed at the surface of the cells. Examples of methods for determining or measuring the level of expression of nucleic acids in cells are known in the Art.

[0123] Non-exhaustively, such methods may comprise determining or measuring the presence of a coding RNA, in particular a messenger RNA (mRNA), which is susceptible to encode the one or more of the antigen(s); in particular the one or more antigen(s) expressed at the surface of the cells.

[0124] Hence, according to some non-mutually exclusive embodiments, the efficacy of the input immunoconjugate(s) (10) is determined based on (i) the organoid response data (21) and (ii) a level of expression of one or more antigen(s) expressed at the surface of the cells (i.e expression data (20)).

[0125] For example, the level of expression of antigen(s) at the surface of the cells is determined by a method selected from the group consisting of immunohistochemistry, mass cytometry, or combinations thereof. In a non-exhaustive manner, mass cytometry may refer to Cytometry by time of flight (CyTOF) and Imaging Mass Cytometry (IMC).

[0126] According to some embodiments, the computer-implemented method further comprises providing reference expression data (24) indicative of levels of expression of the antigens expressed by reference non tumour cells

[0127] According to some non-mutually exclusive embodiments, the level of expression of the one or more antigen(s) is a relative expression determined by:the level of expression of said antigen(s) by the organoid-derived cell, andthe level of expression of said antigen(s) by a reference cell, for example a non-tumor cell.

[0128] According to some non-mutually exclusive embodiments, the level of expression of the one or more antigen(s) is a relative expression determined by:- the level of expression of said antigen(s) at the surface of the organoid-derived cell, and- the level of expression of said antigen(s) at the surface of a reference cell, for example a non-tumor cell.

[0129] According to some non-mutually exclusive embodiment, the level of expression of the one or more antigen(s) expressed by corresponding organoid-derived cells may be distinguished based on or more thresholds; for example a low-value threshold and a high-value threshold over a given reference. In a non-exhausive manner, the level of expression of the antigen(s) may thus be defined as “low” or “high” when compared to a reference value; for example a first threshold being at least 2x-fold, 5x-fold, 1 Ox-fold or 100-fold higher than a reference value, and / or a second threshold being at least 2x-fold, 5x-fold, 1 Ox-fold or 100-fold lower than a reference value.

[0130] According to some non-mutually exclusive embodiments, the organoid response data is selected from the group consisting of proteomic data, genomic data, epigenomic data, metabolic data, microbiome data, imaging data, histological data, cytotoxic data, cytostatis data, or a combination thereof. Alternatively or additionally, the organoid response data may result from a chemical test, for example an ATP test.

[0131] According to some non-mutually exclusive embodiments, the expression data is selected from the group consisting of proteomic data, genomic data, epigenomic data, metabolic data, microbiome data, imaging data, histological data, cytotoxic data, cytostatis data, or combination thereof.

[0132] In some embodiment, the organoid response data comprises cytotoxicity data. Alternatively, cytotoxicity data can be named viability data. Cytotoxicity data may correspond to a measure of IC50 of an input payload on a plurality of PDOs.

[0133] According to some non-mutually exclusive embodiments all or part of the immunoconjugate(s) for which efficacy is predicted is / are recovered or manufactured; in particular wherein the payload and the antigen-binding site, or a combination thereof, is recovered or manufactured.

[0134] According to some embodiments, the method relates to a computer-implemented method for determining the efficacy of an input immunoconjugate (10), in particular an input antibody-drug conjugate (ADC), with respect to PDOs comprising at least an input antigen-binding site (11) and an input payload (12); the method comprising the following steps:- providing cytotoxicity data (alternatively viability data) (21) indicative of efficacies of the input payload (12) on a respective plurality of patient-derived organoids, PDOs, of a patient population, and derived by bringing the input payload (12) into contact with the respective PDOs;- providing expression data (20) indicative of levels of expression of antigens expressed by organoid-derived cells of the respective PDOs; and wherein the antigens are capable of binding to the input antigen-binding site (11);determining the efficacy of the input immunoconjugate (10) based on the cytotoxicity data (alternatively viability data) and the expression data (20).

[0135] Advantageously, the computer-implemented method for determining the efficacy of immunoconjugate(s) is also suitable for comparing the efficacy of a plurality of immunoconjugates, the immunoconjugates comprising at least one antigen-binding site and at least one payload.

[0136] Hence, according to said particular embodiment, the invention further relates to a computer-implemented method for comparing the efficacy of a plurality of input immunoconjugates, the input immunoconjugates comprising at least one input antigenbinding site (11) and at least one input payload (12), the method comprising the steps ofa) determining the efficacy of at least a first input immunoconjugate(s) (10) and of at least a second input immunoconjugate(s) according to the method of the invention, andb) comparing the efficacy of the first and second input immunoconjugate(s) determined at step a).

[0137] According to another embodiment, the invention relates to a computer-implemented method for selecting an optimal immunoconjugate from a set of input immunoconjugates, comprisingdetermining the efficacy of the set of input immunoconjugates according to the method according to the invention,selecting the optimal immunoconjugate based on the efficacy of the set of input immunoconjugates.

[0138] According to another embodiment, the invention relates to a device for determining the efficacy of an input immunoconjugate (10) against PDOs comprising at least an input antigen-binding site (11) and an input payload (12); the device comprising:at least one input configured to receive:• an organoid response data (21) indicative of efficacies of the input payload on a respective plurality of patient-derived organoids, PDOs, of a patient population, and derived by bringing the input payload (12) into contact with the respective PDOs;an expression data (20) indicative of levels of expression of antigens expressed by organoid-derived cells of the respective PDOs; and wherein the antigens are capable of binding to the input antigen-binding site (11); at least one processor configured to:• determine the efficacy of the input immunoconjugate (10) against PDOs based on the organoid response data (21) and the expression data (20). At least one output configured to provide the efficacy of the input immunoconjugate (10) against PDOs.

[0139] It will be understood herein that the processor is advantageously configured to obtain a prediction of the efficacy of the input immunoconjugate using a (first) output vector; wherein said prediction is obtained according to the method for predicting an efficacy of an input immunoconjugate, or for comparing the efficacy of two input immunoconjugate(s), as previously disclosed. Accordingly, an output vector may comprise or consist of a selection of matched patient-derived organoids (PDO); thereby providing a plurality of PDOS on which the efficacy of input immunoconjugates can then be determined.

[0140] According to said embodiment, the invention also relates to the device for predicting an efficacy of an input immunoconjugate, or for comparing the efficacy of two input immunoconjugate(s), the device comprising: at least one processor configured to obtain at least a Progression-Free Survival (PFS) and / or an Overall Survival (OS) for the patient using the first output vector.

[0141] According to some embodiments, the device comprises one or more of the features described in the following embodiments, taken alone or in any possible combination.

[0142] According to one embodiment, the processor is configured to use a first ensemble of one or more learning models for the determining comprising at least one Cox model and, optionally, one or more models of the following types: XGBoost, Random Survival Forest, Ridge Cox, and Lasso Cox.

[0143] According to one embodiment, is the processor is further configured to receive at least one second ensemble of one or more trained learning models each configured to receive as input, with respect to said patient, one or more of said at least one patient feature and one or more of said at least one organoid response feature, and to provide a second output vector, said second ensemble of one or more trained learning models comprising at least one logistic regression model; and said least one processor is further configured to calculate the second output vector by providing said at least one patient feature and at least one organoid response feature as input to said second ensemble of one or more trained learning models, to obtain a prediction of the efficacy of theimmunoconjugate using the second output vector and optionally, to obtain an Overall Response (OR) for the patient using the second output vector.

[0144] According to one embodiment, each received one or more of said at least one patient feature has a statistical significance higher than a first threshold, and each received one or more of said at least one organoid response feature has a statistical significance higher than a second threshold.

[0145] According to one embodiment, the second ensemble of one or more trained learning models further comprises at least one or more models of the following types: XGBoost, logistic regression, and random forest.

[0146] According to one embodiment, when said second ensemble of one or more trained learning models comprises two or more trained learning models, predictions obtained from each trained learning models are combined according to a voting classifier to obtain the second output vector.

[0147] According to a further embodiment, the invention relates to a computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the computer-implemented method for determining the efficacy of immunoconjugate(s), and / or or for comparing the efficacy of two immunoconjugate(s) according to the above-described embodiments.

[0148] According to a further embodiment, the invention relates to a computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the computer-implemented method for determining the efficacy of immunoconjugate(s), and / or for comparing the efficacy of two immunoconjugate(s) according to the above-described embodiments.

[0149] The present description illustrates the principles of the present disclosure. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described or shown herein, embody the principles of the disclosure and are included within its scope.

[0150] All examples and conditional language recited herein are intended for educational purposes to aid the reader in understanding the principles of the disclosure and the concepts contributed by the inventor to furthering the art, and are to be construed as being without limitation to such specifically recited examples and conditions.

[0151] For example, while the embodiments discussed herein are illustrated in particular application to the field of oncology and cancer treatments, it will be appreciated by those skilled in the art that the principles of the disclosure may be applied to any other disease and its respective treatments.

[0152] For example, while the embodiments discussed herein are illustrated in particular application to the field of antibody-drug conjugate (ADC) development, it will be appreciated by those skilled in the art that the principles of the disclosure may be applied to any other immunoconjugates.

[0153] Moreover, all statements herein reciting principles, aspects, and embodiments of the disclosure, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents as well as equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure.

[0154] Thus, for example, it will be appreciated by those skilled in the art that the block diagrams presented herein may represent conceptual views of illustrative circuitry embodying the principles of the disclosure. Similarly, it will be appreciated that any flow charts, flow diagrams, and the like represent various processes which may be substantially represented in computer readable media and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.

[0155] The functions of the various elements shown in the figures may be provided through the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, a single shared processor, or a plurality of individual processors, some of which may be shared.

[0156] It should be understood that the elements shown in the figures may be implemented in various forms of hardware, software or combinations thereof Preferably, these elements are implemented in a combination of hardware and software on one or more appropriately programmed general-purpose devices, which may include a processor, memory and input / output interfaces.EXAMPLESExample 1 : General method

[0157] Using a large cohort of matched Patient-Derived Organoids (PDOs), longitudinal clinical data, transcriptomic and WES data, Inventors explore the use of large PDOs collections to characterize ADC efficacy and affinity with the targeted cells through systematic mapping of the relationship between target antigen abundance and payload efficacy.

[0158] A proof-of concept is provided, based on the workflow established in Figures 1 to 4, for determining the efficacy of antibody-drug conjugates (ADC) targeting the HER2 (Human Epidermal growth factor Receptor 2) antigen, and the Tumor-associated calcium signal transducer 2 (TROP2) antigen.

[0159] This yields an immunoconjugate (e.g. ADC) efficacy quadrant where one can map individual patient-derived organoid models as a function of the antigen abundance in this specific model and the sensitivity to the payload in the same model. Put together, this allows to identify four sub-groups : PDO profiles expressing low levels of the antigen and low payload efficacy, PDO profiles with high antigen expression but low payload sensitivity, PDO profiles with low antigen expression and high payload sensitivity and finally PDO profiles with high antigen expression and high payload sensitivity (see Figure 1).

[0160] The workflow exemplified in Figure 3 shows two steps. In the first step, a target antigen, in particular a target antigen expressed at the cell surface. For example, for a given set of antigen(s), the RNA abundance in the PDO is compared with the RNA abundance in healthy tissues and in non-adjacent tissues from other parts of the body. The relevance of the insights gathered from this step are ensured by the improved correspondence between the transcriptomic profiles of patients and of patient-derived organoid models compared to usual preclinical models. The result of this work is a list of candidate antigen receptors that over-express the antigen.

[0161] In parallel, a plurality of candidate payloads, in an isolated form, are screened on the same collection of PDOs. This yields drug sensitivities for every PDO and for every payload candidate. Of note, for some payloads, drugs of the same family or with the samemechanism of action, may also be assessed in parallel. The workflow is summarized in Figure 4.

[0162] This yields an immunoconjugate (e.g. ADC) efficacy quadrant (see Figure 1) where individual patient-derived organoid models can be mapped as a function of the antigen abundance in this specific model and the sensitivity / efficacy to the payload in the same model; taking advantage of the fact that the PDOs are associated to patient avatars.

[0163] Put together, this enables the identification of a plurality of sub-groups, depending on the number of threshold values characterizing the antigen abundance and efficacy / sensitivity of the pay load, for example four sub-groups : PDO profiles expressing low levels of the antigen and low payload efficacy, PDO profiles with high antigen expression but low payload sensitivity, PDO profiles with low antigen expression and high payload sensitivity and finally PDO profiles with high antigen expression and high payload sensitivity.

[0164] This latter category corresponds to patient profiles that are expected to respond in the clinic to the input immunoconjugate(s), in particular ADCs. By measuring the percentage of patients in the fourth category (“likely responders” we have a robust estimation of the responder population to a given antigen-drug target.

[0165] The inventors validated this approach with specific antigen-payload couples corresponding to existing ADCs for PDOs for which the data had been collected. More specifically, for a series of ADCs the inventors have identified the target antigen and the payload, and for multiple PDOs, the inventors mapped the expected efficacy based on antigen abundance and payload sensitivity. Then, the inventors verified that the predicted response of the PDO to the ADC corresponded to the predicted response.

[0166] To mine for new target-payload couples, the process of identifying the percentage of likely responders across a range of target input and payload inputs was repeated. This systematic mapping yields a heatmap with the percentage of likely responders for each target and each payload. This percentage of likely responders is highly predictive of the efficacy of the immunoconjugates.Example 2: Application of the general methodMaterial and Methods

[0167] PDO transcriptomic profiling

[0168] A collection of PDOs was assembled with matched transcriptomic data from 135 colorectal (CRC) and pancreatic (PDAC) cancer patients, generated from primary tumour samples and metastatic biopsies. Patients ranged from 31 to 89 years old and had experienced an average of 2.6 lines of treatment prior to the PDO-generating tissue sampling.

[0169] mRNAs from PDOs were extracted and sequenced to quantify expression levels of well-known target antigens such as CEACAM5, HER2, and TROP2 as proxies for protein concentration — while accounting for differences in expression between tumors and adjacent tissues.

[0170] Cell viability assay

[0171] Fully grown patient-derived organoids were dissociated using TiypLE (Thermo Fisher Scientific; 12605010), filtered to obtain a single-cell suspension, and pelleted. Cells were resuspended in culture medium containing 2% Matrigel (Coming; 356231) and lOpM Y-27632 (Bio-Techne; 1254 / 10), and dispensed into a 384 well plate (Revvity; 6057802) at a density of 1500 cells in 50 pl per well. Culture plates were incubated at 37°C with 5% CO2 for two days to allow small organoids to form. On day 2 post-seeding, 25 pl of fresh culture medium and drugs were added to each well. On day 7, 25 pl of CellTiter-Glo (Promega; G7570) were added, culture plates were shaken 5 min on an orbital shaker and incubated at room temperature for 20 min. Luminescence was read using the Agilent Biotek Synergy LX reader.

[0172] To ensure clinical relevance, the maximal ADC concentration (MedchemExpress; HY-141598) was set according to the maximal plasma concentration (Cmax) observed in patients, and a dilution series of 12 concentrations covering six orders of magnitude was used. An ADC isotype (MedchemExpress; HY-164152) was used with the same concentration range. Pay load concentrations (MedchemExpress; HY- 1363 ID) were calculated from the ADC concentration range using the drug-to-antibody ratio(DAR), such that the tested payload concentrations were molar equivalents of the ADC-delivered payload.

[0173] The solvent for the ADC and isotype was PBS-0.3% Tween20, while the payload solvent was DMSO. Therefore, wells containing either PBS-0.3% Tween20 or DMSO were also dispensed as negative controls. Each condition was dispensed in triplicates using the D-300e digital dispenser (Tecan).

[0174] The mean luminescence for each concentration was normalised to the corresponding solvent control well to calculate percentage viability. Percentage viability was then plotted as a function of drug concentration, and a sigmoidal dose-response curve was fitted for each compound to calculate EC50 and IC50 values.Results

[0175] Combined with PDO responses to a subset of common payloads, the Inventors used gene expression levels of well-known target antigens to explore the relationship between target antigen abundance and payload efficacy. This allows identification of the relationship between two of the three main vectors of ADC efficacy.

[0176] Data-driven hypothesis were empirically validated on a pool of CRC and PDAC PDOs for certain couples in particular. For instance, a Sacituzumab-govitecan conjugate, predicted as efficient by the above-described method, was confirmed to be cytotoxic for PDOs, with an IC50 ranging from 0.0000759 to 0.000126 mg / mL (Figure 5A-B).

[0177] In contrast, an ADC predicted as inefficient e.g. a famtrastuzumab-deruxtecan conjugate did not display any cytotoxicity for a PDO (Figure 6).

[0178] mRNA expression levels of TROP2 were quantified as proxies for protein concentration to classify PDO lines as high- or low-expressing lines (Figure 7). Based on this stratification, a TROP2-targeting ADC with a cleavable linker (Datopotamab-deruxtecan) was selected to assess ADC efficacy in PDO lines classified as TROP2-high (PDO-A) and TROP2-low (PDO-B).

[0179] Two PDO lines with high and low TROP2 expression level respectively displayed high (PDO-A: IC50<10'8M, Figure 8A) and low (PDO-B: IC50>10'6M, Figure 8B) sensitivity to this TROP2-targeting ADC, while being similarly sensitive tothe payload alone (IC50 <10'8M) or conjugated to an isotype control (IC50 >10'6M) (Figure 8A-B).

[0180] Therefore, these experimental results are consistent with computational predictions on ADCs predicted as ineffective or predicted as effective (e.g. below an IC50 of 0.001 mg / mL or below 10 nM, which is compatible for drug development).

Claims

46CLAIMS1. A computer-implemented method for determining the efficacy of an input immunoconjugate (10) with respect to PDOs comprising at least an input antigen-binding site (11) and an input payload (12); the method comprising the following steps:- providing organoid response data (21) indicative of efficacies of the input payload (12) on a respective plurality of patient-derived organoids, PDOs, of a patient population, and derived by bringing the input pay load (12) into contact with the respective PDOs;- providing expression data (20) indicative of levels of expression of antigens expressed by organoid-derived cells of the respective PDOs; and wherein the antigens are capable of binding to the input antigen-binding site (11); determining the efficacy of the input immunoconjugate (10) based on the organoid response data (21) and the expression data (20).

2. The computer-implemented method according to claim 1, wherein the efficacy of the input immunoconjugate (10) is compared with the efficacy of a reference immunoconjugate (14).

3. The computer-implemented method according to claim 1 or claim 2, wherein the immunoconjugates (10) also comprises at least an input linker (13), and the method comprises providing linker data (23), indicative of efficacies of the input linker (13).

4. The computer-implemented method according to any of the preceding claims further comprising, providing internalization rate data and / or recycling rate data (22) indicative of the efficacies of the input antigen-binding site (11).

5. The computer-implemented method according to any of the preceding claims, wherein the step of determining the efficacy further comprises determining the input immunoconjugate (10) as efficient when at least a predetermined portion (34) of the patient population shows respective levels of expression above a minimum reference47level of expression (35) and shows an efficacy of the input payload (12) above a minimum efficacy threshold (36).

6. The computer-implemented method according to any of the preceding claims, wherein the efficacy of the input immunoconjugate (10) is measured by bringing the input immunoconjugate (10) into contact with the respective PDOs and testing the level of expression of the input antigen and the level of viability of the input payload (12).

7. The computer-implemented method any of the preceding claims further comprising:- providing a data store comprising expression data (20), organoid response data (21), and subject data associated with PDOs from subjects;- selecting the patient population as a subset of the PDOs that are affected by the same or similar tumour.

8. The computer-implemented method according to any of the preceding claims, wherein the patient-derived organoids (PDO) are selected from a group consisting of cerebral organoids, gastrointestinal organoids, lingual organoids, tooth organoids, thyroid organoids, thymic organoids, testicular organoids, prostate organoids, hepatic organoids, pancreatic organoids, epithelial organoid, lung organoids, kidney organoids, gastruloid organoids, blastoid organoids, endometrial organoids, cardiac organoids, retinal organoids, breast cancer organoids, colorectal cancer organoids, glioblastoma organoids, neuroendocrine tumor organoids, myelin organoids, blood-brain barrier (BBB) organoids, and ovarian organoids.

9. The computer-implemented method according to any of the preceding claims; wherein the levels of expression is measured on antigens expressed at a surface of the organoid-derived cells; and, for example, wherein the level of expression of antigens is determined by a method selected from the group consisting of immunohistochemistry, mass cytometry, and combinations thereof.4810. The computer-implemented method according to any of the preceding claims; wherein the expression data (20) comprises at least one of the following data selected from the group consisting of a proteomic data, a transcriptomic data, a genomic data, an epigenomic data, a metabolic data, a microbiome data, an imaging data, a histological data, a cytotoxic data, a cytostatis data, and a combination thereof11. The computer-implemented method according to any of the preceding claims further comprises, providing reference expression data (24) indicative of levels of expression of the antigens expressed by reference non tumour cells.

12. The computer-implemented method according to any of the preceding claims, wherein the organoid response data (21) comprises cytotoxicity data.

13. The computer-implemented method according to any of the preceding claims, wherein the input immunoconjugate (10) is recovered or manufactured completely or in part.

14. A computer-implemented method for selecting an optimal immunoconjugate from a set of input immunoconjugates, comprisingdetermining the efficacy of the set of input immunoconjugates according to the method according to claims 1 to 13,selecting the optimal immunoconjugate based on the efficacy of the set of input immunoconjugates.

15. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the methods according to claims 1 to 14.

16. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the methods according to claims 1 to 14.

17. A device for determining the efficacy of an input immunoconjugate (10) against PDOs comprising at least an input antigen-binding site (11) and an input payload (12); the device comprising:at least one input configured to receive:• an organoid response data (21) indicative of efficacies of the input payload (12) on a respective plurality of patient-derived organoids, PDOs, of a patient population, and derived by bringing the input payload (12) into contact with the respective PDOs;• an expression data (20) indicative of levels of expression of antigens expressed by organoid-derived cells of the respective PDOs; and wherein the antigens are capable of binding to the input antigen-binding site (11); at least one processor configured to:• determine the efficacy of the input immunoconjugate (10) against PDOs based on the organoid response data (20) and the expression data. - At least one output configured to provide the efficacy of the input immunoconjugate (10) against PDOs.

Images