T cell quantitative systems pharmacology model

Abstract

The method can include determining, by at least one data processor, a distribution of the plurality of T cell phenotypes over time based on a set of cell kinetic parameters corresponding to the plurality of T cell phenotypes and transport rates of the plurality of T cell phenotypes between a peripheral tissue and lymph node compartment (i.e., a healthy tissue compartment), a blood compartment, a tumor-draining lymph node compartment, and a tumor compartment. The plurality of T cell phenotypes can include stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells. Related methods and articles of manufacture are also disclosed.

Description

[Technical Field]

[0001] cross reference This application claims priority to U.S. Provisional Patent Application No. 63 / 415,649, filed October 12, 2022, the disclosure of which is incorporated herein by reference in its entirety.

[0002] The present disclosure relates generally to quantitative systems pharmacology models, and more particularly to T cell quantitative systems pharmacology models. [Background technology]

[0003] T cell receptor (TCR)-engineered T cell therapy is an emerging cancer treatment strategy that has shown evidence of antitumor activity in both solid tumors and hematological cancers. Despite initial promise, many challenges remain in understanding and characterizing the unique cellular dynamics, including trafficking, proliferation, apoptosis, and persistence, of TCR-engineered T cells, among other T cells and T cell therapies, following infusion into patients. Because TCR-engineered T cell therapies are composed of live, phenotypically diverse T cells, the pharmacological properties of these therapies differ from molecular therapies, and monitoring and predicting the pharmacokinetics and resulting pharmacodynamics of these therapies poses distinct challenges. Furthermore, it is not yet known how the cell phenotype and composition of the infused product may affect cellular pharmacokinetics and subsequent efficacy and safety. Summary of the Invention

[0004] Methods, systems, and articles of manufacture, including computer program products, are provided for a T cell quantitative systems pharmacology model. In one aspect, a method is provided, the method comprising: generating a model that simulates the distribution of each of a plurality of T cell phenotypes in a plurality of physiological compartments over time after delivery of a T cell targeted (TCT) product to a patient; generating a plurality of digital twins with the model, each of the plurality of digital twins representing the distribution of the plurality of T cell phenotypes in the plurality of physiological compartments over time associated with a corresponding patient, the corresponding patient having a set of patient characteristics; receiving the set of characteristics associated with a target patient; and selecting a subset of digital twins from the plurality of digital twins that resemble the target patient. To achieve this, the method may include matching target patient characteristics with patient characteristics associated with the digital twin; predicting the target patient response to multiple virtual deliveries of a TCT product with different doses and compositions of T cell phenotypes using a selected subset of the digital twin for the target patient; and generating a treatment plan for the target patient based on the predicted response provided by the subset of the digital twin, wherein the treatment plan is predicted to provide a T cell persistence level above a predetermined threshold over time in at least one of a plurality of physiological compartments of the target patient, and the treatment plan includes doses and compositions of the multiple T cell phenotypes of the treatment.

[0005] In some variations, the method further includes identifying sources of inter-patient variability by visualizing the parameter space for a second subset of the plurality of digital twins, and modifying the model to account for the sources of inter-patient variability.

[0006] In some variations, the method further includes generating a plurality of simulations across different T cell phenotypic compositions and dose levels with the model, and visualizing a distribution of each of the plurality of T cell phenotypes over time associated with the plurality of simulations, wherein the visualization shows the effect of T cell phenotypic composition on T cell persistence levels over time.

[0007] In some variations, the set of patient characteristics includes patient biometrics, patient medical history, and baseline biomarker data.

[0008] In some variations, each of the multiple digital twins further represents a response to a dose of a composition of the TCT product within the multiple compartments.

[0009] In some variations, each of the plurality of digital twins comprises a set of cellular kinetic parameters comprising at least one of abundance, proliferation rate, trafficking rate, apoptosis rate, and differentiation rate of a plurality of T cell phenotypes in at least one of a plurality of physiological compartments of a corresponding patient over a period of time.

[0010] In some variations, the TCT product composition includes an initial amount of each of multiple T cell phenotypes.

[0011] In some variations, the multiple T cell phenotypes include at least two of stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells.

[0012] In some variations, the multiple physiological compartments include a peripheral tissue and lymph node compartment, a blood compartment, a tumor-draining lymph node compartment, and a tumor compartment.

[0013] Methods, systems, and articles of manufacture, including computer program products, are provided for T cell quantitative systems pharmacology models. In one aspect, a method is provided. The method may include determining, by at least one data processor, a set of cell kinetic parameters corresponding to a plurality of T cell phenotypes in a peripheral tissue and lymph node compartment (i.e., a healthy tissue compartment), a blood compartment, a tumor-draining lymph node compartment, and a tumor compartment of a patient after delivery of a T cell-targeted (TCT) product. The plurality of T cell phenotypes include stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells. The method may include determining, by the at least one data processor, a first transport rate of the plurality of T cell phenotypes between the peripheral tissue and lymph node compartment and the blood compartment of the patient. The method may include determining, by the at least one data processor, a set of cell kinetic parameters corresponding to the plurality of T cell phenotypes in the blood compartment. The method may include determining, by the at least one data processor, a second transport rate of the plurality of T cell phenotypes between the blood compartment and the tumor-draining lymph node compartment of the patient. The method may include determining, by at least one data processor, a set of cell kinetic parameters corresponding to a plurality of T cell phenotypes in a tumor-draining lymph node compartment. The method may include determining, by at least one data processor, a third transport rate of effector memory T cells and effector T cells from the blood compartment to a tumor compartment of the patient. The method may include determining, by at least one data processor, a set of cell kinetic parameters corresponding to effector memory T cells and effector T cells in the tumor compartment. The method may include determining, by the at least one data processor, a distribution of each of the plurality of T cell phenotypes over time in the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment), the blood compartment, the tumor-draining lymph node compartment, and the tumor compartment based on the set of cell kinetic parameters, the first transport rate, the second transport rate, and the third transport rate.

[0014] In some variations, the method can optionally include one or more features disclosed herein, including the following features, in any operable combination: In some variations, the method includes determining, by at least one data processor, a differentiation rate parameter corresponding to a rate of differentiation of effector memory T cells into effector T cells within the tumor compartment; Determining the distribution is further based on the differentiation rate parameter.

[0015] In some variations, the method includes determining, by at least one data processor, differentiation rate parameters corresponding to differentiation of stem-like memory T cells into central memory T cells, differentiation of central memory T cells into effector memory T cells, and differentiation of effector memory T cells into effector T cells within the tumor-draining lymph node compartment, wherein determining the distribution is further based on the differentiation rate parameters.

[0016] In some variations, the method includes determining a T cell therapy for treating the tumor, wherein the T cell therapy includes a dose of a multi-T cell phenotypic composition of TCT products.

[0017] In some variations, the T cell therapy is at least one of T cell receptor (TCR) engineered T cell therapy, autologous T cell therapy, allogeneic T cell therapy, iPSC-derived T cell therapy, and CAR T cell therapy.

[0018] In some variations, the TCT product includes doses of compositions of multiple T cell phenotypes.

[0019] In some variations, the tumor compartment comprises the tumor site of the tumor.

[0020] In some variations, the distributions include a first distribution corresponding to doses of the TCT product composition administered to the first patient and a second distribution corresponding to doses of the TCT product composition administered to the second patient. The method includes determining, by at least one data processor, a response of a third patient to a T cell therapy comprising the doses of the TCT product composition based on at least the first distribution and the second distribution.

[0021] In some variations, the method includes determining a treatment plan for the third patient based at least on the third patient's response to the T cell therapy.

[0022] In some variations, the dose of the TCT product composition administered to a first patient is administered to the first patient after administering a lymphocyte-depleting regimen to the first patient, and the dose of the TCT product composition administered to a second patient is administered to the second patient after administering another lymphocyte-depleting regimen.

[0023] In some variations, the set of cell kinetic parameters includes at least one of the following: abundance, proliferation rate, apoptosis rate, and differentiation rate of a plurality of T cell phenotypes.

[0024] In one aspect, a method includes determining, by at least one data processor, a first patient profile representing a first response to a first dose of a first composition of a T cell targeted (TCT) product within multiple physiological compartments of a first patient. The method includes determining, by at least one data processor, a second patient profile representing a second response to a second dose of a second composition of a TCT product within multiple physiological compartments of a second patient. The method includes generating, by the at least one data processor, an output indicative of a pharmacokinetic-pharmacodynamic (PKPD) relationship for the TCT product based on at least the first patient profile and the second patient profile. The method includes determining, by the at least one data processor, a third patient's response to a T cell therapy including a third dose of a third composition of a TCT product based at least on the output. The method includes determining a treatment plan for the third patient based on the third patient's response to a T cell therapy including at least a third dose of the third composition of a TCT product.

[0025] In some variations, the first patient profile includes a first set of cell kinetic parameters including at least one of a first amount, a first proliferation rate, a first trafficking rate, a first apoptosis rate, and a first differentiation rate of a plurality of T cell phenotypes in at least one of a plurality of physiological compartments of the first patient over a period of time, and the second patient profile includes a second set of cell kinetic parameters including at least one of a second amount, a second proliferation rate, a second trafficking rate, a second apoptosis rate, and a second differentiation rate of a plurality of T cell phenotypes in at least one of a plurality of physiological compartments of the second patient over a period of time.

[0026] In some variations, the first composition comprises a first amount of each of a plurality of T cell phenotypes, and the second composition comprises a second amount of each of the plurality of T cell phenotypes.

[0027] In some variations, the first patient profile and the second patient profile each include a response corresponding to each of a plurality of T cell phenotypes.

[0028] In some variations, the multiple T cell phenotypes include at least two of stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells.

[0029] In some variations, the multiple physiological compartments include a peripheral tissue and lymph node compartment, a blood compartment, a tumor-draining lymph node compartment, and a tumor compartment.

[0030] In some variations, the T cell therapy is at least one of T cell receptor (TCR) engineered T cell therapy, autologous T cell therapy, allogeneic T cell therapy, iPSC-derived T cell therapy, and CAR T cell therapy.

[0031] In some variations, a linear increase from a first dose to a second dose results in a non-linear change between the first response and the second response.

[0032] In some variations, the first composition is different from the second composition. The first dose is different from the second dose.

[0033] In some variations, determining the response in the third patient to the T cell therapy includes simulating a plurality of responses to a plurality of doses of a plurality of compositions of the TCT product in a plurality of simulated patients based at least on the output, wherein the response in the third patient is one of the plurality of simulated responses.

[0034] In some variations, the first patient profile is determined after administering a first lymphocyte-depleting regimen to a first patient, and the second patient profile is determined after administering a second lymphocyte-depleting regimen to a second patient.

[0035] In some variations, the response in the third patient is determined by varying at least one of the first dose, the first composition, the second dose, and the second composition.

[0036] In some variations, simulating multiple responses is based on the application of a lymphocyte depletion regimen to multiple simulated patients.

[0037] In one aspect, a system is provided. The system may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, result in operations. The operations may include determining a set of cell kinetic parameters corresponding to a plurality of T cell phenotypes in a peripheral tissue and lymph node compartment (i.e., a healthy tissue compartment) of a patient after delivery of a T cell targeted (TCT) product. The plurality of T cell phenotypes may include stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells. The operations may include determining a first transport rate of the plurality of T cell phenotypes between the peripheral tissue and lymph node compartment (i.e., the healthy tissue compartment) and a blood compartment of the patient. The operations may include determining a set of cell kinetic parameters corresponding to the plurality of T cell phenotypes within the blood compartment. The operations may include determining a second transport rate of the plurality of T cell phenotypes between the blood compartment and a tumor-draining lymph node compartment of the patient. The operations may include determining a set of cell kinetic parameters corresponding to a plurality of T cell phenotypes within a tumor-draining lymph node compartment. The operations may include determining a third transport rate of effector memory T cells and effector T cells from the patient's blood compartment to a tumor compartment. The operations may include determining a set of cell kinetic parameters corresponding to effector memory T cells and effector T cells within the tumor compartment. The operations may include determining, by at least one data processor, a distribution of each of the plurality of T cell phenotypes over time in the peripheral tissue and lymph node compartment, the blood compartment, the tumor-draining lymph node compartment, and the tumor compartment based on the set of cell kinetic parameters, the first transport rate, the second transport rate, and the third transport rate.

[0038] In another aspect, a system is provided. The system may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, result in operations. The operations may include determining, by the at least one data processor, a first patient profile representing a first response to a first dose of a first composition of a T cell targeted (TCT) product within a plurality of physiological compartments of a first patient. The operations include determining, by the at least one data processor, a second patient profile representing a second response to a second dose of a second composition of the TCT product within a plurality of physiological compartments of a second patient. The operations include generating, by the at least one data processor, an output indicative of a pharmacokinetic-pharmacodynamic (PKPD) relationship for the TCT product based at least on the first patient profile and the second patient profile. The operations include determining, by the at least one data processor, a response of a third patient to a T cell therapy including a third dose of a third composition of the TCT product based at least on the output. The method includes determining a treatment regimen for the third patient based on the third patient's response to a T cell therapy comprising at least a third dose of a third composition of TCT products.

[0039] In another aspect, a computer program product is provided that includes a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium may include program code that, when executed by at least one processor, causes operations. The operations may include determining a set of cell kinetic parameters corresponding to a plurality of T cell phenotypes in a peripheral tissue and lymph node compartment (i.e., a healthy tissue compartment) of a patient after delivery of a T cell-targeted (TCT) product. The plurality of T cell phenotypes may include stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells. The operations may include determining a first transport rate of the plurality of T cell phenotypes between the peripheral tissue and lymph node compartment (i.e., the healthy tissue compartment) and a blood compartment of the patient. The operations may include determining a set of cell kinetic parameters corresponding to the plurality of T cell phenotypes within the blood compartment. The operations may include determining a second transport rate of the plurality of T cell phenotypes between the blood compartment and a tumor-draining lymph node compartment of the patient. The operations may include determining a set of cell kinetic parameters corresponding to a plurality of T cell phenotypes within a tumor-draining lymph node compartment. The operations may include determining a third transport rate of effector memory T cells and effector T cells from the patient's blood compartment to a tumor compartment. The operations may include determining a set of cell kinetic parameters corresponding to effector memory T cells and effector T cells within the tumor compartment. The operations may include determining, by at least one data processor, a distribution of each of the plurality of T cell phenotypes over time in the peripheral tissue and lymph node compartment, the blood compartment, the tumor-draining lymph node compartment, and the tumor compartment based on the set of cell kinetic parameters, the first transport rate, the second transport rate, and the third transport rate.

[0040] In another aspect, a computer program product is provided that includes a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium may include program code that, when executed by at least one processor, causes operations to be performed. The operations may include determining, by at least one data processor, a first patient profile representing a first response to a first dose of a first composition of a T cell targeted (TCT) product within a plurality of physiological compartments of a first patient. The operations include determining, by the at least one data processor, a second patient profile representing a second response to a second dose of a second composition of the TCT product within a plurality of physiological compartments of a second patient. The operations include generating, by the at least one data processor, an output indicative of a pharmacokinetic-pharmacodynamic (PKPD) relationship for the TCT product based at least on the first patient profile and the second patient profile. The operations include determining, by the at least one data processor, a response of a third patient to a T cell therapy including a third dose of a third composition of the TCT product based at least on the output. The method includes determining a treatment regimen for the third patient based on the third patient's response to a T cell therapy comprising at least a third dose of a third composition of TCT products.

[0041] Implementations of the present subject matter may include methods according to the descriptions provided herein, as well as articles comprising tangibly embodied machine-readable media operable to cause one or more machines (e.g., computers, etc.) to perform operations that implement one or more of the described features. Similarly, computer systems are described, which may include one or more processors and one or more memories coupled to the one or more processors. The memory, which may include a non-transitory computer-readable or machine-readable storage medium, may include, encode, or store one or more programs that cause the one or more processors to perform one or more of the operations described herein. Computer-implemented methods consistent with one or more implementations of the present subject matter may be implemented by one or more processors in a single computing system or in multiple computing systems. Such multiple computing systems may be connected, for example, via one or more connections, including connections over a network (e.g., the Internet, a wireless wide area network, a local area network, a wide area network, a wired network, etc.), via a direct connection between one or more of the multiple computing systems, and may exchange data and / or instructions or other instructions, etc.

[0042] Details of one or more variations of the subject matter described herein are set forth in the accompanying drawings and the description below. Other features and advantages of the subject matter described herein will be apparent from the description and drawings, and from the claims. While certain features of the presently disclosed subject matter are described for illustrative purposes in connection with a T cell quantitative systems pharmacology model, it should be readily understood that such features are not intended to be limiting. The claims following this disclosure define the scope of the protected subject matter. [Brief explanation of the drawings]

[0043] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate certain aspects of the subject matter disclosed herein and, together with the description, serve to explain certain of the principles associated with the disclosed implementations.

[0044] [Figure 1] 1 depicts an exemplary T cell phenotyping system consistent with implementations of the present subject matter.

[0045] [Figure 2] 1 depicts an exemplary architecture of a QSP model consistent with implementation of the present subject matter.

[0046] [Figure 3] 1 depicts an exemplary graph showing cell count over time, consistent with implementations of the present subject matter.

[0047] [Figure 4] 1 depicts an exemplary graph showing a set of cellular kinetic parameters across T cell phenotypes, e.g., a single simulation showing a model that captures known dynamics of T cell therapy over time, consistent with implementation of the present subject matter.

[0048] [Figure 5] 1 depicts an exemplary comparison of cellular pharmacokinetics in blood consistent with implementation of the present subject matter.

[0049] [Figures 6A-6B] 1 depicts an exemplary comparison of cellular pharmacokinetics in blood across various dose groups and patients consistent with implementation of the present subject matter.

[0050] [Figure 7] 1 depicts an exemplary graph showing changes in cellular pharmacokinetics based on phenotypic composition and dose level of a T cell product consistent with implementations of the present subject matter.

[0051] [Figure 8]1 depicts an exemplary graph showing changes in cellular pharmacokinetics based on phenotypic composition and dose level of a T cell product consistent with implementations of the present subject matter.

[0052] [Figure 9] 1 depicts a flowchart showing an example of a process for generating a distribution of T cell phenotypes consistent with implementations of the present subject matter.

[0053] [Figure 10] 1 depicts a flowchart showing an example of a process for determining patient response to T cell therapy consistent with implementation of the present subject matter.

[0054] [Figure 11] 1 depicts a block diagram illustrating an example of a computing system consistent with implementations of the present subject matter.

[0055] [Figure 12] 1 depicts exemplary parameter space ridge plots and principal component analysis of the digital twin revealing sources of variation between patients and across dose groups (A-O).

[0056] [Figure 13] Illustrates exemplary biological variations (patient-specific) affecting cellular dynamics of TCR-engineered T cell therapy resulting in sustained or non-sustained outcomes.

[0057] [Figure 14] 1 depicts an exemplary graph showing that dose composition affects cell kinetics of TCR-engineered T cells.

[0058] [Figure 15] 1 depicts exemplary predictive simulations of a digital twin demonstrating alignment with profiles observed in patients with available clinical data.

[0059] [Figure 16]16 depicts a flowchart illustrating an example of a process 1600 for generating a personalized treatment plan using a model and a digital twin, consistent with implementations of the present subject matter.

[0060] In practice, like labels are used in the drawings to refer to the same or similar items. DETAILED DESCRIPTION OF THE INVENTION

[0061] As mentioned, there are many challenges in understanding and characterizing the unique cellular dynamics of T cell therapies, such as when T cell therapies are unconventional in that they contain live, phenotypically diverse T cells, their pharmacological properties differ from those of molecular therapies, and monitoring and predicting their pharmacokinetics and resulting pharmacodynamics presents distinct challenges. In such cases, traditional pharmacokinetic-pharmacodynamic (PKPD) approaches may not be sufficient. Furthermore, it is not yet known how the cellular phenotype and composition of the infused product may affect cellular pharmacokinetics (e.g., cell kinetics) and subsequent efficacy and safety. For example, there is currently a lack of appropriate animal models and preclinical data to understand the translation of T cell therapies, such as engineered TCR T cell therapies. Furthermore, the cellular dynamics of such T cell therapies can be significantly affected by variations in the administered dose, composition, and patient characteristics and / or makeup.

[0062] To address these challenges, the quantitative systems pharmacology (QSP) model architecture provided herein, consistent with implementations of the present subject matter, accounts for variations in dose and / or composition of T cell therapy, as well as patient variability, to accurately predict the distribution of multiple T cell phenotypes and / or patient response in various physiological compartments over time. For example, the QSP model described herein analyzes stem-like memory T cells (T cells) across physiological compartments, including tissue and lymph node compartments, blood compartments, tumor-draining lymph node compartments, and tumor compartments. scm ), central memory T cells (T cm ), effector memory T cells (T em), effector T cells (T eff The QSP model architecture can track T cell pharmacokinetics of different subsets of TCR-engineered T cells, such as endogenous T cells (T cells), and / or endogenous T cells. Each subpopulation undergoes homeostatic proliferation, antigen-driven proliferation and differentiation, apoptosis, margination and trafficking between and / or within physiological compartments. Furthermore, as described herein, the architecture of the QSP model captures effector memory T cells and effector T cell populations that infiltrate the tumor compartment, where the effector T cells can kill tumor cells in the tumor compartment. The architecture of the QSP model described herein can be used to ... endogenous T cells (T cells), and / or endogenous T cells. endo By incorporating the proliferation of TCR-engineered T cells, potential competition between immune cells following infusion of TCR-engineered T cells can also be captured. As described herein, the QSP model was calibrated to phase I clinical trial data of TCR-engineered T cells targeting E7 in epithelial cancer patients. The QSP model described herein can be adapted to other cancer antigens and tumor types. Thus, the QSP model described herein accurately predicts the impact of various doses and T cell phenotype ratios (e.g., compositions) of various T cell therapies.

[0063] 1 is a block diagram depicting an example system 100 comprising a client-server architecture and network configured to perform various methods described herein. A platform (e.g., machines and software interoperating, possibly via a series of network connections, protocols, application-level interfaces, etc.) in the form of a server platform 120 provides server-side functionality to one or more client nodes 102 and / or 106 via a communications network 114 (e.g., the Internet or other type of wide area network (WAN), such as a wireless or private network with additional security appropriate to the tasks performed by the users).

[0064] A client node (e.g., client node 102 and / or client node 106) may be, for example, a user device (e.g., a portable electronic device, a stationary electronic device, etc.). A client node may be associated with and / or accessible to a user. In another example, a client node may be a computing device (e.g., a server) accessible to and / or associated with an individual or entity. A node may include a network module (e.g., a network adapter) configured to transmit and / or receive data. Multiple users and / or servers may communicate and exchange data via nodes in a computer network. In some embodiments, a client node may facilitate transmission of patient data to platform 120 for further processing.

[0065] 1, the client node 102 hosts a web extension 104, thus allowing a user to access functionality provided by the server platform 120, e.g., receiving one or more visualizations of a treatment plan from the server platform 120. The web extension 104 may be compatible with any web browser application used by a user of the client node. Additionally, FIG. 1 shows another client node 106, e.g., hosting a mobile application 108, thus allowing a user to access functionality provided by the server platform 120, e.g., receiving one or more visualizations of a treatment plan from the server platform 120. The delivery of the visualizations may be via wired or wireless communication modes.

[0066] In at least some examples, server platform 120 may be one or more computing devices or systems, storage devices, and other components that include or facilitate the operation of the various executive modules shown in Figure 1. These modules may include, for example, model generation engine 122, digital twin generation engine 124, matching engine 126, treatment plan generation engine 128, data access module 142, analytics engine 110, and data storage 150. Each of these modules is described in more detail below.

[0067] The model generation engine 122 may generate a model that may capture the biomedical mechanisms of the TCT product after injection into a patient's body. In some embodiments, the model generation engine 122 may generate the model in cooperation with the analysis engine 110. The model generation operations performed by the analysis engine 110 and the model generation engine 122 are described in further detail elsewhere herein. The digital twin generation engine 124 may facilitate the generation of a family of digital twins. In some embodiments, the digital twin generation engine 124 may generate a family of digital twins in cooperation with the analysis engine 110. In some embodiments, the digital twins may be designed in a manner such that each patient from a clinical trial is matched to a specific subset of digital twins, for example, by the matching engine 126. For example, using the model 1202, the digital twin generation engine 124 may generate hundreds, thousands, or more digital twins. In some embodiments, each of the multiple digital twins may represent the distribution of multiple T cell phenotypes in multiple physiological compartments over time associated with a corresponding patient, and the corresponding patient may be associated with a set of patient characteristics. Matching engine 126 may receive a set of target patient characteristics from one or more of client nodes 102 or 106 and may match the target patient characteristics with patient characteristics associated with the digital twins to select a subset of digital twins from multiple digital twins that are similar to the target patient. Treatment plan generation engine 128 may cooperate with other engines / modules of server platform 120 to generate treatment plans for one or more target patients. The treatment plan generation operations performed by treatment plan generation engine 128 are described in further detail elsewhere herein.

[0068] The data access module 142 may facilitate access to the data storage 150 of the server platform 120 by any of the remaining modules / engines 110, 122, 124, 126, and 128 of the server platform 120. In one example, one or more of the data access modules 142 may be a database access module or any type of data access module that can store data in and / or retrieve data from the data storage 150 according to the needs of the particular module 110, 122, 124, 126, and 128 that uses the data access module 142 to access the data storage 150. Examples of the data storage 150 include, but are not limited to, one or more data storage components, such as magnetic disk drives, optical disk drives, solid-state disk (SSD) drives, and other forms of non-volatile and volatile memory components.

[0069] Data storage 150 may store input clinical data and / or one or more decisions / models / digital twins created and / or generated by the remaining modules / engines 110, 122, 124, 126, and 128 of server platform 120. Data storage 150 may comprise a graph database, a time series database, a relational database, or a combination thereof to efficiently manage and organize a large amount of data related to T cell therapy. This data encompasses various aspects of patient response to TCT products, including a comprehensive set of patient profiles / digital twins representing the distribution of T cell phenotypes within different physiological compartments over time.

[0070] In some embodiments, graph databases may be utilized due to their inherent ability to efficiently model and manage intricate relationships within complex datasets. For example, as discussed elsewhere herein, T cell therapy studies may involve a complex web of relationships between various data entities, including different T cell phenotypes, the distribution and / or concentrations of different T cell phenotypes in different physiological compartments, and cell kinetic parameters. A graph database may be able to represent this interconnected data structure. In some embodiments, a graph database may be utilized by using nodes and edges to represent entities and their relationships, thus modeling the interconnectivity of different components in the T cell therapy domain. For example, nodes may represent T cell phenotypes, compartments, and parameters, while edges indicate the relationships and interactions between them. Additionally or alternatively, graph databases may provide flexibility in querying and traversing relationships. For example, a graph database may visualize how different T cell phenotypes affect each other, how different T cell phenotypes move between compartments, and how changes in cell kinetic parameters affect distribution over time. This may provide insight into the dynamics of T cell responses.

[0071] In some embodiments, a time series database may be useful for processing data that evolves over time, such as the distribution of multiple T cell phenotypes in multiple physiological compartments over time. A time series database may capture the temporal and dynamic aspects of patient response, allowing visibility into changes in T cell phenotypes within physiological compartments as treatment progresses. For example, a time series database may be utilized to store and query time-stamped data points, which may facilitate tracking of cell kinetic parameters over time. In some embodiments, a relational database may be utilized to store structured clinical data and metadata associated with T cell therapy, facilitating linking patient profiles to specific treatment regimens, laboratory results, and patient demographics. A relational database may improve data integrity and enable direct retrieval of specific information.

[0072] FIG. 2 schematically depicts an exemplary architecture 200 of a QSP model 1202 consistent with implementations of the present subject matter. As previously described, the QSP model 1202 depicts the in vivo kinetics of lymphocyte (e.g., T cell) proliferation following treatment with a T cell-targeted (TCT) product consistent with embodiments of the present subject matter. In other words, the architecture 200 of the QSP model 1202 may represent the pharmacokinetic-pharmacodynamic relationship of a TCT product. The TCT product may additionally and / or alternatively increase anti-tumor activity in both solid tumors and hematological cancers. For example, the TCT product may be delivered to a patient as part of a T cell therapy to treat tumors, cancer, or the like. The T cell therapy may be at least one of T cell receptor (TCR)-engineered T cell therapy, autologous T cell therapy, allogeneic T cell therapy, induced pluripotent stem cell ("iPSC")-derived T cell therapy, and chimeric antigen receptor ("CAR") T cell therapy.

[0073] A TCT product may be composed of multiple T cell phenotypes that proliferate within various physiological compartments of a patient, as described herein. For example, a TCT product may include a dose of a composition of multiple T cell phenotypes. Referring to Figure 2, multiple T cell phenotypes may include stem-like memory T cells (T scm)210, central memory T cells (T cm )212, effector memory T cells (T em ) 214, effector T cells (T eff ) 216, and / or endogenous T cells 219. The particular T cell phenotypes of the plurality of T cell phenotypes included in architecture 200 were selected to improve the accuracy of predictions generated based on QSP model 1202, more accurately track T cell phenotypic behavior within and between physiological compartments, and more accurately determine the distribution of T cell phenotypes. Thus, in some implementations, the plurality of T cell phenotypes included in QSP model 1202 are only and / or all of stem-like memory T cells 210, central memory T cells 212, effector memory T cells 214, effector T cells 216, and / or endogenous T cells 219. In other implementations, the plurality of T cell phenotypes included in QSP model 1202 are only and / or all of stem-like memory T cells 210, central memory T cells 212, effector memory T cells 214, and effector T cells 216.

[0074] In some implementations, the TCT product comprises a live drug. In other words, the TCT product comprises live, phenotypically diverse T cells. Thus, the actual composition and / or proportion of each of the multiple T cell phenotypes comprising the TCT product may vary from patient to patient and / or between doses delivered to the same patient. The TCT product delivered to each patient may have a dose containing a certain amount of T cells. The dose of the TCT product may vary based on the patient. Additionally and / or alternatively, a dose of the TCT product may be delivered multiple times to a patient as part of a treatment regimen. Each dose delivered to a patient may be the same and / or may vary. In other words, each dose may have a dose level containing the same amount of T cells or a different amount of T cells. For example, a dose may be about 10 9 T cells, 10x10 9 T cells, 100x10 9 The doses may include doses of T cells, etc. Thus, there may be a 10-fold difference in the amount of T cells included in each dose level corresponding to each dose.

[0075] The TCT product can be delivered to a patient after the patient has been administered a lymphocyte-depleting regimen, which can help prolong the persistence of multiple T cell phenotypes in the TCT product delivered to the patient and enhance the efficacy of the delivered T cell therapy.

[0076] 2 , architecture 200 of QSP model 1202 includes multiple physiological compartments. For example, architecture 200 includes peripheral tissue and lymph node compartments (i.e., healthy tissue compartments) 202, blood compartments 204, tumor-draining lymph node compartments 208, and tumor compartments 206. Tumor compartment 206 may include a tumor site of a tumor configured to be treated with a TCT product. The specific physiological compartments included in architecture 200 were selected to improve the accuracy of predictions generated based on QSP model 1202, more accurately track T cell behavior within and between multiple physiological compartments, and more accurately determine the distribution of multiple T cell phenotypes. Thus, in some implementations, the multiple physiological compartments included in QSP model 1202 are only and / or all of peripheral tissue and lymph node compartments (i.e., healthy tissue compartments) 202, blood compartment 204, tumor-draining lymph node compartments 208, and tumor compartment 206.

[0077] Upon delivery of the TCT product to a patient, multiple T cell phenotypes migrate to each of multiple physiological compartments. For example, in some embodiments, at least some or all of the multiple physiological compartments comprise multiple T cell phenotypes. For example, the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202, the blood compartment 204, the tumor-draining lymph node compartment 208, and the tumor compartment 206 may comprise stem-like memory T cells 210, central memory T cells 212, effector memory T cells 214, effector T cells 216, and / or endogenous T cells 219.

[0078] In some embodiments, each T cell phenotype of the plurality of T cell phenotypes is associated with a set of cell kinetic parameters that describe the behavior of the T cell phenotype after delivery of the TCT product to the patient. The set of cell kinetic parameters includes one or more cell kinetic parameters. For example, the set of cell kinetic parameters includes at least one of the amount, proliferation (e.g., expansion) rate, apoptosis rate, and differentiation rate of the plurality of T cell phenotypes. The values ​​of each cell kinetic parameter of the one or more cell kinetic parameters corresponding to each T cell phenotype of the plurality of T cell phenotypes within each physiological compartment may be different from each other.

[0079] As one example, FIG. 3 illustrates a graph 300 showing T cell counts over time, consistent with implementation of the present subject matter. As shown in graph 300, after delivery of a TCT product to a patient, the amount of T cells declines before rapidly expanding. After reaching a maximum amount of T cells, the amount of T cells declines slowly over time. As another example, FIG. 4 depicts an exemplary graph 400 showing the effect of a set of cell kinetic parameters on T cell phenotype over time, consistent with implementation of the present subject matter. In particular, graph 400 shows the concentration of effector T cells over time after delivery of a TCT product to a patient. For example, graph 400 shows T cell behavior, including trafficking (e.g., margination), proliferation (e.g., expansion), peak proliferation (e.g., peak expansion), apoptosis and redistribution, and persistence over time, of effector T cells in physiological compartments after delivery of a TCT product in a patient. A set of cell kinetic parameters in a T cell phenotype affects the behavior of the corresponding T cell phenotype over time, as illustrated by graph 400. While graph 400 shows T cell behavior over time corresponding to an effector T cell, any of the T cell phenotypes described herein can also experience T cell behavior including trafficking (e.g., margination), proliferation (e.g., expansion), peak proliferation (e.g., peak expansion), apoptosis and redistribution, and persistence, which can be represented by a set of cell kinetic parameters (e.g., changes in a set of cell kinetic parameters over time).

[0080] By way of example, referring to graph 400, after delivery of the TCT product, the effector T cells undergo margination 402, followed by rapid proliferation 404, as the concentration of effector T cells increases before reaching peak expansion 406. In some implementations, the rate of proliferation, peak expansion 406, etc., may indicate the effectiveness of the T cell therapy. After reaching peak expansion 406, the effector T cells undergo apoptosis and redistribution 408 as the effector T cells migrate between physiological compartments and undergo apoptosis. Finally, the effector T cells experience persistence 410. Persistence 410 (e.g., length of persistence, etc.) may additionally and / or alternatively indicate the effectiveness of the T cell therapy.

[0081] As mentioned, the set of cell kinetic parameters can include differentiation of multiple T cell phenotypes. In some implementations, as shown in FIG. 2 , architecture 200 includes differentiation of effector memory T cells 214 into effector T cells 216 at 240 after the effector memory T cells 214 migrate to tumor compartment 206. Differentiation of effector memory T cells 214 into effector T cells 216 within the tumor compartment can increase the amount of effector T cells for infiltrating tumor 218 at 242. Additionally, architecture 200 can include differentiation of multiple T cell phenotypes within tumor-draining lymph node compartment 208. Within the tumor-draining lymph node compartment 208, architecture 200 includes antigen-driven differentiation of stem-like memory T cells 210 into central memory T cells 212 at 244, differentiation of central memory T cells 212 into effector memory T cells 214 at 246, and differentiation of effector memory T cells 214 into effector T cells 216 at 248. Activation of multiple T cell phenotypes within the tumor-draining lymph node compartment 208 can be triggered by antigen-presenting cells that present tumor-derived antigens. The antigen-presenting cells activate multiple T cell phenotypes within the tumor-draining lymph node compartment 208, which enables effector T cells 216 to recognize and infiltrate the tumor 218.

[0082] 2, architecture 200 of QSP model 1202 may include the transport of multiple T cell phenotypes between each of multiple physiological compartments. For example, architecture 200 may include the transport (e.g., via transport or margination rates) of multiple T cell phenotypes between peripheral tissue and lymph node compartments (i.e., healthy tissue compartments) 202 and blood compartments 204, between blood compartments 204 and tumor-draining lymph node compartments 208, and between blood compartments 204 and tumor compartments 206. In other words, architecture 200 may include transport of multiple T cell phenotypes from peripheral tissue and lymph node compartments (i.e., healthy tissue compartments) 202 to blood compartments 204 (e.g., via transport rate or margination rate), from blood compartments 204 to peripheral tissue and lymph node compartments (i.e., healthy tissue compartments) 202, from blood compartments 204 to tumor-draining lymph node compartments 208, from tumor-draining lymph node compartments to blood compartments 204, and / or from blood compartments 204 to tumor compartments 206.

[0083] 2 , architecture 200 includes transport of stem-like memory T cells 210 between a peripheral tissue and lymph node compartment (i.e., a healthy tissue compartment) 202 and a blood compartment 204 at 220, and transport of stem-like memory T cells 210 between the blood compartment 204 and a tumor-draining lymph node compartment 208 at 222. Architecture 200 may additionally and / or alternatively include transport of central memory T cells 212 between the peripheral tissue and lymph node compartment (i.e., a healthy tissue compartment) 202 and the blood compartment 204 at 224, and transport of central memory T cells 214 between the blood compartment 204 and a tumor-draining lymph node compartment 208 at 226. Architecture 200 may additionally and / or alternatively include transport of effector memory T cells 214 between the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202 and the blood compartment 204 at 228, transport of effector memory T cells 214 between the blood compartment 204 and the tumor-draining lymph node compartment 208 at 230, and transport of effector memory T cells 214 between the blood compartment 204 and the tumor compartment 206 at 230. Architecture 200 may additionally and / or alternatively include transport of effector T cells 216 between the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202 and the blood compartment 204, and transport of effector T cells 216 from the blood compartment 204 to the tumor compartment 206 and from the tumor-draining lymph node compartment 208 to the blood compartment 204 at 232. As shown in FIG. 2, architecture 200 may further include and / or include transport of endogenous T cells 219 between peripheral tissue and lymph node compartments (i.e., healthy tissue compartments) 202 and blood compartments 204, and transport of endogenous T cells 219 between blood compartments 204 and tumor-draining lymph node compartments 208.

[0084] Referring again to FIG. 2 , architecture 200 includes proliferation (e.g., proliferation rates, e.g., antigen-driven proliferation rates and / or homeostatic proliferation rates) for each of multiple T cell phenotypes within each physiological compartment (e.g., peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202, blood compartment 204, tumor compartment 206, and tumor-draining lymph node compartment 208).

[0085] For example, within the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202, architecture 200 includes proliferation rates (e.g., homeostatic proliferation rates) 250 corresponding to stem-like memory T cells 210, proliferation rates (e.g., homeostatic proliferation rates) 252 corresponding to central memory T cells 212, proliferation rates (e.g., homeostatic proliferation rates) 254 corresponding to effector memory T cells 214, and proliferation rates (e.g., homeostatic proliferation rates) 256 corresponding to effector T cells 216. Within the blood compartment, architecture 200 includes proliferation rates (e.g., homeostatic proliferation rates) 258 corresponding to stem-like memory T cells 210, proliferation rates (e.g., homeostatic proliferation rates) 260 corresponding to central memory T cells 212, proliferation rates (e.g., homeostatic proliferation rates) 262 corresponding to effector memory T cells 214, and proliferation rates (e.g., homeostatic proliferation rates) 264 corresponding to effector T cells 216. Within tumor-draining lymph node compartment 208, architecture 200 includes proliferation rates (e.g., antigen-driven proliferation rates) 266 corresponding to stem-like memory T cells 210, proliferation rates (e.g., antigen-driven proliferation rates) 268 corresponding to central memory T cells 212, proliferation rates (e.g., antigen-driven proliferation rates) 270 corresponding to effector memory T cells 214, and proliferation rates (e.g., homeostatic proliferation rates) 272 corresponding to effector T cells 216. Within tumor compartment 206, architecture 200 includes proliferation rates (e.g., antigen-driven proliferation rates) 274 corresponding to effector memory T cells 214 and proliferation rates (e.g., homeostatic proliferation rates) 276 corresponding to stem-like memory T cells 216.

[0086] Referring again to FIG. 2 , architecture 200 includes apoptosis (e.g., apoptosis rates) for each of multiple T cell phenotypes within each physiological compartment (e.g., peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202, blood compartment 204, tumor compartment 206, and tumor-draining lymph node compartment 208). For example, within peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202, architecture 200 includes apoptosis rate 278 corresponding to stem-like memory T cells 210, apoptosis rate 280 corresponding to central memory T cells 212, apoptosis rate 282 corresponding to effector memory T cells 214, and apoptosis rate 284 corresponding to effector T cells 216. Within the blood compartment, architecture 200 includes apoptosis rate 286 corresponding to stem-like memory T cells 210, apoptosis rate 288 corresponding to central memory T cells 212, apoptosis rate 290 corresponding to effector memory T cells 214, and apoptosis rate 292 corresponding to effector T cells 216. Within tumor-draining lymph node compartment 208, architecture 200 includes apoptosis rate 294 corresponding to stem-like memory T cells 210, apoptosis rate 296 corresponding to central memory T cells 212, apoptosis rate 298 corresponding to effector memory T cells 214, and apoptosis rate 299 corresponding to effector T cells 216. Within tumor compartment 206, architecture 200 includes apoptosis rate 297 corresponding to effector memory T cells 214 and apoptosis rate 295 corresponding to stem-like memory T cells 216.

[0087] As previously mentioned, the QSP model 1202 may be calibrated based on clinical data. In some implementations, the cell kinetic parameters described herein, such as the amount, proliferation rate, apoptosis rate, differentiation rate, and trafficking rate, may be determined based on at least clinical data. The calibrated QSP model 1202 is used to measure the amount of cytotoxicity at three different dose levels (approximately 10 9 , about 10 10 , about 10 11We were able to reproduce the dynamics of multiple T cell phenotypes targeting E7 in epithelial cancer patients administered E7 at doses of 10 ...

[0088] In some implementations, the model generation engine 122 may determine the behavior of multiple T cell phenotypes within each of multiple physiological compartments (e.g., peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202, blood compartment 204, tumor compartment 206, and tumor-draining lymph node compartment 208) at various time points, at various doses (e.g., dose levels) of the TCT product, and / or at various compositions of the TCT product. In some implementations, the model generation engine 122 may determine a corresponding T cell phenotypic distribution of the multiple T cell phenotypes in each of the physiological compartments over time, such as based at least on the determined behavior of the multiple T cell phenotypes within each of the multiple physiological compartments at various time points and between the multiple physiological compartments at various time points. In some implementations, the model generation engine 122 may determine the persistence of a corresponding T cell phenotype of the plurality of T cell phenotypes in each of the plurality of physiological compartments over time, for example, based at least on the determined behavior of the plurality of T cell phenotypes within each of the plurality of physiological compartments at various times and the determined behavior of the plurality of T cell phenotypes between the plurality of physiological compartments at various times and / or based on the determined distribution.

[0089] For example, the analysis engine 110 or the model generation engine 122 can determine a distribution or patient profile (including a set of cell kinetic parameters, patient response, etc.) for each patient based on the determined behavior of multiple T cell phenotypes within and between physiological compartments. Figure 5 depicts an exemplary comparison of cellular pharmacokinetics in blood consistent with an implementation of the present subject matter. In particular, Figure 5 shows a comparison of the behavior of multiple T cell phenotypes over time (captured at various time points) across three doses and various patients. As described herein, the composition of the TCT product delivered to each patient was kept constant; however, a constant composition of the TCT product may have some variability (e.g., within a threshold range) because, at least, the TCT product may contain live drug. Furthermore, variation in patient type or patient composition may also affect the variability of the results (as shown in Figure 5). However, by varying the dose and incorporating a constant composition, the impact of dose level on cellular pharmacokinetics can be captured and determined by the analysis engine 110 or the model generation engine 122, for example, based on architecture 200. The architecture 200 may also account for such variations in the "constant" composition of a patient's TCT product and / or composition.

[0090] For example, in Figure 5, the first row of the graph corresponds to a low dose of TCT product delivered to a first group of patients, including two patients. 9 The second row of the graph corresponds to the medium dose of TCT product delivered to a second group of patients, which included three patients. The medium dose was approximately 10x10 9 The third row of the graph corresponds to a high dose of TCT product delivered to a third group of patients, which included five patients. The high dose was approximately 100x10 9Each TCT product contained a composition of T cells. Thus, various dose levels (e.g., low dose, medium dose, high dose) correspond to the amount and / or concentration of T cells in the TCT product delivered to the patient. Each line on the graph corresponds to a different patient. As previously discussed, the variation in the behavior of each T cell in each patient in response to delivery of the TCT product at each dose level is due, at least in part, to the constant composition of the TCT product and the variation in the specific makeup of the patient. In Figure 5, the first column corresponds to the behavior of stem-like memory T cells 210, the second column corresponds to the behavior of central memory T cells 212, the third column corresponds to the behavior of effector memory T cells 214, and the fourth column corresponds to the behavior of effector T cells 216.

[0091] The distributions in each patient at each dose level may allow for the determination of important drivers of cell population values ​​over time, which may then be used by the analysis engine 110 or the model generation engine 122 and / or as part of the architecture 200 to accurately determine the dose and / or composition of TCT products for treating a particular patient as part of a T cell therapy, such as T cell receptor (TCR)-engineered T cell therapy, autologous T cell therapy, allogeneic T cell therapy, iPSC-derived T cell therapy, CAR T cell therapy, or engineered T cell therapy w / TCRs targeting various tumor antigens. For example, the determined distributions and / or patient profiles described herein may allow for the determination of the impact of variations in one or more of the set of cell kinetic parameters and / or trafficking described herein on effector T cell numbers, which may determine the efficacy of a particular treatment. In some implementations, cell kinetic parameters such as proliferation and trafficking rates of stem-like memory T cells 210 and / or trafficking, proliferation, and differentiation rates of effector memory T cells 214 can be important drivers for accurately determining and capturing patient profiles, such as those with high levels of cell proliferation or cytoreduction. Thus, based at least on architecture 200, analysis engine 110 or model generation engine 122 can predict blood and tissue distribution and pharmacokinetics of T cells (e.g., multiple T cell phenotypes) following administration of T cell therapy in patients with solid tumors. Analysis engine 110 or model generation engine 122 can also leverage architecture 200 to capture observed pharmacokinetics and efficacy of T cells (e.g., multiple T cell phenotypes) in preclinical animal species, such as mice, to aid in translational efforts of the T cell therapies described herein.

[0092] In some embodiments, as described elsewhere herein in connection with Figures 2-5, a QSP model 1202 may be developed to simulate the dynamic profiles of various T cell phenotypes in the context of cell therapy treatment. In some embodiments, the QSP model 1202 may be carefully designed to align with the relevant theoretical foundations of biological processes. Thus, the QSP model 1202 may encapsulate fundamental biomedical mechanisms governing T cell biology in physiologically relevant compartments, encompassing both natural T cells within a patient and TCR-engineered T cells introduced as part of a treatment. These compartments include key regions within a patient's body, including the central blood compartment, healthy tissue, tumor compartment, and tumor-draining lymph node compartment. The model 1202 may integrate essential biological mechanisms inherent in T cell behavior, such as antigen-induced proliferation and differentiation, cell migration and trafficking, and the crucial processes of homeostatic proliferation and apoptosis. The QSP model 1202 may be formulated using a system of ordinary differential equations (ODEs), which serve as a mathematical basis for describing and predicting the in vivo dynamics of different T cell phenotypes. In some embodiments, these phenotypes are consistent with the endogenous T cells (T) naturally present in the patient's system throughout the treatment regimen. endo ), as well as stem-like memory T cells (T) derived from TCR-engineered T cell therapy. scm ), central memory T cells (T cm ), effector memory T cells (Tem), and effector T cells (T eff ).

[0093] In some embodiments, the digital twin generation engine 124 may generate a digital twin with the aid of the QSP model 1202. For example, by utilizing the QSP model 1202, the digital twin generation engine 124 may generate a patient-specific digital twin that replicates the circulating T cell dynamics demonstrated in a clinical trial involving TCR-engineered T cells against E7 in patients suffering from metastatic HPV-associated epithelial cancer. In some embodiments, analysis of the key factors influencing the cell dynamics and differences between these digital twins may provide the following conclusions: 1. Stem cell-like memory T cells (Tscm ) may serve as a crucial determinant of both T cell expansion and persistence. scm Differences associated with T may play a substantial role in explaining the observed variability in cell dynamics between patients. Details of analysis using digital twins are described in more detail elsewhere herein. By conducting virtual clinical trials through computational modeling using digital twins, it is possible to determine the T in administered products. scm It is anticipated that increasing the presence of KRAS G12D will improve the durability of engineered T cells, allowing for the utilization of lower dose levels. Additionally or alternatively, the present disclosure explores the broader applicability of QSP model 1202, digital twins, and T cell therapy by predicting the kinetics of two patients with pancreatic cancer who were subjected to KRAS G12D-targeted T cell therapy. scm Validate insights into the significance of enrichment.

[0094] In some embodiments, QSP model 1202 characterizes the cellular dynamics of T cells within the bloodstream, various tissues, tumor-draining lymph nodes, and tumor environment, encompassing five different T cell phenotypes. In some embodiments, model calibration was performed using clinical data from E7-targeted TCR-engineered T cells to generate a reference virtual patient. Subsequently, in some embodiments, utilizing the same dataset, digital twin generation engine 124 can facilitate the generation of a group of digital twins, with each patient from a clinical trial being matched to a specific subset of digital twins, e.g., by matching engine 126. For example, using model 1202, digital twin generation engine 124 can generate hundreds, thousands, or more digital twins. In some embodiments, each of the multiple digital twins may represent the distribution of multiple T cell phenotypes in multiple physiological compartments over time associated with a corresponding patient, which may be associated with a set of patient characteristics. In some embodiments, matching engine 126 may receive a set of target patient characteristics and may match the target patient characteristics with patient characteristics associated with digital twins to select a subset of digital twins from the plurality of digital twins that are similar to the target patient. In some embodiments, matching engine 126 may calculate a matching score for each digital twin in the collection for the target patient. Matching engine 126 may then select the top N digital twins with the highest scores. This may facilitate the selection of a subset of digital twins from the collection of digital twins that are similar to the target patient among the available digital twins.

[0095] In some embodiments, a set of patient characteristics, including patient biometrics, patient medical history, and baseline biomarker data, may be used as the basis for calculating a matching score. This matching score is used to evaluate and determine the similarity between a target patient and a group of digital twins. Calculating the matching score may include different methods, such as biometric-driven scoring, medical history-driven scoring, biomarker data-driven scoring, integrated scoring that considers all characteristics, weighted scoring based on characteristic relevance, dynamic adjustment of scoring criteria, and threshold-based scoring. These examples may provide a universal approach to facilitate the selection of digital twins that closely match the characteristics of a target patient across a range of clinical scenarios, enabling predictive simulation of patient response.

[0096] In some embodiments, for each target patient, the matching engine 126 may select a small number of digital twin virtual patients, e.g., 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, etc. digital twin virtual patients. In some embodiments, each of these selected digital twins may be created to represent a distinct combination of underlying biological parameter values ​​with the goal of closely replicating the patient's observed kinetics when receiving the corresponding clinically administered dose and composition. Given the inherent uncertainty in biological parameters, the use of multiple digital twins per patient may offer numerous advantages, such as allowing for explicit consideration of alternative parameterizations of the underlying biology that may remain consistent with the observed data. Additionally or alternatively, this approach may accommodate scenarios in which different predictions may occur under untested protocols, different dosing regimens, or variations in the cellular composition of the administered product.

[0097] These digital twins have proven useful, for example, for generating predictions regarding alternative dosing strategies via the treatment plan generation engine 128. In some embodiments, a selected subset of the digital twin for a target patient is utilized to generate predictions regarding the target patient's response to multiple virtual deliveries of TCT products. These virtual deliveries encompass a variety of different doses and compositions of T cell phenotypes. Through this approach, comprehensive insights into potential responses and outcomes can be generated, aiding in informed decision-making for treatment strategies. Furthermore, in some embodiments, parametric analysis is performed to gain insights into the biological mechanisms that may drive divergence between individual patients or sustained cell persistence over time. In some embodiments, the digital twins have been subjected to simulations including alternative dosing strategies, thus enabling predictions regarding the impact of dose composition and dosage on cellular kinetics. In some embodiments, predictions generated by the digital twin have been validated through a distinct TCR-engineered T cell therapy scenario, specifically targeting KRAS G12D in patients diagnosed with pancreatic cancer.

[0098] 6A and 6B depict exemplary comparisons of cell pharmacokinetics in blood across various dose groups and patients, consistent with implementations of the present subject matter. In particular, FIGS. 6A and 6B show individualized virtual patients characterizing variability across dose levels, TCT treatment compositions, and patients. In FIGS. 6A and 6B, each row corresponds to a different patient, and each column of graphs corresponds to a particular T cell phenotype. For example, the first column of graphs corresponds to stem-like memory T cells 210, the second column of graphs corresponds to central memory T cells 212, the third column of graphs corresponds to effector memory T cells 214, the fourth column of graphs corresponds to effector T cells 216, and the fifth column of graphs corresponds to endogenous T cells. Each line on each graph corresponds to a different dose of TCT product (e.g., amount and / or concentration of T cells) delivered to each patient, and the composition of the TCT product (e.g., specific ratios or proportions of various T cell phenotypes, e.g., stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells) delivered to each patient.

[0099] Based on the architecture 200 of the QSP model 1202, the analysis engine 110 or the model generation engine 122 may determine various distributions (e.g., graphs representing patient profiles corresponding to each of multiple T cell phenotypes) as shown in FIGS. 6A and 6B. Based on the determined distributions (e.g., patient profiles), the analysis engine 110 or the model generation engine 122 may predict the dynamics of T cell subsets or T cell phenotypic composition in the TCT product and the impact of each composition or subset on anti-tumor efficacy over time. Thus, the architecture 200 may enable virtual clinical trials and / or experiments to be conducted with the aid of a digital twin. Additionally and / or alternatively, the architecture 200 may enable the analysis engine 110 or the model generation engine 122 to determine the impact of variations in lymphodepletion efficiency and / or protocol on anti-tumor efficacy, T cell phenotypic composition of the TCT product, and / or dose and regimen of T cell therapy administration based on the architecture 200.

[0100] Additionally and / or alternatively, based on at least architecture 200 and / or the determined patient profiles of at least one, two, or more patients, analysis engine 110 or treatment plan generation engine 128 may predict an individual patient response (or range thereof) to a given TCT product by simulation based on cellular kinetic parameters determined by individual patient baseline biomarker data (data collected by genomic, transcriptomic, and / or flow cytometric methods). For example, analysis engine 110 may determine a response in a target patient to a T cell therapy including a given TCT product by at least simulating multiple responses to multiple doses of a composition (or multiple compositions) of the TCT product in multiple simulated patients (as shown in FIGS. 6A and 6B ) based on at least architecture 200 and / or the determined patient profiles (e.g., distributions). The multiple responses may be simulated by at least varying the dose, composition, etc. of the TCT product.

[0101] The response in the target patient may be one of multiple simulated responses. For example, at least one simulated patient profile (e.g., distribution, response, etc.) corresponding to a patient having similar characteristics, configuration, diagnosis, tumor, cancer, etc. to the target patient may be selected (e.g., by the analysis engine 110 or the matching engine 126). Based on the selected simulated patient profile, an effective T cell therapy, including a dose and / or composition of a TCT product, may be determined and / or administered to the target patient to treat the tumor and / or cancer. This allows for the determination of an ideal dose and / or composition of a TCT product for treating the tumor and / or cancer of the target patient. In some implementations, the simulated patient profile can be used to assess both the safety, such as with respect to cytokine release syndrome (CRS), and the efficacy of a particular T cell therapy, including a particular dose and / or composition of a TCT product and regimen.

[0102] Furthermore, as shown in Figures 6A and 6B, the digital twin generation engine 124 can capture the different cell kinetics exhibited by each patient in response to various doses and compositions of TCR-engineered T cells (Figures 6A and 6B, patient IDs 2 and 11 are excluded from the dataset due to unavailability of data). Digital twins resembling reference patients reproduce the disparities observed between different patient cohorts. Notably, the digital twins can reproduce the limited cell expansion in patients belonging to the low-dose group (e.g., IDs 1 and 3) and the more substantial expansion of Tscm and Teff cell populations in patients within the high-dose cohort (e.g., IDs 7, 8, 9, 10, and 12). Furthermore, the digital twins can capture intra-group variability. For example, within the mid-dose group, the simulation emulates the variation in cell number and kinetic profiles between patient 5 and patients 4 and 6. Initial composition differences can be identified as a source of variability in predicted patient responses.

[0103] As shown in Figures 6A and 6B, the digital twin forms a virtual population representing 10 patients. This virtual population may be used to predict cellular behavior for these 10 patients under various TCR-engineered T cell therapy regimens and compositions, thereby allowing for detailed insight into the biological mechanisms behind the observed and predicted behaviors and inter-patient variability. Each column of data provided represents a T scm ,T cm ,T em , or T effThe experimental measurements in blood, including TCR-engineered T cells and endogenous T cells (Tendo), are displayed. Each row represents the cell measurements in cells / milliliter (cells / mL) for an individual patient. As shown in Figures 6A and 6B, patients 1 and 3 received 10^9 cells, patients 4, 5, and 6 received 10^10 cells, and all other patients received 10^11 cells. While the data points correspond to experimental measurements over time, the curves represent the 10 digital twins that most closely match each patient's experimental data. The digital twins shown in Figures 6A and 6B capture the variability in cell dynamics between patients receiving treatment with TCR-engineered T cells. Notably, the digital twins reproduce the multiphasic cell dynamics observed in blood after administration of HPV-16 E7-targeting TCR-engineered T cells.

[0104] FIG. 12 depicts an exemplary parameter space ridge plot and principal component analysis of digital twins revealing sources of variation between patients and across dose groups (A-O). As shown in FIG. 12, the ridge plot can visualize the distribution for 14 parameters (one subplot per parameter) across the digital twins of each clinical patient (i.e., target patient). As shown in FIG. 12, for each subplot (i.e., represented by FIGS. 12A-12O), individual patient IDs are listed along the y-axis and the range of potential parameter values ​​is listed along the x-axis. FIG. 12P shows variation across the principal components (PC1 vs. PC2) of parameter space. In some embodiments, each dot in FIG. 12P may represent an individual digital twin, for a total of 100 digital twins.

[0105] Digital twins, featuring explicitly inferred parameterizations of the underlying biological processes governing cellular dynamics, are useful for this analysis. Ridge plots, such as those shown in Figures 12A-12O and principal component analysis (PCA) (Figure 12P), help visualize the parameter space of the digital twin associated with each patient and facilitate cross-patient comparisons.

[0106] As shown in Figure 12, several parameters show consistent distribution across patients in the ridge plots (Figures 12A-12O). For example, antigen-driven T em Proliferation (Fig. 12C) and T em Parameters associated with transformation (Figure 12N) showed similar distributions, suggesting that these processes do not significantly contribute to interpatient differences in cell kinetics. scm The rate constants governing antigen-driven proliferation of cells (i.e., kprolif_atg_scm subplot - Figure 12A) show significant variation both within and between dose groups. Within the mid-dose group, the digital twin of patient 5 exhibited a higher T, similar to values ​​observed in the high-dose cohort. scm Conversely, the digital twins of patients 4 and 6 exhibited lower T scm This finding is consistent with the finding that patient 5 exhibited higher cell numbers compared to patients 4 and 6, suggesting antigen-driven T scm This suggests that proliferation trends play an important role in interpatient variability in T cell counts.

[0107] Alternatively or additionally, the specific parameters and processes may be scm ,T cm , and T em (Figures 12F, 12G, and 12H), as well as the transport rate constants of T cmThe data show substantial inter-patient variability, including the proliferation rate constants of the endothelial cells (Figure 12B) and endogenous cells (Figure 12E). These findings may highlight the importance of these biological processes as key determinants of cell dynamics and heterogeneity. Alternatively or additionally, the analysis reveals clear inter-cohort variability among specific parameters. This is further illustrated in the PCA plot (Figure 12P), where minimal overlap is observed along the first principal component (PC1) between the low- and high-dose groups. In the PCA, parameters governing T cell proliferation and trafficking emerge as influential contributors to the variability among dose cohorts, as demonstrated by contribution analysis. These trends suggest a complex interplay between dose level and early memory T cell proliferation, demonstrating effects beyond those explicitly incorporated into the model.

[0108] In some embodiments, as shown in Figure 12, identifying sources of inter-patient variability can be achieved by visualizing the parameter space of a subset of multiple digital twins. This analysis allows for a comprehensive assessment of potential underlying factors contributing to different kinetic profiles between patients. Furthermore, model 1202 can be adapted and / or modified to account for these identified sources of inter-patient variability.

[0109] FIG. 13 depicts exemplary biological variations (patient-specific) affecting the cellular dynamics of TCR-engineered T cell therapy resulting in sustained or non-sustained outcomes. As shown in FIG. 13, a group of digital twins generated by the digital twin generation engine 124 can be analyzed in 10 10 dose of TCR-engineered T cells and 1% T scm ,3% T cm ,48% T em , and 48% T effThe cell dose composition (compositions shown in FIG. 13A) can be simulated. FIGS. 13B-13E show the cell kinetics of each T cell phenotype over time. In some embodiments, the bands shown on the graphs may represent the 25th and 75th percentiles of each subgroup, with the solid lines representing the median values ​​for each T cell phenotype. Additionally or alternatively, FIG. 13G shows a bar plot showing partial rank correlation coefficient (PRCC) scores. Each parameter is represented along the y-axis, and its relationship with outcome (persistence vs. non-persistence) is quantified. An empty bar may mean a parameter that is not statistically significant in this context.

[0110] As shown in FIG. 13, stratification of digital twins within a virtual clinical trial provides insight into key determinants affecting T cell persistence. For example, as shown in FIG. 13, the aforementioned analysis can provide visualization of the biological variation inherent across the digital twin population. In some embodiments, the digital twin can enable assessment of how inter-patient biological variation affects T cell persistence over time. As shown in the example depicted in FIG. 13, all digital twins exhibited a 1% T scm ,3% T cm ,48% T em , and 48% T eff A virtual clinical trial was simulated, reflecting a representative dose composition from the E7 clinical trial, receiving a medium dose containing 10 TCR-engineered T cells with a composition consisting of 10 TCR-engineered T cells (Figure 13A). The simulated dynamics (median and interquartile range) of various T cell phenotypes across the virtual population are depicted in Figures 13B-13E. At this constant dose and composition, T scm Cells and T eff Both cells (Fig. 13B, Fig. 13E) show an initial distribution and expansion, which then decreases across all digital twins. In contrast, T cm and T em Cells (Figures 13C, 13D) are predicted to expand less. Notably, predicted responses span nearly an order of magnitude for each T cell phenotype, highlighting the expected interpatient variability given consistent dose composition.

[0111] Given the correlation between cell expansion in the first few months and treatment efficacy in hematological malignancies, we performed stratification of digital twins based on the frequency of TCR-engineered T cells in the blood at day 50. Digital twins showing a predominant presence of TCR-engineered T cells in the blood, indicating persistence over time, were classified as "persistent" under this treatment regimen, while the remainder, showing a decline over time, were classified as "non-persistent" (Figure 13F).

[0112] A global sensitivity analysis may be performed to dig into the underlying biological processes driving sustained versus non-sustained outcomes, as shown in Figure 13. In some embodiments, T scm Transport and proliferation, T endo proliferation, and T cm Rate constants governing proliferation and transport were identified as important determinants of persistence or non-sustainability outcome (partial rank correlation coefficient values ​​are shown in Figure 13G). Notably, these parameters were also identified in previous analyses of inter-individual variability. Therefore, non-linear relationships may exist between parameters. For example, the overall T scm Because expansion depends on both transport to and proliferation within tumor-draining lymph nodes, T scm A digital twin with a lower transport rate constant should have a higher T to indicate persistence. scm The proliferation rate constant may be required, and vice versa. These differences in parameters between persistence and non-persistence become more apparent in bivariate plots. In some embodiments, these parameters appear as important contributors to inter-patient variability in cell persistence, which may indicate the need to effectively modulate these parameters and optimize treatment regimens, doses, and / or compositions to achieve the desired level of T cell persistence.

[0113] FIG. 14 shows an exemplary graph illustrating the effect of dose composition on the cytokinetics of TCR-engineered T cells. The figure shows two different sets of digital twins: one, Matched Patient 1 (i.e., associated with FIG. 14A ), and the other, Matched Patient 6 (i.e., associated with FIG. 14B ). In some embodiments, these digital twins may be re-simulated with alternative dose compositions and various dose amounts. Each row in the figure may correspond to a different dose composition or dose amount undergoing simulation. Each column may show a representation of the dose composition or cytokinetics of a TCR-engineered T cell subset. For ease of comparison, the cytokinetics of the digital twin treated with the original dose composition are replotted on the corresponding curve to provide a baseline for the cytokinetics obtained from the alternative dose simulation.

[0114] In some aspects, beyond dosage and inherent biological patient variability, dose composition can affect T cell expansion and persistence. To illustrate the implications of alternative dose compositions, simulations can be performed using different compositions for representative patients drawn from both the low-dose (patient 1) and mid-dose (patient 6) cohorts (shown in Figures 6A and 6B). Specifically, T scm Based on the hypothesis that enrichment may promote more substantial expansion and long-term persistence, T scm Cell-enriched composition (85% T scm ,5% T cm ,5% T em , and 5% T eff As shown in Figure 14, the cell kinetics after treatment with the original dose composition was compared with that of T at two different dose levels. scm Cell kinetics are compared to that predicted after treatment with concentrated doses.

[0115] The predicted cell kinetics is scm This supports the idea that treatment based on TCR-engineered T cells results in improved overall expansion and persistence of TCR-engineered T cells. As shown in Figure 14, in patient 6, T cells at a 10-fold lower dose (10^9 cells) were scm The simulated injection of concentrated product was mainly T emCells and T eff This is expected to result in comparable overall cell expansion and persistence compared to the injection of 10^10 cells of the original dose material. For patient 1, the original dose level of 10^9 cells was scm The simulated injection of concentrated material was mainly em Cells and T eff We predict an approximately 100-fold increase in T cell numbers in the bloodstream compared to the original administered material consisting of cells. Higher T scm This simulation demonstrates that the composition corresponds to a greater T cell abundance in these patients, and demonstrates the role of T cells in cell dynamics and persistence formation. scm This is consistent with previous findings regarding the important role of parameters. Notably, these insights go beyond what can be gleaned from a mere correlation analysis of the E7 clinical trial. Thus, in some embodiments, the treatment plan generation engine 128 may utilize the QSP model 1202 and the digital twin to perform optimization of dose compositions and amounts for clinical patients (i.e., target patients), ultimately aiming to enhance T cell persistence in clinical applications.

[0116] In some embodiments, after exploring optimized dose compositions and amounts using the QSP model and digital twins, these simulations and findings may be leveraged in real-world clinical scenarios. For example, the system may predict a target patient's response to multiple hypothetical deliveries of a TCT product, each delivery characterized by a distinct combination of T cell dose and composition. To predict this target patient response, a subset of digital twins resembling the characteristics of the target patient may be selected. These selected digital twins provide a predictive framework for estimating how a patient's T cell dynamics will respond to various treatment scenarios, thereby enabling a comprehensive assessment of potential outcomes.

[0117] In some embodiments, the treatment plan generation engine 128 may utilize the predicted target patient response provided by a subset of the digital twin. In this manner, the treatment plan generation engine 128 may develop a personalized treatment plan for the target patient, taking into account predicted cell dynamics and persistence levels in various physiological compartments over time. This treatment plan may be designed, for example, by the treatment plan generation engine 128 to optimize T cell persistence within the target patient, which may consistently exceed a predetermined threshold across multiple physiological compartments. In some embodiments, the treatment plan determines the specific dose and composition of the TCT product for the patient's treatment, aligning it with the expected goal of maintaining T cell levels above a defined threshold. Leveraging insights derived from the QSP model 1202 and the digital twin, this application establishes a framework for data-driven, personalized decision-making within the field of TCR-engineered T cell therapy, thereby introducing the ability to fine-tune the treatment approach for each patient, focusing on optimizing both the composition and amount of the therapeutic T cell dose. The goal is to achieve and maintain targeted T cell persistence levels, ultimately enhancing the precision and efficacy of clinical interventions in this specialty.

[0118] FIG. 15 depicts an exemplary predictive simulation of a digital twin demonstrating alignment with observed profiles in patients with available clinical data. In some embodiments, clinical data for patients with pancreatic cancer treated with TCR-engineered T cells targeting KRAS G12D may be utilized. As shown in FIG. 15, the percentage of TCR-engineered T cells relative to total T cells in the blood is shown over time for patient #1 in FIG. 15A and patient #2 in FIG. 15B. As shown in FIG. 15, 100 digital twins (represented by the lighter color curves) may be subjected to re-simulation at the dosages and compositions specified in clinical trials for KRAS G12D. The darker color curves in FIG. 15 may represent the single digital twin that most accurately reproduces each patient's observed data, as determined by the average root mean square error (RMSE).

[0119] As shown in Figure 15, the predictive ability of the digital twin in predicting T cell profiles in patients undergoing KRAS G12D-targeted TCR-engineered T cell therapy is demonstrated. In some embodiments, to evaluate the applicability of the QSP model structure and digital twin to other TCR-engineered T cell therapies and medical indications, simulations were performed using a 100-member digital twin virtual population with the same dose and composition parameters used in a clinical trial involving TCR-engineered T cells targeting KRAS G12D in patients with metastatic pancreatic cancer. The selection of the KRAS G12D clinical dataset for this evaluation may be based on its public availability, providing insight into the dose composition, administration, and dynamics of this treatment across different patients. In the context of the KRAS G12D clinical trial, both patients received a mid-dose of approximately 10 cells, composed primarily of Tem cells, although patient #2 received a larger proportion of early memory cells. Therefore, the variability in the simulation curves may reflect the inherent inter- and intra-patient variability encapsulated within the digital twin. Clinical patient trajectories match the range of predictions made by the digital twin, and in each case the best-fitting digital twin converges with the densest cluster of simulated profiles. Figure 15 further validates model 1202 and the digital twin as it demonstrates their ability to accurately predict the cellular dynamics of TCR-engineered T cells in patients with pancreatic cancer undergoing KRAS G12D-targeted therapy. This alignment between predicted and observed profiles highlights the robustness and generalizability of the model and digital twin, extending their utility beyond the specific treatment and patient cohort studied and suggesting their potential applicability for predicting T cell behavior in diverse TCR-engineered therapies and medical settings.

[0120] FIG. 7 depicts an exemplary graph showing changes in cellular pharmacokinetics based on the phenotypic composition and dose level of a T cell product, consistent with implementations of the present subject matter. As previously described, architecture 200 enables the determination of the impact of variations in dose level and phenotypic composition of a TCT product on cellular pharmacokinetics of multiple T cell phenotypes, as well as the distribution and persistence of T cell therapy over time. FIG. 7 includes a graph showing a re-simulation of a high-dose group of patients, e.g., the example shown by the third row of FIG. 5. The graph includes the previously determined behavior of T cell phenotypes (solid line) and a tested dose of a TCT product comprising a composition including a greater number of stem-like memory T cells than effector T cells (dashed line). As shown in FIG. 7, a TCT product containing a majority of stem-like memory T cells is likely to result in a higher number of total T cells remaining in the patient 200 days post-infusion compared to a simulation of a TCT product containing a majority of effector T cells.

[0121] FIG. 8 depicts an exemplary graph showing the change in cellular pharmacokinetics based on the phenotypic composition and dose level of a T cell product, consistent with an implementation of the present subject matter. In particular, the graph in FIG. 8 shows a set of simulations performed based on the QSP model 1202. The simulations shown in FIG. 8 were performed at three different dose levels (10 9 cells, 10 10 cells, and 10 11This included resimulating the cellular pharmacokinetics of patient 6 shown in the cellular pharmacokinetic comparison of FIG. 6A based on different compositions of the TCT product (cells). FIG. 9 depicts a flowchart illustrating an example of a process 900 for determining the distribution of multiple T cell phenotypes over time after delivery of a TCT product, consistent with implementations of the present subject matter. With reference to FIG. 9 , process 900 may be performed by the analysis engine 110 to generate a treatment plan, determine a T cell therapy including a TCT product dose and / or TCT product composition for treating a tumor, predict patient response to a TCT product composition dose, etc. For example, the analysis engine 110 may implement the QSP model 1202 to determine a patient profile including the distribution of T cell phenotypes of the TCT product in the patient's physiological compartments over time, for a TCT product dose and / or composition, etc. A T cell therapy, treatment plan, etc. may be determined based at least on the determined distributions. Based on architecture 200, analysis engine 110 can capture cell dynamics of multiple T cell phenotypes after administration of T cell therapy for solid tumors, etc. Consistent with implementations of the present subject matter, process 900 references architecture 200 shown in FIG.

[0122] At 902, the analysis engine 110 (e.g., at least one data processor) may determine a set of cell kinetic parameters corresponding to a plurality of T cell phenotypes within a patient's peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) (e.g., peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202). The plurality of T cell phenotypes may include stem-like memory T cells (e.g., stem-like memory T cells 210), central memory T cells (e.g., central memory T cells 212), effector memory T cells (e.g., effector memory T cells 214), effector T cells (e.g., effector T cells 216), and endogenous T cells. In some implementations, the plurality of T cell phenotypes include stem-like memory T cells 210, central memory T cells 212, effector memory T cells 214, and effector T cells 216. The TCT product may include doses of compositions of the plurality of T cell phenotypes. For example, a dose can include the total amount of T cells in the TCT product. The TCT product composition can include an amount of each T cell phenotype of multiple T cell phenotypes. In other words, the TCT product composition can include a specific proportion (or range of proportions) of each T cell phenotype of the multiple phenotypes that make up the TCT product.

[0123] The set of cell kinetic parameters includes at least one of the amount, proliferation rate, apoptosis rate, and differentiation rate of a plurality of T cell phenotypes. The set of cell kinetic parameters corresponding to a plurality of T cell phenotypes in the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202 may be determined at a plurality of time points (e.g., a first time point, a second time point, a third time point, etc.). The values ​​of the cell kinetic parameters in the set of cell kinetic parameters may be different for each T cell phenotype of the plurality of T cell phenotypes in the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202. Thus, the analysis engine 110 may track the behavior of a plurality of T cell phenotypes in the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202 at various time points.

[0124] At 904, the analysis engine 110 (e.g., at least one data processor) may determine a first transport rate of a plurality of T cell phenotypes between the patient's peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202 and a blood compartment (e.g., blood compartment 204). The first transport rate may correspond to the transport of a plurality of T cell phenotypes in lines 220, 224, 228, and 232 shown in architecture 200 of FIG.

[0125] At 906, the analysis engine 110 (e.g., at least one data processor) may determine a set of cell kinetic parameters corresponding to the plurality of T cell phenotypes in the blood compartment 204. The set of cell kinetic parameters includes at least one of the amount, proliferation rate, apoptosis rate, and differentiation rate of the plurality of T cell phenotypes in the blood compartment 204. The set of cell kinetic parameters corresponding to the plurality of T cell phenotypes in the blood compartment 204 may be determined at multiple time points (e.g., a first time point, a second time point, a third time point, etc.). The values ​​of the cell kinetic parameters in the set of cell kinetic parameters may be different for each T cell phenotype of the plurality of T cell phenotypes in the blood compartment 204. Thus, the analysis engine 110 may track the behavior of the plurality of T cell phenotypes in the blood compartment 204 at various time points.

[0126] At 908, platform 120 (e.g., at least one data processor) may determine a second transport rate of the plurality of T cell phenotypes between the patient's blood compartment 204 and a tumor-draining lymph node compartment (e.g., tumor-draining lymph node compartment 208). The second transport rate may correspond to the transport of the plurality of T cell phenotypes in lines 222, 226, 230, and 234 shown in architecture 200 of FIG.

[0127] At 910, the analysis engine 110 (e.g., at least one data processor) may determine a set of cell kinetic parameters corresponding to a plurality of T cell phenotypes in the tumor-draining lymph node compartment 208. The set of cell kinetic parameters includes at least one of quantity, proliferation, apoptosis, and differentiation of the plurality of T cell phenotypes in the tumor-draining lymph node compartment 208. Values ​​of cell kinetic parameters in the set of cell kinetic parameters may be different for each T cell phenotype of the plurality of T cell phenotypes in the tumor-draining lymph node compartment 208. In some implementations, the analysis engine 110 determines differentiation rate parameters (e.g., the set of cell kinetic parameters) corresponding to the differentiation of stem-like memory T cells 210 into central memory T cells 212, the differentiation of central memory T cells 212 into effector memory T cells 214, and the differentiation of effector memory T cells 214 into effector T cells 216 in the tumor-draining lymph node compartment 208. A set of cell kinetic parameters corresponding to multiple T cell phenotypes within the tumor-draining lymph node compartment 208 may be determined at multiple time points (e.g., a first time point, a second time point, a third time point, etc.) Thus, the analysis engine 110 may track the behavior of multiple T cell phenotypes within the tumor-draining lymph node compartment 208 at various time points.

[0128] At 912, the analysis engine 110 (e.g., at least one data processor) may determine a third transport rate of effector memory T cells 214 and effector T cells 216 from the patient's blood compartment to a tumor compartment (e.g., tumor compartment 206). The third transport rate may correspond to the transport of multiple T cell phenotypes in lines 236, 238 shown in architecture 200 of FIG. 2.

[0129] At 914, the analysis engine 110 (e.g., at least one data processor) may determine a set of cell kinetic parameters corresponding to effector memory T cells 214 and effector T cells 216 in the tumor compartment 206. The values ​​of the cell kinetic parameters of the set of cell kinetic parameters may be different for each T cell phenotype of the multiple T cell phenotypes in the tumor compartment 206. The set of cell kinetic parameters includes at least one of the amount, proliferation rate, apoptosis rate, and differentiation rate of the effector memory T cells 214 and effector T cells 216 in the tumor compartment 206. In some implementations, the analysis engine 110 determines a differentiation rate parameter (e.g., the set of cell kinetic parameters) corresponding to the differentiation rate of effector memory T cells 214 into effector T cells 216 in the tumor compartment 206. The sets of cell kinetic parameters corresponding to effector memory T cells 214 and effector T cells 216 in the tumor compartment 206 may be determined at multiple time points (e.g., a first time point, a second time point, a third time point, etc.). Thus, the analysis engine 110 can track the behavior of effector memory T cells 214 and effector T cells 216 within the tumor compartment 206 at various times.

[0130] At 916, the analysis engine 110 (e.g., at least one data processor) may determine a distribution over time of each of a plurality of T cell phenotypes in the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202, the blood compartment 204, the tumor-draining lymph node compartment 208, and the tumor compartment 206 based at least on the set of cell kinetic parameters, the first transport rate, the second transport rate, and the third transport rate. In some implementations, the analysis engine 110 may determine a persistence of a plurality of T cell phenotypes over time in the peripheral tissue and lymph node compartment (i.e., healthy tissue compartment) 202, the blood compartment 204, the tumor-draining lymph node compartment 208, and the tumor compartment 206 based at least on the set of cell kinetic parameters, the first transport rate, the second transport rate, and the third transport rate, and / or based on the determined distributions.

[0131] The distributions may include a first distribution corresponding to a dose of the TCT product composition administered to a first patient and a second distribution corresponding to a dose of the TCT product composition administered to a second patient. Based on at least the first distribution corresponding to the first patient and / or the second distribution corresponding to the second patient (among other distributions and patients), the analysis engine 110 may determine the response of another (e.g., a third) patient to a T cell therapy including the dose and composition of the TCT product. Based on at least the determined response of the third patient to the T cell therapy, the analysis engine 110 may determine a treatment plan for the third patient to treat the third patient's tumor and / or cancer. In other words, the treatment plan may be determined based at least on the predicted distributions (e.g., the first distribution and / or the second distribution) corresponding to a particular dose and / or composition of the TCT product. The T cell therapy may be administered to the patient according to the treatment plan including the determined dose and composition of the TCT product. In some implementations, a dose of the TCT product composition is administered to a first patient after administering a lymphocyte-depleting regimen to the first patient, and a dose of the TCT product composition is administered to a second patient after administering a lymphocyte-depleting regimen to the second patient. The lymphocyte-depleting regimens administered to each patient may be the same or different. Thus, in some implementations, the analysis engine 110 determines a T cell therapy for treating a tumor, including doses of the TCT product composition of multiple T cell phenotypes. For example, in some implementations, the T cell therapy can be determined based at least on a predicted distribution (e.g., a first distribution and / or a second distribution) corresponding to a particular dose and / or composition of the TCT product. As described herein, the T cell therapy can include at least one of T cell receptor (TCR)-engineered T cell therapy, autologous T cell therapy, allogeneic T cell therapy, iPSC-derived T cell therapy, and CAR T cell therapy.

[0132] FIG. 10 depicts a flowchart illustrating an example of a process 1000 for determining a patient treatment plan, consistent with implementations of the present subject matter. With reference to FIG. 10 , process 1000 may be executed by analysis engine 110 to generate a treatment plan, determine a T cell therapy including a TCT product dose and / or a TCT product composition for treating a tumor, predict a patient response to a TCT product composition dose, etc. For example, analysis engine 110 may implement QSP model 1202 to determine a patient profile including a distribution of TCT product T cell phenotypes in the patient's physiological compartments over time, for a TCT product dose and / or composition, etc. A T cell therapy, treatment plan, etc. may be determined based at least on the determined distribution. Consistent with implementations of the present subject matter, process 1000 references architecture 200 shown in FIG. 2 . Furthermore, consistent with implementations of the present subject matter, process 1000 may include one or more steps of process 900, and process 900 may include one or more steps of process 1000.

[0133] At 1002, the analysis engine 110 (e.g., at least one data processor) may determine a first patient profile representing a first response to a first dose of a first composition of a T cell target (TCT) product within a plurality of physiological compartments of a first patient. The first patient profile includes a first set of cell kinetic parameters including at least one of a first amount of a plurality of T cell phenotypes, a first proliferation rate, a first trafficking rate, a first apoptosis rate, and a first differentiation rate within at least one of the plurality of physiological compartments of the first patient over a period of time. Thus, the first patient profile may include one or more cell kinetic parameters (e.g., the first set of cell kinetic parameters) of the set of cell kinetic parameters as described herein. Further, consistent with implementations of the present subject matter, the plurality of physiological compartments include a peripheral tissue and lymph node compartment (i.e., a healthy tissue compartment) 202, a blood compartment 204, a tumor-draining lymph node compartment 208, and a tumor compartment 206.

[0134] The plurality of T cell phenotypes includes at least two of stem-like memory T cells 210, central memory T cells 212, effector memory T cells 214, effector T cells 216, and endogenous T cells. The first composition of the TCT product can include a first amount of each of the plurality of T cell phenotypes. The first patient profile includes a first response corresponding to each of the plurality of T cell phenotypes.

[0135] At 1004, the analysis engine 110 (e.g., at least one data processor) may determine a second patient profile representing a second response to a second dose of a second composition of T cell targeted (TCT) product within a plurality of physiological compartments of a second patient. The second patient profile includes a second set of cell kinetic parameters including at least one of a second amount of a plurality of T cell phenotypes, a second proliferation rate, a second trafficking rate, a second apoptosis rate, and a second differentiation rate within at least one of the plurality of physiological compartments of the second patient over a period of time. Thus, the second patient profile may include one or more cell kinetic parameters (e.g., the second set of cell kinetic parameters) of the set of cell kinetic parameters as described herein.

[0136] The plurality of T cell phenotypes includes at least two of stem-like memory T cells 210, central memory T cells 212, effector memory T cells 214, effector T cells 216, and endogenous T cells. The second composition of the TCT product can include a second amount of each of the plurality of T cell phenotypes. The second patient profile includes a second response corresponding to each of the plurality of T cell phenotypes.

[0137] In some implementations, the first composition is different from the second composition and the first dose is different from the second dose, and a linear increase from the first dose to the second dose can result in a non-linear change between the first response and the second response.

[0138] At 1006, the analytical engine 110 (e.g., at least one data processor) generates an output indicative of a pharmacokinetic-pharmacodynamic (PKPD) relationship of the TCT product based on at least the first patient profile and the second patient profile. The PKPD relationship may be illustrated generally as the architecture 200 of the QSP model 1202. The PKPD relationship may represent at least the distribution of each of a plurality of T cell phenotypes in physiological compartments of the patient over time.

[0139] At 1008, the analysis engine 110 (e.g., at least one data processor) determines a response for the third patient to a T cell therapy including a third dose of a third composition of the TCT product. As described herein, the T cell therapy is at least one of T cell receptor (TCR)-engineered T cell therapy, autologous T cell therapy, allogeneic T cell therapy, iPSC-derived T cell therapy, and CAR T cell therapy. In some implementations, determining the third patient's response to the T cell therapy includes simulating multiple responses to multiple doses of the multiple compositions of the TCT product in the multiple simulated patients based at least on the output. In some implementations, simulating the multiple responses is based on the application of a lymphocyte-depleting regimen to the multiple simulated patients. The response in the third patient can be one of the multiple simulated responses. This enables determination of an ideal dose and / or composition of the TCT product for treating the tumor and / or cancer of the third patient. Additionally, a third patient may be determined by varying at least one of the first dose, the first composition, the second dose, and the second composition, which may also allow for determination of an ideal dose and / or composition of the TCT product for treating the tumor and / or cancer of the third patient.

[0140] At 1010, the analysis engine 110 (e.g., at least one data processor) determines a treatment plan for a third patient based at least on the third patient's response to a T cell therapy including a third dose of a third composition of the TCT product. In other words, the treatment plan may be determined based on at least the predicted PKPD relationship, the first patient profile, the second patient profile, a predicted response of the third patient, etc. The T cell therapy may be administered to the third patient according to the treatment plan.

[0141] FIG. 16 depicts a flowchart illustrating an example of a process 1600 for generating a personalized treatment plan using a model and a digital twin, consistent with implementations of the present subject matter. Referring to FIG. 16 , process 1600 may be executed by platform 120 to generate a treatment plan, determine a T cell therapy including a dose of a TCT product and / or a composition of a TCT product to treat a tumor, predict a patient response to a dose of a TCT product composition, etc. For example, platform 120 may generate a model in operation 1602, which may be a QSP model, such as a QSP model generated by model generation engine 122 using architecture 200 shown in FIG. 2 . In some embodiments, the generated model may simulate the distribution of each of multiple T cell phenotypes in multiple physiological compartments over time after delivery of a T cell targeted (TCT) product to a patient. In some embodiments, process 1600 may proceed to operation 1604, where digital twin generation engine 124 may generate multiple digital twins according to the model, e.g., according to QSP model 1202. In some embodiments, each of the multiple digital twins may represent the distribution of multiple T cell phenotypes in multiple physiological compartments over time associated with a corresponding patient, the corresponding patient having a set of patient characteristics. In some embodiments, the patient characteristics may include patient biometrics, the patient's medical history, and baseline biomarker data. In some embodiments, the patient characteristics may include the type of cancer / tumor the corresponding patient has. Process 1600 may then proceed to operation 1606, where matching engine 126 may receive a set of characteristics associated with the target patient. In some embodiments, the set of characteristics may be received from client node 102 or 106, as shown in FIG. 1 . Matching engine 126 may then match the target patient characteristics with patient characteristics associated with the digital twins to select a subset of digital twins from the multiple digital twins that are similar to the target patient in operation 1608. In some embodiments, matching engine 126 may calculate a matching score for each digital twin in the group for the target patient.The matching engine 126 may then select the top N digital twins with the highest scores. This may facilitate the selection of a subset of digital twins from the group of available digital twins that are similar to the target patient. Process 1600 may then proceed to operation 1610, where the platform 120 may use the selected subset of digital twins for the target patient to predict the target patient's response to multiple virtual deliveries of TCT products with different doses and compositions of T cell phenotypes. The treatment plan generation engine 128 may then generate a treatment plan for the target patient based on the predicted responses provided by the subset of digital twins in operation 1612. In some embodiments, the treatment plan may be predicted to provide a T cell persistence level above a predetermined threshold over time in multiple physiological compartments of the target patient, where the treatment plan provides / includes multiple T cell phenotypic doses and compositions of treatment. In some embodiments, the treatment plan may be predicted to provide a T cell persistence level above a predetermined threshold over time in at least one of the multiple physiological compartments of the target patient. In some embodiments, a treatment regimen may be predicted to provide persistent levels of T cells above a predetermined threshold over time in at least the tumor compartment of a target patient.

[0142] 11 depicts a block diagram of a computing system 1100 consistent with implementations of the present subject matter. Referring to FIGS. 1-10, computing system 1100 can be used to implement analysis engine 110, QSP model 1202, and / or any components therein.

[0143] 11, computing system 1100 may include a processor 1110, a memory 1120, a storage device 1130, and an input / output device 1140. The processor 1110, the memory 1120, the storage device 1130, and the input / output device 1140 may be interconnected via a system bus 1150. The computing system 1100 may additionally or alternatively include a graphics processing unit (GPU), such as for image processing, and / or associated memory for the GPU. The GPU and / or associated memory for the GPU may be interconnected with the processor 1110, the memory 1120, the storage device 1130, and the input / output device 1140 via the system bus 1150. The memory associated with the GPU may store one or more images described herein, and the GPU may process one or more of the images described herein. The GPU may be coupled to and / or form part of the processor 1110. The processor 1110 is capable of processing instructions for execution within the computing system 1100. Such executed instructions may implement one or more components, such as, for example, the analysis engine 110, the QSP model 1202, etc. In some implementations of the present subject matter, the processor 1110 may be a single-threaded processor. Alternatively, the processor 1110 may be a multi-threaded processor. The processor 1110 is capable of processing instructions stored in the memory 1120 and / or in the storage device 1130 to display graphical information for a user interface provided via the input / output device 1140.

[0144] Memory 1120 is a computer-readable medium, such as a volatile or non-volatile medium, that stores information within computing system 1100. Memory 1120 may store, for example, data structures representing a configuration object database. Storage device 1130 may provide persistent storage for computing system 1100. Storage device 1130 may be a floppy disk drive, a hard disk drive, an optical disk drive, a tape drive, or other suitable persistent storage means. Input / output device 1140 performs input / output operations for computing system 1100. In some implementations of the present subject matter, input / output device 1140 includes a keyboard and / or a pointing device. In various implementations, input / output device 1140 includes a display unit for displaying a graphical user interface.

[0145] According to some implementations of the present subject matter, input / output devices 1140 can perform input / output operations for network devices. For example, input / output devices 1140 can include Ethernet ports or other networking ports for communicating with one or more wired and / or wireless networks (e.g., a local area network (LAN), a wide area network (WAN), the Internet).

[0146] In some implementations of the present subject matter, computing system 1100 may be used to execute various interactive computer software applications (e.g., Microsoft Excel® and / or other types of software) that may be used for organizing, analyzing, and / or storing data in various (e.g., tabular) formats. Alternatively, computing system 1100 may be used to execute any type of software application. These applications may be used to perform various functions, such as planning functions (e.g., creating, managing, editing spreadsheet documents, word processing documents, and / or any other objects), computing functions, communication functions, etc. Applications may include various add-in functions or may be standalone computing products and / or functions. When active within an application, functionality may be used to generate a user interface that is provided via input / output devices 1140. The user interface may be generated and presented to a user by computing system 1100 (e.g., on a computer screen monitor, etc.).

[0147] One or more aspects or features of the subject matter described herein may be implemented in digital electronic circuitry, integrated circuits, specially designed ASICs, field programmable gate array (FPGA) computer hardware, firmware, software, and / or combinations thereof. These various aspects or features may include implementation in one or more computer programs executable and / or interpretable on a programmable system including at least one programmable processor, which may be special-purpose or general-purpose, coupled to receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device. The programmable system or computing system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0148] These computer programs, which may also be referred to as programs, software, software applications, applications, components, or code, contain machine instructions for a programmable processor and may be implemented in a high-level procedural and / or object-oriented programming language and / or assembly / machine language. As used herein, the term “machine-readable medium” refers to any computer program product, apparatus, and / or device used to provide machine instructions and / or data to a programmable processor, such as, for example, magnetic disks, optical disks, memory, and programmable logic devices (PLDs), and includes a machine-readable medium that receives machine instructions as a machine-readable signal. The term “machine-readable signal” refers to any signal used to provide machine instructions and / or data to a programmable processor. A machine-readable medium may non-transitory store such machine instructions, such as, for example, a non-transitory solid-state memory, a magnetic hard drive, or any equivalent storage medium. Alternatively or additionally, a machine-readable medium may temporarily store such machine instructions, such as, for example, a processor cache or other random access memory associated with one or more physical processor cores.

[0149] To provide for user interaction, one or more aspects or features of the subject matter described herein may be implemented on a computer having, for example, a display device, such as a cathode ray tube (CRT) or liquid crystal display (LCD) or light-emitting diode (LED) monitor, for displaying information to a user, and a keyboard and pointing device, such as a mouse or trackball, through which the user may provide input to the computer. Other types of devices may also be used to provide for user interaction. For example, feedback provided to the user may be any form of sensory feedback, such as visual feedback, auditory feedback, or tactile feedback, and input from the user may be received in any form, including acoustic, speech, or tactile input. Other possible input devices include touchscreens or other touch-sensitive devices, such as single-point or multi-point resistive or capacitive trackpads, voice recognition hardware and software, optical scanners, optical pointers, digital image capture devices and associated interpretation software.

[0150] The subject matter described herein may be embodied in systems, devices, methods, and / or articles, depending on the desired configuration. The implementations set forth in the above description do not represent all implementations of the subject matter described herein. Instead, these implementations are merely some examples consistent with aspects associated with the described subject matter. While some variations have been described in detail above, other modifications or additions are possible. In particular, additional features and / or variations may be provided in addition to those described herein. For example, the implementations described above may be directed to various combinations and subcombinations of the disclosed features and / or combinations and subcombinations of certain additional features described above. Furthermore, the logical flows illustrated in the accompanying drawings and / or described herein do not necessarily require the particular order shown, or sequential order, to achieve desirable results. For example, the logical flows may include different and / or additional operations from those shown without departing from the scope of the present disclosure. One or more operations of the logical flows may be repeated and / or omitted without departing from the scope of the present disclosure. Other implementations may be within the scope of the following claims.

Claims

1. generating a model that simulates the distribution of each of a plurality of T cell phenotypes in a plurality of physiological compartments over time after delivery of a T cell targeted (TCT) product to a patient; generating a plurality of digital twins with the model, each of the plurality of digital twins representing a distribution of the plurality of T cell phenotypes in the plurality of physiological compartments over time associated with a corresponding patient, the corresponding patient having a set of patient characteristics; receiving a set of characteristics associated with the target patient; matching target patient characteristics with patient characteristics associated with digital twins to select a subset of digital twins from the plurality of digital twins that are similar to the target patient; predicting target patient response to multiple virtual deliveries of the TCT product with different doses and compositions of T cell phenotypes using the selected subset of digital twins for the target patient; generating a treatment plan for the target patient based on the predicted response provided by the subset of the digital twin, the treatment plan predicted to provide a T cell persistence level above a predetermined threshold over time in at least one of the plurality of physiological compartments of the target patient, the treatment plan including a dose and composition of the plurality of T cell phenotypes of the treatment; A method comprising:

2. identifying sources of inter-patient variability by visualizing parameter space for a second subset of the plurality of digital twins; modifying the model to account for sources of inter-patient variability; The method of claim 1 further comprising:

3. generating a plurality of simulations across different T cell phenotype compositions and dose levels with the model; visualizing a distribution of each of the plurality of T cell phenotypes over time associated with the plurality of simulations, wherein the visualization shows an effect of T cell phenotype composition on T cell persistence levels over time; The method of claim 1 further comprising:

4. The method of claim 1 , wherein the set of patient characteristics includes patient biometrics, patient medical history, and baseline biomarker data.

5. 10. The method of claim 1, wherein each of the plurality of digital twins further represents a response to a dose of a composition of the TCT product within the plurality of compartments.

6. 6. The method of claim 5, wherein each of the plurality of digital twins comprises a set of cellular kinetic parameters comprising at least one of abundance, proliferation rate, trafficking rate, apoptosis rate, and differentiation rate of the plurality of T cell phenotypes within at least one of the plurality of physiological compartments of the corresponding patient over a period of time.

7. 7. The method of claim 6, wherein the composition of TCT products comprises an initial amount of each of the plurality of T cell phenotypes.

8. 2. The method of claim 1, wherein the plurality of T cell phenotypes comprises at least two of stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells.

9. The method of claim 1 , wherein the plurality of physiological compartments comprises a peripheral tissue and lymph node compartment, a blood compartment, a tumor-draining lymph node compartment, and a tumor compartment.

10. generating the model determining, with at least one data processor, a set of cell kinetic parameters corresponding to the plurality of T cell phenotypes in the patient's peripheral tissue and lymph node compartment following delivery of a T cell targeted (TCT) product, wherein the plurality of T cell phenotypes include stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells; determining, with the at least one data processor, a first transport rate of the plurality of T cell phenotypes between the peripheral tissue and lymph node compartment and the blood compartment of the patient; determining, by the at least one data processor, the set of cell kinetic parameters corresponding to the plurality of T cell phenotypes in the blood compartment; determining, with the at least one data processor, a second transport rate of the plurality of T cell phenotypes between the blood compartment and a tumor-draining lymph node compartment of the patient; determining, with the at least one data processor, the set of cell kinetic parameters corresponding to the plurality of T cell phenotypes in the tumor-draining lymph node compartment; determining, with the at least one data processor, a third transport rate of the effector memory T cells and the effector T cells from the blood compartment to a tumor compartment of the patient; determining, with the at least one data processor, the set of cell kinetic parameters corresponding to the effector memory T cells and the effector T cells in the tumor compartment; determining, with the at least one data processor, a distribution of each of the plurality of T cell phenotypes over time in the peripheral tissue and lymph node compartment, the blood compartment, the tumor-draining lymph node compartment, and the tumor compartment based on the set of cell kinetic parameters, the first transport rate, the second transport rate, and the third transport rate; The method of claim 1 further comprising:

11. 11. The method of claim 10, further comprising determining, by the at least one data processor, a differentiation rate parameter corresponding to differentiation of the effector memory T cells into the effector T cells within the tumor compartment, wherein the determining of the distribution is further based on the differentiation rate parameter.

12. 11. The method of claim 10, further comprising determining, by the at least one data processor, differentiation rate parameters corresponding to differentiation of the stem-like memory T cells into the central memory T cells, differentiation of the central memory T cells into the effector memory T cells, and differentiation of the effector memory T cells into the effector T cells within the tumor-draining lymph node compartment, wherein said determining said distribution is further based on said differentiation rate parameters.

13. determining a T cell therapy to treat the tumor; the T cell therapy comprises a dose of the composition of the multiple T cell phenotypes of the TCT product, and the T cell therapy is at least one of T cell receptor (TCR) engineered T cell therapy, autologous T cell therapy, allogeneic T cell therapy, iPSC-derived T cell therapy, and CAR T cell therapy. The method of claim 10.

14. determining, with at least one data processor, a first patient profile representative of a first response to a first dose of a first composition of T cell targeted (TCT) product within a plurality of physiological compartments of the first patient; determining, with the at least one data processor, a second patient profile representative of a second response to a second dose of the second composition of the TCT product in the plurality of physiological compartments of a second patient; generating, with the at least one data processor, an output indicative of a pharmacokinetic-pharmacodynamic (PKPD) relationship for the TCT product based on at least the first patient profile and the second patient profile; determining, by the at least one data processor, a third patient's response to a T cell therapy comprising a third dose of the third composition of TCT products based at least on the output; and determining a treatment plan for the third patient based at least on the response of the third patient to the T cell therapy comprising the third dose of the third composition of the TCT product; A method comprising:

15. the first patient profile comprises a first set of cellular kinetic parameters comprising at least one of a first amount, a first proliferation rate, a first trafficking rate, a first apoptosis rate, and a first differentiation rate of a plurality of T cell phenotypes in at least one of the plurality of physiological compartments of the first patient over a period of time; the second patient profile comprises a second set of cell kinetic parameters comprising at least one of a second amount, a second proliferation rate, a second trafficking rate, a second apoptosis rate, and a second differentiation rate of the plurality of T cell phenotypes in at least one of the plurality of physiological compartments of the second patient over the period of time; 15. The method of claim 14.

16. 16. The method of claim 15, wherein the first composition comprises a first amount of each of a plurality of T cell phenotypes and the second composition comprises a second amount of each of the plurality of T cell phenotypes.

17. 17. The method of claim 16, wherein the first patient profile and the second patient profile each comprise a response corresponding to each of the plurality of T cell phenotypes.

18. 18. The method of any one of claims 15-17, wherein the multiple T cell phenotypes comprise at least two of stem-like memory T cells, central memory T cells, effector memory T cells, effector T cells, and endogenous T cells.

19. 15. The method of claim 14, wherein the plurality of physiological compartments comprises a peripheral tissue and lymph node compartment, a blood compartment, a tumor-draining lymph node compartment, and a tumor compartment.

20. 15. The method of claim 14, wherein the T cell therapy is at least one of T cell receptor (TCR) engineered T cell therapy, autologous T cell therapy, allogeneic T cell therapy, iPSC-derived T cell therapy, and CAR T cell therapy.

21. 15. The method of claim 14, wherein a linear increase from the first dose to the second dose results in a non-linear change between the first response and the second response.

22. 15. The method of claim 14, wherein the first composition is different from the second composition and the first dose is different from the second dose.

23. 15. The method of claim 14, wherein determining the response of the third patient to the T cell therapy comprises simulating a plurality of responses to a plurality of doses of a plurality of compositions of the T cell therapy product in a plurality of simulated patients based at least on the output, and the response of the third patient is one of a plurality of simulated responses.

24. 15. The method of claim 14, wherein the first patient profile is determined after administration of a first lymphocyte depleting regimen to the first patient and the second patient profile is determined after administration of a second lymphocyte depleting regimen to the second patient.

25. 15. The method of claim 14, wherein the response of the third patient is determined by at least varying at least one of the first dose, the first composition, the second dose, and the second composition.

26. 24. The method of claim 23, wherein said simulating said plurality of responses is based on the application of a lymphocyte depletion regimen to said plurality of simulated patients.

27. at least one data processor; at least one memory storing instructions that, when executed by said at least one data processor, result in operations comprising the method of any one of claims 14 to 26; A system comprising:

28. 27. A non-transitory computer readable medium storing instructions that, when executed by at least one data processor, result in operations comprising the method of any one of claims 14 to 26.

29. maintaining a plurality of patient profiles in a database, the plurality of patient profiles indicating patient response to a dose of the TCT product composition in the form of a distribution of a plurality of T cell phenotypes in a plurality of physiological compartments over time; receiving, by at least one data processor, a baseline biomarker dataset associated with the target patient; determining, by the at least one data processor, a set of cell kinetic parameters based on the baseline biomarker dataset in the target patient; generating, with the at least one data processor, a plurality of simulations associated with the target patient based at least in part on the cell kinetic parameters; selecting, by the at least one data processor, a subset of simulations from the plurality of simulations based on similarity between the target patient and a patient; generating a treatment plan for the target patient based on a set of predicted responses generated by the subset of simulations; A method comprising:

Images