Means and methods for determining cellular avidity

Abstract

The dynamics of cell-cell interaction and the formation of a cell bonds involves many interaction partners, a multitude of ligands and receptors including e.g. a specific interaction between e.g. a CAR and a target antigen. This interaction is believed to be highly complex and dependent on many factors such as receptor and antigen density, steric interactions, membrane and receptor mobility, etc. The present inventors delineated from a mathmetical model which best describes the cell-cell interaction and formation of cell-cell bonds. This model most accurately describes the relation between cell fraction bounds and incubation time and allows for highly improving cellular avidity measurements, in particular when performing high throughput measurements, and, provides for a highly useful cellular avidity metric.

Description

[0001] Title: Means and methods for determining cellular avidity

[0002] Introduction

[0003] Despite tremendous clinical success of T cell therapy, including CAR-T cell therapy, for haematological malignancies, many obstacles still remain for such therapies in treatment of a wider range of cancers. Predicting success of T-cell therapy using in vitro data alone is still a major challenge. Affinity of a binding molecule used for CAR generation towards the target antigen is frequently considered during the in vitro testing, but this is generally understood to not define CAR-T cell activity. With the increasing complexity of constructing novel methodologies to further improve the clinical outcome of immunotherapies, screening methods to identify the best lead candidates is becoming even more essential. It is understood in the art that merely measuring the affinity between a CAR molecule and its target molecule with which it is to interact may not accurately predict in vitro nor in vivo outcomes. Determining cellular avidity, i.e. the overall cellular binding strength between a T-cell and its target cell, is to provide a more complete and physiologically relevant measurement that reflects the bona fide interaction between T cells and target cells which is believed to more accurately predict in vivo outcomes.

[0004] Cellular avidity measurements may drive better, more informed decisions at earlier stages for drug selection and has the potential to improve clinical outcomes. One of the main obstacles in the process of measuring cellular avidity was the lack of fast and accurate tools to assess cellular avidity. The z-Movi® Cell Avidity Analyzer, a platform for measuring cell-cell binding strength facilitates a direct analysis of CAR- T cells, and the like, e.g. against surface immobilised antigens or a monolayer of target cells. Using this system, cell-cell interactions are perturbed using resonant sound waves generated by a piezoelectric element and a cell tracking system is employed to measure cell-cell disruption events in response to applied force. This way, cellular avidity measurements can be done which provide for a cellular avidity score which allows to compare e.g. different candidate receptors.

[0005] The current inventors, in working with cellular avidity were now looking for means and methods to further improve cellular avidity measurements. Summary of the invention

[0006] In the art, cellular avidity measurements methods are known, such as methods utilizing a z-Movi® Cell Avidity Analyzer, or utilizing centrifugation forces (WO2023232881), or the like. In these measurements, under specific experimental conditions, cell-cell interactions are allowed to occur, and by exerting a force, cells are moved away from each other. The number of cells that have moved away after a defined force exerted is understood to be a relevant measure of cellular avidity. This way, different intra- experimental measurement outcomes are compared e.g. comparing different CAR- receptors. However, measurements and values representing cellular avidity are understood to be not well comparable between experiments, nor between different laboratories. Hence, in the art, there is a need to provide for a parameter representing cellular avidity, i.e. to provide for a cellular avidity metric, which allows for a more objective measurement of cellular avidity which can be well compared between different experiments, e.g. conducted at the same laboratory or conducted at different laboratories.

[0007] When a T-cell provided with a CAR, or the like, interacts with a target cell, there are many interactions that take place before the T-cell becomes effectively bound with the target cell. Cell-cell interaction may be initiated via interaction between a first ligand and receptor, such as between an antigen and a CAR, respectively presented on on the target cell and an engineered T-cell and. Such a cell-cell interaction may result in an initial binding, followed by a subsequent detachment. Subsequent cumulative interactions, between the cell surfaces, i.e. between further receptors and ligands presented on the T-cell, and conversely, further receptors and ligands presented on the target cells, may lead to further and stronger binding, ultimately resulting in the formation of a strong bond (e.g. a synapse or other bond able to resists a significant force). Cells may need time to interact to allow a bond form. Moreover, it may also be that not all engineered T-cells have the capacity to interact with a target cell and form a cell-cell bond, i.e. a pool of engineered T-cells may contain a fraction of cells that is highly active in forming bond. Lastly, (engineered) T-cells may continuously probe target cells to find a target, and once a target is found, binds therewith, promptly forming a strong bond. Hence, the dynamics of cell-cell interaction and the formation of a cell bonds involves many interaction partners, i.e. a multitude of ligands and receptors including a specific interaction between a CAR (or a TCR) and a target antigen (or an MHC presented antigen peptide). This interaction is believed to be highly complex and dependent on many factors such as receptor and antigen density, steric interactions, membrane and receptor mobility, etc. Hence, the present inventors sought to delineate from cellular avidity measurements a mathmetical model which best describes the cell-cell interaction and formation of cellcell bonds. The present inventors highly surprisingly found that a constant binding rate model most accurately describes the relation between Fraction_bound and incubation time, i.e. a defined time of contacting, in minutes. The mathemical formula is as follows:

[0008] Fraction_bound[t] = 1 - exp(-k_on * t)

[0009] This formula can be rewritten to express k_on: k_on = - log (1 - Fraction_bound[t]) 1 1

[0010] This mathmetical model thus allows to provide for a k_on rate for a given cell cell interaction, such as an effector cell and a target cell. This k_on rate is independent of the time of contacting and represents a metric characterizing cellular avidity, i.e. representing an intrinsic more objective property representative of the capacity of a given effector cell to form a bond with a target cell. Comparing k_on rates of different effector cells allows for a more meaningful comparison from a biological perspective, instead of the simply measuring the number of cells that remain bound after exerting a defined force at a single defined time. In addition, by including in cellular avidity measurements the k_on rate, this allows for less experimental work, i.e. a lower number of measurements.

[0011] In addition to improving cellular avidity measurements by determining k_on rates, the present inventors also observed that measurements may be advantageously corrected, e.g. by a correction factor alpha. Such a correction factor being used to correct for scenarios wherein e.g. not all effector cells were transduced with a viral vector encoding a CAR (e.g. when the transduction efficiency is 75%, alpha is 0.75), or, not all cells are viable (e.g. when the fraction of viable cells is 60%, alpha is 0.6). Such a correction factor may be determined by characterizing the cells (e.g. with staining for dead cells or by determing transduction efficiency). Such a correction factor alpha may also be empirically determined, e.g. by measuring the cell fraction bound after relatively long contacting times, (e.g. longer than 2.5 x 1 / k_on), the Fraction_bound after such long contacting times is a good estimate for alpha. This way, one can correct cellular avidity measurements to represent a k_on value as determined from 100% active cells. Such a correction factor is advantageous for determining k_on, but is not necessarily restricted to that. Standard cellular avidity measurements may also apply such a correction factor alpha, by dividing the measured value by alpha.

[0012] Figures

[0013] Figure 1. Schematic showing target cells and effector cells carrying a receptor and cellular avidity, showing the complex interaction of an effector cell (6) carrying a receptor (4) at its cell surface and a target cell (2) having a ligand (5), e.g. a target molecule, at its cell surface. Target cells (2) are provided on a surface (1), which as depicted is in this case a flat surface. The target cell expresses many ligands and receptors, likewise the effector cells expresses ligands and receptors as well. The specific ligand-receptor interaction (4 and 5) combined with multiple further ligandreceptor interactions are understood to be involved in cell-cell binding (3). To rupture cell-cell binding, a force (7) is exerted on the cell (6) away from the target cell (2), which can be in the z-axis direction e.g. when the flat surface is defined as being in the x-y plane. Alternatively, this can also be in the x or y-axis direction. When the cellcell bond is ruptured, the cell moves away from the cell surface and I or the target cell and this event can be detected and / or detached cells can be collected and quantified and / or further analyzed. Unbound effector cells also move away when the force is exerted.

[0014] Figure 2. The fraction of bound cells (%) was plotted against incubation time (min). In A) datapoints were simply connected. In B) the datapoints were fitted to the formula for Fraction_bound[t] listed above using the least squares method. The lines shown from top to bottom represent 4G7, B4, FMC and UNT.

[0015] Figure 3. Plot showing the theoretical estimate of the error in the binding rate using the equation for k_on that describes the interaction between effector cells and target cells, against incubation time, assuming a standard deviation of 2% between technical repeats in avidity measurements. Figure 4. Plot showing a strong correlation between Fraction of dead cells and measured Fractions bound.

[0016] Figure 5. Plot showing dose response curve of C19Ab, on Fraction_bound cells of different CAR expressing effector cells (FMC, B4, and 4G7) and control cells (UNT).

[0017] Figure 6. Schematic depicting effects of settling time in certain embodiments in methods in accordance with the invention. 6A schematically shown is a suitable well configuration (e.g. as configured in suitable multi-well configurations) in upright and upside down position, with a ceiling for attachment of target cells and a bottom to which no cells can attach, such as suitable for cellular avidity measurements as outlined herein, and as further shown in Figures 6B and 6C and Figure 7A and B. 6B. A well is shown with a ceiling surface coated for attachment of target cells, wherein target cells were added to the well, and the avidity sample plate is placed in the upside down position and the target cells have settled and have attached to the ceiling (1). The avidity sample plate is placed in upright position and effector cells in suspension are introduced in the well (2). The avidity sample plate is placed in upside down position (3). The effector cells settle down on the target cells (4). The avidity sample plate is placed in upright position (5) after a defined time and centrifuged, wherein upon exertion of sufficient force exerted effector cells can become unbound and move away from the target cells, wherein the fraction of effector cells that remain bound (or that are detached) is determined. 6C. Shows the same steps as depicted in 6A, with a slight variation of step 2, i.e. after effector cells are introduced in the well (2a), cells are first allowed to settle to the bottom (2b). Figure 6D shows a plot of the fraction of effector cells that have landed I settled at the ceiling surface with target cells

[0018] Figure 7. Schematic outlining embodiments of spin assisted settling and effects on cellular avidiy measurements. 7A and 7B depicts the same process as depicted in Figures 6B and C respectively, with spin assisted settling at step (3), i.e. a centrifugal I external force is exerted with the aim to reduce the time to have the effector cells settle on the target cells, i.e. an active settling force is exerted on the effector cells instead of allowing the effector cells to settle under the influence of gravity. 7C lists parameter values related to effector cell settling, and 7D shows the effect of spin assisted settling (spin down) when exerting a force of 5 x g for 45 seconds. The median arrival time and 98 percentile arrival time with spin assisted settling is 27 seconds and 84 seconds, respectively. Without spin assisted settling, these are respectively 76 seconds and 264 seconds. 7E When calculating the mean error in avidity measurements, the error is significantly reduced with spin assisted settling with incubation times of up to 5 minutes.

[0019] Detailed description

[0020] Hence, as said, the current inventors through meticulously analysing data obtained from cellular avidity measurements, now provide for a further method highly useful for assessing cellular avidity, which involves determining the cellular avidity metric k_on. The cellular avidity metric k_on may also be referred to as cellular avidity binding rate k_on.ln one embodiment, the present invention provides for a method for determining cellular avidity of effector cells carrying a receptor to target cells, the method comprising the steps of: a) providing effector cells carrying a receptor; b) providing target cells; wherein the target cells are attached to a surface; c) contacting the effector cells with the target cells to allow the effector cells to bind with the target cells, wherein in the contacting step, the target cells attached to the surface are in excess as compared with the effector cells, and wherein the contacting step is for a defined time t ; d) applying a force, wherein the force is in a direction away from the attached target cells, such that at least part of the effector cells in contact and / or bound to the target cells attached to the surface move away therefrom; e) detecting effector cells that have remained bound with the target cells and attached to the surface after applying the force; f) determining the fraction of the effector cells (Fraction_bound[t]), that were contacted with the target cells that have remained bound; g) optionally, perform steps c), d), e) and f) one or more times; and h) calculating the cellular avidity metric k_on from - determined fraction(s) that remained bound (Fraction_bound[t]) and defined contacting time(s) t; with the following formula: k_on = - log(1-Fraction_bound[t]) 1 1 to therewith provide the cellular avidity metric k_on.

[0021] In step a) effector cells carrying a receptor are provided and in step b) target cells, wherein the target cells are attached to a surface. Target cells and effector cells carrying a receptor are provided. Target cells in accordance with the invention are the cells on which the effector cells are to exert an effect, bind therewith and e.g. trigger an immune reaction thereto. Target cells include cancer cells presenting an antigen. An antigen may be presented by MHC, i.e. HLA in humans, which are specialized receptors that present peptides e.g. derived from digested proteins expressed by the cell (e.g. usually 8-11 amino acids in length for MHCI). An antigen may also be a protein or other biomolecule that is presented on the surface of a cell, e.g. epidermal growth factor receptors or checkpoint proteins, which in the case of cancer cells are overexpressed therewith providing a differentiating feature. Target cells may also include cells expressing auto-antigens, e.g. known to be involved in autoimmunity diseases or cells infected with a pathogen, e.g. a virus. In accordance with the invention, effector cells carrying a receptor include effector cells of the immune system that can exert an effect, via the receptor. For example, a T cell carrying a T cell receptor can bind an antigen on a cancer cell, upon which it can e.g. exert a cytotoxic effect and kill the target cell. Effector cells can be derived from nature, e.g. obtained from a host, and can also include genetically modified cells wherein e.g. a receptor in particular useful is provided to an effector cell. Effector cells may be T cells that are provided with a Chimeric Antigen Receptor (CAR). Effector cells can include NK cells.

[0022] The target cells are attached to a surface. It is understood that surfaces for attaching cells may be any surface suitable for attaching cells. Suitable surfaces for attaching cells include plastic or glass surfaces. These surfaces may be coated e.g. with a protein to attach cells to the surface, such as poly-L-lysine or the like. The attachment of the cells to the surface is such that the strength of the binding to the surface is sufficient to have the cells remaining attached when applying a suitable force. The effector cell carrying the receptor is capable of binding target cells, or is studied for its capability of binding target cells. It is understood this capacity of binding target cells can include a specific interaction, i.e. an interaction between a receptor (e.g. CAR) and a target antigen (e.g. a cancer antigen such as CD19 as shown in the examples) that can optionally induce synapse formation, i.e. the effector cell is to bind to a target cell and exert an effect thereon, e.g. induce target cell killing.

[0023] In the next step c), the effector cells are contacted with the target cells to allow the effector cells to bind with the target cells, wherein in the contacting step, the target cells attached to the surface are in excess as compared with the effector cells, and wherein the contacting step is for a defined time t. This step is such that the effector cells will have sufficient time to be in contact with target cells and can form a cell-cell bond. It is understood that a cell-cell interaction or contact may not always result in a cell-cell bond. The contacting step is such that cell-cell bonds can be formed and appropriate conditions therefore are selected. This contacting step is to be performed for a defined time t. In this contacting step, the target cells attached to the surface are in excess as compared with the effector cells. An excess of effector cells means that the number of effector cells is e.g. at least 2 fold, at least 3 fold, at least 4 fold, at least 5 fold, at least 10 fold, at least 15 fold, at least 20 fold, at least 25 fold, or at least 30 fold or more, of the number of target cells.

[0024] Once the effector cells have had contact with the target cells for a defined time, and thus have had the opportunity, if possible, to bind with the target cells, in a subsequent step d) a force is applied away from the cells attached to the surface, such that at least part of the cells bound to the cells attached to the surface, and unbound cells as well, move away from the cells attached to the surface. This way, target cells are obtained with effector cells bound thereto, which are attached to the surface, wherein the exerted force was not sufficient to break cell-cell bonds. Of course, cells that are attached to the surface, to which no subsequent cells are bound, remain as well.

[0025] As said, the force is applied in a direction away from the attached cells. It is understood that in this step, the force applied may be perpendicular (in the direction of z-axis) to the surface (x,y) to which cells are attached, for example when a centrifugal force or acoustic force is applied. The force may also be lateral (in the direction of the x-axis or y-axis relative to the surface), for example when a shear force is applied. In any case, the force is applied and is controlled such that a defined force is exerted on the effector cells that interacted with the attached cells. It is understood that the force that is exerted on the effector cells interacting with attached cells is to be substantially equal, such can be achieved e.g. when using a flat surface. Other suitable surface shapes may be used (e.g. a tube with exerted concentrical force or laminar flow force in the direction of the length of the tube), as long as the force exerted can be substantially equal at a defined surface area to which cells are attached, such a surface shape may be contemplated.

[0026] Subsequently, step e) comprises detecting effector cells that have remained bound with the target cells and attached to the surface after applying the force. The cells that are attached are preferably attached to a glass or plastic surface, preferably a surface in a chip or in well, which allows for detection of cells e.g. via microscopy or other means. The applied force required to move a cell away from an attached cell preferably can be detected, e.g. via microscopy or other means, to which may be referred to as a cell detachment event or cells moving away. This way, cells moving away can be monitored and counted. It may be advantageous and convenient to use microscopy, with which bound cells can be identified and quantified and cells moving away can be likewise monitored and quantified, also allowing e.g. to detect markers. For example, the z-Movi® device as available from Lumicks which applies an acoustic force may be well equipped to do so. Likewise, similar devices may be provided which allow microscopy or other means to quantify cells, detect markers and bound cells and / or cells moving away or that have moved away, and also utilizing e.g. shear force or centrifugal forces instead of acoustic force. A chip or multiwell plate, in which target cells attached can be contacted with effectors, allows for applying centrifugation forces and microscopy. Before and after centrifugation, bound and unbound effector cells can be easily detected and quantified in a chip or multiwell plate utilizing microscopy, as shown in the examples herein.

[0027] In any case, effector cells carrying a receptor that have remained bound with the target cells and have withstood the applied force, and thus remained attached to the surface, after the effector cells and target cells have interacted for a defined time, are detected. Next, step f) comprises determining the fraction of the effector cells (Fraction_bound[t]), that were contacted with the target cells that have remained bound. It is understood that the Fraction_bound can be determined based on the number of effector cells that have moved away and / or remain bound after the force has been exerted, as the amount of effector cells initially provided can be controlled. A Fraction_bound can for example be calculated by determining the ratio of the number of effector cells that remained bound with the target cells to the number of effector cells that was initially provided or to the number of effector cells that moved away. A Fraction_bound can also for example be calculated by determining the ratio of the number of effector cells that remained bound with the target cells to the number of effector cells that was in contact with the target cells before force application. As is understood herein, the ratio’s determined may be based on absolute numbers, or may be calculated based on numbers determined of fractions thereof. For example, one may determine of a defined surface with target cells attached, the number of subsequent effector cells that interact therewith and detect effector cells that move away from the attached cells and / or determine the number of cells at the defined surface that remain after the force has been exerted. One may represent calculated ratio’s, i.e. Fraction_bound as a number. In any case, determining the number of cells that have moved away and / or remain bound after the force has been exerted allows one to provide for the Fraction_bound determined after a defined time t, i.e. Fraction_bound[t], of contacting.

[0028] Next, optionally, in step g) the steps c), d), e) and f) are performed one or more times, and, in step h) subsequently, is performed comprising calculating the cellular avidity metric k_on from the determined fraction(s) that remained bound (Fraction_bound[t]) and defined time(s) t of the contacting step; with the following formula: k_on = - log(1-Fraction_bound[t]) 1 1 to therewith provide the cellular avidity metric k_on.

[0029] It is understood that the mathematical formula allows for calculating the k_on already from a single measurement. However, as shown in the examples herein, whether or not such a single measurement provides for a reliable k_on value depends on the incubation time, i.e, the defined time of contacting, selected. Hence, in case one has a reasonable estimate of the expected k_on value, e.g. from an earlier experiment, one can select a suitable incubation time with which one can determine the Fraction_bound[t] and calculate the k_on value. However, as outlined in the examples herein, the present invention by utilizing the k_on value calculated from Fraction_bound[t] and defined time t of contacting now allows for carrying out cellular avidity measurements in a more efficient manner to therewith provide for k_on values which represent a cellular avidity metric of effector cell - target cell interaction. The k_on value being more reflective of the inherent biological property of the effector cell - target cell interaction as opposed to merely Fraction_bound.

[0030] As outlined herein, the steps c), d), e) and f) are performed one or more times. It is understood that one or more times means that the exact same defined time may be used in step c) when performing the steps one or more times, or that different defined times are used in step c) each time when performing the steps one more times. It is understood that one or more times also can mean that one or more different defined times are used and the steps c), d), e) and f) for each different defined time are performed one or more times. Determining more Fraction_bound[t] values aids in improving the reliability of determining the k_on value. Hence, it is understood that in the methods as outlined herein, the defined time of contacting in step c) may vary, and may also advantageously be adjusted in an iterative process to allow for more reliable measurements to aid in more efficiently determining k_on, as outlined herein below and as shown in the examples.

[0031] In particular as shown in the examples, in one embodiment, step g), comprises the steps c), d), e) and f) which are performed one or more times at different defined contacting times. Such further different defined times of contacting in step c) preferably are selected from from 0 minutes to 60 minutes. The number of different defined times of contacting preferably includes 3 different times, or more. In yet a further embodiment, such further defined contacting times include defined times of between 0 minutes and 10 minutes, between 10 minutes and 20 minutes, between 20 minutes and 40 minutes, between 40 minutes and 60 minutes. Hence, it is understood that preferably, the different defined times are distributed (relatively) evenly over a defined range. Having the Fraction_bound determined for different contacting times allows to calculate the k_on by using a least squares fitting method. Such least squares fitting method can of course include multiple datapoints for a single defined time of the contacting step. As shown in the examples herein, having different defined times of the contacting step may allows to provide for more accurate k_on calculations as opposed to having a single defined contacting time.

[0032] Nevertheless, it may be contemplated of course to have, in the method in accordance with the invention, step g), the steps c), d), e) and f) performed one or more times at the same defined contacting time. Calculating the k_on than involves calculating averages and potentially standard deviations (e.g. to estimate errors). As is shown in the examples herein, having some knowledge of the k_on value for a certain effector cell - target cell allows to select for a defined time of contacting which, from an error rate perspective, provides for the most accurate measurements. Hence, one may, in further embodiments, first determine a k_on_initial value (based on one or more determined Fraction_bound values and corresponding defined times) and, subsequently, based in such an initial value of k_on, select a defined contacting time with which to perform further measurements, based on which, with relatively a low number of measurements, k_on can be accurately determined. Such subsequent defined contacting time based on a k_on_initial, is about to 1 / k_on_initial, for example plus or minus 20% of 1 / k_on_initial. Preferably, the subsequent defined contacting is 1 / k_on_initial. For example, in case the k_on initial is 0.2 min-1, that means that 1 / k_on initial is 5 min, the defined time for contacting subsequently selected can be from 4 minutes to 6 minutes, preferably the time selected is than 5 minutes.

[0033] Accordingly, in another embodiment, in the methods as outlined, in step h) k_on_initial is determined, and, subsequently, step i) is performed, wherein in step i) subsequently steps steps c), d), e) and f) are performed one or more times; at a defined time of contacting in step c) of 1 / k_on_initial, and subsequently k_on is calculated based on the determined fraction(s) that remained bound (Fraction_bound(t)) and defined contacting time(s) t obtained in step i) using the formula provided in step h).

[0034] In any case, as outlined above, in methods in accordance with the invention, determining k_on, including an k_on_initial, can highly improve cellular avidity measurements, as it allows for selecting defined times for contacting that results in measurements of bound fractions (i.e. Fraction_bound) that are least variable and or most representative for the cell-cell interactions. This way, different effector cells, e.g. against the same target cell, can be compared. Conversely, the same effector cells can be compared against different target cells. Of course, one can also test different effector cells and different target cells (e.g. differentiating respectively with regard to receptor, e.g. CAR, and, target antigen).

[0035] Accordingly, in another embodiment, in the method in accordance with the present invention, effector cells with different receptors are provided and for each effector cell with a different receptor the method is performed. Moreover, instead of defining for each condition selected a k_on_inital, and subsequent defined contacting time, one can also select an optimal time for the different conditions tested, e.g. comparing different effector cells and / or different CAR receptors provided to effector cells. Accordingly, as shown in the examples herein, in another embodiument, in the method in accordance with the present invention, effector cells with different receptors are provided and for each effector cell with a different receptor the method is performed and in step h) a k_on_initial is calculated, and, subsequently, the highest and lowest k_on_initial determined (k_on_init_h and k_on_init_l, respectively), and subsequently, step i) is performed, wherein in step i) steps c), d), e) and f) are subsequently performed one or more times; at an optimal defined time of contacting in step c) of: t = log (k_on_init_h / k_on_init_l) I ((k_on_init_h - k_on_init_l)) and subsequently k_on is calculated based on the determined fraction(s) that remained bound (Fraction_bound(t)) and defined contacting time(s) (t) obtained in step i) using the formula provided in step h). This way, an optimal contacting time can be selected when multiple effector cells are to be compared. This may be advantageous when for examples avidity measurements are carried out in a multiwell plate, wherein contacting times most conveniently are the same for all wells. As shown in the examples herein, for different effector cells it may not be possible to arrive at an optimal incubation time that applies to all different effector cells. In such a scenario, it is contemplated in accordance with the invention to determine optimal contacting times for subgroups. Hence, in such a scenario, different effector cells of a subgroup can be tested in one multiwell plate or chip, and different effector cells of another subgroup can be tested in another multiwell plate. Hence, in a further embodiument, in the method in accordance with the present invention, wherein effector cells with different receptors are provided, optimal contacting times are determined for subgroups of different effector cells, using the formula above. Accordingly, in the method in accordance with the invention, in a further embodiment, highest and lowest k_on_initial are determined for subgroups of different effector cells, and for each subgroup step i) is performed with their corresponding optimal defined time in step c).

[0036] It is understood that the methods in accordance with the invention are highly preferably performed in a controlled fashion as that is useful for repeatability of experiments, as experiments with biological materials such as cells, may be sensitive to experimental conditions. Hence, highly preferably, the methods are performed at defined temperatures and using defined culture medium. It is also understood that the experiments may be performed with one type of target cells. That is advantageous as one can than compare different effector cells and select an optimal effector cell. It may also be advantageous to perform experiments with different types of target cells, that way, one can test e.g. the effect of effector cells of varying target cells, having the same target at its surface, but having different expression levels and / or presented on different types of cells. Accordingly, in one embodiment, the method in accordance with the invention is performed at a defined temperature and in a defined cell culture medium. In another embodiment, the method in accordance with the invention is performed with one type of target cell or with different target cells.

[0037] With regard to the amounts of target cells and effector cells that are to interact, it is understood that the target cells are to be in excess as compared with the effector cells. That way, effector cells do not compete for target cells and can contact target cells. The excess of target cells can be 2-fold, 3-fold or more. Preferably, the excess of target cells is about 10-fold or more. Hence, in one embodiment, the target cells attached to the surface are in at least a 10-fold excess as compared with the effector cells carrying a receptor. Preferably, the target cells attached to the surface are at high confluency, presenting a monolayer of target cells.

[0038] As outlined herein and as shown in the examples, target cells present a cell surface molecule, and the effector cells are expressing a receptor at their cell surface a receptor molecule that is capable of binding said cell surface molecule. It is understood that of course the method can also be utilized to test the capability of binding of an effector cell to a target cell. Reference is made herein to effector cells and target cells, wherein effector cells include cells of the immune system such as T- cells that are to bind with a suitable target cell, i.e. a cell with which the immune cell is to interact and subsequently e.g. eradicate. Hence, it is understood that preferably, in accordance with the invention, effector cells are cells of the immune system, which may be primary cells or cell lines. Likewise it is understood that target cells may preferably include suitable target cells for the immune system, such as cancer cells, cells infected with a virus or other cells expressing an antigen of interest. Nevertheless, without being bound by theory, the present methods as outlined herein with regard to determining k_on may also apply to other types of cells that are to specifically interact, and hence may include any cell that is to interact with another cell, via surface molecules (e.g. a receptor) on one cell, and target molecules on another cell, e.g. respectively a receptor and a ligand for that receptor. Accordingly, in one embodiment, in the methods in accordance with the invention, the target cells present a cell surface molecule and the receptor of the effector cells is capable of binding said cell surface molecule.

[0039] In yet a further embodiment, in the methods as outlined herein, the target cells present a cell surface molecule, and the contacting is performed in the presence of a cell engager, wherein the cell engager has one binding arm capable of binding with the cell surface molecule of the target cell and another binding arm capable of binding said receptor of the effector cells. Determining the k_on for an effector cell and target cell may also apply to cell engagers, i.e. in such a scenario, the k_on determined is representative for the combination of a cell engager an effector cell and target cells. For example, an effector cell may be first incubated with a cell engager prior to the contacting step. This way effector cells are obtained with cell engagers bound therewith. Subsequently, the effector cells with cell engagers can be contacted with the target cells as outlined above. Of course, the cell engager may also be included in the medium, or may be contacted first with the target cells instead.

[0040] As outlined above, the target cells are to be attached to a surface. As shown in the examples herein, this allows for a convenient way of determining cellular avidity, as most often target cells (e.g. cancer cell lines) are readily available and often include adherent cells, and the amount of effector cells (e.g. primary T cells) available may be limited or are cumbersome to prepare (e.g. because these need to be transduced with a vector encoding a CAR or the like). Nevertheless, in an alternative embodiment, it may be contemplated to not have the target cells attached to a surface. As long as a method is provided in accordance with the invention, in which the effector cells are contacted with the target cells for a defined time, allowing effector cells to bind with the target cells, and, subsequently exerting a suitable force, and determining the fraction of effector cells and target cells that remain bound to each other thereafter, k_on values can be likewise determined as outlined above. Methods that do not require cells attached to a surface may involve differential forces, such as when applied during pipetting, which can break up effector cell - target cell pairs when neither is bound to a surface. Also, it may also be contemplated in an alternative embodiment in accordance with the invention to not have the target cells attached to a surface, but have the effector cells attached to a surface. In such a scenario, in the methods as outlined above, the “effector cells” and “target cells” are reversed, but the principles underlying the present invention still apply and one can also calculate the k_on, though in the latter scenario the Fraction_bound relates to target cells bound instead. Hence, in one alternative embodiment, the means and methods in accordance with invention are performed with the target cells not attached to a surface. In another alternative embodiment, the target cells are not attached to a surface, but the effector cells are attached to a surface, instead. The skilled person is well capable of performing the means and methods in such an inverse scenario.

[0041] Hence, in such an alternative embodiment the invention provides for a method for determining cellular avidity of effector cells carrying a receptor to target cells, the method comprising the steps of: a) providing effector cells carrying a receptor; b) providing target cells; wherein the effector cells are attached to a surface; c) contacting the target cells with the effector cells to allow the target cells to bind with the effector cells, wherein in the contacting step, the effector cells attached to the surface are in excess as compared with the target cells, and wherein the interaction is for a defined time t ; d) applying a force, wherein the force is in a direction away from the attached effector cells, such that at least part of the target cells in contact and / or bound to the effector cells attached to the surface move away therefrom; e) detecting target cells that have remained bound with the effector cells and attached to the surface after applying the force; f) determining the fraction of the target cells (Fraction_bound[t]), that were contacted with the effector cells that have remained bound; g) optionally, perform steps c), d), e) and f) one or more times; and h) calculating the cellular avidity metric k_on from

[0042] - determined fraction(s) that remained bound (Fraction_bound[t]) and defined contacting time(s) t; with the following formula: k_on = - log(1-Fraction_bound[t]) 1 1 to therewith provide the cellular avidity metric k_on.

[0043] Accordingly, further embodiments as outlined above may be appropriately amended taking into account the reverse configuration. Accordingly, further embodiments may equally be appropriately provided which include configurations in which neither of the effector cells or target cells are attached to a surface. In another further alternative embodiment, instead of utilizing target cells attached to a surface, a surface functionalized with a target molecule, or mixture of molecules comprising a target molecule, or multiple different target molecules, is provided. In yet another further alternative embodiment, instead of providing effector cells with a receptor, beads functionalized with a receptor are provided instead. Such beads can for example be fluorescent beads coated with one or more receptor molecules, co-receptor molecules or other molecules involved in typical cell-cell interactions or as can be present on a cell surface. Hence, such functionalized beads and surfaces can advantageously comprise complex mixtures of different molecules, which may mimick cellular surfaces. Such alternative embodiments also allow to determine highly useful k_on values. An advantage of using receptor-coated beads instead of effector cells carrying a receptor may be that such beads and the surface molecules they present may be more easily controlled. This may reduce variability related to live cell-cell interactions while still capturing important aspects of the cellcell bonding mechanism when the beads are allowed to interact with target cells. Using such artificial beads may allow systematic changing of different concentrations of surface bound molecules thereby studying and / or controlling different aspects of the cell-cell recognition and binding mechanism. Similarly, using live effector cells but using a functionalized surface instead of a target cell layer may enable improved control over the relative concentrations of different molecules on the functionalized surface. To assess the interaction between effector cells and a protein or target molecule of interest, instead of providing cells attached to a surface, the cells expressing the protein or target molecule of interest at its cell surface, one can also provide a surface that is functionalized with said protein or target molecule of interest instead. Functionalization may be performed using either physical or chemical methods (e.g. physisorption, chemical crosslinking, or high affinity molecular interactions e.g. Avidin-biotin). A functionalized surface may comprise one or more types of interaction moiety types, in particular comprising at least one of antibodies, peptides, MHC-complexes, biological tissue factors, biological tissue portions, bacteria, antigens, proteins, ligands, tissues, viruses, (synthetic) drug compounds, lipid (bi)layers, fibronectin, cellulose, nucleic acids, RNA, small molecules, allosteric modulators, (bacterial) biofilms, organ-on-a-chip.

[0044] To assess the interaction between effector cells and a protein or target molecule of interest, instead of providing effector cells carrying a receptor, one can also provide a bead that is functionalized with said receptor. Fluorescently labelled beads of sufficient size and density are available that enable acoustic, centrifugal or fluidic (shear) flow forces to be applied (e.g. polystyrene particles of 3-20 pm diameter). Use of fluorescent beads may also simplify or enhance the ability to multiplex experiments for example by using optical barcoding. When using multi color detection (e.g. 4 color fluorescence detection) fluorescent beads can easily be distinguished based on different ratios of fluorescent dyes allowing many more than 4 different color combinations. By coating each unique set of beads (barcoded with a controlled ratio of different colors) with different (concentrations and or combinations of surface molecules) many different cell-bead interactions can be measured in the same cell avidity experiment. By separating the different types of beads based on optical barcodes during analysis, avidity metrics related to the different concentrations and or combinations of surface molecules may be obtained. Thus, multiple surface molecules / antibodies, or concentrations may be combined in a single channel / well. For convenient attachment of a receptor molecule to the beads appropriate methods therefore are known in the art, and can include common conjugation methods and / or including of appropriate linker molecules, or other means that allow for (covalently) attaching receptor molecules to beads. Accordingly, in a further alternative embodiment, in the means and methods as provided herein, the target cells attached to a surface are replaced with a surface functionalized with a target molecule. In another further embodiment, the effector cells with a receptor are replaced by beads functionalized with a receptor.

[0045] Advantageously, providing a system capable of determining and / or utilizing the avidity metric k_on in a more or less fully automated manner may enable higher throughput (e.g. measuring more samples) with less operator hands-on time. This is highly important for such methods to be adopted in industry. Also automating (parts of) the methods as outlined above and herein, reduces the chance for mistakes and or variations in time, force application, volumes and concentrations used, pipetting speeds, mechanical shocks during handling of sample holders, imaging artefacts, inconsistencies in cell counts, etc. Thereby improving the quality and reproducibility of results thus obtained.

[0046] Hence, the methods as described herein are highly useful for high throughput cellular avidity measurements. Hence, the methods as outlined herein are preferably applied as a high-throughut method, e.g. utilizing high throughput systems. Such high throughput cellular avidity measurements can be performed utilizing e.g. multiwell or multichannel formats, such as 48-well, 96-well formats, or the like, which may be automatically handled, measuring multiple plates or chips as well. The methods as outlined herein may be performed in such systems and thus advantageously may include computer implemented method. Such high throughput methods may be advantageously executed in an automated manner, e.g. applying the methods as outlined herein in a computer implemented method.

[0047] Corrections for cell viability, or for active cell fraction

[0048] In addition to the above related to k_on, it was also found that cell viability can have an effect on the quality of cellular avidity measurement. Hence, the present invention further provides in addition a method for determining cellular avidity of effector cells carrying a receptor to target cells, wherein the method comprises contacting the effector cells with the target cells and allowing the effector cells to bind with the target cells, and, subsequently exerting a force, and determining the fraction of effector cells and target cells that remain bound thereafter, wherein the fraction of cells that remains bound after exerting the force is corrected for cell viability. In one embodiment, the fraction of cells that remains bound after exerting the force, Fraction_bound, is corrected for cell viability, as described above in the methods utilizing k_on. Such a correction preferably is performed by dividing the fraction of cells that remain bound (e.g. Fraction_bound) by the viable fraction value.

[0049] Determining cell viability is commonly done in the art, and various means and methods are commercially available for determining cell viability, such as tryplan blue, propidium iodide, DAPI, and the like. Hence, this aspect is not limited by the means of determining cell viability, and any means may suffice in accordance with the invention. The present inventors found that cell viability may cause variation when conducting cellular avidity measurements, and found that it may allow to correct for such variation thereby further improving cellular avidity measurements in general, which includes methods that focus on determining k_on as outlined herein. Cell viability, in particular of the effector cells, correlated with the values determined for fractions bound. Hence, in one embodiment, in a method for measuring cellular avidity, the fraction of the cells that is viable is calculated and the Fraction_bound is corrected for cell viability by division by the viable fraction value. It was observed that cellular viability may be particularly important when cells need to be thawed prior to a cellular avidity measurement, e.g. when providing effector cells carrying a receptor that have been aliquoted in frozen vials (e.g. derived from primary cells). Furthermore, as shown in the examples herein, when effector cells were allowed to recover prior to measuring cellular viability, and / or cellular avidity was determined immediately after thawing, the best correlation between cell viability and cellular fraction bound was observed. Accordingly, in a further embodiment, the effector cells are thawed prior to performing a cellular avidity measurement, and cell viability is determined. In a further embodiment, wherein the effector cells are thawed prior to performing a cellular avidity measurement, cell viability is determined after allowing the cells to recover from thawing. In yet a further embodiment, the cells are allowed to recover for about 20 - 28 hours prior to measuring cell viability, preferably 24 hours after thawing. In still a further embodiment, cellular avidity is determined immediately after thawing the effector cells. It is understood that immediately means as soon as practically feasible, i.e. cells are thawed, need to be provided in an appropriate medium and need to dispensed e.g. in a chip or multiwell plate for a cellular avidity measurement. Immediately thus means e.g. within an hour after thawing, or within half an hour after thawing. With regard to allowing the cells to recover, such usually involves thawing the cells on one day, and using the cells on the next day, in an experimental setting. Hence, allowing the cells to recover includes culturing the cells from about 14 hours to 32 hours after thawing. Preferably, the cells are allowed to recover for about 20 to 28 hours, more preferably about 24 hours.

[0050] The fraction of viable cells may be defined as alpha. By dividing the fraction bound cells by alpha, a corrected value of the fraction of bound cells is obtained that represent the k_on as would have been determined when the cells would have been 100% viable. This means that by such correction, one corrects for the active fraction of cells, i.e. providing a value as if 100% of the cells would have been active. Hence, one can also likewise correct for any type of active fraction as can be determined. For example, in case the percentage of transduction is about 70%, one can correct by the fraction of transduction, i.e. an alpha of 0.7. When applied in means and methods that utilize k_on, this translates to a corrected Fraction_bound, divided by alpha. Hence, in a further embodiment, wherein a fraction alpha of the effector cells is active, the k_on rate is corrected therefor by using instead of the formula provided in step h), the following formula: k_on = - log(1 - (Fraction_bound[t]) I alpha) 1 1

[0051] One can provide alpha by determing the fraction of viable cells and / or the percentage of transduced cells. Instead, one can also provide alpha by determining the Fraction_bound at a defined incubation time which well exceeds 1 / k_on. Preferably, alpha is determined by measuring the fraction_bound at a defined incubation time of 2.5 x 1 / k_on, or more, wherein alpha corresponds with the fraction bound at that incubation time. Accordingly, in a further embodiment, alpha is provided by determining the Fraction_bound at a contacting time of 2.5 x 1 / k_on or more.

[0052] Use of an active settling force - spin assisted cell settling

[0053] In certain embodiments in means and methods in accordance with the invention, and the like, i.e. in methods for determining cellular avidity wherein a layer of attached cells (e.g. target cells, attached to a surface), is contacted with cells in suspension (e.g. effector cells) such as described i.a. in WO2023232881 , in which cells subsequently settle down to than subsequently have contact with the attached cells), it was observed that it may take a certain amount of time for cells to arrive at the monolayer, before the effector cells and target cells can interact. This means that the actual contacting time of each cell that was in suspension and that arrived at and has had contact with the attached cells, may vary and may differ from the incubation time I defined contacting time. This is because the distance and / or time spent by each cell settling down from the suspension under the influence of gravity to the layer of attached cells may vary, i.e. the arrival time for each cell in suspension to reach the attached cells may vary. Of course, one may assume that the average arrival time combined for all cells in a suspension of cells is substantially the same, for example when comparing different assays with similar types of cells, and / or between experiments. To a large extent such arrival time variation may be substantially averaged out, and, in means and methods as outlined herein, this aspect may be a constant factor which is part of the defined contacting time I incubation time having a relatively minor effect on calculations made.

[0054] Hence, one can use in the calculation of cellular avidity parameters such as outlined herein the incubation time or defined contacting time, e.g. defined as the time from the moment in time upon initiation of the contacting step up until the moment of exerting a force away from the attached cells (i.e. defined contacting time t = t_exerting force - tjnitating contact). For example, in a scenario wherein effector cells in suspension are introduced in a well with attached target cells at the ceiling, with the ceiling side up (upright position), the defined contacting time (corresponding to the incubation time) starts from the moment in time of placing the multi-well plate in the upside down position (Fig. 6B and C, (3)), up until the moment in time of placing the well (e.g. as part of a multi-well plate) in the upright position again and exerting a centrifugal force away from the attached cells (Fig. 6B and C, (5)). Positioning in the upright position and exerting a centrifugal force away from the attached cells occur directly consecutively. See Fig. 6A, 6B and 6C, and i.a. WO2023232881 , outlining a suitable workflow for carrying out methods in accordance with the invention. Such a defined contacting time can appropriately be used in calculation of cellular avidity parameters as outlined herein.

[0055] Alternatively, one may use in the calculation of cellular avidity parameters such as outlined herein instead the incubation time from the moment in time upon placing a well in the upside down position up until the moment of placing the well in the upright position and exerting a force away from the attached cells corrected by a factor related to the time it takes for the median of cells to settle. The latter corrected defined contacting time or corrected incubation time may more accurately be used to calculate cellular avidity parameters as defined herein. Hence, in any of the calculations as outlined herein, instead of the defined contacting time, the defined contacting time is corrected with the time the median of the cells have settled. This can be e.g. an estimated median settling time, or calculated based on physical parameters, or, this can be empirically determined for any preparation of cells, i.e. the corrected defined contacting time t = t_exert force - (t_initate contact + t_settling_median).

[0056] In means and methods in accordance with the invention, the effector cells in suspension highly preferably have substantially all settled on the target cells attached to a surface. The time it takes for substantially all cells to settle, may be referred to as settling time herein, i.e. settling time may be defined as the time it takes 98% or more of the cells in suspension to have settled down / arrived at the attached target cells such that the initially suspended effector cells are in contact with the target cells attached to the surface (or functionalized wall)). The time it takes for 50% of the cells to settle is referred to herein as a median settling time. In accordance with the present invention, even though considerations and scenarios as outlined above allow to select for incubation times for use in means and methods in accordance with the invention with which cellular avidity parameters can be well determined, the present inventors sought to further improve upon methods for determining cellular avidity parameters, in particular with regard to the aspect of the defined contacting time. The present inventors advantageously determined that by introducing spin assisted settling, i.e. by exerting a relatively low (centrifugal) force at the effector cells in suspension (referred to herein as an active settling force), the settling time can be significantly reduced as compared with settling times solely based on a gravity therewith advantageously improving accuracy of cellular avidity methods, such as outlined herein. An active settling force as defined in these embodiments refers to the force actively exerted on the effector cells in the direction of the attached cells. An active settling force does not refer to the force exerted on the effector cells in the direction away from the attached cells after the contacting step, such as described elsewhere in embodiments for determing cellular avidity, and as opposed to the gravitational force. Such means and methods with an active settling force allow for more accurate determination of cellular avidity parameters, in particular in scenarios wherein cells have or are suspected to have high k_on rates. In such scenarios, short defined contacting times may be preferred and advantageously, using corrected defined contacting times, more in particular corrected defined contacting times obtained by exerting an active settling force with spin assisted settling methods or the like, allows for more accurate cellular avidity determinations.

[0057] Hence, in the means and methods in accordance with the invention as outlined herein, in further embodiments, step c) of contacting the effector cells with the target cells, comprises providing the effector cells in suspension, and, bringing the suspension in contact with the target cells attached to a surface, and subsequently exerting an active settling force on the effector in the direction of the target cells attached to the surface.

[0058] In another embodiment, step c) of contacting the effector cells with the target cells, comprises providing the effector cells in suspension in a well with a ceiling surface with target cells attached thereto and a bottom surface, optionally the bottom surface coated with antifouling, and, subsequently exerting an active settling force on the effector cells in the direction of the target cells attached to the surface. Preferably, in a further embodiment, after the effector cells are introduced in a well in the upright position, the effector cells are first allowed to settle to the bottom of the well, and subsequently, the well is positioned in the upside down position and an active settling force is exerted to allow the effector cells to settle on the target cells attached to the ceiling of the well. Such embodiments are depicted respectively in Figure 7A and 7B.

[0059] In a preferred embodiment, the active settling force exerted is a centrifugal force. Preferably, the active settling force exerted is in the range of 2 - 20 x g (i.e. g represents the standard gravity of 9.8 m / s2), more preferably about 2 - 10 x g. The duration in which the active settling force is exerted is preferably in the range of 10 seconds to 1 minute, preferably in the range of 30 seconds to 1 minute, more preferably in the range of 40 seconds to 50 seconds. In another embodiment, the maximum distance to the monalyer of cells in suspension is 1 mm or less, more preferably 0.5 mm or less.lt is understood that in any of the means and methods of embodiments with an active settling force exerted as outlined herein, the defined contacting time preferably is corrected by the t_settling_median, i.e. the corrected defined contacting time is defined as the time period from t_settling_median until the moment in time of exerting the force away from the attached target cells. In embodiments involving methods such as outlined in Figure 7A and B (with contacting steps in wells with the effector cells in suspension in a well with a ceiling surface with target cells attached thereto and a bottom surface), the corrected defined contacting time can be defined as the time period from t_settling_median until the moment in time of positioning in the upright position and exerting the force away from the attached target cells. In any of the calculations related to cellular avidity as outlined herein, t_settling_median can be calculated or empirically determined. The t_settling median can also be set in case of exerting an active settling force for 45 seconds or more, to be 22.5 seconds.

[0060] Blocking cell surface molecules

[0061] In addition to the above related to k_on, and active fraction correction, the present invention further provides in addition a method for determining cellular avidity of effector cells carrying a receptor to target cells, wherein the method comprises contacting the effector cells with the target cells and allowing the effector cells to bind with the target cells, and, subsequently exerting a force, and determining the fraction of effector cells and target cells that remain bound thereafter, wherein the target cells present a cell surface molecule and the receptor of the effector cells is capable of binding said cell surface molecule, wherein the amount of available cell surface molecule at the target cells is modulated by providing a blocking molecule capable of binding said cell surface molecule, and said blocking molecule is capable of blocking binding of the receptor to the cell surface molecule of the target cell.

[0062] As shown in the examples herein, by providing such a blocking molecule, and including it in the contacting step the amount of available target molecule can be advantageously modulated. It is understood that including a blocking molecule in the contacting step may include first incubating the target cells with the blocking molecule and subsequently add the effector cells. It is understood that modulating target cell availability can be advantageous in any type of cellular avidity measurement involving contacting effector cells and target cells for a defined time, and subsequently exerting a force, after which the fraction of effector cells that remained bound to the target cells is determined. Such cellular avidity measurements, preferably include determining k_on and / or corrections for active fractions, as outlined herein.

[0063] As shown in the examples herein, said blocking molecule can comprise an antigen binding domain. It is understood that the receptor of the effector cell may comprises an antigen binding domain as well (e.g. as comprised in a CAR or the like). These antigen binding domains may be selected to be the same. Of course, one may also opt for a different blocking molecule instead, as long as the blocking molecule is capable of blocking the binding of the receptor of the effector cell to the cell surface target molecule presented, such a blocking molecule can be contemplated. Accordingly, in one embodiment, the said blocking molecule comprises an antigen binding domain capable of binding said cell surface molecule. In a further embodiment, the receptor of the effector cell comprises an antigen binding domain capable of binding said cell surface molecule. In a further embodiment, the antigen binding domain of said blocking molecule corresponds with the antigen binding domain comprised in the receptor of the effector cells.

[0064] The blocking molecule may also be used to determine the amount of cell surface molecules presented by a target cell. This can be done e.g. by providing the blocking molecule with a quantifiable marker, providing an excess of blocking molecule with the quantifiable marker and allowing it to interact with the target cells and subsequently quantify the amount of quantifiable marker per cell. This way, the amount of cell surface molecule, i.e. target molecule, presented by a target cell can be determined. This allows for example to modulate the amount of available cell surface molecule per cell in a quantifiable manner, and e.g. advantageously allows for example to have different target cells provide for the same amount of available cell surface target molecules per cell. For example, one can provide predetermined high, medium and low amounts of a target antigen on different cell types. This may also be useful as it may allow to compare different effector cells targeting different surface molecules, as one may select the different surface molecules available to be the same.

[0065] Hence, in one embodiment, provided is a method for determining cellular avidity, comprising determining the amount of cell surface molecules presented by the target cells with said blocking molecule, prior to performing cellular avidity measurements. In another or further embodiment, in the method of determining cellular avidity, this is performed with a defined amount of available cell surface molecule at the target cells. In yet another or further embodiment, the amount of available cell surface molecule at the target cells is modulated in accordance with any of claims, wherein k_on is determined based on a defined amount of cell surface molecule at the target cells.

[0066] In particular, by modulating the amount of cell surface molecules available, EC50 values can be determined. These EC50 values can be determined for the amount of cell surface molecules, if determined. Alternatively, EC50 values can be determined for the blocking molecule concentration. In any case, EC50 values or the like, are highly useful as outlined in the example herein. Hence, in a further or another embodiment, when modulating the amount of cell surface molecule available at a target cell with a blocking molecule, the cellular avidity EC50 is determined of the amount of cell surface molecules available, or of the blocking molecule concentration.

[0067] EMBODIMENTS

[0068] 1. A method for determining cellular avidity of effector cells carrying a receptor to target cells, the method comprising the steps of: a) providing effector cells carrying a receptor; b) providing target cells; wherein the target cells are attached to a surface; c) contacting the effector cells with the target cells to allow the effector cells to bind with the target cells, wherein in the contacting step, the target cells attached to the surface are in excess as compared with the effector cells, and wherein the interaction is for a defined time t ; d) applying a force, wherein the force is in a direction away from the attached target cells, such that at least part of the effector cells in contact and / or bound to the target cells attached to the surface move away therefrom; e) detecting effector cells that have remained bound with the target cells and attached to the surface after applying the force; f) determining the fraction of the effector cells (Fraction_bound[t]), that were contacted with the target cells that have remained bound; g) optionally, perform steps c), d), e) and f) one or more times; and h) calculating the cellular avidity metric k_on from

[0069] - determined fraction(s) that remained bound (Fraction_bound[t]) and defined contacting time(s) t; with the following formula: k_on = - log(1-Fraction_bound[t]) 1 1 to therewith provide the cellular avidity metric k_on.

[0070] 2. The method in accordance with embodiment 1 , wherein in step g), the steps c), d), e) and f) are performed one ore more times at different defined contacting times.

[0071] 3. The method in accordance with embodiment 2, wherein the different defined times are selected from 0 minutes to 60 minutes.

[0072] 4. The method in accordance with embodiment 2 or embodiment 3, wherein the defined times include defined times of between 0 minutes and 10 minutes, between 10 minutes and 20 minutes, between 20 minutes and 40 minutes, between 40 minutes and 60 minutes.

[0073] 5. The method in accordance with any of embodiments 2-4, wherein the k_on is calculated by using a least squares fitting method. 6. The method in accordance with embodiment 1 , wherein in step g), the steps c), d), e) and f) are performed one or more times at the same initially defined contacting time.

[0074] 7. The method in accordance with any of embodiments 1-6, wherein in step h) k_on_initial is determined, and, subsequently, step i) is performed, wherein in step i) subsequently steps c), d), e) and f) are performed one or more times; at a defined time in step c) of 1 / k_on_initial, and subsequently k_on is calculated based on the determined fraction(s) that remained bound (Fraction_bound(t)) and defined contacting time(s) t obtained in step i) using the formula provided in step h).

[0075] 8. The method in accordance with any one of embodiments 1-7, wherein effector cells with different receptors are provided and for each effector cell with a different receptor the method is performed.

[0076] 9. The method in accordance with any of embodiments 1-6, wherein effector cells with different receptors are provided and for each effector cell with a different receptor the method is performed and in step h) a k_on_initial is calculated, and, subsequently, the highest and lowest k_on_initial determined (k_on_init_h and k_on_init_l, respectively), and subsequently, step i) is performed, wherein in step i) steps c), d), e) and f) are subsequently performed one or more times; at an optimal defined time in step c) of t = log (k_on_init_h / k_on_init_l) I ((k_on_init_h - k_on_init_l)) and subsequently k_on is calculated based on the determined fraction(s) that remained bound (Fraction_bound(t)) and defined contacting time(s) (t) obtained in step i) using the formula provided in step h).

[0077] 10. The method in accordance with embodiment 9, wherein highest and lowest k_on_initial are determined for subgroups of different effector cells, and for each subgroup step i) is performed with their corresponding optimal defined time in step c). 11.The method in accordance with any of embodiments 1-10, wherein the method is performed at a defined temperature and in a defined cell culture medium.

[0078] 12. The method in accordance with any of embodiments 1-11 , with one type of target cell or with different target cells.

[0079] 13. Method in accordance with any of embodiments 1-12, wherein the target cells attached to the surface are in at least a 10-fold excess as compared with the effector cells carrying a receptor.

[0080] 14. Method in accordance with any of embodiments 1-12, wherein the target cells present a cell surface molecule and the receptor of the effector cells is capable of binding said cell surface molecule.

[0081] 15. The method in accordance with any of embodiments 1-14, wherein the target cells present a cell surface molecule, and the contacting step c) is performed in the presence of a cell engager, wherein the cell engager has one binding arm capable of binding with the cell surface molecule of the target cell and another binding arm capable of binding said receptor of the effector cells.

[0082] 16. The method in accordance with any of embodiments 1-13, wherein instead of target cells attached to a surface, a functionalized surface is provided comprising a cell surface molecule with which the receptor of the effector cells is capable to bind, and the method is performed with the functionalized surface instead of the target cells attached to the surface.

[0083] 17. The method in accordance with any of embodiments 1-13, wherein instead of effector cells carrying a receptor, beads functionalized with receptors at their surface are provided, and the method is performed with the beads instead of the effector cells.

[0084] 18. A method for determining cellular avidity of effector cells carrying a receptor to target cells, wherein the method comprises contacting the effector cells with the target cells and allowing the effector cells to bind with the target cells, and, subsequently exerting a force, and determining the fraction of effector cells and target cells that remain bound thereafter, wherein the fraction of cells that remains bound after exerting the force is corrected for cell viability. The method in accordance with any of embodiments 1-17, wherein the fraction of cells that remains bound after exerting the force, Fraction_bound, is corrected for cell viability. The method in accordance with embodiment 18 or embodiment 19, wherein the fraction of the cells that is viable is calculated and the Fraction_bound is corrected for cell viability by division by the viable fraction value. The method in accordance with any of embodiments 18-20, wherein the effector cells are thawed prior to performing a cellular avidity measurement, and cell viability is determined. The method in accordance with embodiment 21 , wherein the effector cells are thawed prior to performing a cellular avidity measurement, and cell viability is determined after allowing the cells to recover from thawing. The method in accordance with embodiment 21 or 22, wherein the cells are allowed to recover for about 20 - 28 hours prior to measuring cell viability, preferably 24 hours after thawing. The method in accordance with any of embodiments 21-23, wherein cellular avidity is determined immediately after thawing the effector cells. The method in accordance with any of embodiments 1-17, wherein a fraction alpha of the effector cells is active, and wherein the k_on rate is corrected therefor by using instead of the formula provided in step h), the following formula: k_on = - log(1 - (Fraction_bound[t]) / alpha) 1 1 The method in accordance with embodiment 25, wherein alpha is provided by determining the Fraction_bound at a contacting time t of 2.5 x 1 / k_on or more. method for determining cellular avidity of effector cells carrying a receptor to target cells, wherein the method comprises contacting the effector cells with the target cells and allowing the effector cells to bind with the target cells, and, subsequently exerting a force, and determining the fraction of effector cells and target cells that remain bound thereafter, wherein the target cells present a cell surface molecule and the receptor of the effector cells is capable of binding said cell surface molecule, wherein the amount of available cell surface molecule at the target cells is modulated by providing a blocking molecule capable of binding said cell surface molecule, and said blocking molecule is capable of blocking binding of the receptor to the cell surface molecule of the target cell. The method in accordance with any of embodiments 1-26, wherein the target cells present a cell surface molecule and the receptor of the effector cells is capable of binding said cell surface molecule, and, wherein the amount of available cell surface molecules at the target cells is modulated by providing a blocking molecule capable of binding said cell surface molecule, and said blocking molecule is capable of blocking binding of the receptor to the cell surface molecule of the target cell. The method in accordance with embodiment 27 or embodiment 28, wherein the said blocking molecule comprises an antigen binding domain capable of binding said cell surface molecule. The method in accordance with embodiment 29, wherein the receptor of the effector cell comprises an antigen binding domain capable of binding said cell surface molecule. 31. The method in accordance with any of embodiments 29 or 30, wherein the antigen binding domain of said blocking molecule corresponds with the antigen binding domain comprised in the receptor of the effector cells.

[0085] 32. The method in accordance with any of embodiments 27-31 , wherein the method comprises determining the amount of cell surface molecules presented by the target cells with said blocking molecule, prior to performing cellular avidity measurements.

[0086] 33. The method in accordance with any of embodiments 27-32, wherein in the method of determining cellular avidity, this is performed with a defined amount of available cell surface molecules at the target cells.

[0087] 34. The method in accordance with any of embodiments 28-33, wherein the amount of available cell surface molecules at the target cells is modulated in accordance with any of embodiments, wherein k_on is determined based on a defined amount of cell surface molecules at the target cells.

[0088] 35. The method in accordance with embodiment 27-34, wherein the cellular avidity EC50 is determined of the amount of cell surface molecules, or of the blocking molecule concentration.

[0089] 36. The method in accordance with any of embodiments 1-35, wherein the method is implemented as a high throughput method.

[0090] 37. A method for determining cellular avidity of effector cells carrying a receptor to target cells, the method comprising the steps of: a) providing effector cells carrying a receptor; b) providing target cells; wherein the target cells are attached to a surface; c) contacting the effector cells with the target cells to allow the effector cells to bind with the target cells; wherein the effector cells are provided in a cell suspension and wherein in the contacting step an active settling force is exerted on the effector cells in the direction of the attached cells; wherein in the contacting step, the target cells attached to the surface are in excess as compared with the effector cells, and wherein the contacting is for a defined time t; d) applying a force, wherein the force is in a direction away from the attached target cells, such that at least part of the effector cells in contact with and / or bound to the target cells attached to the surface move away therefrom; and e) detecting effector cells that have remained bound with the target cells and attached to the surface after applying the force.

[0091] 38. The method in accordanc with embodiment 37, wherein step c) of contacting the effector cells with the target cells, comprises providing the effector cells in suspension in a well with a ceiling surface with target cells attached thereto, and a bottom surface, optionally the bottom surface coated with antifouling, and, subsequently exerting the active settling force on the effector cells in the direction of the target cells attached to the ceiling surface.

[0092] 39. The method in accordance with embodiment 38, wherein after the effector cells are introduced in the well in the upright position, the effector cells are first allowed to settle to the bottom of the well, and subsequently, the well is positioned in the upside down position and the active settling force is exerted to allow the effector cells to settle on the target cells attached to the ceiling of the well.

[0093] 40. The method in accordance with any of embodiments 37-39, wherein the active settling force exerted is a centrifugal force, preferably wherein the force is in the range of 2 x g - 20 x g, more preferably wherein the force is in the range of 2x g - 10x g.

[0094] 41. The method in accordance with any of embodiments 37-40, wherein the active settling force is exerted for 10 - 60 seconds, more preferably from 30 - 50 seconds, most preferably for 45 seconds. 42. The method in accordance with any of embodiments 37-41 , wherein the method is further in accordance with any of embodiments 1-36.

[0095] 43. The method in accordance with embodiment 42, wherein the defined contacting time is corrected by t_settling median.

[0096] 44. Methods in accordance with any of embodiments 1-43, wherein methods are performed in a multiwell format, such as 48-well or 96-well format.

[0097] EXAMPLES

[0098] Materials and methods and measurements

[0099] Experiments as described in the examples herein were performed, wherein effector cells and target cells were allowed to interact, and subsequently a force exerted thereon, to determine cell fractions bound after defined interaction times. The experimental set up was as follows.

[0100] Effector cells used in the examples herein are Jurkat cells, which were transduced with CD19 targeting CARs FMC, 4G7 or B4 (Kramer, Anne Marijn. “Delineating the Impact of Binding-Domain Affinity and Kinetic Properties on Chimeric Antigen Receptor T Cell Function.” PhD Thesis, University College London, April 30 2017), as control cells, untransduced (UNT) Jurkat cells were used. As target cells, Nalm6 cells were used in the examples herein. Where in the examples or figures CARs are refered to as FMC these refer to FMC63 cars as described in the Kramer reference above.

[0101] 1 day before an experiment, Ibidi chips (Ibidi #cat 80606) were coated with 30pL of Poly-L-Lysine (Sigma-Aldrich, catalog number: P4707) diluted 1 :5 in PBS (Thermo Fisher, catalog number: 10010023) and incubated at RT for 30 minutes, after which liquid was removed with a syringe and the chips left to dry overnight at 37°C.

[0102] On the day of an experiment, target cells were obtained from cell culture and resuspended in culture medium at, a previously determined, cell line specific concentration. Holding the ibidi chip upside down, 30pL of target cell suspension was pipetted into each channel. For Nalm6, the concentration of cells was 30 million cells per mL, hence 9 x 10A5 cells were pipetted into each channel. This amount of cells results in a highly confluent monolayer.The target cells were kept at 37°C with inlets down until the cells adhered to the channel ceiling, and medium was exchanged halfway.

[0103] To test monolayer integrity 60pl medium was added to each channel and ibidi chips centrifuged with inlets positioned up for 2 minutes at 1000 x g. Monolayer integrity was confirmed by comparing images collected before and after centrifugation.

[0104] To prepare effector cells, frozen aliquots of effector cells were thawed and resuspended at a concentration of 1x10A6 / mL in T cell culture medium: RPMI 1640 + GlutaMAX, 1X Penicillin-Streptomycin, 10% FBS. 35,000 cells were transferred per channel to 96-well plate in 100pL of CellTrace™ Far Red dye (C34564, Invitrogen) diluted 1 :2000 in PBS. The cells were kept at 37°C for 15 minutes. Thereafter, cells were washed with PBS and resuspended in T-cell medium at a concentration of 0.875x10A6 cells / mL.

[0105] To perform a cellular avidity measurement assay, 40pL of effector cells was added to each channel of the ibidi chip with the inlets facing up. Hence, the ratio of the number of effector cells added to the number of Nalm6 added was ( 3,5 x 10A4 : 9 x 10A5 ~ 1 : 25). Once all channels were filled, the ibidi chip was flipped along the longitudinal axis such that the inlets were facing down. The effector cells need to settle down to come into contact with the cell monolayer, which takes about 1 minute. This means contacting times (t) correspond with the time after flipping minus 1 minute (to which may be referred herein throughout interchangeably as incubation time or (defined) time of contacting). During contacting, each channel was imaged and the location of each effector cell was recorded. When the incubation time, i.e. the contacting step, was completed, the ibidi chip was flipped again such that inlets were facing up, and 60pL of T-cell medium was added to each channel and the chips were immediately centrifuged for 2 minutes at 1000 x g. After centrifugation, each channel was imaged and of each remaining effector cell the location was recorded.

[0106] To quantify cell avidity, per channel, in a typical measurement, about 500 cells were assessed in a field of view. The Fraction_bound was calculated by determining the number of cells that remained attached to the monolayer after centrifugation (number of bound cells), and determining the number of cells on the monolayer prior to centrifugation (total number of cells), and dividing the number of bound cells by the total number of cells. This Fraction_bound calculated number was multiplied by 100% to provide for a percentage of bound cells. This percentage represents a value related to cellular avidity, per channel.

[0107] Example 1

[0108] The interaction between effector cells and target cells is a complex interaction in which multiple receptors and ligands at the surface of each cell can interact, ultimately resulting in binding of effector cells and target cells (Figure 1). Cell-cell interaction is thus understood to involve multiple interactions between ligands and receptors between cells, ultimately resulting in cell-cell binding. Cell-cell interaction also is understood to possibly involve association and dissociation between receptors and ligands. The present inventors now performed experiments and an extensive in depth analysis of the data to determine whether there could be a mathmetical model that could accurately describe the fraction bound cells after exerting a force after a defined contacting time.

[0109] Multiple measurements were performed for binding of effector cells to target cells. Four different types of effector cells (Jurkat cells transduced with different CAR encoding lentiviral vectors) were used in the measurements, each capable of targeting the same target cell (Nalm6). The effector cells were incubated with the target cells adhered to a surface (i.e. brought into contact), for a defined time t (in minutes), after which a force away from the target cells was exerted. This way, the fraction of bound cells (Fraction_bound) remaining after the contacting time t was determined. Contacting times selected were up to 60 minutes. In Figure 2, the Fraction_bound (in the range of 0-1) is depicted as a percentage, i.e. the Fraction_bound x 100%. The data points were plotted in Figure 2, i.e. (contacting time, t (min), on the x-axis, against bound cells, Fraction_bound (%), y-axis. Extensive analysis revealed that, highly surprisingly given the complex nature of cell-cell interaction, a constant binding rate model most accurately describes the relation between Fraction_bound and contacting time in minutes. The formula that describes this interaction is: Fraction_bound[t] = 1 - exp(-k_on * t)

[0110] In Figure 2B, the data points were fitted using this equation using least squares fitting (From the Python SciPy package v1.11.2, using the curve_fit function in scipy. optimize. All function parameters left at default values. Apart from an initial guess for k_on = 0.1). The formula above can be rewritten to express k_on: k_on = - log (1 - Fraction_bound[t]) 1 1

[0111] This mathmetical model thus allows to provide for a k_on rate for a given cell cell interaction, such as an effector cell and a target cell. This k_on rate is independent of contacting time and represents a metric characterizing cellular avidity.

[0112] This cellular avidity metric is representative of the mechanism of action of binding. The higher the k_on rate, the faster effector cells bind with the target cells. This metric also allows to quantify differences more appropriately, instead of a difference, one can now express this difference, e.g. in a ratio or difference of different k_on rates. For example a given effector cell may have a k_on rate which is twice faster than a reference effector cell. By determing the cellular avidity metric k_on, optionally including a reference (e.g. a reference effector cell) in an experiment, one can now more accurately compare different experiments, between different laboratories and / or performed on different points in time.

[0113] Example 2

[0114] The use of the cellular avidity metric in experiments allows for conducting fewer measurements to achieve statistical significant results, in particular when comparing e.g. different effector cells and / or different target cells, as compared with measurements conducted at a defined contacting time.

[0115] This is because experimental differences in percentage of cells bound, i.e. Fraction_bound, between two different conditions (e.g. different effector cells) depends strongly on the chosen contacting time. This is exemplified by the plots shown in Figure 3. In these plots it is shown what the effect of different k_on rates and standard deviation is. These plots assume different k_on rates and a standard deviation of 5% at the contacting time t = 1 / k_on. As is clear from the plot, when it is desirable to measure k_on with an error close to 5%, e.g. below 10%, it is important to select a contacting time close to 1 / k_on (vertical gray dashed lines). If k_on rates differ, this means that it would require more measurements to reach the same statistical power as compared with measurements performed for each condition at a defined time of contacting close to 1 / k_on. The plot also shows that it may not be possible to obtain accurate mearurements of konat a single defined time of contacting for experiments on cells that show substantial differences between binding rates.

[0116] In the table below, the number of measurements (sample size) required is shown for estimating k_on with an error less than 10%. These calculations are based on a k_on = 0.1 min-1with a standard deviation in technical repeats of 2%. This calculations show that for estimating k_on with an incubation time (i.e. defined time for the contacting step) very different from 1 / k_on, the required sample size to keep the error below 10% becomes prohibitively high. For example, with an incubation time of 40 minutes, 8 measurements need to be performed to keep the error below 10%.

[0117] Table 1. Sample size requirements with different incubation times.

[0118] This can be resolved by picking multiple incubation times (equally) spaced in time, e.g. between 1 - 60 minutes (t), determine fractions bound (Fraction_bount) and calculating the binding rate by fitting the equation describing the fraction bound[t] given above for each construct using a least squares fitting method (e.g. by using the Python SciPy package v1 .11.2, using the curve_fit function in scipy. optimize.)

[0119] In particular, this becomes even more problematic when many different constructs need to be compared. When comparing constructs (with different bindig rates), choosing an optimal incubation time optimal for all constructs is highly challenging, requiring a large number of measurements. This can be resolved by picking multiple incubation times (equally) spaced in time, e.g. between 1 - 60 minutes (t), determine fractions bound (Fraction_bount) and calculating the binding rate by fitting the equation describing the fraction bound[t] given above for each construct using a least squares fitting method (From the Python SciPy package v1 .11 .2, using the curve_fit function in scipy. optimize. All function parameters left at default values, apart from an initial guess for k_on = 0.1). The binding rates obtained can be directly compared as they are independent of the time of contacting. Based on an initially determined k_on rate (k_on_intial) for a given condition, one can also perform subsequent further incubations within a suitable range of an a determined time point (1 / k_on_initial), to further improve statistical power of an k_on rate. Based on an initially determined k_on rate (k_on_intial) for a given condition, one can also perform subsequent further incubations at a determined time point (at 1 / k_on_initial), to further improve statistical power of the k_on rate. Hence, an iterative process allows for reducing the number of measurements even further.

[0120] In addition, when using a single defined time of contacting, one could first measure the Fraction_bound for all or a subset of all the constructs in the study at a fixed time (step 1). Subseqently one makes an initial calculation for k_on for each construct using the equation above. By taking the highest and lowest initial binding rates obtained, k_on_high (k1_on) and k_on_low (k3_on) respectively, an optimal time can be defined for contacting using the equation: optimal contacting time t= Iog(k1_on / k2_on) / (k1_on-k2_on) and a repeat avidity measurement may be performed again at this optimal incubation time. This may optionally be an iterative process, after each round of measurements calculating the improved k_on rates.

[0121] However, when comparing constructs with very different binding rates, choosing a time for contacting that is optimal for all constructs is not possible (as e.g. shown above). Hence, the above can also be done for subgroups having similar binding rates. For subgroups, optimal times for contacting can be selected by the mathematical formula above. Hence, one would first perform step 1 , as described above, to find a rough estimate for konfor each construct, and then group the constructs into groups with comparable binding rates; e.g. classify as either fast, medium or slow binding candidates. The cellular avidity metric k_on for each member of a subgroups can then be measured at a contacting time that is the median 1 / k_on for each group.

[0122] The table 2 below shows that using the binding rate obtained by fitting the % of bound cells at 4 different incubation times gives a better significance of separation between different constructs as compared to measuring the % bound cells at a single optimal incubation time (optimal incubation time as determined in the equation above). Here we compared 4 constructs, 2 of which have high binding rates (constructs 1 and 2) and 2 have low binding rates (construct 3 and 4). To show how well we can separate construct 1 from construct 2 and construct 3 from construct 4, we give the signal to noise ratio (SnR) in the table. The SnR is defined as the difference in the mean between 2 constructs (comparing the fast binding constructs 1 and 2 and the slow binding constructs 3 and 4) devided by the combined standard errors (SEM) of the 2 constructs. We calculate the SEM as the square root of the sum of the variances of the construct divided by the sarnie size N of each construct. We calculate the SnR for the % bound cells in Method 1 and for the binding rates in method 2. Clearly, using a fit to obtain k_on allows you to separate constructs with higher significance than using the % bound cells taken at a single optimal incubation time.

[0123] For method 1 , the mean and std of the % bound cells for the different binding rate are obtained from 100%*Fraction_bound[t] and a 2% std in technical repeats. For method 2, the mean and std of the binding rates were obtained from 500 simulations by fitting the model to simulated data points defined as 100%*Fraction_bound[t] + N(0, 2%). We used 4 incubation times of 1 , 10, 30 and 60 minutes. N(0, 2%) is the normal distribution with zero mean and a std of 2%. For a fair comparison between methods, we use a sample size of 4 for each construct in method 1 and 4 time points with sample size 1 for each construct in method 2.

[0124] In all, the result of integrating the k_on metric in the experimental setup and workflow is that the results provided are highly improved, i.e. resulting in a meaningful cellular avidity metric, i.e. k_on, which also requires a lower number of measurements. This provides for more robust and sensitive measurements as opposed to cellular avidity measurements as commonly performed at a single predefined incubation time.

[0125] Example 3

[0126] Target cell binding is understood to be dependent of initial (slow) effector cell probing behaviour which bring the effector and target cell membranes into close proximity and allow for an initial contact in cell-cell binding. It was hypothesized that in the absence of effector cell probing, interaction between the receptor (of the effector cell) and target antigen (on the target cell) would not be possible or at least highly impaired, resulting in the absence of observing cell-cell binding. Only viable effector cells would be capable of probing behavior. In conducting cellular avidity measurements (i.e. determining cell fractions bound after defined incubation times) it was observed that the fraction of dead cells in an effector cell sample was influenced by cell preparation and handling and could be highly variable and difficult to control. The relationship between effector cell viability and Fraction_bound was determined. It was observed that there was a linear relationship between between effector cell viability and the fraction bound. This indicates that in order to improve cellular viability measurements, i.e. determining the Fraction_bound of a certain effector cell, after defined incubation times, as described above, this determined Fraction_bound can be corrected for cell viability to further improve accuracy of measurements.

[0127] It was also observed that cellular avidity measurements are generally preferred early after thawing to reduce variability. Strikingly and highly surprisingly, it was observed that the correlation between the cellular avidity measurements (i.e. of fractions bound after defined incubation times) and viability measurements of thawed cells, was best when viability measurements were carried out about 24 hours after thawing.

[0128] An exemplary result of a plot showing the correlating between cell viability (depicted as % dead cells) to cellular avidity measured (depicted as the % of bound cells to target cells remaining after extering a force after a defined incubation time) is shown in Figure 4, wherein the cellular avidity was measured shortly after thawing as described above and cell viability was determined about 24 hours post thawing. Cell viability was determined via AnnexinV staining (Invitrogen, Ref.No. R37174). Similar results are observed whether avidity is measured using the centrifuge method described above or using the LUMICKS The z-Movi® Cell Avidity Analyzer.

[0129] In contrast, viability measured immediately after thawing, at the same time as the cellular avidity measurements provided was shown to result in a correlation as well, but this was less strong (not shown). Combined, this indicates that cellular avidity measurements, i.e. determining the fraction of cells that remain bound, are preferably conducted shortly after thawing, and such measurements should be corrected for the percentage of viable / dead cells of the sample, which is preferably the percentage of viable / dead cells as determined about 24 hours post thawing.

[0130] Hence, this shows that cellular avidity measurements can be improved by correcting for the fraction of viable cells in the measured cell population. Given the strong correlation shown in Figure 4, it should be possible to separately measure cell viability and correct avidity data by linearly projecting the measured avidity, to that of a 100% viable sample.

[0131] Likewise, a similar correction may be performed with the percentage of transduced cells, e.g. in case the percentage of transduced cells is reduced, e.g. is 75%, that would mean that the cellular avidity as measured is underestimated as well, and one could correct for that.

[0132] Estimating binding rates when not all cells bind

[0133] The equation for the binding rate kondefined above assumes all cells in the population bind with an average rate kon. However, in case there is a compounding variable that inhibits binding for a subset of cells (e.g. the viability of the samples is reduced as describe above, or, the percentage of transduced cells, e.g. the percentage of cells expression a CAR is lower than 100%) the binding rate will be underestimated in the samples since not all cells have the capacity to bind. The mathematical equation describing the k_on rate for the case when only a fraction alpha of cells in the population are capable of binding and the remaining cells are not capable of binding is as follows: k_on = - log (1 - Fraction_bound [t] I alpha) 1 1

[0134] To calculate k_on we would need to know the fraction of active cells in the population, alpha. In case the k_on is corrected for cell viability, alpha can be the fraction of cells that is viable, that would correct the Fraction_bound [t] to a number corresponding with the Fraction_bound as observed in case 100% of the cells would have been viable in the measurement In case of the percentage of transduced cells would be a compounding variable, the fraction of transduced cells can be taken as alpha.

[0135] To find it, we measure the % bound cells at an incubation time at least 2.5 times longer than 1 / kon. We can use this % bound cells as an estimate for alpha and calculate kon. We choose the condition incubation time > 2.5 / konsuch that the error in our estimate for alpha is less than 10%.

[0136] Alternatively, one could use the methods described above and fit the equation for konwith the active cell fraction alpha to the data obtained at different incubation times. Provided the longest incubation time is larger than 2.5 times 1 / kon, this should give a good estimate for kon.

[0137] This way, alpha corresponds with the Fraction_bound at an incubation time which is well in excess of the 1 I k_on, and corresponds with the plateau reached. So, in case the plateau reached is 90%, this means alpha is 0.9. This is an alternative method to correct for cell viability or for the percentage of transduced cells, or the like.

[0138] By combining the constant binding rate model with the active cell fraction model one can calculate the correct binding rates even in cases where not all cells are active (e.g. where the fraction of viable cells or the purity of the cells [the fraction of cells that carry the receptor] is lower than one or where there is a population of non-binding cells because some cells are in a different state). In this case there is a plateau fraction of bound cells lower than one.

[0139] Example 4

[0140] In this example, cellular avidity measurements were performed in the presence of an antibody. An amount of antibody was titrated in, i.e. a concentration series of the antibody and the effect on the fraction of cells that remains bound was determined (See Figure 5). During monolayer incubation, blocking antibody targeting CD19 (PE anti-human CD19 Antibody, Clone HIB19, obtained from Biolegend), was introduced to the channel for at least 30 minutes prior to cellular avidity measurements. After monolayer / antibody incubation, avidity measurements were conducted with 4 effector samples (an untransduced negative control and 3 CARs of different expression levels). Blocking antibody was kept present throughout the experiment to ensure the extent of blocking remained consistent during target : effector cell incubation and force application.

[0141] The data shows that including an antibody which binds the tumor associated antigen which obscures a target, e.g. a target antigen such as a CAR binding epitope, the number of available antigen on the target cell surface can be controlled. This allows for testing the effector cell “sensitivity”, i.e. the cellular avidity response to variations in antigen density.

[0142] This advantageously allows for modulation of e.g. antigen density on target cells. This way, by modulating antigen density on target cells, different cell types can be modulated such that these present similar amounts of antigen, or multiple antigen densities can be assessed within the context of the same monolayer type.

[0143] The latter, also allows to determine an Avidity EC50 by titrating down from high- enough density to baseline where the EC50 is the available antigen density where binding is halfway between the plateau and the background binding. This can be done with all other experiment parameters (e.g. incubation time) fixed. One can also titrate down until binding becomes close to becoming insignificant (EC05 or less) indicating the antigen density where binding become unmeasurable. EC50 and EC05 could prove meaningful for CAR binding dynamics. Actual antigen density, number of available antigen, per concentration step should be determined for each measured concentration and cellular avidity EC50 assessed. A low EC50 means that the cell avidity of an effector cell is high even at low antigen densities. Therefore this effector cell is sensitive to the specific antigen on the target cells. This may be an advantage in case of e.g. tumors presenting low antigen densities. A high sensitivity may also indicate the effector cell could have a high on-target, off-tumor reactivity which may be unwanted. A high value of EC50 means that the effector cell avidity is high only at high antigen densities. This may mean there is a lower chance of off-tumor reactivity but may also mean the effector cell is less effective against tumors expressing low antigen densities.

[0144] Modulating availability of target molecule, such as target antigens (e.g. MHC presented antigen or a membrane protein presented on the surface of a target cell, such as CD19), allows for assessing difficult to target cells (because of low amounts present on the surface of cells) or assessing the risk for off target effects (because of e.g. relatively low amounts of target molecules on healthy cells).

[0145] Example 5

[0146] To probe the interaction between effector cells and a protein or target molecule of interest experiments may also be performed by functionalizing a surface directly with such a molecule instead of using a layer of target cells. Functionalization may be performed using either physical or chemical methods (e.g. physisorption, chemical crosslinking, or high affinity molecular interactions e.g. Avidin-biotin). The protocol below describes the functionalization using avidin and the subsequent attachment of a biotinylated target molecule to create a surface for avidity measurements.

[0147] Activation and coating of z-Movi chips z-Movi chips were activated by flusing the channel with 150 pL of 1M NaOH of a clean z-Movi chip, and incubation of the chip for 15 minutes at room temperature. Channels were rinsed with 2 flushes of dH20 and 2 time PBS (pull 0.15 mL each time, leaving each time 10 pL in the reservoir). Subsequently 100 pL undiluted Poly-L-Lysine (Sigma #P4707-50ML) was flushed in the channel and left to incubate for 10 minutes at room temperature. The Chip was rinsed with 3x flushes of PBS before crosslinking of avidin. BS(PEG)9 was dissolved in anhydrous DMSO to obtain a stock concentration of 250 mM and a final concentration of 100mM was made by further dilution in PBS. This crosslinking solution was flushed in the channel and left to incubate for 20 minutes at room temperature. After rinsing with PBS an Avidin solution (0.5 mg / mL) in PBS was introduced and incubated for 20 minutes at room temperature, followed by another 3 rinses with PBS. The chips were now ready to bind the biotinylated protein or molecule.

[0148] Attachment was performed by introducing the biotinylated target at a final concentration of 1-50 pg / mL in PBS and incubating for 20 minutes at room temperature before rinsing it out with 3 flushes of PBS. The chip was then ready to perform an avidity experiment of effector cell toward a the target molecule bound on the chip surface.

[0149] Example 6

[0150] Avidity between cell monolayer and receptor coated beads

[0151] Avidity experiments can also be performed between a monolayer of cells having a target molecule present on their surface and beads coated with the receptor or other interacting molecule of interest. Compared to affinity assays (e.g. SPR or tetramer binding assays) this allows for measuring the avidity of receptor molecules in the biologically relevant context that includes the steric accessibility and mobility of the target molecules presented by the cell monolayer. A bead-cell avidity experiment can be performed with minor changes to cell-cell avidity protocols using fluorescently labelled beads of sufficient size and density to enable acoustic, centrifugal or fluidic (shear) flow forces to be applied (e.g. polystyrene particles of 3-20 pm diameter). For convenient attachment of the interacting molecule (receptor or antibody) beads with appropriate functionalization may be used (e.g. avidin, protein A, protein G, anti-Flag). Alternatively interacting molecules may be covalently attached to the beads e.g. by first coating beads with Poly-L-Lysine and subsequent crosslinking of the interacting molecules using BS(PEG)9. For ease of detection in the avidity experiments it is preferable to use either beads that have a fluorescent label in the bead itself, or incubate the beads with a low concentration of fluorescent protein that has a functional group that enables attachment to the bead but has no interaction with the target cells (e.g. biotinylated GFP, non-reactive fluorescently labelled antibody).

[0152] Example 7

[0153] Spin assisted settling

[0154] It was observed that when having effector cells in suspension, it can take considerable time for cells to land and come into contact with target cells, i.e. in a scenario with a well in which the effector cells are to settle at a surface with target cells (see e.g. Fig. 6). Cells were allowed to settle directly from the suspension (like depicted in Figure 6B) or were allowed to first settle at the bottom and then subsequently allowed to settle at the ceiling having the target cells attached (like depicted in Figure 6C). In the results depicted in Figure 6D, effector cells were primary T cells, which were either untransduced (UNT) or transduced with a lentiviral vector encoding a CAR (FMC), carried out at 26 or 37 degrees centigrade. Plotted in Figure 6D is the fraction of cells that have landed I arrived at the surface with the target cells. As can be seen, different conditions (i.a. different temperatures / types I sizes of cells) results in different arrival times and a spread in arrival times I time to have all the cells landed. As observed, time for substantially all the cells to have settled can range from about 200 - 300 seconds. A relatively large time difference between the first and the last cells settling decreases the resolution of and control over the settling time and thereby limits the time resolution of the k_on determination

[0155] Accordingly, methods were sought to reduce the effect of settling time in means and methods for determining cellular avidity. Advantageously, by reducing settling time, it was found that cellular avidity measurements could be much improved. Settling time and median settling time of the effector cells could be substantially reduced by introducing a step of spin assisted settling, i.e. exerting an active settling force, as depicted in Figure 7A and B. When modelling spin assisted settling, based on parameters as listed in Figure 7C, it was shown that settling time was much reduced, see Figure 7D. By reducing settling times two effects were achieved: firstly, the variation in contact time between individual cells in a single experiment is reduced (see standard deviation stdev in Fig. 7E, and table 3). The data such as depicted in Figure 7E and table 3 was generated, and statistical calculations made, for cellular avidity determinations (measured as % of cells bound after a defined contacting time and an applied detachment force, i.e. the force exterted away from the attached target cells) to determine the mean error in cellular avidity measurements. Secondly the absolute differences in (median) settling times between different experiments (e.g. caused by slight variations in temperature or cell size / density) are also reduced. The effect of the use of spin assisted settling, i.e. the use of an active settling force to reduce settling time was that accuracy of cellular avidity determined was highly improved. Most improvement was obtained with the use of a settling force and by correcting the defined contacting time with the median settling time. The use of an active settling force improves the resolution and accuracy of the defined contacting time, corrected by the median settling time.

[0156] Table 3. Listed are the data depicted in Figure 7E.

Claims

49CLAIMS1. A method for determining cellular avidity of effector cells carrying a receptor to target cells, the method comprising the steps of: a) providing effector cells carrying a receptor; b) providing target cells; wherein the target cells are attached to a surface; c) contacting the effector cells with the target cells to allow the effector cells to bind with the target cells, wherein in the contacting step, the target cells attached to the surface are in excess as compared with the effector cells, and wherein the interaction is for a defined time t ; d) applying a force, wherein the force is in a direction away from the attached target cells, such that at least part of the effector cells in contact and / or bound to the target cells attached to the surface move away therefrom; e) detecting effector cells that have remained bound with the target cells and attached to the surface after applying the force; f) determining the fraction of the effector cells (Fraction_bound[t]), that were contacted with the target cells that have remained bound; g) optionally, perform steps c), d), e) and f) one or more times; and h) calculating the cellular avidity metric k_on from- determined fraction(s) that remained bound (Fraction_bound[t]) and defined contacting time(s) t; with the following formula: k_on = - log(1-Fraction_bound[t]) 1 1 to therewith provide the cellular avidity metric k_on.

2. The method in accordance with claim 1 , wherein in step g), the steps c), d), e) and f) are performed one or more times at different defined contacting times, preferably selected from 0 minutes to 60 minutes, such as defined times of50 between 0 minutes and 10 minutes, between 10 minutes and 20 minutes, between 20 minutes and 40 minutes, between 40 minutes and 60 minutes.

3. The method in accordance with any of claims 2, wherein the k_on is calculated by using a least squares fitting method.

4. The method in accordance with claim 1 , wherein in step g), the steps c), d), e) and f) are performed one or more times at the same initially defined contacting time.

5. The method in accordance with any of claims 1-4, wherein in step h) k_on_initial is determined, and, subsequently, step i) is performed, wherein in step i) subsequently steps c), d), e) and f) are performed one or more times; at a defined time in step c) of 1 / k_on_initial, and subsequently k_on is calculated based on the determined fraction(s) that remained bound (Fraction_bound(t)) and defined contacting time(s) t obtained in step i) using the formula provided in step h).

6. The method in accordance with any one of claims 1-5, wherein effector cells with different receptors are provided and for each effector cell with a different receptor the method is performed.

7. The method in accordance with any of claims 1-6, wherein effector cells with different receptors are provided and for each effector cell with a different receptor the method is performed and in step h) a k_on_initial is calculated, and, subsequently, the highest and lowest k_on_initial determined (k_on_init_h and k_on_init_l, respectively), and subsequently, step i) is performed, wherein in step i) steps c), d), e) and f) are subsequently performed one or more times; at an optimal defined time in step c) of t = log (k_on_init_h / k_on_init_l) I ((k_on_init_h - k_on_init_l)) and subsequently k_on is calculated based on the determined fraction(s) that remained bound (Fraction_bound(t)) and defined contacting time(s) (t) obtained51 in step i) using the formula provided in step h), or, wherein, highest and lowest k_on_initial are determined for subgroups of different effector cells, and for each subgroup step i) is performed with their corresponding optimal defined time in step c) and subsequently k_on is calculated based on the determined fraction(s) that remained bound.

8. The method in accordance with any of claims 1-7, wherein the method is performed: at a defined temperature and in a defined cell culture medium;- with one type of target cell or with different target cells; and / or- with the target cells attached to the surface in at least a 10-fold excess as compared with the effector cells carrying a receptor.

9. The method in accordance with any of claims 1-8, wherein the target cells present a cell surface molecule, and the contacting step c) is performed in the presence of a cell engager, wherein the cell engager has one binding arm capable of binding with the cell surface molecule of the target cell and another binding arm capable of binding said receptor of the effector cells.

10. The method in accordance with any of claims 1-8, wherein instead of target cells attached to a surface, a functionalized surface is provided comprising a cell surface molecule with which the receptor of the effector cells is capable to bind, and the method is performed with the functionalized surface instead of the target cells attached to the surface.

11. The method in accordance with any of claims 1-8, wherein instead of effector cells carrying a receptor, beads are provided, which beads are functionalized with receptors at their surface, and the method is performed with the beads instead of the effector cells.

12. The method in accordance with any of claims 1-11 , wherein the Fraction_bound value is corrected for effector cell viability, by division with the viable fraction value.5213. The method in accordance with claim 12, wherein the effector cells are thawed prior to performing a cellular avidity measurement, and cell viability is determined after allowing the cells to recover from thawing, preferably for about 20-28 hours after thawing, and, wherein cellular avidity is determined immediately after thawing the effector cells, preferably within about an hour after thawing.

14. The method in accordance with any of claims 1-13, wherein a fraction alpha of the effector cells is active, and wherein the k_on rate is corrected therefor by using instead of the formula provided in step h), the following formula: k_on = - log(1 - (Fraction_bound[t]) I alpha) 1 1 optionally, wherein alpha is provided by determining the Fraction_bound at a contacting time t of 2.5 x 1 / k_on or more.

15. The method in accordance with any of claims 1-14, wherein the method is implemented as a high throughput method.

Images