RU2831767C1 - Methods and system for validating measurement results obtained by flow cytometry - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: invention relates to methods of performing validation for assessing reliability and reproducibility of measurement results. Method for validating measurements using an analytical instrument based on measuring fluorescence with respect to a population of coloured target cells in a test sample involves: (a) negative staining in a first portion of a reference sample containing target cells expressing a target cell marker, where the negative staining provides for the introduction into the first part of the reference sample of a fluorescent dye to exclude dead cells, a non-specific antibody conjugated with a fluorochrome, and a specific antibody capable of specifically binding a target marker on a target cell; (b) positive staining in the second part of the reference sample, where the positive staining in the second part of the reference sample in (b) includes the introduction into the second part of the reference sample of a fluorescent dye eliminating dead cells, and specific antibody conjugated with fluorochrome; (c) analysing the first portion and the second portion of the reference sample using the instrument to obtain a fluorescence measurement result indicative of the concentration of target cells; (d) based on the results obtained in (c), diluting the positively stained target cells in the second part of the reference sample with negatively stained target cells in a first part of a reference sample to obtain a diluted sample having the highest concentration of target cells, from which a series of dilutions is obtained, where each diluted sample is characterized by a nominal concentration of target cells, each nominal concentration of cells exceeds the concentration of target cells, which is indicated by the fluorescence measurement result of the negatively coloured first part in (a), where the nominal concentration in each diluted sample differs from the nominal concentration in each of the remaining diluted samples, and where the highest concentration of target cells in the dilution sample is 2%, 1%, 0.5%, 0.2%, 0.1%, 0.05% or 0.02%; (e) analysing a series of diluted samples from (d) using an instrument to obtain a series of fluorescence measurement results; (f) comparing the nominal cell concentration from (d) and measuring the fluorescence from (e) for each diluted sample to quantify the effectiveness of the staining method using the instrument, where the comparison in (f) involves performing a statistical calculation of the difference between the nominal cell concentration and the fluorescence measurement for each diluted sample with the determination of at least one of the linearity, the range, accuracy, precision, limit of detection (LOD) and lower limit of quantification (LLOQ) for the instrument and method of staining, and where determination of LLOQ involves identification of concentration of target cells, associated with a predetermined precision criterion and a predetermined accuracy criterion.

EFFECT: improved methods of validating analytical methods and systems.

17 cl, 9 dwg, 14 tbl

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[1] This application claims the benefit of priority to U.S. Provisional Patent Application No. 63/013,098, filed April 21, 2020, the contents of which are incorporated herein by reference in its entirety.

FIELDS OF TECHNOLOGY TO WHICH THE INVENTION RELATES

[2] The present invention relates to analytical methods and systems based on fluorescence measurement, and in particular to methods for performing validation to assess the reliability and reproducibility of measurement results obtained using an analytical instrument based on fluorescence measurement.

BACKGROUND OF THE INVENTION

[3] Fluorescence-based cell analysis systems such as flow cytometry and fluorescence-activated cell sorting (FACS) are powerful tools for measuring the abundance of various cellular elements in a test sample, such as cell surface receptors and intracellular components. FACS methods use antibody-conjugated fluorochromes (fluorescent dyes) to stain samples, thereby providing a signal for the presence and abundance of elements such as cellular receptors or other cellular markers, peptides, and nucleic acids. The development of several different fluorochromes, each with a unique, identifiable emission spectrum, allows the use of more than one fluorochrome simultaneously. Because of their combined power and versatility, flow cytometry and FACS are commonly used in drug discovery, such as for immunophenotyping, receptor expression or occupancy analysis, and other functional assays.

[4] Despite their power, flow cytometry and FACS methods are generally more challenging to validate than other analytical methods for a variety of reasons. These include the lack of standardization of protocols, the absence of standardized reference materials, and the use of complex and sensitive instruments. The monoclonal and polyclonal antibodies commonly used for staining are not standardized and can vary widely in quality and performance. Thus, the investigator must carefully evaluate whether the reagents used are performing as intended. However, there is little guidance for the investigator to conduct method and instrument validation. The US FDA, for example, defines validation simply as an assessment of the suitability of the method for the intended use, and recommends that the validation testing process should ensure that the assay meets predetermined standards, performs reliably, and is suitable for its intended use. FDA guidelines acknowledge that there is no universal regulation of validation and do not provide specific guidance for validation assays. The lack of standard or accepted normative guidelines and references for validation is a problem that is unfortunately not well described in the literature. The problem of method validation is even more challenging when the target cell or marker of interest is considered to be present in the sample at very low concentrations. There is a need for improved validation methods to validate the performance of fluorescence-based assays and systems.

ESSENCE OF THE INVENTION

[5] Among various aspects of the present invention, there is provided a method for validating measurements by an assay instrument based on measuring fluorescence with respect to a population of stained target cells in a test sample, the method comprising: (a) negative staining or having negatively stained cells in a first portion of a reference sample containing target cells expressing a target cell marker; (b) positive staining or having positively stained target cells in a second portion of the reference sample; (c) analyzing the first portion of the reference sample with the instrument to obtain a fluorescence measurement indicative of a concentration of target cells in the first portion of the reference sample, and analyzing a second portion of the reference sample with the instrument to obtain a fluorescence measurement indicative of a concentration of target cells in the second portion of the reference sample; (d) based on the fluorescence measurement results obtained in (c), obtaining a dilution series comprising a plurality of diluted samples, wherein each diluted sample has a nominal concentration of target cells, each nominal concentration of cells exceeding the concentration of target cells indicated by the fluorescence measurement result of the negatively stained first part in (a), wherein the nominal concentration in each diluted sample is different from the nominal concentration in each of the other diluted samples; (e) analyzing the series of diluted samples from (d) with an instrument to obtain a series of fluorescence measurements comprising a fluorescence measurement of each diluted sample, and (f) comparing the nominal cell concentration from (d) for each diluted sample and the fluorescence measurement result from (e) to quantify the effectiveness of the staining method with the instrument. The instrument may be, for example, a flow cytometer.

[6] In another aspect, the present invention provides a method for validating measurements by a flow cytometer of a population of stained target cells in a test sample, the method comprising: (a) negative staining or having negatively stained cells in a first portion of a reference sample containing target cells expressing a target cell marker; (b) positive staining or having positively stained target cells in a second portion of the reference sample; (c) analyzing both the first portion and the second portion of the reference sample by a flow cytometer to obtain fluorescence measurements indicative of a concentration of target cells in each of the first portion of the reference sample and the second portion of the reference sample; (d) based on the fluorescence measurements obtained in (c), obtaining a series of diluted samples, each having a nominal target cell concentration that varies systematically in the dilution series, wherein the nominal cell concentration in at least one diluted sample exceeds the target cell concentration indicated by the fluorescence measurement of the negatively stained first part in (a); (e) analyzing the series of diluted samples from (d) using a flow cytometer to obtain a series of fluorescence measurements that measure the fluorescence of each diluted sample, and (f) comparing the nominal cell concentration from (d) for each diluted sample and the fluorescence measurement from (e) to quantify the performance of the staining method using the flow cytometer.

[7] Any of the disclosed methods may include one or more of any of the following features. The comparison in (f) may include, for example, performing a statistical calculation of the difference between the nominal cell concentration and the fluorescence measurement for each diluted sample to determine at least one of linearity, range, accuracy, precision, limit of detection (LOD), and lower limit of quantification (LLOQ) for the instrument and staining method. Determining the LLOQ may include, for example, identifying a target cell concentration associated with a predetermined precision criterion and a predetermined accuracy criterion. Negative staining of target cells in the first portion of the reference sample in (a) may involve introducing into the first portion of the reference sample (i) a fluorescent dye to exclude dead cells, (ii) a non-specific antibody conjugated to a fluorochrome, such as an isotype control antibody, and (iii) a specific antibody capable of specifically binding a target marker (antigen) on the target cell, unconjugated to a fluorochrome.

The non-specific antibody which is conjugated to a fluorochrome for negative staining may comprise an antibody which does not have the ability to specifically bind to a cellular antigen, such as, for example, IgD, IgG, IgA, IgM or IgE. The negatively stained first part may be obtained in the absence of a cell depletion step. Positive staining of the target cells in the second part of the reference sample in (b) may comprise introducing into the second part of the reference sample (i) a fluorescent dye to exclude dead cells (the same dye used for negative staining) and (ii) the same specific antibody which was used for negative staining, but not unconjugated to a fluorochrome, but rather conjugated to the same fluorochrome which was used for negative staining (in which the fluorochrome is conjugated to a non-specific antibody). In any of the methods, the fluorescent dye for excluding dead cells used for staining may be selected from a nucleic acid binding dye, propidium iodide, DAPI, DRAQ7, 7-AAD, T0-PR0-3, and an amino group-reactive dye. The fluorochrome used can be selected from known fluorochromes, such as, but not limited to, allophycocyanin (APC), APC C750, APC AF700, brilliant violet (BV)421, BV510, hylitol 7 (H7) BV605, BV650, PE CF594, fluorescein isothiocyanate (FITC), R-phycoerythrin (PE or R-PE), PE-Cy7 (PE linked to the cyanine dye Cy7), APC-Cy7 (APC linked to the cyanine dye Cy7), APC-H7 (APC linked to hylitol 7, which is an analogue of Cy (H7)). The target cell marker can be selected, for example, from CD3, CD4 and CD8 as a T cell marker, CD19 as a B cell marker, CD235a as a red blood cell marker, CD56 as a natural killer (NK) cell marker, CD14 as a monocyte marker and CD66b as a granulocyte marker. In one aspect, the target cell population is a T cell population, and the target cell marker is CD3.

[8] The method may further comprise multiplexing by negative staining and positive staining of two or more different target markers of target cells or potentially staining two or more different target markers of two or more target cells. Negative staining may further comprise introducing into a first portion of the reference sample two or more specific antibodies, wherein each specific antibody is capable of specifically binding one of the two or more target markers and is not conjugated to a fluorochrome. Positive staining may further comprise introducing into a second portion of the reference sample two or more specific antibodies, each of which is conjugated to a fluorochrome.

[9] The calculation in (f) may be performed using a processor connected to the instrument or flow cytometer. The test sample may contain a target cell present at a concentration of no more than 2%, 1%, 0.5%, 0.2%, 0.1%, 0.05%, or 0.02%. The dilution series may provide a diluted sample having the highest concentration of target cells, where the highest concentration is 2%, 1%, 0.5%, 0.2%, 0.1%, 0.05%, or 0.02%. In either method, the diluted sample series may contain at least two, three, four, five, six, seven, eight, nine, or ten diluted samples.

[10] In another aspect, the present invention provides a non-transitory computer-readable medium comprising instructions for a computer processor to perform the comparison in step (f) of any of the disclosed methods or to perform any statistical calculations to determine any one or more of the linearity, range, accuracy, precision, limit of detection (LOD), and lowest limit of quantification (LLOQ) for the staining device and method.

[11] In another aspect, the present invention provides a system for validating fluorescence measurements in a test sample obtained using a fluorescence-based instrument, comprising a fluorescence-based instrument and a computer connected to the fluorescence-based instrument and comprising a computer-readable medium containing instructions for the computer processor described above. In the system, the fluorescence-based instrument may be a flow cytometer.

[12] In another aspect, the present invention provides a kit comprising reagents for staining cells in a reference sample for validating an assay based on fluorescence measurement for analyzing target cells in a test sample, the kit comprising: (i) negative staining reagents comprising: (a) a fluorescent dye for excluding dead cells, (b) a non-specific antibody conjugated to a fluorochrome, and (c) an unconjugated specific antibody capable of specifically binding a target marker (antigen) on the target cells; (11) positive staining reagents comprising: (a) a fluorescent dye for excluding dead cells and (b) a specific antibody conjugated to a fluorochrome, and instructions for (a) negatively staining a first portion of a reference sample, (b) positively staining a second portion of the reference sample, and (c) producing a dilution series comprising a plurality of diluted samples, each having a nominal target cell concentration that varies systematically in the dilution series, wherein the nominal cell concentration in at least one diluted sample is greater than the target cell concentration indicated by a fluorescence measurement of the negatively stained first portion.

[13] Other aspects and features of the present invention are described in detail below.

LINK TO COLOR GRAPHIC MATERIALS

[14] The application file shall contain at least one photograph taken in color. Copies of this patent application publication with color photographs will be provided by the Patent Office upon request and payment of the required fee.

BRIEF DESCRIPTION OF GRAPHIC MATERIALS

[15] Fig. 1 is a schematic representation of the method according to the present invention, which proposes an alternative method of using a cell sorter to solve the technical problem of obtaining a cell preparation characterized by a cell concentration close to 0.00%.

[16] Figure 2 is a FACS plot showing typical results obtained with a negatively stained sample. Events shown were gated on viable cells based on forward and side scatter and PI (to exclude dead cells).

[17] Figure 3 is a FACS plot showing typical results obtained with a positively stained sample. Events shown were gated on viable cells based on forward and side scatter and PI (to exclude dead cells).

[18] Fig. 4 shows the sequence of procedures for obtaining samples for linearity assessment.

[19] Fig. 5 shows the sequence of procedures for obtaining a dilution series.

[20] Fig. 6 shows the performance of the method qualification by two operators in three different cases indicated in Example 2.

[21] Fig. 7 is a plot of the residuals for untransformed nominal %CD3 and calculated %CD3 (result) for a) untransformed data and b)-f) various log-transformed data.

[22] Figure 8 shows the percentage (%) of calculated %CD3 compared to nominal %CD3. The dotted lines indicate 100+30% recovery.

[23] Fig. 9 shows the regression analysis of samples with estimated linearity.

DETAILED DESCRIPTION

[24] Validation of an assay is critical to assessing the reliability and reproducibility of fluorescence-based assays. To be valuable, such assays must be reproducible. For example, fluorescence-activated cell sorting (FACS) is a specialized type of fluorescence-based assay and is a powerful way to sort a mixture of biological cells into distinct populations based on the fluorescent signals of each cell. Unlike flow cytometry, which is used to simply count and sort cells, FACS enables both qualitative and quantitative analysis using the data obtained by flow cytometry. It is important that assays based on fluorescence, such as those performed using FACS, are properly validated due to variations in the quality and performance of the antibodies used in such assays.

[25] One approach to validation is to generate a dilution series from multiple spiked dilutions or samples, each with a known or nominal concentration of the target cell or target marker. For example, serial dilutions are performed by adding known amounts of the target cell to generate multiple samples with known (nominal) cell concentrations that vary systematically from a higher concentration, such as 50%, to a lower concentration, such as 1% of the cells. The spiked samples are analyzed using a cytometer and fluorescence measurements are obtained for each spiked sample. The spiked samples are compared with the measured values obtained using the cytometer. The accuracy of the instrument and the staining method are determined by statistical calculations, such as calculating the correlation of nominal and measured values, determining the slope of the dependence, determining R-sq, etc. As is known, the FACS limit of detection (LOD) and/or limit of quantification (LOQ) can be calculated by staining and obtaining a replicate sample(s) spiked with the analyte and analyzing it with a cytometer to obtain measurements of positive events. Validated assays can then be used to detect and quantify the concentration of cells in the sample.

[26] To accurately generate a series of spiked samples, at least two samples are required: one or more samples each containing a known (nominal) concentration of the target cell or target marker, and a sample that is substantially free of the target cell or target marker (approximately 0 concentration). More than one spiked sample is typically used. For example, validation of an instrument and staining method using five (5) levels would involve generating a series of spiked samples containing five different nominal concentrations of the target cell or target marker (e.g., 2%, 4%, 6%, 8%, and 10%), plus one sample containing 0%. To generate a series of spiked samples, a reference material containing a known, relatively higher concentration of the target cell or target marker of interest is sequentially and systematically diluted to generate a series of spiked samples having lower concentrations. However, in many cases, generating spiked samples is difficult due to the unavailability of a readily available starting reference material. For example, if the analyte of interest is a cell, such as a CD3 ⁺ cell (T cell), no such standard reference material exists. In other words, one cannot simply purchase a standard reference material having a known concentration, such as 10% or 0% CD3 ⁺ cells, that can be diluted to generate spiked samples having lower concentrations of 8, 6, 4, and 2%. In theory, commercially available cell sorters could be used to generate a series of spiked samples with successively lower concentrations. However, at low target cell or target marker concentrations, such as approximately 0.1% or lower, down to approximately 0.01%, this approach is expensive and time consuming.

[27] The present invention solves the technical problem of providing negative reference samples and samples with a spiked analyte for a situation where there is no standard reference material and providing samples with a spiked analyte at a concentration below 0.1%, which is useful for validation if the target cell or target marker is present at a concentration below 0.1%. For example, depending on the instrument and staining used, the LLOQ can be below 0.01%. Accordingly, the present invention relates to methods and systems for validating measurement results obtained using an analytical instrument based on fluorescence measurement, such as a flow cytometer, of a stained population of target cells in a test sample.

Analytical instruments based on fluorescence measurement

[28] A fluorescence-based analytical instrument refers to any analytical instrument capable of measuring a fluorescent signal as an indicator of the presence of a target analyte or target cell type in a sample. Depending on the equipment and methods used, the instrument may be capable of quantifying the target analyte or target cell type in a test sample based on the intensity of the fluorescent signal. One example of a fluorescence-based analytical instrument is a flow cytometer. FACS (fluorescence-activated cell sorting) is a different but related technique based on flow cytometry. Flow cytometry is a technique that uses a flow cytometer to examine and determine the expression of molecules both inside and outside a cell and to identify and characterize individual cell types. Flow cytometry can also be used to determine other parameters such as cell size, volume, and purity of a sample of isolated cells. FACS is a flow cytometry technique that uses specific antibodies labeled with fluorochromes to obtain expression data and sort cell samples according to a number of variables.

[29] FACS can be performed using any of a number of types of flow cytometers, including traditional flow cytometers, acoustic focusing cytometers, cell sorters, or imaging flow cytometers. The most common traditional cytometers use a sheath fluid to focus the sample flow and common lasers such as 488 nm (blue), 405 nm (violet), 532 nm (green), 552 nm (green), 561 nm (green-yellow), 640 nm (red), and 355 nm (ultraviolet) lasers. Acoustic focusing cytometers use ultrasound waves to focus cells for analysis, which prevents sample clogging and allows more sample to be injected. Cell sorters are a type of traditional flow cytometer that allow the user to collect samples after processing. Cells that are positive for a desired parameter can be separated from cells that are negative for those parameters. Imaging cytometers are traditional cytometers combined with fluorescence microscopy, all designed to rapidly analyze a sample for morphology and multiparameter fluorescence at both the single-cell and population levels. Flow cytometers are typically connected to a computer containing a processor that controls the operation of the instrument. The methods and systems described herein can be used with any flow cytometer described herein.

Methods

[30] The methods described in this document solve the technical problem of providing a negative standard and samples with low levels of added analyte when no standard reference material exists, thereby allowing reliable validation of fluorescence-based measurements, such as those made by flow cytometry (FACS), in situations where the target cell or target marker is present in the sample at concentrations below 0.1% or even less.

[31] Fig. 1 is a schematic illustration of the method. Positively and negatively stained reference samples are prepared using an antibody capable of specifically binding a target cell marker (specific antibody) and an "isotypic antibody" (non-specific antibody), and a fluorochrome conjugated to the specific antibody or to the non-specific antibody, depending on whether negative or positive staining is performed. More specifically, a reference sample is prepared and divided into two parts. The first part of the reference sample is negatively stained with (a) a fluorescent dye to exclude dead cells, (ii) a non-specific antibody conjugated to a fluorochrome, such as an isotype control antibody, and (iii) a specific antibody capable of specifically binding a target marker (antigen) on the target cell. For negative staining, the specific antibody is not conjugated to a fluorochrome. A non-specific antibody conjugated to a fluorochrome for negative staining is an antibody capable of binding to a cell that does not contain an epitope specific for the antibody, i.e., it binds to the cell but not by specific binding, as when an antibody binds to a specific epitope on the cell, and thus the non-specific antibody also has a high dissociation constant (Kd) for the marker, as described in more detail below. Any of the non-specific immunoglobulins IgD, IgG, IgA, IgM or IgE can be used as a non-specific antibody.

[32] The second part of the reference sample is positively stained with a fluorescent dye to exclude dead cells (the same dye used in the negative staining step) and the same specific antibody used for negative staining, but not unconjugated, but rather conjugated to the same fluorochrome used for negative staining (in which the fluorochrome is conjugated to a non-specific antibody).

[33] The target cell marker may be any cellular marker, such as any known T cell, B cell, NK cell, red blood cell, granulocyte, or monocyte marker. T cell markers, for example, include but are not limited to CD3, CD4, CD8, CD69, CD71, and CD25. B cell markers include but are not limited to IL-b, CD19, CD25, CD30, CD27, CD38, CD78, CD138, and CD319. In one aspect, for example, the target marker may be selected from CD3 as a T cell marker, CD19 as a B cell marker, CD235a as a red blood cell marker, CD56 as a natural killer (NK) cell marker, CD14 as a monocyte marker, and CD66b as a granulocyte marker. It should be understood that validation for more than one cell population marker can be easily accomplished by applying the method to multiple samples from a single reference material, using in each case a suitable specific antibody. As a non-limiting example, for validation of two different staining protocols for two different T cell population markers, the method can be performed a first time using an antibody that specifically binds CD3, and a second time performed using an antibody that specifically binds CD4.

[34] It should be understood that the choice of target marker determines the choice of specific staining. Specific binding of an antibody to a target marker refers to the ability of the antibody to recognize the epitope(s) of the marker such that the selected antibody exhibits high binding affinity for the marker and preferably low binding affinity for other cross-reacting species. Typically, the binding affinity of a molecule is described using the dissociation (or equilibrium) constant (Kd). The lower the dissociation constant, the higher the binding affinity or extent of binding of antibody A to analyte (marker) B. Thus, an antibody that is specific for a marker exhibits a low dissociation constant (Kd) for the marker. Methods for determining the Kd of a given antibody are very well described in the literature, as further noted below. Antibodies that generate a poor signal at low concentrations will have high variability (noise) and will therefore be difficult to distinguish from antibodies with higher measured cross-reactivity. As used herein, the term "specific antibody" refers to an antibody that has a Kd for a target marker of no more than 10 nM or lower, no more than 7 nM or lower, no more than 5 nM or lower, no more than 1 nM or lower, no more than 0.5 nM or lower, no more than 0.15 nM or lower, or no more than 0.1 nM or lower. As a non-limiting example, a specific antibody can be selected that has a specific binding to a target marker with a Kd of 5 nM or lower, such as 2 nM or lower, preferably 1 nM or lower, and more preferably 0.7 nM or lower. The Kd value, which describes the binding affinity of an antibody considered specific for a target marker, is provided by commercial suppliers of commercially available antibodies or can be determined by well-known methods, including but not limited to fluorescence titration, competitive ELISA, calorimetric methods such as isothermal titration calorimetry (ITC), flow cytometric titration analysis (FACS titration), and surface plasmon resonance (BIAcore). Such methods are well described in the literature. (See, for example, Goodrich & Kugel, 2007, Binding and Kinetics for Molecular Biologists, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.; De Jong, L. A. A. et al., J. Chromatogr. B 829(1-2):1-25 (2005); Heinrich, L. et al. J. Immunol. Methods 352 (1-2):13-22 (2010); Williams, M. A. & Daviter, T. (Eds.) 2013, Protein-Ligand Interactions, Methods and Applications, Springer, New York, N.Y.).

[35] Accordingly, a specific antibody in the present invention is any antibody having a Kd of no more than 1 nM, no more than 0.5 nM, no more than 0.15 nM, or no more than 0.1 nM with respect to any of the target cell markers, including but not limited to any of the cell markers described herein: CD3, CD4, CD8, CD69, CD71, and CD25, IL-6, CD19, CD25, CD30, CD27, CD38, CD78, CD138, and CD319. CD235a, CD56, CD14, and CD66b. Specific antibodies may be polyclonal, monoclonal, chimeric, humanized, or fully human antibodies. They may be single-chain or multichain antibodies. Although suitable specific antibodies can be and are typically purchased from commercial suppliers, monoclonal antibodies can also be produced from hybridomas or in host cells containing vectors containing nucleic acid sequences encoding antibodies, which are then expressed by the cells. Thus, the specific antibodies described herein can be produced by transforming a host cell with at least one nucleic acid molecule encoding a specific antibody; expressing the nucleic acid molecule in the host cell; and isolating the specific antibody, all of which steps can be accomplished using materials and techniques well known in the art. Commercial suppliers of specific antibodies to a wide range of cell markers include, but are not limited to, BD Biosciences, Takara Bio USA (e.g., offers monoclonal antibody to CD3 (OKT3)); BioX Cell (Lebanon, NH; e.g., offers monoclonal antibody to human CD19); Thermo-Fisher Scientific Inc.; BioLegend Inc.; Bio-Rad Antibodies.

[36] The fluorescent dye for excluding dead cells used in both the negative staining steps and the positive staining steps may be any dye that can distinguish viable cells from dead or dying cells, and these dyes are also known as viability dyes. For example, viability dyes include dyes that are not membrane-permeable, that is, they do not pass through cell membranes. For example, nucleic acid-binding dyes that do not permeate cell membranes will selectively stain accessible nucleic acid (e.g., DNA) of dead and dying cells and will not stain inaccessible nucleic acid within viable cells with intact membranes. Alternatively, the dead cell-exclusion dye or viability dye may be a protein-binding dye (a dye that reacts with amine groups) rather than a nucleic acid-binding dye. Protein binding dyes bind to both live and dead cells, but dead and dying cells with disrupted membranes are more intensely stained by a dye that has access to more intracellular protein, and thus dead and dying cells exhibit higher fluorescence intensity. Thus, using either a nucleic acid binding dye or an amine-reactive dye, dead cells can be excluded by gating on the less intensely stained population that contains live cells. Dyes for excluding dead cells include numerous commercially available nucleic acid binding dyes, amine-reactive dyes such as, but not limited to, propidium iodide, DAPI, DRAQ7, 7-AAD, T0-PR0-3, live/dead cell fix dyes, eFluor fix dyes, Horizon dyes, Biolegend Zombie dyes, and Ghost dyes. The dye that reacts with amino groups includes M1420MP, M1410, D6105, P130, P6114, A10168, M10165, D10161, D374, D126, D1421, B30250, H185, H1428, H1193, P30253, A30000, A30100, C10164, P10163, S6110, D2184, D2183, D3834 and others.

[37] The fluorochrome may be any fluorochrome that has a characteristic emission spectrum in the visible region, such as, but not limited to, any of the many known fluorochromes currently commercially available, such as allophycocyanin (APC), APC C750, APC AF700, brilliant violet (BV)421, BV510, hylitol 7 (H7) BV605, BV650, PE CF594, fluorescein isothiocyanate (FITC), R-phycoerythrin (PE or R-PE), PE-Cy7 (PE linked to the cyanine dye Cy7), APC-Cy7 (APC linked to the cyanine dye Cy7), APC-H7 (APC linked to hylitol 7, which is an analogue of Cy (H7)).

[38] The target marker binding sites in a negatively stained sample (a sample stained with an isotype antibody) are blocked using an unconjugated version of the specific antibody, i.e., a specific antibody to which the fluorochrome is not conjugated or covalently linked. The positively stained sample is then used to generate at least two, and typically more than two, spiked samples by adding the analyte to nominal concentrations of target cells using negatively stained cells. The series of spiked samples are then analyzed in a flow cytometer, with the measured concentration of each spiked sample determined by the instrument. The data (nominal and measured concentration of each spiked sample) are plotted and statistically analyzed for correlation and to calculate validation parameters such as linearity, accuracy, precision, LOD and/or LLOQ. It should be understood that in any of the methods, the analysis of the comparison of the data on the nominal and measured cell concentration for the quantitative evaluation of the efficiency of the staining method in the device can be carried out using a computer processor connected to the device, such as a processor configured according to instructions to perform one or more standard statistical tests on the differences between each nominal cell concentration and the fluorescence measurement result for each diluted sample, to determine at least one of the linearity, range, accuracy, precision, LOD and LLOQ for the device and the staining method. Determining the LLOQ can include, for example, identifying the concentration of target cells associated with a predetermined precision criterion and a predetermined accuracy criterion.

[39] Example 1 below illustrates methods used to determine the concentration of viable T cells in a cell product sample using propidium iodide as a dead cell exclusion dye, an anti-CD3 antibody, IgG as a non-specific antibody, and allophycocyanin as a fluorochrome. However, it should be understood that any of the dead cell exclusion dye, the cell marker (CD3) and specific antibody (anti-CD3 antibody) pair, the non-specific antibody (IgG), and/or the fluorochrome (APC) in the example may be substituted as described herein. Following negative and positive staining, the positively stained sample is diluted to the planned highest desired level of added analyte (the highest nominal concentration of target cells) using negatively stained cells, and serial dilutions are then made to produce diluted samples with systematically varying nominal cell concentrations. The dilution series may involve any multiple. In theory, even two diluted samples can be used, but in practice at least three diluted samples are prepared, each with a varying nominal concentration, preferably to systematically facilitate the analysis. A dilution series may include three, four, five, six, seven, eight, nine, ten, eleven, twelve or more diluted (spiked) samples. As a non-limiting example, the nominal concentration of each diluted (spiked) sample may vary from 0.1% to 0.003%, but it should be understood that the choice of nominal values will vary depending on a number of factors, such as the expected concentration of the target cell or target marker in the sample, the number of spiked samples used, the instrumentation, the staining protocol used, etc.

[40] Furthermore, it should be understood that the methods and systems of the present invention solve the problem of providing a negative control sample for a test sample that may contain a target cell present at a concentration of no more than 2%, 1.5%, 1%, 0.8%, 0.7%, 0.6%, 0.5%, 0.4%, 0.3%, 0.25%, 0.2%, 0.15%, 0.1%, 0.05%, 0.02%, or no more than 0.01%. For comparison, a dilution series may provide a diluted sample having the highest concentration of target cells, wherein the highest concentration is 2%, 1.5%, 1%, 0.8%, 0.7%, 0.6%, 0.5%, 0.4%, 0.3%, 0.25%, 0.2%, 0.15%, 0.1%, 0.05%, 0.02%, or 0.01%. For example, the methods provided herein can be used to determine whether a target cell or target marker is considered a residue or impurity in a sample. For example, in natural killer cells, CD3+ cells or T cells may be considered a residue or impurity. The method of the present invention may allow the determination of a residue or impurity at a very low level, such as less than 2%, 1.5%, 1%, 0.8%, 0.7%, 0.6%, 0.5%, 0.4%, 0.3%, 0.25%, 0.2%, 0.15%, 0.1%, 0.05%, 0.02% or 0.01%.

Systems

[41] The present invention also provides systems and devices comprising any one or more fluorescence-based devices and corresponding system components for implementing the disclosed methods and various aspects and features thereof. The systems may include, for example, one or more fluorescence-based devices, one or more computer processors coupled to the device(s), and optionally provide one or more samples of any cells, dyes, fluorochromes, cell markers, antibodies, or any combination thereof. For example, the present invention provides a non-transitory computer-readable medium containing instructions executable by a processor for performing any data analysis or statistical calculations by the processor using data generated by the device, as described herein. A system for validating fluorescence measurements obtained using a fluorescence-based instrument may include a fluorescence-based instrument, a computer connected thereto, and a machine-readable medium, the contents of which can be read by the computer, containing instructions executable by a processor. As described in detail above in this document, the instrument may be a flow cytometer that can be used to perform FACS.

Sets

[42] The present invention also provides kits and devices comprising any one or more of cell samples, dyes, fluorochromes, cell markers, specific and/or non-specific antibodies, buffers, diluents, or any combination thereof for performing any of the validation methods described herein. A kit may comprise, for example, any one or more reagents in amounts necessary for negative and/or positive staining of cells in a reference sample for validating an assay based on fluorescence measurement for analyzing target cells in a test sample. The kit may comprise components necessary for negative staining, including any one or more of a fluorescent dye for excluding dead cells, a non-specific antibody conjugated to a fluorochrome, and an unconjugated specific antibody described herein that is capable of specifically binding a target marker (antigen) on target cells. The kit may contain components necessary for positive staining, including any one or more of the fluorescent dyes for excluding dead cells and a specific antibody conjugated to a fluorochrome described herein.

[43] It should be understood that the specific antibody for both negative staining and positive staining will be the same, due to the target cell marker selected. Likewise, the fluorochrome in the kit for both negative staining and positive staining is the same. The kit may contain written instructions in printed or electronic format for negative and positive staining of the reference sample and for preparing the dilution series described herein above. The instructions may further include any one or more instructions for using the kit to perform any of the validation methods described herein; to prepare the reference sample; and/or to read, analyze, and interpret the results of the validation method. The kit may conveniently contain one or more containers for receiving and/or storing any of the reagents and components for labeling them.

DEFINITIONS

[44] Since various changes can be made in the above-described apparatus, kits and methods without departing from the scope of the present invention, it is intended that all matter contained in the above description and in the examples given below be interpreted as illustrative and not in a limiting sense.

[45] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The following references

provide the skilled person with a general definition of many of the terms used in the present invention: The Cambridge Dictionary of Science and Technology (Walker ed., 1988); The Glossary of Genetics, 5th Ed., R. Rieger et al. (eds.), Springer Verlag (1991), and Hale & Marham, The Harper Collins Dictionary of Biology (1991). The following terms used herein have the meanings assigned to them unless otherwise noted.

[46] When referring to elements of the present invention or preferred aspects thereof, the singular forms, plural forms and "said" are intended to mean the presence of one or more elements. The terms "comprising," "providing," and "having" are intended to be inclusive and mean that additional elements other than the listed elements may be present.

[47] The term "comprising" means "including, but not necessarily limited to"; it specifically indicates an open inclusion in, or membership in, the combination, group, series, etc. so described. As used herein, the terms "comprising" and "comprising" are inclusive and/or open-ended and do not exclude additional, unrecited elements or process processes. The term "consisting substantially of" is more restrictive than "comprising" but not as restrictive as "consisting of." In particular, the term "consisting substantially of" limits the recited materials or steps to those materials or steps that do not materially affect the important characteristics of the claimed invention.

EXAMPLES

[48] Any publications discussed herein are provided solely for their disclosure prior to the filing date of this application. Nothing herein should be construed as an admission that the present invention is not precluded from being opposed to such disclosure by virtue of prior invention.

[49] The following examples are included to demonstrate the present invention. It will be understood by those skilled in the art that the techniques disclosed in the following examples represent techniques developed by the inventors of the present invention for properly practicing the present invention. However, it will be understood by those skilled in the art, in light of the present invention, that many changes can be made in the present invention and still obtain a like or similar result without departing from the spirit and scope of the present invention, and therefore all material set forth herein should be interpreted as illustrative and not in a limiting sense.

Example 1: Obtaining spiked samples for flow cytometric assay validation

[50] In this example, samples spiked with an analyte were obtained to validate a flow cytometric assay of a “rare event target” to determine the LLOQ (lowest level of quantification) of a residue or impurity present in the sample.

Way

[51] Fig. 1 is a schematic illustration of the method. Positive and negatively stained cell samples were prepared using a "fluorochrome-conjugated target marker" (a specific antibody) and an "isotypic antibody" (a non-specific antibody), respectively. The binding sites of the target marker in the negatively stained sample (the sample stained with the isotype antibody) were blocked using an unconjugated version of the specific antibody, i.e., an antibody to which no fluorochrome was covalently linked. The positively stained sample was then added to the target cells at a nominal percentage using negatively stained cells. The spiked samples were analyzed in a flow cytometer to obtain measured percentages. The data (nominal and measured values of the percentage of samples spiked with analyte) were then used for statistical analysis to calculate validation parameters such as linearity, accuracy, precision, and LLOQ.

Sample collection procedures

[52] This example describes the validation of the percentage of viable T cells in a cell product using the following staining panel:

PI=propidium iodide, and APC=allophycocyanin

Viable T cells are identified as PI-CD3+.

a) Obtaining a negatively stained sample

[53] The cellular test sample was divided into two parts. One part was stained according to the manufacturer's protocols using a developed staining panel: a combination of a fluorochrome-conjugated antibody and a dead cell exclusion dye, necessary for gating on the viable cell population of interest. In the case of the target cell population, an isotype control was used in the staining mixture. In this example, the following staining mixture was used:

PI=propidium iodide, IqG=immunoglobulin G, and

APC=allophycocyanin

[54] After staining, unconjugated antibody to the target marker was added to the final cell solution. In this example, the unconjugated antibody was purified anti-CD3 antibody. As shown in Fig. 2, the FACS plot shows the result obtained with a negatively stained sample. The events shown in this example were gated on viable cells based on forward and side scatter and PI (to exclude dead cells).

b) Obtaining a positively stained sample

[55] From the cell test sample in stage 1, another portion of the cells were stained according to the manufacturer's protocol using the developed staining panel:

[56] As shown in Fig. 3, the FACS plot shows the result obtained with a positively stained sample. The events shown in this example were gated on viable cells based on forward and side scatter and PI (to exclude dead cells).

c) Generation of samples with low levels of spiked T cells The positively stained sample was diluted to the planned highest level of spiked analyte (nominal target cell concentration) using negatively stained cells from step 1 and then serial dilutions were made to generate diluted samples with systematically varying nominal T cell percentages.

Sample analysis

[57] Spiked samples were prepared as described above and used to generate flow cytometric fluorescence measurements, with each measurement representing a nominal CD3+ cell concentration in the spiked sample. Each nominal concentration and corresponding flow cytometric fluorescence measurement were subjected to routine statistical analyses as described elsewhere in this document to assess the accuracy and reliability of the measurements.

[58] This example describes a new preparation of negatively stained cells using an isotype antibody conjugated to a fluorochrome and an unconjugated antibody. Typically, cells depleted of a target cell population using magnetic beads or a cell sorter are used to generate a negative control population to generate spiked samples. However, these depletion procedures rarely deplete up to 99.9% of the target population. As a consequence, the negative control population will contribute at least 0.1% of the background, so that spiked samples below 0.1% are not possible. Validation of a fluorescence-based assay such as FACS with an LLOQ below 0.1% is challenging. This example demonstrates the efficient production of a negative control population having a target cell concentration below 0.1% and provides samples spiked with the analyte below 0.1% for validation. Depending on the instruments and staining panel used, the methods of the present invention are capable of providing LLOQs below 0.01%.

Example 2. Qualification protocol for a method for the quantitative determination of T cells in a medicinal product based on K-NK

[59] This method qualification protocol provides a procedure for qualifying a T cell quantification method (protocol reference: ATM-3343) for quantifying CD3+ expressing cells in an expanded natural killer cell composition (K-NK drug product (DP)), such as the compositions disclosed in PCT publication WO 2018160673 A1, the disclosure of which is incorporated herein by reference in its entirety. The K-NK DP was a CD3+ depleted donor lymphocyte preparation from which NK cells were expanded to high numbers and densities in vitro using membrane particles derived from plasma membrane 21-41bbl (PM21), such as the particles described in WO 2018160673 A1. The primary contaminant in K-NK-based DP was residual CD3+ expressing cells, which are capable of causing graft-versus-host disease when administered to patients above the clinically relevant limit. To detect and provide quantification of low concentrations of residual CD3+ expressing cells in K-NK-based DP, the FACS-based method described herein was developed to detect and/or quantify the content of exclusively viable CD3+ cells. The definitions presented in Table 1 and Table 2 are used throughout the following examples.

[60] Healthy donor PBMCs (RM-PBMC 225398) were used as “positive control samples” to verify CD3+ staining and to set the CD3+ gate for flow cytometric analysis. These PBMCs were obtained by density gradient centrifugation procedures. This positive control had no associated acceptance criteria but was used to set the gate and to verify the addition and reactivity of antibodies to CD3-APC.

[61] The test sample used for this qualification was cryopreserved NK cell sample “20035 d14”. 20035 d14 samples were obtained by further processing of NK cells expanded using the PM21 particle platform and were obtained after day 7 and were representative of the K-NK based product. Following purchase, these cells were cultured in the laboratory for seven days and cryopreserved at 5 x 10 ⁷ cells per vial on day 14. This material was formulated cryopreserved in CS10 and although it did not contain plasmalyte or HSA, samples were washed in cell staining buffer (CSB) and resuspended in CSB prior to staining in the assay. Thus, the matrix of all test samples was standardized prior to the assay, regardless of starting composition. Through performance testing during method development, a relative mean viable CD3+ cell count of 1.4% (n=8) was determined for sample 20035 d14.

[62] The equipment used was a classic AttuneTM acoustic focusing flow cytometer, and the software used for data acquisition and processing was Attune Cytometric Software v2.1 and JMP v14.

[63] An assay and procedure were developed to detect and quantify CD3+ expressing cells at low concentrations. Briefly, a high cell concentration sample was washed several times in FACS staining buffer and then stained with CD3-APC antibody to detect CD3+ expressing cells and propidium iodide (PI) to detect dead cells. The sample was then resuspended to a fixed concentration and acquired using an Attune flow cytometer. Flow cytometric data were processed using Attune Cytometric Software v2.1. To obtain resolution with precise and accurate quantification at a cell content of 0.01%, 1× ¹⁰⁶ viable MNCs were obtained. The flow cytometry gating strategy for ATM-3343 was defined as MNC-*single cells-*viable MNC-*viable CD3+.

The qualification was developed to demonstrate the suitability of the method for detecting and quantifying the percentage of cells expressing CD3+. Parameters reported include % viable CD3+ cells.

[64] Preparation of test samples. For linearity, accuracy/truthfulness, and precision analysis, cryopreserved test sample 20035 d14 was split into two preparations. One aliquot (1 x 10 ⁸ cells) was stained with IgG-APC and unlabeled anti-CD3 antibody to generate a CD3 mock sample, and the second aliquot (2 x 10 ⁷ cells) was stained with anti-CD3-APC antibody. The sample stained with the isotype antibody (IgG-APC) served as a diluent and matrix sample for addition to the fully stained sample to assess the linearity and relative accuracy of the method. Unlabeled anti-CD3 antibody was used to block CD3 binding sites due to any carryover of anti-CD3-APC antibody from the added sample. Prior to preparation, the cell density was adjusted to achieve an event rate of 1000 events/sec according to ATM-3343-0. The concentration of the initial positively stained CD3+ spiked sample and the negatively stained IgG-APC stained sample were used to generate the dilution schedule. The CD3-APC stained samples were split into two samples; 1) for %CD3+ determination and 2) for use as the spiked sample. Similarly, one aliquot (one-tenth of the preparation) of the IgG-APC stained sample was used for %CD3+ determination and the remainder of the sample was used as a diluent in the linearity schedule.

[65] Sample preparation. Samples were prepared according to ATM-3343 and scaled up as described in Table 3, including staining of positive control samples, IgG-APC stained samples, and anti-CD3-APC stained samples. Since 1 × 10 ⁸ cells were stained with IgG-APC, the amount of reagents used was increased 10-fold. In addition, molar equivalents of unlabeled anti-CD3 antibody were added to block CD3 binding sites. Anti-CD3-APC staining was performed using 2 × 10 ⁷ cells, so the amount of reagents used according to ATM-3343 was doubled. No additional reserve tubes were used. Each serial dilution sample had the same cell concentration of 2 × 10 ⁶ /mL (minimum volume 2 mL). The procedural flow for preparing both the linearity evaluation samples and the dilution series is illustrated in Fig. 4 and Fig. 5, respectively. A 4-fold dilution series to five levels was prepared and included a dilution range from 1.00% to 0.004%. The spiked analyte samples were prepared according to Table 4. For each case, one positive control sample, one IgG-APC stained sample, and one CD3-APC stained sample were prepared and used.

[66] The data set for evaluating the qualification parameters, namely linearity, accuracy/truth, precision and specificity, was obtained from 3 cases performed by 2 operators as shown in Fig. 6. One test sample for each case was processed according to the above dilution scheme and qualification was carried out according to the scheme illustrated in Table 5 below. The samples were processed according to ATM-3343 (serial dilution was equivalent to the test samples). Dilution level 5 was obtained first, then dilution level 4 and sequentially up to the first dilution level. One data collection point per level was created for each case. For this qualification testing, three measurements were made at each level for each operator for a total of six data collection points per dilution level.

[67] Relative precision/truthfulness. Relative precision or truthfulness was estimated using the mean value obtained for the test material relative to the corresponding nominal relative percentage of CD3+ per dilution level. Intermediate precision was determined for the case and the analyst depending on the dilution level.

[68] The range of the procedure was determined by dilution linearity testing. The lower limit of quantification was determined by linearity testing as the lowest dilution that had an appropriate level of precision and relative accuracy.

[69] The antibodies and viability dye used for ATM-3343 are listed in Table 6. The mean of 6 measurements (one per case for each operator) for each sample type was used for comparison.

[70] The range of the procedure was determined by dilution linearity testing. The lower limit of quantification was determined by linearity testing as the lowest dilution that had an appropriate level of precision and relative accuracy.

[71] Linearity: Linear regression analysis was performed to determine the R2 value by fitting the measured relative percentage from all analyses to the nominal relative percentage as a fixed effect in a linear mixed model using JMP.

The recovery for each test sample level from the linearity assessment was compared with the nominal relative percentages in a linear mixed model with the interaction of case and operator specified as random effects using JMP.

[72] Precision (interassay variation): The estimate of interassay precision variation was calculated by fitting the relative percentage in a linear mixed mode in JMP, where chance and operator are set as random effects and the nominal relative percentage is set as a fixed effect.

[73] The analytical procedure range was determined by precision and accuracy analysis using a linear mixed model generated using JMP. The lower limit of quantification was determined by precision and accuracy analysis using a linear mixed model generated using JMP. The LOQ was defined as the lowest sample concentration that met the appropriate precision and accuracy conditions.

[74] No acceptance criteria were established for the qualification of the method, but target criteria are listed in Table 7.

Example 3: Quantitative analysis of CD3+ T cells in a K-NK-based medicinal product

[75] Objective: This example presents the results of the qualification of the ATM-3343 method, where accuracy, precision, linearity, limit of quantification, and range were tested. The qualification parameters and procedures were outlined in Example 2. These procedures were developed to evaluate the sensitive analytical results obtained by the ATM-3343 method and to determine the limit of quantification of % residual CD3+ T cell impurity in a K-NK-based DP. This example also summarizes the method qualification data obtained by two operators on three occasions, as specified in Example 2. The results of the regression analysis to compare observed and predicted target measurement results are used to determine the qualification parameter determination results for Example 2.

[76] Materials and reagents. The materials and reagents described in ATM-3343 were used to perform the qualification protocol as detailed in Example 2. The materials and reagents used are listed in Table 8.

[77] The equipment used was a classical AttuneTM acoustic focusing cytometer and centrifuge (Thermo Scientific) (. The software used was Attune Cytometric Software v2.1 and JMP v. 14.3.0.

[78] Cryopreserved PBMC-225398 samples were used as a positive control to verify CD3+ gating and to set the CD3+ gate for flow cytometric analysis. The preparation of PBMC-225398 is described in Example 2.

[79] Test sample material. Cryopreserved NK cells from lot 20035 dl4 were used as test samples for the qualification of this method. The preparation of 20035 d14 cells for the qualification of this method is described in Example 2. 20035 d14 cells were expanded using CD3+ depleted PM21 particles and used in the standardized sample matrix in this assay and are therefore representative of the K-NK-based DP. Lot 20035 d14 has an average CD3+ % content of 1.389%, which was determined as part of the assay performance testing using a draft version of ATM-3343 (test sample assayed in 2 instances, with 2 sample vials and 2 replicates for n=8). 1, 38 9% CD3+ represents the expected (nominal) value for the test sample.

[80] Arrows from the top to the bottom section of the table) and subsequently included in the accompanying statistical analysis.

[81] Procedure: The method qualification for linearity, accuracy/truthfulness, and precision testing is described in Example 2. Briefly, cryopreserved NK 20035 d14 samples were stained with isotype-APC antibody and anti-CD3-APC antibody according to ATM-3343. The isotype-stained cells were also mock-stained to prepare the sample matrix as a diluent for CD3+ T cell addition. These mock-stained cells were treated with unlabeled purified anti-CD3 antibody to block any binding of CD3-APC due to any carryover of antibody solution during CD3+ T cell addition. Thus, serial dilutions for assessing the linearity of CD3+ T cell detection were prepared in the K-NK-based DP sample matrix and according to the requirements set forth in Example 2 (Figure 6). Data analysis was performed in JMP V14.3.0 statistical analysis software. Nominal (predicted) and estimated (results) %CD3+ values were compared from two analysts' reports, with three cases each. The test sample used for spiking required ATM-3343 analysis for CD3+ content in spiked analyte preparations with sham staining, and measurement was performed for each case. Therefore, 6 measurements were performed according to the test method, these data collection points were included in the method qualification data analysis as an additional level (level 0) used to assess linearity, accuracy, precision, and range. Plots of residuals of untransformed and logarithmically (Log10, Log, logistic regression, logit percent, and logit regression) transformed nominal %CD3 and estimated %CD3 were examined to select the regression model (Figure 7). When analyzing the residual plot, the untransformed data showed a tendency for the data to be scattered, with a small scatter at the lower level and a wider scatter at the higher level. Therefore, the data were transformed for further analysis. Log10 transformation was the simplest transformation, which resulted in a better distribution of the data in the residual plot. Therefore, the "nominal" and "estimated" data with Log10 transformation of %CD3 to three decimal places were selected for further analysis. The results shown in the tables are presented to three decimal places.

[82] Results: Accuracy was analyzed using a linear mixed model with "operator" and "chance" as random factors. Recovery was calculated using the equation below:

[83] The target acceptance criteria for relative precision outlined in Example 2 were 10 0+30%. The % recovery values for dilution levels with nominal %CD3, 0.016-1.389%, were within the 10 0+30% acceptance criteria.

The nominal %CD3 at a CD3+ T cell level of 0.004% did not meet the acceptance criteria for accuracy (Table 10 and Figure 8).

[84] Precision. Precision was analyzed using a linear mixed model in JMP V14.3.0 with “operator” and “chance” as random factors. As specified in Example 2, subsequent replicate measurements on the same samples were not performed due to time constraints. Since the acceptance criteria for precision/true were not met at level 5 (nominal CD3+ T cell count of 0.004%), nominal and estimated data obtained at this level were excluded from the precision analysis. The acceptance criteria outlined in Example 2 were to report %CV for precision and <20% CV for intermediate precision. The %CV results for the operator, case, residual, and intermediate precision were 0.288, 2.386, 4.819, and 5.386%, respectively (Table 11). Thus, levels 0–4 (1.389–0.016% CD3+ T cells) met the acceptance criteria for the precision analysis.

[85] Linearity. To determine the linearity of CD3+ quantification, cells stained with anti-CD3-APC antibody were added and serially diluted in the K-NK-based DP sample matrix as shown in Figure 5 to predict five levels of %CD3+ T cell score (levels 1-5 or 1.000-0.004% CD3+ T cells). The sample matrix consisted of mock cells treated with isotype-APC antibody and purified unlabeled anti-human CD3 antibody (negative staining). Unlabeled (unconjugated) anti-CD3 antibody was included to block any CD3 staining due to carryover of anti-CD3-APC antibody from the spiked sample into the sample matrix diluent. The undiluted test sample data obtained when the analyte was added were included as an additional level (level 0) for linearity assessment; thus, linearity was assessed from 1.389% to 0.004% of CD3+ T cells. Since the assessed results at the nominal level of 5 (0.004% of CD3+ T cells) did not meet the acceptance criteria for precision, this level was excluded from the linearity analysis in the regression model. The regression plot and the residual from the predicted plot are shown in (Figure 9). The target criteria for linearity outlined in Example 2 were R2>0.8. The ^R2 value observed in the regression analysis was 0.999 (Table 12). Thus, linearity for levels 0-4, 1.389-0.016% of CD3+ T cells, met the target acceptance criteria.

[86] Range. The range of %CD3+ T cell quantification was determined by accuracy and precision analysis using a linear mixed model generated in JMP V14.3.0. Accuracy was analyzed for five dilution levels (1-5) with expected CD3+ T cell contents of 1.000-0.004% according to a four-fold dilution scheme within the K-NK-based DP sample matrix. The inclusion of undiluted test sample resulted in the addition of an additional level increasing the upper range to 1.389% CD3+ T cells. The predicted %CD3+ T cell quantification result for levels (0-4), 1.389-0.016% CD3+ T cells, met the acceptance criteria for accuracy (100%+30%). Level 5 (0.004% CD3+ T cells) did not meet the acceptance criteria for accuracy. Therefore, the range for this method was defined as 1.389% to 0.016% of CD3+ T cells for quantification. %CD3+ T cells below this quantification range should be reported as BLOQ (lower limit of quantification), while %CD3+ T cells above this quantification range should be reported as above ULOQ (upper limit of quantification). Assay validity criteria of 1 x ¹⁰⁶ viable single cells must be met to report results within this quantification range.

[87] K-NK-based DP consisted of NK cells that were expanded to high densities in vitro from a CD3+ T cell depleted PBMC population using PM21 particles. The contaminant in K-NK-based DP may be residual CD3+ T cells after depletion, which may persist at levels below certain detection limits of previous assays. Therefore, a flow cytometric-based assay, ATM-3343, was developed to detect low (<0.3% CD3+ T cells) levels of residual CD3+ T cells as a potential contaminant in K-NK-based DP. ATM-3343 combines blocking and washing steps to specifically stain and recover CD3+ T cells. In this method, reagents consisting of human TruStain FcX, CD3-APC and isotype antibody-APC were used to specifically detect and quantify %CD3+ T cells in K-NK-based DP. A reference population of PBMCs was used as a positive control. Given the requirement for low-level CD3+ quantification according to ATM-3343, a yield of 1× ¹⁰⁶ viable single cells from K-NK-based DP samples was set to meet statistical expectations.

[88] Qualification parameter data including accuracy, precision, linearity, and range were obtained with two operators on three occasions each. In each case, reference material and cells stained with isotype-APC or anti-CD3-APC antibody were obtained in addition to obtaining five levels of predetermined % CD3+ cell concentration values (Table 8). In each case, these dilution levels were obtained by spiking CD3+ T cells into a matrix of mock (isotype- and unlabeled anti-CD3 antibody-treated) cell samples from a K-NK-based DP. Flow cytometry acquisition parameters were set to obtain 1× ¹⁰⁶ viable single cells from each sample based on the sequential gating strategy (total events > MNC > single cells > viable cells > CD3+) outlined in ATM-3343 and Example 2.

[89] Brief Description: This method was qualified according to the procedures described in Example 2, "Method Qualification Protocol for T Cell Quantification Using ATM-3343". The K-NK-based drug product (DP) consisted of NK cells expanded to high densities in vitro from a CD3+ T cell depleted population of donor-derived PBMCs. ATM-3343 was designed to detect and quantify low frequencies of residual CD3+ T cells within a population of 1 x 10 ⁶ viable single cells by flow cytometry. The data evaluated in this method qualification report were generated by two analysts with three cases each. The evaluation of accuracy, precision, and linearity parameters described in Example 2 demonstrates the suitability of ATM-3343 for its intended use in the detection and quantification of residual CD3+ T cells in a K-NK-based DP. ATM-3343 was accurate and precise in measuring levels ranging from 0.016% to 1.389% CD3+ T cell contamination in K-NK-based DP (Table 14). The ^R2 value of 0.999 for %CD3+ quantification in this range met the acceptance criteria for K-NK-based DP sample matrix linearity. Thus, the method established the detection of 0.016% CD3+ T cells as the limit of quantification (LOQ). CD3+ T cell detection by ATM-3343 is specific and can reliably distinguish positive events from background noise due to isotype antibody staining. ATM-3343 is required to recover 1× ¹⁰⁶ viable single cells from K-NK-based DP to meet the acceptance criteria. Overall, the protocol used is suitable for the determination of CD3+ T cell admixture in K-NK based DP. The % CD3+ T cells below or above the quantification range of 0.016% to 1.38 9% should be reported as BLOQ (lower limit of quantification) or above ULOQ (upper limit of quantification), respectively. This quantification range, obtained by obtaining 1×10 ⁶ viable individual cells, defined in this qualification, can be established as the validity and suitability criteria of the system.

*Level 0: Undiluted test sample from lot 20035 d14, also referred to as CD3-APC (with added analyte) in Table 5.

[90] Conclusion: ATM-3343 was qualified for the quantification of low (>0.016%) CD3+ T cell % in K-NK-based DP. The accuracy, precision, and linearity of the CD3+ T cell % quantification in the range of 0.016% to 1.389% were determined to be acceptable according to the qualification parameters. The validity of the ATM-3343 assay and system suitability criteria were required to reliably measure CD3+ T cell contaminants in K-NK-based DP.

Claims (32)

1. A method for validating measurements using a fluorescence-based assay instrument on a population of stained target cells in a test sample, the method comprising: (a) negative staining or the presence of negatively stained target cells in a first portion of a reference sample containing target cells expressing a target cell marker, wherein the negative staining of target cells in the first portion of the reference sample in (a) comprises introducing into the first portion of the reference sample (i) a fluorescent dye to exclude dead cells, (ii) a non-specific antibody conjugated to a fluorochrome, and (iii) a specific antibody capable of specifically binding the target marker on the target cell; (b) positive staining or the presence of positively stained target cells in a second portion of the reference sample, wherein the positive staining of target cells in the second portion of the reference sample in (b) comprises introducing into the second portion of the reference sample (i) a fluorescent dye that excludes dead cells and (ii) a specific antibody conjugated to a fluorochrome; (c) analyzing a first portion of the reference sample with the instrument to obtain a fluorescence measurement result indicative of the concentration of target cells in the first portion of the reference sample, and analyzing a second portion of the reference sample with the instrument to obtain a fluorescence measurement result indicative of the concentration of target cells in the second portion of the reference sample; (d) based on the fluorescence measurement results obtained in (c), diluting the positively stained target cells in the second portion of the reference sample with the negatively stained target cells in the first portion of the reference sample to obtain a diluted sample having the highest concentration of target cells, from which a dilution series is obtained comprising a plurality of diluted samples, each diluted sample having a nominal concentration of target cells, each nominal cell concentration being greater than the concentration of target cells indicated by the fluorescence measurement result of the negatively stained first portion in (a), wherein the nominal concentration in each diluted sample is different from the nominal concentration in each of the remaining diluted samples, and wherein the highest concentration of target cells in the dilution sample having the highest concentration of target cells in the dilution series is 2%, 1%, 0.5%, 0.2%, 0.1%, 0.05%, or 0.02%; (e) analysis of the series of diluted samples from (d) using an instrument to produce a series of fluorescence measurements that include a measurement of the fluorescence of each diluted sample; (f) comparing the nominal cell concentration of (d) and the fluorescence measurement of (e) for each diluted sample to quantify the performance of the staining method using the instrument, where the comparison in (f) comprises performing a statistical calculation of the difference between the nominal cell concentration and the fluorescence measurement for each diluted sample to determine at least one of linearity, range, accuracy, precision, limit of detection (LOD), and lower limit of quantification (LLOQ) for the instrument and staining method, and where the determination of LLOQ comprises identifying a target cell concentration associated with a predetermined precision criterion and a predetermined accuracy criterion. 2. The method according to claim 1, wherein the fluorescent dye for excluding dead cells is selected from a nucleic acid binding dye, propidium iodide, DAPI, DRAQ7, 7-AAD, TO-PRO-3 and a dye that reacts with amino groups. 3. The method according to claim 1, wherein the non-specific antibody conjugated to the fluorochrome comprises an antibody that does not have the ability to specifically bind to a cellular antigen. 4. The method of claim 3, wherein the non-specific antibody conjugated to a fluorochrome comprises any of IgD, IgG, IgA, IgM or IgE conjugated to a fluorochrome. 5. The method according to claim 1, wherein the fluorochrome is selected from allophycocyanin (APC), APC C750, APC AF700, brilliant violet (BV)421, BV510, hylitol 7 (H7) BV605, BV650, PE CF594, fluorescein isothiocyanate (FITC), R-phycoerythrin (PE or R-PE), PE-Cy7 (PE linked to the cyanine dye Cy7), APC-Cy7 (APC linked to the cyanine dye Cy7), APC-H7 (APC linked to hylitol 7, which is an analogue of Cy (H7)). 6. The method according to claim 1, wherein the target cell marker is selected from CD3 as a T cell marker, CD19 as a B cell marker, CD235a as a red blood cell marker, CD56 as a natural killer (NK) cell marker, CD14 as a monocyte marker, and CD66b as a granulocyte marker. 7. The method of claim 1, wherein the target cell population is a population of T cells and the target cell marker is CD3. 8. The method of claim 1, wherein the test sample contains a target cell present at a concentration of no more than 2%, 1%, 0.5%, 0.2%, 0.1%, 0.05%, or 0.02%. 9. The method according to claim 1, wherein the negatively stained first portion of the test sample is obtained in the absence of a cell depletion step. 10. The method according to claim 1, wherein the device based on measuring fluorescence is a flow cytometer. 11. A method for validating measurements using a flow cytometer on a population of stained target cells in a test sample, the method comprising: (a) negative staining or the presence of negatively stained target cells in a first portion of a reference sample containing target cells expressing a target cell marker, wherein the negative staining of target cells in the first portion of the reference sample in (a) comprises introducing into the first portion of the reference sample (i) a fluorescent dye to exclude dead cells, (ii) a non-specific antibody conjugated to a fluorochrome, and (iii) a specific antibody capable of specifically binding the target marker on the target cell; (b) positive staining or the presence of positively stained target cells in a second portion of the reference sample, wherein the positive staining of target cells in the second portion of the reference sample in (b) comprises introducing into the second portion of the reference sample (i) a fluorescent dye that excludes dead cells and (ii) a specific antibody conjugated to a fluorochrome; (c) analyzing both the first portion and the second portion of the reference sample using a flow cytometer to obtain fluorescence measurements indicative of the concentration of target cells in each of the first portion of the reference sample and the second portion of the reference sample; (d) based on the fluorescence measurement results obtained in (c), diluting the positively stained target cells in the second portion of the reference sample with the negatively stained target cells in the first portion of the reference sample to obtain a diluted sample having the highest concentration of target cells, from which a series of diluted samples are obtained, each of which has a nominal concentration of target cells that varies systematically in the dilution series, wherein the nominal concentration of cells in at least one diluted sample is greater than the concentration of target cells indicated by the fluorescence measurement result of the negatively stained first portion in (a), and wherein the highest concentration of target cells in the dilution sample having the highest concentration of target cells in the dilution series is 2%, 1%, 0.5%, 0.2%, 0.1%, 0.05%, or 0.02%; (e) analysis of the series of diluted samples from (d) using a flow cytometer to obtain a series of fluorescence measurements comprising measuring the fluorescence of each diluted sample; (f) comparing the nominal cell concentration of (d) and the fluorescence measurement of (e) for each diluted sample to quantify the performance of the staining method using a flow cytometer, where the comparison in (f) comprises performing a statistical calculation of the difference between the nominal cell concentration and the fluorescence measurement for each diluted sample to determine at least one of linearity, range, accuracy, precision, limit of detection (LOD), and lower limit of quantification (LLOQ) for the instrument and the staining method, and where the determination of LLOQ comprises identifying a target cell concentration associated with a predetermined precision criterion and a predetermined accuracy criterion. 12. The method according to claim 11, wherein the target cell marker is selected from CD3 as a T cell marker, CD19 as a B cell marker, CD235a as a red blood cell marker, CD56 as a natural killer (NK) cell marker, CD14 as a monocyte marker, and CD66b as a granulocyte marker. 13. The method of claim 11, wherein the test sample contains a target cell present at a concentration of no more than 2%, 1%, 0.5%, 0.2%, 0.1%, 0.05%, or 0.02%. 14. The method of claim 1 or 11, wherein the series of diluted samples comprises at least two, three, four, five, six, seven, eight, nine or ten diluted samples. 15. A non-volatile computer-readable medium containing instructions for a computer processor to perform the comparison step (f) in paragraph 1 or 11. 16. A system for validating the results of fluorescence measurements in a test sample obtained using a device based on fluorescence measurement, comprising a device based on fluorescence measurement and a computer connected to the device based on fluorescence measurement and comprising a machine-readable medium according to claim 15, where the device based on fluorescence measurement is a flow cytometer. 17. A kit containing reagents for staining cells in a reference sample for validating an analytical method based on fluorescence measurement for analyzing target cells in a test sample, the kit comprising: negative staining reagents comprising: (a) a fluorescent dye to exclude dead cells, (b) a non-specific antibody conjugated to a fluorochrome, and (c) a specific antibody capable of specifically binding a target marker on target cells, not conjugated to a fluorochrome; positive staining reagents comprising: (a) a fluorescent dye to exclude dead cells and (b) a specific antibody conjugated to a fluorochrome, and instructions for (a) negatively staining a first portion of the reference sample, (b) positively staining a second portion of the reference sample, and (c) providing a dilution series comprising a plurality of diluted samples, each of which has a nominal target cell concentration that varies across the dilution series, wherein the nominal cell concentration in at least one diluted sample is greater than the target cell concentration indicated by a fluorescence measurement of the negatively stained first portion.

Images