Patent Application
20250052985
The present disclosure is directed to a quantitative phase imaging system for determining multiple response parameters of therapeutic treatments.
Precision oncology can be useful for improving outcomes in cancer patients by tailoring effective therapies to an individual patient's tumor while minimizing toxic side effects from ineffective drugs. In some implementations, biomarker-driven personalized cancer treatment can improve response rates and extend progression-free survival. Sequencing studies using large oncogene panels in advanced cancers find an actionable DNA mutation in 5-35% of cases, depending on associated tumor histology. Although some mutations are exceptional responders to targeted therapy, rarely do advanced cancer patients with a candidate “targetable” mutation exhibit long-term survival. Thus, there is a need in precision oncology to implement functional cell-based assays to complement genomic panels.
Recent advances in tumor cell expansion allow for the development of ex vivo patient-derived models of cancer that faithfully recapitulate clinical behavior in terms of drug response. Ex vivo testing is also amenable to clinical testing since ex vivo can be multiplexed and completed within weeks of tumor sample collection. Considering the fast turnaround time from tumor sample collection, ex vivo testing also allows many more drugs to be screened at a lower cost, and on a timescale with potential for informing patient care. A variety of analytic methods for measuring the response of cultured cells to drug exposure are presently employed. Cell culture-based drug-screening assays vary from simple cell counts and determination of live: dead ratios with stains, to metabolic assays (e.g., assays for determining the release of ATP or lactate12), to measurement of specific programmed cell death effectors such as caspases or BH3-domain activation. CellTiter-Glo (CTG), for example, is an assay that measures cell ATP content as a proxy for cell viability. When used as an endpoint assay post drug exposure, CTG can produce reproducible drug response data, more rapidly and with less bias than cell counting. CTG can also produce greater signal-to-noise ratio than other luminescence assays such as Toxilight and resazurin-based assays. However, the described measures are typically applied as bulk, endpoint assays, and are incapable of capturing the dynamics of single-cell responses to therapy.
In contrast to endpoint assays, real-time assays can elucidate the temporal dynamics of drug response. In some implementations, real-time assays discriminate between a cytostatic response where cell growth is substantially reduced and a cytotoxic response where the therapy induces cell death. For example, incubator-housed microscope systems for measuring real-time cell proliferation (e.g., Incucyte) can yield results concordant to CTG and BH3 profiling. As a longitudinal imaging approach, the Incucyte measures parameters such as population-averaged proliferation rate and cell viability throughout the experimental duration granting insight into changes in cell behavior throughout the course of imaging. An emerging alternative is to use cell mass accumulation rate as a measure of cell growth. For example, suspended microchannel resonators are a highly sensitive tool for measuring changes in cell mass. Microchannel resonators can measure statistically meaningful changes in cell growth from very short duration (˜10 min) measurements, or individual resonators can be used for longitudinal imaging of cell behavior in response to drugs. However, suspended microchannel resonators are limited by the need to flow cells through individual resonators and work best with non-adherent cell types.
Thus, systems and methods that provide real-time imaging of cell-based assays with capturing dynamic single-cell responses, faster response rate determination, and reduced cost would be desirable.
Quantitative phase imaging (QPI) measures the growth rate of individual cells by quantifying changes in mass versus time. In some implementations, breast cancer cell lines MCF-7, BT-474, and MDA-MB-231 can be used to validate QPI as a multiparametric approach for determining response to single-agent therapies. The validation of QPI allows for rapid determination of drug sensitivity, cytotoxicity, heterogeneity, and time of response for about 100,000 individual cells or small clusters in a single experiment. In some examples, QPI half maximal effective concentration (EC50) values are concordant with CTG, a gold standard metabolic endpoint assay. Additionally, by applying multiparametric QPI, cytostatic/cytotoxic and rapid/slow responses can be characterized, and the emergence of resistant subpopulations can be tracked. Thus, QPI reveals dynamic changes in response heterogeneity in addition to average population responses, a key advantage over endpoint viability or metabolic assays. Overall, multiparametric QPI reveals a rich picture of cell growth by capturing the dynamics of single-cell responses to candidate therapies.
In one aspect, the disclosure provides a microscope imaging system including a quantitative phase imaging microscope. The quantitative phase imaging microscope includes a stage configured to hold and move a well plate having a plurality of therapy treatment samples, an array of light emitting diodes (LEDs) configured to illuminate the plurality of therapy treatment samples, and a camera configured to capture images of the plurality of therapy treatment samples over a period of time using an amount of illumination provided by the array of LEDs. The microscope imaging system also includes an electronic controller communicatively coupled to the quantitative phase imaging microscope. The electronic controller is configured to receive a plurality of images of the plurality of therapy treatment samples over the period of time from the camera, determine a cell mass for each therapy treatment sample of the plurality of therapy treatment samples based on each image of the plurality of images, track the cell mass over the period of time for each therapy treatment sample of the plurality of therapy treatment samples, determine a plurality of response parameters for each therapy treatment sample of the plurality of therapy treatment samples based on the cell mass over the period of time, and determine a treatment response for each therapy treatment sample of the plurality of therapy treatment samples based on the plurality of response parameters.
In some aspects, the plurality of response parameters include at least one selected from the group consisting of a specific growth rate, a half maximal effective concentration, a depth of response, a time of response, and a standard deviation of response.
In some aspects, each therapy treatment sample of the plurality of therapy treatment samples includes live cancer cells exposed to a different therapeutic drug.
In some aspects, the cell mass is a mass of the live cancer cells exposed to the different therapeutic drugs.
In some aspects, the quantitative phase imaging microscope further includes a lens configured to focus the camera on the plurality of therapy treatment samples.
In some aspects, when determining the cell mass for each therapy treatment sample of the plurality of therapy treatment samples, the electronic controller is further configured to determine a phase shift of the amount of illumination passing through each therapy treatment sample of the plurality of therapy treatment samples.
In some aspects, when tracking the cell mass over the period of time for each therapy treatment sample of the plurality of therapy treatment samples, the electronic controller is further configured to determine a mass accumulation rate for each therapy treatment sample of the plurality of therapy treatment samples based on the determined cell mass and the period of time.
In another aspect, the disclosure provides a method for determining treatment response parameters with a microscope imaging system. The method includes receiving, by an electronic controller, a plurality of images of a plurality of therapy treatment samples over a period of time from a camera, determining, via the electronic controller, a cell mass for each therapy treatment sample of the plurality of therapy treatment samples based on each image of the plurality of images, tracking, via the electronic controller, the cell mass over the period of time for each therapy treatment sample of the plurality of therapy treatment samples, and determining, via the electronic controller, a plurality of response parameters for each therapy treatment sample of the plurality of therapy treatment samples based on the cell mass over the period of time.
In some aspects, the method includes determining, via the electronic controller, a treatment response for each therapy treatment sample of the plurality of therapy treatment samples based on the plurality of response parameters.
In some aspects, the plurality of response parameters include at least one selected from the group consisting of a specific growth rate, a half maximal effective concentration, a depth of response, a time of response, and a standard deviation of response.
In some aspects, each therapy treatment sample of the plurality of therapy treatment samples includes live cancer cells exposed to a different therapeutic drug.
In some aspects, the cell mass is a mass of the live cancer cells exposed to the different therapeutic drugs.
In some aspects, the method includes determining, via the electronic controller, a phase shift of the amount of illumination passing through each therapy treatment sample of the plurality of therapy treatment samples.
In some aspects, the method includes determining, via the electronic controller, a mass accumulation rate for each therapy treatment sample of the plurality of therapy treatment samples based on the determined cell mass and the period of time.
In another aspect, the disclosure provides a quantitative phase imaging microscope including a camera configured to capture images of a plurality of therapy treatment samples over a period of time and an electronic controller communicatively coupled to the quantitative phase imaging microscope. The electronic controller is configured to receive a plurality of images of the plurality of therapy treatment samples over the period of time from the camera, determine a cell mass for each therapy treatment sample of the plurality of therapy treatment samples based on each image of the plurality of images, track the cell mass over the period of time for each therapy treatment sample of the plurality of therapy treatment samples, and determine a plurality of response parameters for each therapy treatment sample of the plurality of therapy treatment samples based on the cell mass over the period of time.
In some aspects, the electronic controller is configured to determine a treatment response for each therapy treatment sample of the plurality of therapy treatment samples based on the plurality of response parameters.
In some aspects, the electronic controller is configured to determine the plurality of response parameters for each therapy treatment sample of the plurality of therapy treatment samples in cells from a patient derived organoid.
In some aspects, the electronic controller is configured to determine a difference in the plurality of response parameters between each therapy treatment sample of the plurality of treatment samples.
In some aspects, the difference in the plurality of response parameters is determined from a tumor organoid.
In some aspects, the electronic controller is configured to determine a tumor cell from a non-tumor cell for each therapy treatment sample of the plurality of therapy treatment samples based on the cell mass over the period of time.
Other aspects of various embodiments will become apparent by consideration of the detailed description and accompanying drawings.
The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.
Before any embodiments of the invention are explained in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways.
Quantitative phase imaging (QPI) is a real-time, label-free technique for determining growth of individual cells by measuring a phase shift of light as it passes through a transparent sample, such as, for example, a cell. The phase shift of light is directly proportional to cell mass, which increases due to cell growth, such as, for example, during progression through a cell cycle. In some instances, cell mass is a mass of live cancer cells exposed to a therapeutic drug. QPI is a real-time, high throughput tool for measuring growth responses of individual cells to therapy. Previous applications of QPI have utilized microfluidics to study cell response to fluidic shear stress or identify resistant cell populations using up to 20 different QPI-derived features such as, for example, area and shape. However, these and other studies of QPI have narrowly focused on measuring the overall sensitivity or toxicity of potential therapies with only limited studies of a heterogeneity of response. Previous work with CellTiter-Glo (CTG) shows that combining multiple measures of cell response is superior to only measuring drug sensitivity. In some instances, QPI can have a greater impact as a tool for functional medicine by enabling simultaneous measurement of multiple parameters that are indicative of cancer cell response to therapy.
In some instances, QPI can be used as a quantitative multiparametric method to characterize dynamic changes in growth rate, drug sensitivity, drug toxicity, heterogeneity, and ToR (e.g., measured parameters) with sufficient throughput to make QPI suitable for clinical applications. The measured parameters, summarized in Table 1 below, are orthogonal measurements that cannot be derived from traditional measurements, such as mean drug sensitivity alone. In some instances, the measured parameters can also be combined to quantify drug-dependent, dynamic responses of cell populations in terms of time-varying mean and standard deviation of growth rates. Taken together, the measured (e.g., QPI-derived) parameters described herein provide a richer, more complete description of cell response to therapy than conventional drug screening approaches.
In some embodiments, breast cancer cell lines are imaged, representing diverse clinical subtypes, summarized in Table 2 below, using a QPI microscope (
The image 1700 shows a field of view of BT-474 cells treated with DMSO at the beginning of a 72 h experiment. The image 1705 shows a field of view of BT-474 cells at the end of the experiment where the cells have grown into clusters. The image 1710 shows an example of a well of the well-plate 105 containing MCF-7 cells at the beginning of the experiment from the DMSO control. The image 1715 shows at the end of the experiment, the MCF-7 cells have grown to fill most of the field of view. For example,
The image 1720 shows a cluster of 3 BT-474 cells from the DMSO solvent control that grow into a larger cluster of 6 cells, as shown by the image 1725, that are segmented together (green) to measure the growth of the cluster. The graph 1730 shows individual mass measurements for the cluster of images 1720 and 1725 and are plotted showing robust growth. The image 1735 shows a small cluster of MCF-7 cells treated with DMSO at the beginning of the experiment, grows into a larger cluster by the 45 h timepoint as shown in the image 1740. The graph 1745 shows measuring the mass of the cluster of images 1735 and 1740 demonstrating robust growth. The image 1750 is an example of a small cluster of BT-474 cells exposed to 2 μM of doxorubicin at the beginning of the experiment. At the end of the experiment, the image 1755 shows the cluster has not grown noticeably in size. The graph 1760 shows the clusters mass of images 1750 and 1755 is tracked over time showing the cluster is losing mass. The image 1765 shows MCF-7 cells exposed to 20 μM fulvestrant at the 10 h timepoint and the image 1770 shows MCF-7 cells exposed to 20 μM fulvestrant at the 72 h timepoint appear to shrink in size over the course of the experiment. The graph 1775 shows plotting the mass over time for this small cluster of the images 1765 and 1770, that they grow slowly at the beginning and end the experiment with a lower mass than they started with.
For example,
In some embodiments, QPI mass versus time data includes several features that underlie the ability of QPI to distinguish multiple dynamic characteristics of cell responses to drugs. First, the rate of mass accumulation, or cell growth rate (dm/dt), may be used to characterize cell growth. As illustrated in
For proliferating cells, the SGR matches the exponential growth constant measured by cell counting as cells double their mass with each cell cycle, as shown by graphs 1815-1825 of
In some embodiments, a dose-dependent change in growth, as indicated by a change in the rate of mass accumulation or loss, is a parameter that can be extracted from QPI dose-response data. For example,
For conditions with a response, the QPI microscope system 110 determines DoR from the fitted Hill curve as the difference between the asymptotes at the highest and lowest concentrations, normalized by the asymptote at low concentration. The normalization accounts for differences in the control growth rates of each cell line. With reference to
By precisely measuring the mass of individual cells during treatment, the QPI microscope system 110 gives rapid and sensitive insight into the dynamic response of cells to therapy throughout an experiment with high temporal resolution.
Referring now to
Often, the concentration tested just below the EC50 elicits no response. For example, drugs that elicit no response never cross the Hellinger distance response threshold, as shown by the graphs 2400 in
In some instances, the QPI microscope system 110 uses the ToR at the tested concentration just above the calculated EC50 as the nearest approximation of ToR at the calculated EC50 (e.g., ToR at EC50). Comparing ToR to DoR indicates that cytotoxic conditions elicit the fastest response, but even conditions classified as cytostatic often elicit a response in less than 24 h with a moderately negative relationship between ToR and DoR (R=−0.62, p=0.002), as illustrated in
SGR in control populations shows intrinsic heterogeneity as shown by the large standard deviation in growth rates even in the control group, as shown by graphs 2500-2510 of
For example,
For example, as illustrated in
The SD is used at the tested concentration nearest the EC50 (SD at EC50) as a relevant measure of heterogeneity in a responding cell population. With reference to
To further elucidate the relationships among QPI parameters, the QPI microscope system 110 determines the Pearson correlation coefficient between all QPI measured parameters (EC50, DoR, ToR at EC50, and SD at EC50), as well as CTG-based EC50 and DoR in
With reference to
BT-474 cells however, as illustrated in
The graphs 3100 of
The graph 3200 of
Graphs A, B, C, and D of the graphs 3300 of
The application of QPI with the QPI microscope system 110 is a multiparametric, label-free, high-throughput tool for measuring growth response of adherent cells to cancer therapies. QPI predictions of which drugs a given population of cells will not respond to and the concentration at which cells demonstrate sensitivity to therapy (EC50) is strongly concordant with traditional CTG measurements. Additionally, QPI with the QPI microscope system 110 includes additional metrics for characterization of drug response at a single cell level. The DoR measured using QPI with the QPI microscope system 110 is a useful tool for classifying the effect of therapies as either cytostatic or cytotoxic. As a method of tracking growth rates over time, the QPI microscope system 110 measures response dynamics of single cells, including ToR and heterogeneity, and includes tracking of outliers. Considering all of the noted parameters (e.g., EC50, DOR, ToR at EC50, and SD at EC50) shows orthogonality and the dynamic responses of populations over time. A summary of the measured parameters for all studied conditions is included in Table 9 below.
In some instances, the QPI microscope system 110 measures time-averaged growth rate for individual cells/clusters, both on a scale of the entire experiment and the growth rate over smaller time intervals demonstrating how QPI tracks the temporal dynamics of growth and heterogeneity (e.g., as shown in
In some embodiments, several aspects of multiparametric QPI point towards clinical applications, in addition to its concordance with CTG, which is widely used in clinical trials of functional oncology. QPI includes a readout of cell response and is marker free. In addition, QPI is not susceptible to drug-stain interference or false positives from sub-lethal/sub-cytostatic alterations in ATP production. QPI uses relatively few cells, making it amenable for application to clinical samples with limited cell numbers or where expansion is difficult. The examples and embodiments described herein use approximately 200,000 cells to achieve sufficient density for imaging in each of 96-wells of the well-plate 105, with the QPI microscope system 110 measuring the response of up to half of the total cells plated in the well-plate 105. In some embodiments, a different number of cells may be used and the well-plate 105 may include a different number of wells.
In some instances, 72 h QPI experiments are not required for measuring cell sensitivity to therapy. In most instances, 24 h QPI is sufficient to predict treatment response relative to CTG and 24 h QPI results are strongly correlated with 72 h QPI results (
Referring to
In some embodiments, the QPI microscope 120 is coupled to a high-speed xy translation stage (e.g., a translation stage) 145 (e.g., MLS203, manufactured by Thorlabs, U.S.A), set to, for example, an acceleration of 2480 mm/s2 and a maximum velocity of 400 mm/s, to image each location of the well-plate 105 with a temporal resolution of 20 minutes. In some embodiments, the translation stage 145 is configured to hold and move the well-plate 105 having a plurality of samples (e.g., a plurality of therapy treatment samples). In some embodiments, for illumination, the QPI microscope 120 includes an 18 mm square 8×8 light emitting diode (LED) array (e.g., an LED array 150) including a plurality of LEDs, positioned approximately 24 mm above a sample plane of the well-plate 105 for a numerical aperture of illumination (NALED array) of 0.39. In some embodiments, the LED array 150 is configured to illuminate the well-plate 105 including the plurality of samples. In some embodiments, a coherence parameter, denoted as σ, which is a ratio of NALED array: NAobj is 1.5235,36. The LED array 150 is controlled, for example, via an Arduino Metro M4 (e.g., manufactured by Adafruit, U.S.A.) (e.g., an electronic controller 155). In some embodiments, the LED array 150, the translation stage 145, and the camera 135 are each electronically (e.g., communicatively) connected to the electronic controller 155. In other words, the electronic controller 155 may control an operation of each of the LED array 150, the translation stage 145, and the camera 135. For example, the electronic controller 155 transmits a light signal indicative of an illumination level to the LED array 150 and the LED array 150 illuminates the well-plate 105 based on the light signal.
In some embodiments, the electronic controller 155 transmits a movement signal to the translation stage 145 and the translation stage 145 moves the well-plate 105 based on the movement signal. In some embodiments, the electronic controller 155 is electronically connected to a motor 160. The motor 160 is coupled to a drive shaft 165 and configured to drive the drive shaft 165. The drive shaft 165 is coupled to the translation stage 145 and configured to move the translation stage 145. In some embodiments, the electronic controller 155 transmits the movement signal to the motor 160 and the motor 160 drives the drive shaft 165 to move the translation stage 145 (holding the well-plate 105 having the plurality of therapy treatment samples) based on the movement signal. In some embodiments, the electronic controller 155 transmits an image signal to the camera 135 and the camera 135 captures images of the plurality of samples on the well-plate 105 based on the image signal. In some embodiments, the QPI microscope system 110 includes the electronic controller 155 and the electronic controller 155 may be electronically connected to each of the components of the QPI microscope 120. It should be understood that the electronic controller 155 may be configured to calculate or determine values for each of the representations illustrated by
In some embodiments, the electronic controller 155 implements DPC image acquisition and phase retrieval with the QPI microscope 120. In some examples, the QPI microscope 120 captures, via the camera 135, four images with half circle illumination (top, bottom, left, right), via the LED array 150, in less than one second including QPI microscope motion. For example, the electronic controller 155 receives a plurality of images of the plurality of treatment samples from the camera 135 over the period of time. Opposing pairs of images are used to compute the phase gradient in two orthogonal directions, via the electronic controller 155. The QPI microscope 120 computes, via the electronic controller 155, a phase shift of the captured images by deconvolution with an estimated optical transfer function via Tikhonov regularization. In some embodiments, parameters for obtaining phase reconstruction include illumination angles (e.g., 90 and 180 degrees because QPI images are taken along axes normal to each other) and a regularization parameter (e.g., 1×10-3), which is determined experimentally based on system calibration (described below).
The QPI microscope system 110 controls for the effect of drug solvents and phototoxicity using on-plate solvent controls during each experiment. In some embodiments, the on-plate solvent controls are matched to the highest concentrations of solvent used in the experiment (e.g., 0.125%). For example, as illustrated by
The QPI microscope 120 is calibrated prior to imaging, to validate the QPI microscope system 110 and to ensure the measurements are not impacted by misalignment. In some embodiments, a first calibration experiment validates centering of the LED array 150 by comparing a raw intensity from each half of the LED array 150 while imaging an empty sample of the well-plate 105. In some embodiments, after centering and aligning the LED array 150, the intensity from each half of the LED array 150 is equal to about 5% of the mean for all four half circles, as illustrated in a graph 1600 of
In some embodiments, the QPI microscope 120 segments the cells using a Sobel filter, via the electronic controller 155, to find cell edges, and morphological operators to create a mask. Single MDA-MB-231 cells can be further segmented using a watershed algorithm. In some embodiments, cells are masked, using the electronic controller 155, and an 8th order polynomial fit is removed from the background prior to averaging images, using the electronic controller 155, from each experiment to correct for aberrations and optical artifacts. A rolling ball filter, using, for example, a disk structuring element of 100 px, is applied by the electronic controller 155 to remove high spatial frequency noise. In some embodiments, the electronic controller 155 determines cell mass using a cell average specific refractive increment of 1.8×10-4m3/kg23. For example, the electronic controller 155 determines cell mass for each therapy treatment sample of the plurality of treatment samples based on each image of the plurality of images received from the camera 135. The QPI microscope 120 tracks, using the electronic controller 155, segmented cells from frame to frame of captured images based on approximate minimization of the distance between cell objects in successive frames in terms of cell mass and position in x and y directions. For example, the electronic controller 155 tracks cell mass over the period of time for each therapy treatment sample of the plurality of treatment samples.
Prior to operation of the QPI microscope 120, drugs (e.g., treatment therapies) are mixed with a volume of solvent to make a 20 mM stock solution. In some instances, stock solutions are stored in a −20° C. freezer and are not thawed more than five times to preserve the efficacy of the treatment therapies. A stock concentration of the treatment therapies is aliquoted into media to make a 40 μM solution, which is serially diluted on a 96-well plate (e.g., the well-plate 105) with 1 mL wells to make a solution that is double a concentration for the assay. In some instances, the diluted therapies are added to cells 3 h prior to the start of the assay at a 1:1 ratio of drugged media to cell media, to dilute the concentration of therapy to its final concentration.
In some instances, 1500 cells are plated in each well of two 96 well plates (e.g., two well-plates 105) with 100 μL of media to allow space for drugged media to be added. Cells may be incubated in cell culture conditions for 18 h prior to dosing. In some instances, 100 μL of diluted treatment therapies and solvent controls are added to the cells 3 h prior to the start of imaging. Cells may be allowed to incubate on the QPI microscope 120 in cell culture conditions (e.g., 37° C. and 5% CO2) for an hour prior to focusing the center of each well of the well-plate 105. In some embodiments, the electronic controller 155 selects nine imaging positions per well and each position is imaged every 20 minutes via the QPI microscope 120 with a single autofocus before each imaging cycle to account for thermal and z-stage drift. After 24 h of imaging a first well-plate 105, a first CTG assay is performed on a second well-plate 105 (e.g.,
In some embodiments, the electronic controller 155 determines a plurality of response parameters for each therapy treatment sample of the plurality of treatment samples based on the cell mass over the period of time. In some instances, the electronic controller 155 performs cell counting experiments by measuring proliferation throughout the duration of an experiment. For example, the electronic controller 155 counts cells in three different ethanol-treated wells of the well-plate 105 at 0 h, 18 h, 36 h, 48 h, and 72 h to measure the doubling time for MCF-7 and MDA-MB-231 cells. The following equation (Equation 1) may be used to compute the exponential growth constant for each cell line:
Such that tdoubling is the doubling time measured using cell counting for each cell line, ln(2) is the natural logarithm of 2, and SGR is the specific growth rate also known as the exponential growth constant.
In some instances, cell lines are acquired from the American Type Culture Collection (ATCC) and routinely screened for mycoplasma infection using, for example, the Agilent MycoSensor qPCR assay. In some embodiments, MCF-7 cells are cultured in Dulbecco's Modified Eagle Medium F12 supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 1% penicillin/streptomycin (Pen-Strep). In some embodiments, MDA-MB-231 cells are cultured in RPMI medium supplemented with 10% FBS and 1% Pen-Strep. In some embodiments, BT-474 cells are cultured in Hybri-Care Medium 46-X prepared with 18 MΩ deionized water supplemented with 1.3 mM of sodium bicarbonate, 10% heat-inactivated FBS, and 1% Pen-Strep. Cells may be passaged on 10 cm cell culture treated dishes at 37° C./5% CO2 and passaged by washing with Dulbecco's phosphate buffered saline and then incubating with Trypsin at 37° C. with 5% CO2 for 7 min before splitting at a 1:5 ratio.
CTG (e.g., read by the CTG reader 115) is a cell viability assay that quantifies the amount of ATP present at the end of an experiment as an indicator of the number of live cells. The electronic controller 155 measures ATP content in each condition as a readout by a luminescent signal, which is normalized by the luminescence of the control (e.g., a control well) to determine cell viability relative to on-plate controls. In some embodiments, the CTG assay is performed by first removing 100 μL of media from each well on the well-plate 105, which is then replaced with an equal amount of CellTiter-Glo reagent (e.g., Promega, G7572). Assayed plates may be shaken at 500 RPM for 20 minutes and allowed to rest for 10 minutes. In some instances, 100 μL of volume from each well is transferred to a white 96-well plate (e.g., Perkin Elmer, 6005680). The electronic controller 155 collects luminescence data from each well using an Envision plate reader (e.g., Perkin Elmer) and normalized against the solvent control to measure ATP content. The electronic controller 155 fits a 4-parameter hill curve to the dose response to determine the EC50 and depth of response.
The QPI microscope system 110 measures the correlation, using the electronic controller 155, between variables using the Pearson correlation coefficient as implemented in, for example, Matlab which tests the null hypothesis that there is no relationship between the variables. The QPI microscope system 110 may also determine, using the electronic controller 155, Lin's concordance coefficient to measure the concordance between variables. In some instances, the electronic controller 155 determines the confidence interval by bootstrapping based on resampling the observed data 10,000 times and reporting the confidence interval as the minimum and maximum of the middle 95% of these data.
The mixture of resistant and sensitive cells can be simulated by randomly selecting control cells to mix into the drug-treated population at mixing ratios of 50%, 25%, 10%, 5%, 2%, 1%, 0.1%, and 0.01%. The electronic controller 155 uses this mixture to identify resistant cells using threshold growth rates ranging from −0.1 h−1 to 0.15 h−1 and for each threshold, determines precision and recall of the identified resistant cells. The electronic controller 155 parameterizes the precision and recall measurements by threshold-specific growth rates and uses the area under the precision-recall curve (AUPRC). Generally, a higher AUPRC indicates a more effective classifier. To compare AUPRCs across mixing ratios, the electronic controller 155 compares AUPRCs to a no-skill line. In some instances, all AUPRCs reported are normalized by the area above the no skill line, via the electronic controller 155, which represents the best possible performance of the classifier at a specified mixing ratio. To reduce errors due to random sampling of the control population, especially at the lowest mixing ratios, mixing may be repeated about 100 times for each condition and the electronic controller 155 sorts the data based on a measured AUC. In some embodiments, the electronic controller 155 plots a median precision and recall value for each tested threshold and reports the median AUC.
In some examples, the electronic controller 155 median filters cell mass versus time data with a minimum length of 20 frames, with a kernel size of 5 frames, to remove small fluctuations. Data with a mean mass lower than 110 pg is removed from the analysis, as data with a mean mass lower than 110 pg may be debris. The electronic controller 155 time shifts each mass over time plot, such that the first mass measurement of each cell starts at t=0. The electronic controller 155 applies a linear regression to find the slope, which defines the growth rate (pg/h), and the y-intercept, which is used as the initial mass. The electronic controller 155 determines SGR by computing the growth rate divided by the initial mass. The electronic controller 155 uses the standard error of the estimate (sy.x) and normalized slope (specific growth rate; k) to remove outliers. An outlier may be defined as 3 median absolute deviation from the median.
In some embodiments, an overall mass of an imaging location is the summation of the mass of individual pixels after background correction. The electronic controller 155 collects overall mass for each location over time, median filtered, then normalized by the mass at the state of the imaging by the QPI microscope 120. In some embodiments, all locations in all wells for a given condition are averaged. In some instances, the electronic controller 155 determines the standard deviation for each triplicate of wells.
The electronic controller 155 fits average SGR data for individual treatments to both a response (Hill equation) and a no-response (flat line) model. The response model is a four-parameter logistic (Hill equation) function for fitting SGR versus concentration, C:
Such that E0 is asymptote at lowest concentration, Emax is asymptote at maximum concentration, EC50 is the inflection point of the hill curve, and HS is the Hill slope. The no-response model is a flat line parallel to the concentration axis. The residual variance from each fit is compared using an F-test with a p value of 0.01. For responding conditions (logistic fit better than no-response fit as determined by F-test), depth of response (DoR) is computed as:
To measure dynamic changes in SGR over time, the electronic controller 155 breaks up cell mass versus time tracks into overlapping 24 h intervals centered on each imaging time point in the experiment for a total of 144 total intervals. In some embodiments, the electronic controller 155 filters the tracks within each interval for a minimum path length of 20 frames within each interval, a minimum mean mass of 110 pg, and goodness of fit to a linear model, as described above. The electronic controller 155 determines the specific growth rate for each cell in each time interval by time shifting each track to start at t=0, and using a linear regression to find the rate of mass accumulation (e.g., based on cell mass and the period of time) and this slope was normalized by the y-intercept of each regression of the time-shifted data in the interval. In some instances, 8 adjacent intervals are binned together to produce 18 bins throughout the duration of the experiment. The electronic controller 155 determines the kernel density function in each bin using the mean SGR of each cell in the bin.
The electronic controller 155 measures the Hellinger distance between growth rate distributions by fitting a probability density function (PDF) to the distribution of raw growth rates of each cell or cluster normalized by its initial mass such that an integral over the PDF is equal to 1. The Hellinger distance may be defined as:
which is then discretized to:
Such that H is the computed Hellinger distance between probability distribution functions s and t. The electronic controller 155 determines the sum using a histogram bin size of 10−4 h−1 such that the bin size is small enough to capture differences in growth rate for BT-47 cells, the slowest growing cell line tested.
The electronic controller 155 determines the ToR by plotting the Hellinger distance between the drug-treated group and the control against time, and fitting the data to the following equation:
Such that a, b, and c are fit parameters to minimize the sum of least squared residuals. The electronic controller 155 fits Equation 6 to the Hellinger distance measured between the controls to find the Hellinger distance threshold as the maximum distance between controls for each cell line. The electronic controller 155 determines the ToR by fitting the noted model to the Hellinger distance versus time for each therapy and analytically solving for t:
Where H is equal to the threshold Hellinger distance, and a, b, and c are the fitting parameters to minimize the sum of least squared residuals.
In some embodiments, the electronic controller 155 determines a treatment response for each therapy treatment sample of the plurality of treatment samples based on the plurality of response parameters. In some embodiments, the QPI microscope system 110 performs QPI on cells from patient-derived organoids and direct-from-patient samples. For example, the electronic controller 155 analyzes data on cells from a set of 5 organoids representing samples collected from two patients, as illustrated by a workflow 3400 in
In some embodiments, data from a set of samples shows that the QPI microscope system 110 differentiates response based on site of origin, as shown by the graphs 3500 of
While all three PDxO models demonstrate a similar response profile to doxorubicin and docetaxel, the PDxO models did not demonstrate the same response across the entire set of therapies. For example, more progressive disease models (e.g., bone metastasis and pleural effusion) show a stronger response to vinblastine than expected based on data from the primary lesion. For birinapant, while the EC50 is similar for HCI-037 and HCI-039, HCI-038 had no response indicating this model exhibited resistant. Furthermore, even though the EC50 for HCI-037 and HCI-039 is similar, HCI-037 has a much greater depth of response and that HCI-039 has a much faster time of response indicating a differential growth response behavior between the different disease models. Together, the QPI microscope system 110 interrogates the differential drug response between different patient derived models from the same patient, which may be useful for detecting the emergence of resistance. In some embodiments, the QPI microscope system 110 applies multiparametric QPI to primary cells from two clinical samples stored in cell bank at, for example, the Huntsman Cancer Institute plated after thawing from liquid nitrogen and CD45+ depleted to reduce the number of immune cells, as shown in
For example, the primary cells are plated as a 2D monolayer in Clever's complete medium in a 96 well plate (e.g., the well-plate 105) and are allowed to acclimate to cell culture conditions for 18 h prior to dosing with treatments assigned to each patient by a physician. Even after CD45+ depletion, there are still small cells that maintain mass throughout the duration of imaging as shown by the image 3600 of
Without separating the two populations, the fitting code for EC50 and DoR fails to find a response. In some embodiments, both samples are expanded from primary cells in 2D culture using optimized ex vivo culture conditions for 2 weeks, changing media every 3 days to further purify the tumor cells, as shown in a workflow 3700 of
After short-term expansion, the sample is enriched for high mass tumor cells relative to the high content of low mass immune cells in the direct from thaw primary sample, as shown in the image 3705 of
All statements herein reciting principles, aspects, and embodiments of the disclosure, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents as well as equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure.
Various other components may be included and called upon for providing for aspects of the teachings herein. For example, additional materials, combinations of materials and/or omission of materials may be used to provide for added embodiments that are within the scope of the teachings herein. Adequacy of any particular element for practice of the teachings herein is to be judged from the perspective of a designer, manufacturer, seller, user, system operator or other similarly interested party, and such limitations are to be perceived according to the standards of the interested party.
In the disclosure hereof any element expressed as a means for performing a specified function is intended to encompass any way of performing that function including, for example, a) a combination of circuit elements and associated hardware which perform that function or b) software in any form, including, therefore, firmware, microcode or the like as set forth herein, combined with appropriate circuitry for executing that software to perform the function. Applicants thus regard any means which can provide those functionalities as equivalent to those shown herein. No functional language used in claims appended herein is to be construed as invoking 35 U.S.C. § 112 (f) interpretations as “means-plus-function” language unless specifically expressed as such by use of the words “means for” or “steps for” within the respective claim.
When introducing elements of the present invention or the embodiment(s) thereof, the articles “a,” “an,” and “the” are intended to mean that there are one or more of the elements. Similarly, the adjective “another,” when used to introduce an element, is intended to mean one or more elements. The terms “including” and “having” are intended to be inclusive such that there may be additional elements other than the listed elements. The term “exemplary” is not intended to be construed as a superlative example but merely one of many possible examples.
This application is a non-provisional of and claims the benefit of U.S. Provisional Application No. 63/517,976, filed on Aug. 7, 2023, the entire contents of which are incorporated herein by reference.
This invention was made with government support under W81XWH-19-1-0065 and W81XWH-19-1-0066 awarded by the USAMRAA. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 63517976 | Aug 2023 | US |
