Method for prediction of drug-induced arrhythmogenicity during exercise-related cardiac stress

Abstract

A system for testing medications for cardiac activity at elevated heart rates utilizes an array of electrodes connected to a computer and cardiomyocyte samples on the electrodes. Software implements an electrical stimulation protocol in a method of delivering an electrical signal at a baseline frequency to stimulate respective cardiomyocyte samples and calculating a functional refractory period (mISI) for the respective cardiomyocytes at the baseline frequency along with increased respectively applied frequencies of consecutive electrical signals.

Description

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[0001] This application claims priority to and incorporates entirely by reference United States Provisional Patent Application Serial No.63 / 688,596 filed on August 29, 2024. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[0002] This disclosure was made with government support under Grant No. R43HL140920 awarded by the National Institutes of Health. The Government has certain rights in the invention. ^^^^^^^^^^^

[0003] Drug-induced QT prolongation is a critical issue to examine when testing for cardiac safety and arrhythmias during the development of new drug compounds. Many drugs, including those functionally unrelated to cardiac treatments, block the human ether-á-go-go- related gene (hERG) channel, affecting the cardiac action potential repolarization leading to prolongation of the QT interval and arrhythmias [1-5]. While the hERG channel is a highly important channel to consider for drug cardiotoxicity, hERG channel testing can miss drug- induced QT prolongation due to interaction with other ion channels. Adenosine triphosphate (ATP) sensitive potassium (KATP) channels are regulated by intracellular ATP and adenosine diphosphate (ADP) and are typically closed, but open in response to myocardial stress to protect the heart [6]. As the heart rate increases, these ion channels are activated due to changes in the ATP / ADP ratio caused by increased cellular energy usage. The subsequent QT interval shortening serves to enable elevated heart rates without arrythmia, and blocking of these channels prevents this shortening thus leading to a relative QT interval prolongation. Dog telemetry models have been used for QT prolongation studies, especially drug-induced QT prolongation and have demonstrated they can serve as a preclinical predictor of QT prolongation for new human drug compounds. Chaves et al. demonstrated high sensitivity in beagle dog telemetry to detect small but significant increases in QT / QT corrected (QTc) interval with moxifloxacin, haloperidol and MK-499 [7]. Because the QT interval is affected by the heart rate, in both animal and human testing, the QT interval is corrected for the subject’s heart rate during the test (QTc)

[0114] . While a drug compound may pass cardiac safety testing for QT prolongation at a resting heart rate, there may be vastly different results ^for a patient under cardiac stress (e.g., physical exertion). Frequency-dependent (or heart-rate dependent) arrhythmic effects may go undetected during cardiotoxicity screenings. These arrhythmias may occur at elevated heart rates, leading to potential patient death. Lestuzzi et al. has reported on the incidence of cardiotoxicity in patients from capecitabine chemotherapy at rest and during physical exercise. It was observed that many patients were asymptomatic during treatment and that most of the cardiac events occurred during physical exercise and were detected only through an exercise stress test (EST) indicating that moderate physical activities may elicit cardiac arrhythmias [9]. At this time, there is no known in vitro model designed to screen for cardiotoxicity during cardiac stress. There is a need for a human in vitro stress test-on-a-chip for the ability to investigate drug-induced cardiotoxicity on cardiomyocytes not only at resting rate but also under cardiac exercise-related stress. ^^^^^^^^

[0004] In one embodiment, a system for measuring cardiac muscle function may include a computer having a processor and memory connected to a power source, wherein the memory stores software having computer implemented commands. The system utilizes an array of electrodes connected to the computer and cardiomyocyte samples on the electrodes. The software implements an electrical stimulation protocol in a method having steps of delivering an electrical signal at a baseline frequency from the power source to the electrodes to stimulate respective cardiomyocyte samples; measuring, with the computer, a repolarization time for the respective cardiomyocyte samples at the baseline frequency; using the repolarization time, calculating a functional refractory period (mISI) for the respective cardiomyocytes at the baseline frequency; iteratively increasing respectively applied frequencies of consecutive electrical signals stimulating the cardiomyocyte samples; tracking the mISI for the electrical signal at the baseline frequency; tracking the recorded mISI for the consecutive electrical signals at the respectively applied frequencies; and calculating a frequency-dependent mISI dataset for the baseline frequency.

[0005] In another embodiment, a system for measuring cardiac muscle function may include tests conducted in the presence of a compound being tested as a cardiac medication and again using a computer having a processor and memory connected to a power source, wherein the memory stores software having computer implemented commands. The system utilizes an array of electrodes connected to the computer and cardiomyocyte samples on the electrodes. The software implements an electrical stimulation protocol in a method having steps of delivering an electrical signal at a baseline frequency from the power source to the electrodes to stimulate respective cardiomyocyte samples; measuring, with the computer, a ^repolarization time for the respective cardiomyocyte samples at the baseline frequency; using the repolarization time, calculating a functional refractory period (mISI) for the respective cardiomyocytes at the baseline frequency; iteratively increasing respectively applied frequencies of consecutive electrical signals stimulating the cardiomyocyte samples; tracking the mISI for the electrical signal at the baseline frequency; tracking the recorded mISI for the consecutive electrical signals at the respectively applied frequencies; and calculating a frequency-dependent mISI dataset for the baseline frequency; applying a Holzegrefe correction to the recorded mISI for the consecutive electrical signals; calculating a non-drug related frequency-dependent mISI dataset of corrected mISI values (mISIc) for the baseline frequency; and calculating a frequency-dependent response curve, modeling the mISI for respective concentrations of the compound present at the cardiomyocyte samples, for the baseline frequency. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[0006] Reference will now be made to the accompanying drawings, which are not necessarily drawn to scale.

[0007] FIG.1 shows a flow chart of a computer-implemented method according to this disclosure.

[0008] FIG.2 shows a system drawing of an example computer environment that may be used to accomplish the methods and systems described herein.

[0009] FIG.3A illustrates a cardiac custom microelectrode array (cMEA) chip and cardiomyocyte patterning. FIG.3A includes an image of custom MEA device with ten recording electrodes spaced for cardiac electrical measurements.

[0010] FIG.3B illustrates a cardiac custom microelectrode array (cMEA) chip and cardiomyocyte patterning . FIG.3B illustrates a 2-row pattern of human induced pluripotent stem cell cardiomyocytes (hiPSC-CMs) across electrodes, patterned into 200 µm wide lines using patterned silane chemistry.

[0011] FIG.4A illustrates a method to determine field potential duration FPD (which correlates to a minimal inter-spike interval (mISI) at rest and at higher steady state base frequencies. FIG.4A shows example frequency pulse trains: as the frequency is increased, the FPD (mISI) decreases.

[0012] FIG.4B illustrates a single quick pulse at faster and faster frequencies that is sent during baseline pacing frequency to determine if the cell repolarizes. ^

[0013] FIG.5. illustrates pacing at baseline frequency and probing cells at higher frequencies to determine where the cells no longer repolarize and stop pacing. Measured mISI is from the higher frequency stimulation pulses just prior to where it was no longer pacing at the higher frequency stimulation pulses.

[0014] FIG.6A illustrates a status of the cardiomyocytes of this disclosure. FIG.6A is shown without Ivabradine treatment (Phosphate buffered saline (PBS) vehicle control) in which cells spontaneously beat within stimulation pulses at less than 1.25 Hz.

[0015] FIG.6B illustrates a status of the cardiomyocytes of this disclosure. FIG.6B illustrates a system with Ivabradine treatment that allowed fully stimulation-paced cardiomyocytes as low as 0.5 Hz.

[0016] FIG.7A illustrates a relative change in spontaneous beat frequency and mISI. FIG.7A shows a relative change in spontaneous beat frequency after treatment with 1 µM, 3 µM, 5 µM, 10 µM, 15 µM, and 30 µM compared to control at 2-, 4-, 6-, and 24-hours post- treatment with ivabradine.

[0017] FIG.7B also shows a relative change in mISI after treatment with 1 µM, 3 µM, 5 µM, 10 µM, 15 µM, and 30 µM compared to control at 2-, 4-, 6-, and 24-hours post- treatment with ivabradine.

[0018] FIG.8A shows a modeled dose response curve according to this disclosure with dose in micromoles (µM) on the x-axis and 1 / mISI (1 / s) on the y-axis. Terfenadine and E4031 showed a frequency-dependent response due to dose at 4 hours and 24 hours post- dose. Modeled dose response curves to 4 hours post-treatment to terfenadine are shown in FIG.8A. Red dashed lines of FIG.8A indicate that the model predicts that there will be cardiac problems 4 hours post-treatment with ~0.01 µM terfenadine, once cardiac frequency reaches 2 Hz.

[0019] FIG.8B shows a modeled dose response curve according to this disclosure with dose in micromoles (µM) on the x-axis and 1 / mISI (1 / s) on the y-axis.. Terfenadine and E4031 showed a frequency-dependent response due to dose at 4 hours and 24 hours post- dose. Modeled dose response curves to 24 hours post-treatment to terfenadine are shown in FIG.8B. Red dashed lines of FIG.8B indicate that the model predicts that there will be cardiac problems 24 hours post-treatment with ~0.03 µM terfenadine, once cardiac frequency reaches 1.75 Hz

[0020] FIG.8C shows a modeled dose response curve according to this disclosure with dose in micromoles (µM) on the x-axis and 1 / mISI (1 / s) on the y-axis.. Terfenadine and E4031 showed a frequency-dependent response due to dose at 4 hours and 24 hours post- ^dose. Modeled dose response curves to 4 hours post-treatment to E4031 are shown in FIG. 8C. Red dashed lines indicate that the model predicts that there will be cardiac problems 4 hours post-treatment with ~0.03 µM E4031 once cardiac frequency reaches 2 Hz.

[0021] FIG.8D shows a modeled dose response curve according to this disclosure with dose in micromoles (µM) on the x-axis and 1 / mISI (1 / s) on the y-axis. Terfenadine and E4031 showed a frequency-dependent response due to dose at 4 hours and 24 hours post- dose. Modeled dose response curves to 24 hours post-treatment to E4031 are shown in FIG. 8D. Red dashed lines of FIG.8D indicate that the model predicts that there will be cardiac problems 24 hours post-treatment with ~0.034 µM E4031, once cardiac frequency reaches 2 Hz.

[0022] FIG.8E shows a modeled dose response curve according to this disclosure with exposure to glibenclamide for 4 hours. FIG.8E shows a dose in micromoles (µM) on the x-axis and 1 / mISI (1 / s) on the y-axis.

[0023] FIG.8F shows a modeled dose response curve according to this disclosure with exposure to glibenclamide for 24 hours. FIG. 8F shows a dose in micromoles (µM) on the x-axis and 1 / mISI (1 / s) on the y-axis.

[0024]

[0025] FIG.8G shows a modeled dose response curve according to this disclosure with exposure to pinacidil for 4 hours. FIG.8G shows a dose in micromoles (µM) on the x- axis and 1 / mISI (1 / s) on the y-axis.

[0026] FIG.8H shows a modeled dose response curve according to this disclosure with exposure to pinacidil for 24 hours. FIG.8H shows a dose in micromoles (µM) on the x- axis and 1 / mISI (1 / s) on the y-axis.

[0027] FIG.8I shows a legend of values for the modeled dose response curves of FIGS.8A-8H according to this disclosure.

[0028] FIG.8J shows a legend of values for the modeled dose response curves of FIGS.8A-8H according to this disclosure.

[0029] FIG.9A shows modeled values on dose response curves and measurements for 1, 1.5, and 2 Hz frequencies. FIG.9A shows respective results for 4 hours post-treatment of terfenadine.

[0030] FIG.9B shows modeled dose response curves and measurements for 1, 1.5, and 2 Hz frequencies. FIG.9B shows respective results for 24 hours post-treatment of terfenadine. ^

[0031] FIG.9C shows modeled dose response curves and measurements for 1, 1.5, and 2 Hz frequencies. FIG.9C shows respective results for 4 hours post-treatment of E4031.

[0032] FIG.9D shows modeled dose response curves and measurements for 1, 1.5, and 2 Hz frequencies. FIG.9D shows respective results for 24 hours post-treatment E4031.

[0033] FIG.9E shows modeled dose response curves and measurements for 1, 1.5, and 2 Hz frequencies. FIG.9E shows respective results for 4 hours post-treatment of glibenclamide.

[0034] FIG.9F shows modeled dose response curves and measurements for 1, 1.5, and 2 Hz frequencies. FIG.9F shows respective results for 24 hours post-treatment glibenclamide.

[0035] FIG.9G shows modeled dose response curves and measurements for 1, 1.5, and 2 Hz frequencies. FIG.9G shows respective results for 4 hours post-treatment of pinacidil.

[0036] FIG.9H shows modeled dose response curves and measurements for 1, 1.5, and 2 Hz frequencies. FIG.9H shows respective results for 24 hours post-treatment of pinacidil.

[0037] FIG.9I shows a legend of values for the modeled dose response curves of FIGS.8A-8H according to this disclosure.

[0038] FIG.9J shows a legend of values for the modeled dose response curves of FIGS.8A-8H according to this disclosure.

[0039] FIG.10A shows average frequency-dependent responses across a respective tested batch with drug exposure. FIG.10A shows respective results for 4 hours post- treatment of terfenadine.

[0040] FIG.10B shows average frequency-dependent responses across a respective tested batch with drug exposure. FIG.10B shows respective results for 24 hours post- treatment of terfenadine.

[0041] FIG.10C shows average frequency-dependent responses across a respective tested batch with drug exposure. FIG.10C shows respective results for 4 hours post-treatment of E4031.

[0042] FIG.10D shows average frequency-dependent responses across a respective tested batch with drug exposure. FIG.10D shows respective results for 24 hours post- treatment E4031. ^

[0043] FIG.10E shows average frequency-dependent responses across a respective tested batch with drug exposure. FIG.10E shows respective results for 4 hours post-treatment of glibenclamide.

[0044] FIG.10F shows average frequency-dependent responses across a respective tested batch with drug exposure. FIG.10F shows respective results for 24 hours post- treatment glibenclamide.

[0045] FIG.10G shows average frequency-dependent responses across a respective tested batch with drug exposure. FIG.10G shows respective results for 4 hours post- treatment of pinacidil.

[0046] FIG.10H shows average frequency-dependent responses across a respective tested batch with drug exposure. FIG.10H shows respective results for 24 hours post- treatment of pinacidil.

[0047] FIG.11A shows modeled EC50 curves across a respective tested batch with drug exposure. FIG.11A shows respective results for 4 hours and 24 hours post-treatment of terfenadine with a frequency-dependent change in EC50 at 4 hours and 24 hours post treatment.

[0048] FIG.11B shows modeled EC50 curves across a respective tested batch with drug exposure. FIG.11B shows respective results for 4 hours and 24 hours post-treatment of E4031 with a frequency-dependent change in EC50 at 4 hours and 24 hours post treatment.

[0049] FIG.11C shows modeled EC50 curves across a respective tested batch with drug exposure. FIG.11C shows respective results for 4 hours and 24 hours post-treatment of glibenclamide. Glibenclamide did not show a frequency-dependent response in the EC50 at 4 hours and 24 hours post treatment.

[0050] FIG.11D shows modeled EC50 curves across a respective tested batch with drug exposure. FIG.11D shows respective results for 4 hours and 24 hours post-treatment of pinacidil. Pinacidil did not show a frequency-dependent response in the EC50 at 4 hours and 24 hours post treatment.

[0051] FIG.12A shows a summary of model predicted curves where the system is expected to start seeing cardiac problems (arrhythmias / death). Blue lines show the fitted model predicted cutoff. Red triangles indicate where tested cardiac systems stopped pacing. The pink colored areas represent where the model predicts the dose and heart rate (bpm) at which there will be cardiac arrhythmias for 4 hours post-treatment of terfenadine.

[0052] FIG.12B shows a summary of model predicted curves where the system is expected to start seeing cardiac problems (arrhythmias / death). Blue lines show the fitted ^model predicted cutoff. Red triangles indicate where tested cardiac systems stopped pacing. The pink colored areas represent where the model predicts the dose and heart rate (bpm) at which there will be cardiac arrhythmias for 24 hours post-treatment of terfenadine.

[0053] FIG.12C shows a summary of model predicted curves where the system is expected to start seeing cardiac problems (arrhythmias / death). Blue lines show the fitted model predicted cutoff. Red triangles indicate where tested cardiac systems stopped pacing. The pink colored areas represent where the model predicts the dose and heart rate (bpm) at which there will be cardiac arrhythmias for 4 hours post-treatment of E4031.

[0054] FIG.12D shows a summary of model predicted curves where the system is expected to start seeing cardiac problems (arrhythmias / death). Blue lines show the fitted model predicted cutoff. Red triangles indicate where tested cardiac systems stopped pacing. The pink colored areas represent where the model predicts the dose and heart rate (bpm) at which there will be cardiac arrhythmias for 24 hours post-treatment of E4031.

[0055] FIG.12E shows a summary of model predicted curves where the system is expected to start seeing cardiac problems (arrhythmias / death). Blue lines show the fitted model predicted cutoff. Red triangles indicate where tested cardiac systems stopped pacing. The pink colored areas represent where the model predicts the dose and heart rate (bpm) at which there will be cardiac arrhythmias for 4 hours post-treatment of glibenclamide.

[0056] FIG.12F shows a summary of model predicted curves where the system is expected to start seeing cardiac problems (arrhythmias / death). Blue lines show the fitted model predicted cutoff. Red triangles indicate where tested cardiac systems stopped pacing. The pink colored areas represent where the model predicts the dose and heart rate (bpm) at which there will be cardiac arrhythmias for 24 hours post-treatment of glibenclamide.

[0057] FIG.12G shows a summary of model predicted curves where the system is expected to start seeing cardiac problems (arrhythmias / death). Blue lines show the fitted model predicted cutoff. Red triangles indicate where tested cardiac systems stopped pacing. The pink colored areas represent where the model predicts the dose and heart rate (bpm) at which there will be cardiac arrhythmias for 4 hours post-treatment of pinacidil.

[0058] FIG.12H shows a summary of model predicted curves where the system is expected to start seeing cardiac problems (arrhythmias / death). Blue lines show the fitted model predicted cutoff. Red triangles indicate where tested cardiac systems stopped pacing. The pink colored areas represent where the model predicts the dose and heart rate (bpm) at which there will be cardiac arrhythmias for 24 hours post-treatment of pinacidil. ^^^^^^^^^^^^^^^^^^^^^^

[0059] Although example embodiments of the disclosed technology are explained in detail herein, it is to be understood that other embodiments are contemplated. Accordingly, it is not intended that the disclosed technology be limited in its scope to the details of construction and arrangement of components set forth in the following description or illustrated in the drawings. The disclosed technology is capable of other embodiments and of being practiced or carried out in various ways.

[0060] It must also be noted that, as used in the specification and the appended claims, the singular forms “a,” “an” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” or “approximately” one particular value and / or to “about” or “approximately” another particular value. When such a range is expressed, other exemplary embodiments include from the one particular value and / or to the other particular value.

[0061] By “comprising” or “containing” or “including” is meant that at least the named compound, element, particle, or method step is present in the composition or article or method, but does not exclude the presence of other compounds, materials, particles, method steps, even if the other such compounds, material, particles, method steps have the same function as what is named.

[0062] In describing example embodiments, terminology will be resorted to for the sake of clarity. It is intended that each term contemplates its broadest meaning as understood by those skilled in the art and includes all technical equivalents that operate in a similar manner to accomplish a similar purpose. It is also to be understood that the mention of one or more steps of a method does not preclude the presence of additional method steps or intervening method steps between those steps expressly identified. Steps of a method may be performed in a different order than those described herein without departing from the scope of the disclosed technology. Similarly, it is also to be understood that the mention of one or more components in a device or system does not preclude the presence of additional components or intervening components between those components expressly identified.

[0063] As discussed herein, a “subject” (or “patient”) may be any applicable human, animal, or other organism, living or dead, or other biological or molecular structure or chemical environment, and may relate to particular components of the subject, for instance specific organs, tissues, or fluids of a subject, may be in a particular location of the subject, referred to herein as an “area of interest” or a “region of interest.” ^

[0064] Some references, which may include various patents, patent applications, and publications, are cited in a reference list and discussed in the disclosure provided herein. The citation and / or discussion of such references is provided merely to clarify the description of the disclosed technology and is not an admission that any such reference is “prior art” to any aspects of the disclosed technology described herein. In terms of notation, “[n]” corresponds to the nth reference in the list. All references cited and discussed in this specification are incorporated herein by reference in their entireties and to the same extent as if each reference was individually incorporated by reference.

[0065] In the following description, references are made to the accompanying drawings that form a part hereof and that show, by way of illustration, specific embodiments or examples. In referring to the drawings, like numerals represent like elements throughout the several figures.

[0066] FIG.1 is a summary flow chart of a method according to this disclosure that may be implemented with a computer or a computer system shown in FIG. 2.

[0067] FIG.2 is a computer architecture diagram showing a general computing system capable of implementing aspects of the present disclosure in accordance with one or more embodiments described herein. A computer 200 may be configured to perform one or more functions associated with embodiments of this disclosure. For example, the computer 200 may be configured to perform operations of the method as described below. It should be appreciated that the computer 200 may be implemented within a single computing device or a computing system formed with multiple connected computing devices. The computer 200 may be configured to perform various distributed computing tasks, which may distribute processing and / or storage resources among the multiple devices. The data acquisition and display computer 150 and / or operator console 110 of the system shown in FIG.2 may include one or more systems and components of the computer 200.

[0068] As shown, the computer 200 includes a processing unit 202 (“CPU”), a system memory 204, and a system bus 206 that couples the memory 204 to the CPU 202. The computer 200 further includes a mass storage device 212 for storing program modules 214. The program modules 214 may be operable to perform one or more functions associated with embodiments of method as illustrated in one or more of the figures of this disclosure, for example to cause the computer 200 to perform operations of the automated DENSE analysis as described below. The program modules 214 may include an imaging application 218 for performing data acquisition functions as described herein, for example to receive image data corresponding to magnetic resonance imaging of an area of interest. The computer 200 can ^include a data store 220 for storing data that may include imaging-related data 222 such as acquired image data, and a modeling data store 224 for storing image modeling data, or other various types of data utilized in practicing aspects of the present disclosure.

[0069] The mass storage device 212 is connected to the CPU 202 through a mass storage controller (not shown) connected to the bus 206. The mass storage device 212 and its associated computer-storage media provide non-volatile storage for the computer 200. Although the description of computer-storage media contained herein refers to a mass storage device, such as a hard disk or CD-ROM drive, it should be appreciated by those skilled in the art that computer-storage media can be any available computer storage media that can be accessed by the computer 200.

[0070] By way of example, and not limitation, computer-storage media (also referred to herein as a “computer-readable storage medium” or “computer-readable storage media”) may include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-storage instructions, data structures, program modules, or other data. For example, computer storage media includes, but is not limited to, RAM, ROM, EPROM, EEPROM, flash memory or other solid state memory technology, CD-ROM, digital versatile disks (“DVD”), HD-DVD, BLU-RAY, or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by the computer 200. Transitory signals are not “computer-storage media”, “computer-readable storage medium” or “computer-readable storage media” as described herein.

[0071] According to various embodiments, the computer 200 may operate in a networked environment using connections to other local or remote computers through a network 216 via a network interface unit 210 connected to the bus 206. The network interface unit 210 may facilitate connection of the computing device inputs and outputs to one or more suitable networks and / or connections such as a local area network (LAN), a wide area network (WAN), the Internet, a cellular network, a radio frequency network, a Bluetooth-enabled network, a Wi-Fi enabled network, a satellite-based network, or other wired and / or wireless networks for communication with external devices and / or systems. The computer 200 may also include an input / output controller 208 for receiving and processing input from a number of input devices. Input devices may include one or more of keyboards, mice, stylus, touchscreens, microphones, audio capturing devices, or image / video capturing devices. An end user may utilize such input devices to interact with a user ^interface, for example a graphical user interface, for managing various functions performed by the computer 200.

[0072] The bus 206 may enable the processing unit 202 to read code and / or data to / from the mass storage device 212 or other computer-storage media. The computer-storage media may represent apparatus in the form of storage elements that are implemented using any suitable technology, including but not limited to semiconductors, magnetic materials, optics, or the like. The computer-storage media may represent memory components, whether characterized as RAM, ROM, flash, or other types of technology. The computer-storage media may also represent secondary storage, whether implemented as hard drives or otherwise. Hard drive implementations may be characterized as solid state or may include rotating media storing magnetically-encoded information. The program modules 214, which include the imaging application 218, may include instructions that, when loaded into the processing unit 202 and executed, cause the computer 200 to provide functions associated with embodiments illustrated herein. The program modules 214 may also provide various tools or techniques by which the computer 200 may participate within the overall systems or operating environments using the components, flows, and data structures discussed throughout this description.

[0073] In general, the program modules 214 may, when loaded into the processing unit 202 and executed, transform the processing unit 202 and the overall computer 200 from a general-purpose computing system into a special-purpose computing system. The processing unit 202 may be constructed from any number of transistors or other discrete circuit elements, which may individually or collectively assume any number of states. More specifically, the processing unit 202 may operate as a finite-state machine, in response to executable instructions contained within the program modules 214. These computer-executable instructions may transform the processing unit 202 by specifying how the processing unit 202 transitions between states, thereby transforming the transistors or other discrete hardware elements constituting the processing unit 202.

[0074] Encoding the program modules 214 may also transform the physical structure of the computer-storage media. The specific transformation of physical structure may depend on various factors, in different implementations of this description. Examples of such factors may include but are not limited to the technology used to implement the computer-storage media, whether the computer storage media are characterized as primary or secondary storage, and the like. For example, if the computer-storage media are implemented as semiconductor-based memory, the program modules 214 may transform the physical state of ^the semiconductor memory, when the software is encoded therein. For example, the program modules 214 may transform the state of transistors, capacitors, or other discrete circuit elements constituting the semiconductor memory.

[0075] As another example, the computer-storage media may be implemented using magnetic or optical technology. In such implementations, the program modules 214 may transform the physical state of magnetic or optical media, when the software is encoded therein. These transformations may include altering the magnetic characteristics of particular locations within given magnetic media. These transformations may also include altering the physical features or characteristics of particular locations within given optical media, to change the optical characteristics of those locations. Other transformations of physical media are possible without departing from the scope of the present description, with the foregoing examples provided only to facilitate this discussion.

[0076] The computing system can include clients and servers. A client and server are generally remote from each other and generally interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. In embodiments deploying a programmable computing system, it will be appreciated that both hardware and software architectures require consideration. Specifically, it will be appreciated that the choice of whether to implement certain functionality in permanently configured hardware (e.g., an ASIC), in temporarily configured hardware (e.g., a combination of software and a programmable processor), or a combination of permanently and temporarily configured hardware can be a design choice. Below are set out hardware (e.g., machine 300) and software architectures that can be deployed in example embodiments.

[0077] Microelectrode arrays (MEAs) (FIG.3, Ref.315) are highly valuable tools for body-on-a-chip systems for monitoring cardiac muscle function, particularly for drug toxicity and efficacy screening [17, 20, 116, 117]. This technology has been extended for use with human induced pluripotent stem cell (hiPSC)-derived cardiomyocytes (CMs) for drug screening to investigate cardiotoxicity and drug-induced QT prolongation, which can lead to the potentially fatal arrythmia, Torsades de Pointes (TdP). The clinical QT interval is determined from an electrocardiogram, whereas MEAs measure the field potential duration (FPD) or minimum interspike interval (mISI) of cardiomyocyte cultures, which are in vitro analogues for the QT interval. The QT interval is frequency-dependent which is driven by ion channels, especially cardiac potassium channels responding to a variation in heart rate

[0012] . Activation of these channels allows for a faster repolarization while inhibition of these ^channels allows for a slower repolarization. Many of the studies developed for the screening of cardiotoxicity and drug-induced QT prolongation have focused only on the effects of spontaneously beating cardiomyocytes (i.e., resting heart rate) when treated with drug compounds [13-15]. However, very few studies have investigated the effects of how these drug compounds may affect the cardiomyocytes when they are stressed by pacing at faster frequencies. Some progress has been made in investigating the drug-induced QT prolongation spontaneous response of hiPSC-CMs exposed to classic cardiac ion channel blockers using MEA technology. For instance, Kitaguchi et al. reported on the beat rate and FPD of hiPSC- CMs on MEAs treated with ion channel activators or inhibitors

[0015] . However, few reports have studied the response of hiPSC-CMs, under faster pacing conditions, to those same drug compounds. Hinata et al. investigated the relationship between contraction waveform parameters and cardiac cell tissue beating under pacing conditions of 1 – 3 Hz, however contractile response to typical ion channel blockers were only obtained during pacing at 1 Hz

[0016] . Zeng et al. investigated the effects of various pharmacological agents on hiPSC-CMs under E-pacing conditions

[0017] . Wei et al. studied the effects of electrical stimulation on hiPSC-CM responses to ion channel blockers and showed that electrical stimulation was able to stabilize the beat rate and FPD as well as allow for a more accurate assessment of rate- and concentration-dependent drug effects

[0018] .

[0078] This investigation was designed to develop a novel phenotypic assay system utilizing MEAs for functional measurements of cardiac muscle function, specifically during exercise-related stress. This system setup was developed to closely model in vivo cardiac function using hiPSC-CMs. Manipulation of the electrical properties of cardiomyocytes, by the application of an electrical stimulation protocol (FIG.1) to induce an exercise-related stress in the cells, allows for functional measurements of electrical signals to determine if dosing of a certain drug compound affects the frequency-dependent mISI of the cardiomyocytes. This stress test-on-a-chip system consists of hiPSC-CMs patterned on a custom-designed MEA in a serum-free medium. Cell patterning via excimer laser ablation lithography of silane monolayers was used to control the pathway for action potential signal propagation for measurement of refractory periods and for high information content. The cells were stressed using an electrical stimulation protocol that induced pacing at higher baseline frequencies while probing the repolarization time at that baseline frequency. This protocol was used to determine the refractory period (or mISI) across increasing frequencies. While measuring the FPD from the electrical signal can be performed, this mISI method ^measures the functional refractory period, indicating the period during which the cardiomyocytes cannot produce a subsequent heartbeat.

[0079] This work focuses on the frequency-dependent response of hiPSC-CMs dosed with drug compounds that inhibit or activate the hERG channel or frequency-dependent KATP channels, particularly investigating specific frequency ranges in which different drugs affect arrhythmogenicity. By monitoring the drug effects on cardiac tissue refractory period across a range of steady state stimulation frequencies (i.e., heart rates), the resting condition and stressed conditions were studied. Four drug compounds were used to validate this system: terfenadine, E4031, glibenclamide, and pinacidil. Terfenadine, a hERG channel blocker and a KATP channel blocker, was previously used as an antihistamine and has been withdrawn from the market due to drug-induced QT prolongation which can lead to a potentially fatal arrhythmia, Torsades de Pointes (TdP). E4031 is a class III anti-arrhythmic drug, a hERG channel blocker, and is known to cause QT prolongation [9-11]. Glibenclamide is a sulfonylurea drug widely used for treatment of type 2 diabetes mellitus and is a selective KATP channel blocker [5, 22-26]. Pinacidil is an antihypertensive drug that acts as a KATP channel activator which increases channel opening and may shorten the action potential [5, 24, 26-29]. This stress test-on-a-chip developed here aims to assess the effects of drug compounds on stressed cardiomyocytes with the potential to detect any arrhythmogenicity that may be undetected at regular resting heart rates. It has been designed so it can easily be integrated into a multi-organ body-on-a-chip system and has the potential to determine if a new drug compound will fail early in the development process.

[0080] Materials and Methods

[0081] Custom microelectrode array (cMEA) microfabrication

[0082] Custom microelectrode array (cMEA) chips (FIG.31a) were fabricated using standard microfabrication techniques using a previously described protocol

[0024] . The cMEA devices were fabricated by a contract vendor, IMT (IMT Masken Und Teilungen AG, Greifensee, Switzerland), according to the specified design. The design included two rows of five 80 µm diameter recording / stimulating electrodes, two 2000 µm diameter ground electrodes on one side of the stimulation / recording electrodes, and a single 2500 µm diameter ground electrode on the other side. The wires and electrodes, 10 nm titanium (Ti) / 50 nm platinum (Pt), were deposited on a fused silica substrate via electron-beam evaporation and patterned by a liftoff process. A three-stack insulation layer of silicon oxide, silicon nitride, and silicon oxide, 150 nm each layer, was deposited on top of the Ti / Pt wires via plasma- enhanced chemical vapor deposition (PECVD) and etched using reactive ion etching (RIE). ^

[0083] Surface modification and patterning

[0084] The cMEA chips were surface modified and patterned as previously published

[0023] . Briefly, the chips were coated with a poly(ethylene glycol) (PEG)-containing silane as a cytophobic surface. The surfaces were then patterned using a 193-nm ArF excimer laser (Lambda Physik, Santa Clara, CA, USA) through a quartz photomask for ablation of PEG coated areas to form a two-row pattern, each 200 µm wide, aligned over each row of the five electrodes. The two separate, patterned rows of cardiomyocytes allowed for electrical stimulation of two different electrodes, one per row, for more information per system.

[0085] Cell culture on custom MEAs

[0086] Surface modified and patterned cMEAs were sterilized in 70% isopropanol (VWR, cat#BDH1131) for 10 minutes. Once dried, a poly(dimethlysiloxane) (PDMS; Grace Bio-Labs, Bend, OR, USA) barrier was placed on top of the cMEA to enclose the area of the electrode arrays. Chips were incubated with 10 µg / ml human fibronectin (Millipore, Burlington, MA, USA) in phosphate buffered saline (PBS; Thermo Fisher Scientific, Waltham, MA, USA) for 30 min. at 37°C and 5% CO2. The fibronectin solution was aspirated, and chips were washed twice with PBS, after which cryogenically preserved, human induced pluripotent stem cell-derived cardiomyocytes (iCell Cardiomyocytes, Cellular Dynamics International, Madison, WI, USA) were plated onto the fibronectin-coated cMEAs (50,000 cells per MEA) in serum-free medium

[0030] . Chips were then placed in an incubator maintained at 37°C and 5% CO2, and the medium was changed 24 hours after plating and every two days after. The cMEA devices with cultured cardiomyocytes were maintained for 7 days prior to assembly into housings to enable recording. FIG.31b shows an image of the 2- row pattern of cardiomyocytes across the electrodes.

[0087] Cardiac system housing assembly

[0088] System housings and gaskets were manufactured from 6.35 mm thick transparent

[0089] poly(methyl methacrylate) (PMMA; McMaster-Carr, Elmhurst, IL, USA) and 0.5 mm thick PDMS, respectively. Both housings and gaskets were designed in Fusion 360 (Autodesk) software and laser cut with a Boss LS1420 laser cutter (Boss Laser, Sanford, FL, USA). The cardiac cMEA systems were assembled in the following order: PMMA bottom housing, PDMS bottom gasket, cardiac cultured cMEA, PDMS top gasket, and PMMA top housing. Screw thread insert taps were

[0090] used to create a thread inside a drilled hole in the housings and the system assembly was sealed with stainless steel screws (McMaster-Carr, Elmhurst, IL, USA). The ^gaskets sealed the contact pads on the device from the electrode area, which contained 700 µL of serum-free medium. Systems were incubated at 37°C and 5% CO2 overnight with testing beginning the following day.

[0091] Cardiac electrical measurement system setup and electrical stimulation protocol

[0092] For cardiac electrical measurements, a cardiac cMEA system was connected to a custom housing printed circuit board (PCB), via an elastomeric connector, and connected to a custom stimulation / amplifier chip and customized recording software. The testing setup enabled sixteen systems to be measured simultaneously and was performed in an incubator maintained at 37°C, 5% CO2, and 100% relative humidity. Because the cultures contain both atrial and ventricular cardiomyocytes, including pacemaker cells, a funny-channel blocker, ivabradine, was applied to reduce spontaneous beating for a higher level of stimulation- induced control. To determine the optimal dose of ivabradine to reduce spontaneous activity while inducing little to no effect on the mISI, a 30 second spontaneous recording in which no electrical stimulations were applied, and an mISI sweep electrical stimulation protocol were performed. This stimulation sweep protocol consisted of 10 pulses of each frequency, from 0.5 to 4 Hz, with increasing increments of 0.25 Hz. The single measured mISI from this protocol was determined as the smallest interval (highest pulse frequency) at which the cardiomyocytes beat in direct response to stimulation without missing any electrical stimulation-induced beats. In FIG.32a, the QT interval shortening with increased heart rate is shown schematically with frequency pulse trains of a 1 Hz, 1.5 Hz, and 2 Hz stimulation pulse, in which the steady state timing of the T-wave occurs with increasingly shorter period after the QRS complex with increasing frequency. When a consistent stimulation frequency produces a steady state QT interval, a stimulation pulse occurring at a timing earlier than the steady state frequency can probe the steady state refractory period, shown schematically in FIG.32b. When the single pulse occurs earlier than the refractory period is complete, the cardiomyocytes fail to respond to the probing pulse. The baseline frequency is used and applied to establish a steady state ATP / ADP ratio within the cardiomyocytes, which controls the frequency-dependent ion channels. The pulse occurring at a faster interpulse period is used to probe the refractory period for that specific steady state ATP / ADP ratio established by the baseline frequency. A frequency-dependent mISI electrical stimulation protocol was designed to send electrical signals to pace the cardiomyocytes so that they reached a metabolic steady state at various baseline frequencies (0.5 to 3 Hz, with 0.25 Hz increments). Throughout each baseline frequency, the cardiomyocytes were regularly stimulated with two ^increasingly higher frequency electrical pulses (1 to 5 Hz, with 0.1 Hz increments) to probe the cellular refractory period and measure the frequency-dependent mISI at each base frequency (FIG.33).

[0093] Frequency-dependent mISI model

[0094] The relationship between expected mISI and the stimulation frequency for use with testing and analysis was created based on relationships previously described [31, 32]. The relationship defined by Holzgrefe et al. for correcting the QT interval for heart rate (Equation 17) was used as the basis for the non-drug related frequency-dependence of cardiac culture repolarization.

[0095] QTc = QTraw / RRraw^RRref)^

[0096] In Equation 17, QTc is the heart-rate corrected QT, QTraw is the uncorrected QT interval (typically reported in ms), RRraw is the RR interval (typically reported in ms) associated with QTraw, RRref is the reference heart rate (species-specific RR interval achieved by all test subjects), and ^ is the individual species rate-correction coefficient

[0138] . The Holzgrefe relationship follows the Fridericia QT correction

[0033] , but allows for species- specific ^ and an arbitrary RRref to be used.

[0097] As the in vitro system does not measure QT interval, but instead enables measurement of spontaneous and electrically stimulated cardiac syncytium electrical activity – which are both related to the same underlying repolarization dynamics – Equation 17 was adapted for the output measurements of the in vitro system to produce Equation 18. In Equation 18, the QT interval values at a reference heart rate (QTc) and the QT interval at a given heart rate were replaced with the in vitro analogues mISIc (corrected mISI) and mISI(f) (the mISI at a given stimulation frequency, f), respectively. Because the in vitro system enables an adjustable stimulation frequency, while the clinical QT interval is limited to the heart rate the patient currently exhibits, the clinical RR measurement terms are replaced with the inverse of the stimulation frequencies (stimulated RR = 1 / f), where fref represents the reference stimulation frequency to which the mISI is corrected. For the stress test-on-a-chip in vitro system (or any stimulated in vitro system), the stimulation frequency-corrected mISI is given by Equation 18, assuming a Holzgrefe correction. ^^^^^^ mISIc = mISI(^) × (^^ !" #^$^^

[0099] In the stress test-on-a-chip, the focus is to determine changes to this relationship due to pharmaceutical compounds or other QT interval-modifying conditions, therefore Equation 18 was rearranged (Equation 19) for the determination of mISI(f). While ^the fref is arbitrary, the value of 1 Hz was used for consistency with the human clinical data. Because the system uses human iPSC-derived cardiomyocytes, the ^ determined clinically for humans (^ = 0.3)

[0138] was used.

[0100] To isolate the effects of compounds affecting the relationship between frequency and mISI in Equation 19, measurements were performed prior to dosing for paired analysis, and any experimental replicate that did not follow within ±10% mISI vs f relationship based on ^ = 0.3 were rejected and not included in dosing experimentation.

[0101] Ivabradine screening

[0102] Because spontaneous beating due to the small number of atrial cardiomyocytes in the culture would interfere with electrical pacing, the funny channel blocker, ivabradine, was applied. The concentration of ivabradine was determined such that a sufficient reduction of spontaneous beating was achieved to prevent interference with the pacing, while minimizing repolarization effects due to ivabradine. Ivabradine hydrochloride (Sigma- Aldrich, Inc., catalog #SML0281) was diluted in PBS to a stock concentration of 3 mM and then further diluted in cell culture medium. Cardiac cMEA systems were treated with 1, 3, 5, 10, 15, or 30 µM of ivabradine, or vehicle control, to determine an optimal concentration to reduce spontaneous activity without increasing mISI, prior to adding the compounds used for testing frequency-dependent mISI. For all experiments, spontaneous recordings and an mISI sweep electrical stimulation protocol were run at pre-dose baseline and 2-, 4-, 6-, and 24- hours after treatment.

[0103] Drug preparation and treatment

[0104] Prior to drug testing and recordings, cardiac cMEA systems were pre-treated with 3 µM ivabradine hydrochloride, based on identifying the optimal concentration of ivabradine, and were equilibrated in the incubator for 4 hours prior to the pre-dose baseline measurements. Stock solutions of terfenadine (Sigma-Aldrich, Inc., catalog #T9652-5G), glibenclamide (Sigma-Aldrich, Inc., catalog #PHR1287), and pinacidil monohydrate (Enzo Life Sciences, Inc., catalog #ALX-550-285-M050) were diluted in dimethyl sulfoxide (DMSO) and E4031 (MedChemExpress, catalog #HY-15551) was diluted in PBS. For drug experiments, cardiac cMEAs were dosed with terfenadine at 0.01, 0.03, 0.10.3 or 1 µM, E4031 at 0.01, 0.03, 0.1, 0.5, or 1 µM, glibenclamide at 1, 10, 30, 50, or 100 µM, or pinacidil monohydrate at 0.01, 0.1, 0.3, 1, 10, 30, 50, or 100 µM. In all drug experiments, control and drug-dosed systems were subjected to the stimulation protocol at baseline (pre-dose) and 4 hours and 24 hours after treatment.

[0105] Dose response and EC50 curve modeling ^

[0106] The frequency-dependent mISI response to compound doses were modeled based off a two-stage sigmoidal equation, in which the inverse of mISI follows a sigmoidal relationship with respect to concentration in the equation below with a half-maximal effective concentration (EC50) that itself follows a sigmoidal relationship with respect to frequency in the equation below.

[0107] Calculating a frequency dependent half maximum effective concentration EC50(f) for the compound to which the cardiomyocyte samples are exposed by a relationship expressed as:

[0108] EC50(f) = EC50_0 / (1 + (f / f50)h), wherein h is a Hill coefficient corresponding to a sigmoidal shape.

[0109] Calculating a frequency dependent mISI response to doses of the compound wherein the frequency dependent mISI response corresponds to another equation:

[0110] 1 / mISI(f) = (1 / mISI0(f)) / (1 + [A] / EC50(f)), wherein mISI0(f) corresponds to mISI at frequency (f) from the respective non-drug related, frequency- dependent mISI datasets and [A] corresponds to compound concentration.

[0111] In the equations above, the mISI at a given frequency and compound concentration [A] is related to the mISI at that same frequency when no drug is applied (mISI0(f)), and the EC50 at that frequency (EC50(f)). In the next equation, the EC50(f) is related to the EC50 when the pacing frequency is extrapolated to 0 Hz (EC50_0), the frequency at which the frequency effect is half of the maximum (f50), and a Hill coefficient describing the sigmoidal shape (h).

[0112] Data analysis and statistical analysis

[0113] Real-time measurements were recorded using a custom stimulation / amplifier chip and customized recording software. Raw voltage data was converted to pClamp, using a custom python script, to a format that was analyzed using Clampfit 10.7 software (Axon Instruments). All results are presented as the mean, and error bars are the standard error of the mean (SEM).

[0114] Results

[0115] Ivabradine screening

[0116] Due to the cells spontaneously beating often at a frequency greater than 1.25 Hz (FIG.34a), preliminary experiments using cardiac cMEA systems were performed in which the cultures were treated with ivabradine hydrochloride to reduce spontaneous activity of cardiomyocytes without substantially increasing the refractory period mISI. Ivabradine is a pacemaker current inhibitor (funny current, If), and reduces the heart rate by blocking the ion ^channels responsible for If

[0017] without affecting contractility. FIG.34b shows cells pacing at a frequency starting at 0.5 Hz after ivabradine treatment. A concentration of 3 µM ivabradine was selected for future co-dosing experiments based on results showing a decrease in spontaneous beat frequency near 0.5 Hz (FIG.35a), and with minimum effect on the refractory period mISI during an mISI sweep protocol (FIG.35b).

[0117] Frequency-dependent responses of hiPSC-CMs treated with terfenadine, E4031, glibenclamide, and pinacidil .

[0118] The dose response modeled curves of terfenadine, E4031, glibenclamide, and pinacidil at 4 hours and 24 hours post-treatment are shown in FIG.36. From these responses, the heart rate above which arrhythmias would be expected to occur at a given drug dose can be determined by correlating the modeled frequency curve to the 1 / mISI (Hz) value. Above the heart rate corresponding to the stimulation frequency at which the mISI(f) equals the time between stimulation pulses or heart beats (1 / f), cardiac arrhythmias are expected to occur. For example, in the figures, it is expected that at a heart rate of 120 beats per minute (i.e., frequency of 2 Hz), the cardiomyocytes in the heart would begin to have arrythmia at 4 hours after treatment with ~ 0.01 µM terfenadine. This disclosure shows that at a heart rate of 105 beats per minute (i.e., frequency of 1.75 Hz), the cardiomyocytes in the heart would begin to have arrythmia at 24 hours after treatment with ~ 0.03 µM terfenadine. This disclosure shows that at a heart rate of 120 beats per minute (i.e., frequency of 2 Hz), the cardiomyocytes in the heart would begin to have arrythmia at 4 hours after treatment with ~ 0.03 µM E4031This disclosure shows that at a heart rate of 120 beats per minute (i.e., frequency of 2 Hz), the cardiomyocytes in the heart would begin to have arrythmia at 24 hours after treatment with ~ 0.034 µM E4031. At higher heart rates, even lower doses would be expected to cause arrhythmia at both 4-hour and 24-hour time points for both terfenadine and E4031. Glibenclamide does not show a frequency-dependent response due to dose at 4 hours (FIG.36e) and 24 hours post-dose. Pinacidil does not show a frequency-dependent response due to dose at 4 hours (and 24 hours post-dose. In the cases of both glibenclamide and pinacidil treatment, results indicate that even at higher heart rates, there is not an increase in the chance of arrythmias due to increasing doses, until doses are so high that they are clinically irrelevant. The modeled dose responses in addition to the actual measured responses at 4 hours and 24 hours post-treatment with terfenadine, E4031, glibenclamide, and pinacidil, are shown in the figures respectively. The average frequency-dependent response for all systems tested at 4 hours and 24 hours post-treatment with terfenadine, E4031, glibenclamide, and pinacidil, respectively. ^

[0119] EC50 modeling

[0120] The model for frequency-dependent EC50 for each drug compound was fit to Equation 21 and shown in the figures. For terfenadine, the 4-hour post-treatment EC50 declines gradually as the frequency increases, while at 24 hours post-treatment, the EC50 is higher than that of the 4 hours and has a steep decline in the EC50 with respect to frequency, specifically beginning around 1.25 Hz. For E4031, the EC50 at lower frequencies is higher at 24 hours than at 4 hours post-treatment, and the EC50 decreases as the frequency increases for both 4 hours and 24 hours post-treatment. Unlike terfenadine and E4031, glibenclamide and pinacidil do not show a frequency-dependent response on the EC50.

[0121] Model predicts arrhythmogenicity at elevated heart rates

[0122] The results of the model are summarized in the figures, such that the predicted heart rate at which arrhythmia will occur across compound doses is represented. The shaded areas above the model predicted curve represent the dose and heart rate combinations at which arrhythmias and / or death are expected. With terfenadine treatment, 4 hours post- dosing with 0.1 µM terfenadine, the model predicts that arrhythmogenic effects will start to occur as the heart rate reaches ~ 80 bpm. As the dose increases, the model predicts that arrhythmias will occur at lower heart rates. At 24 hours post-treatment, as the dose of terfenadine increases, the model predicts that arrhythmogenic effects will occur) at lower heart rates as the dosing concentration is increased. The case is similar for E4031, in which the model predicts as the dose increases, arrhythmias will occur at lower heart rates for both 4 hours and 24 hours post-treatment. It is important to note that the model still predicts arrhythmogenic effects at elevated heart rates 24 hours post-treatment. At 4 hours post- treatment with glibenclamide, the model predicts arrhythmogenic effects at heart rates higher than 200 bpm, with increasing concentrations, until a high dose of ~ 30 µM is reached and effects start occurring at lower heart rates. At 24 hours post-treatment with glibenclamide, the model predicts arrhythmogenic effects at heart rates higher than 230 bpm, with increasing concentrations, until a high dose of ~ 50 µM is reached and effects start occurring at lower heart rates. At 4 hours post-treatment with pinacidil, the model predicts arrhythmogenic effects at heart rates higher than 180 bpm, with increasing concentrations, until a high dose of ~ 160 µM is reached and effects start occurring at lower heart rates. At 24 hours post- treatment with pinacidil, the model predicts arrhythmogenic effects at heart rates higher than 180 bpm, with increasing concentrations, until a high dose of ~ 75 µM is reached and effects start occurring at lower heart rates. ^

[0123] Few studies have been reported on the cardiotoxicity effects of drug compounds to patients during physical activity. Even fewer in vitro studies have been reported investigating the cardiac response of drug compounds to cardiomyocytes under exercise stress. This in vitro stress test-on-a-chip system enables a high level of control of the simulated heart rate through electrical stimulation of the cardiomyocytes for measuring drug arrhythmogenic effects at elevated heart rates. With this unique capability, this stress test-on- a-chip will serve as a useful tool in the prediction of potential arrhythmic activity of drug candidates using this in vitro system prior to clinical trials. Not only does this system enable screening prior to clinical trials, it also provides data that is not practically obtainable with human, or even animal, subjects. With the ability to test on hiPSC-CMs prior to human trials, this further extends the opportunities for cardiotoxicity screenings to help move new drugs to market faster as well as weed out those that will have safety issues during cardiac stress or physical exertion. This testing may indicate potential problems and may play a significant role in preventing asymptomatic arrhythmias. The differential effects obtained among drugs used in this study demonstrates that resting heart rate (or either spontaneous or a single stimulation frequency) testing is insufficient to fully capture potential arrhythmogenic safety concerns. Specifically, terfenadine and E4031 showed frequency-dependent effects, with predicted arrhythmogenic effects at higher resting heart rates. Glibenclamide and pinacidil did not show frequency-dependent effects for most dosing concentrations, acting as a negative exercise stress test.

[0124] A system for measuring cardiac muscle function includes a computer having a processor and memory connected to a power source, wherein the memory stores software having computer implemented commands. An array of electrodes (318) is connected to the computer, and cardiomyocyte samples (320) are positioned on the electrodes (100), wherein the software implements an electrical stimulation protocol in a method that may include steps of

[0125] delivering an electrical signal at a baseline frequency (130, 400)) from the power source (305) to the electrodes310) to stimulate respective cardiomyocyte samples;

[0126] measuring, with the computer, a repolarization time for the respective cardiomyocyte samples at the baseline frequency;

[0127] using the repolarization time, calculating a functional refractory period (mISI) for the respective cardiomyocytes at the baseline frequency;

[0128] iteratively increasing respectively applied frequencies (140, 405, 500) of consecutive electrical signals stimulating the cardiomyocyte samples; ^

[0129] tracking the mISI (150) for the electrical signal at the baseline frequency;

[0130] tracking the recorded mISI for the consecutive electrical signals at the respectively applied frequencies; and

[0131] calculating a frequency-dependent mISI dataset (160) for the baseline frequency.

[0132] The consecutive electrical signals are electrical pulses added to the electrical signal at the baseline frequency. The method continues by iteratively changing the baseline frequency and calculating a respective frequency-dependent mISI data set for respective baseline frequencies. The method includes stimulating the cardiomyocyte samples to a metabolic steady-state at the baseline frequency. The baseline frequency corresponds to a heartbeat frequency of a human in a resting state and the applied frequencies correspond to elevated heart rates of the human. Iteratively increasing respectively applied frequencies of consecutive electrical signals includes adding a series of pulses of increasingly higher frequencies to the baseline frequency. The recorded mISI for the consecutive electrical signals is statistically corrected relative to the baseline frequency. Correcting the recorded mISI may include applying a Holzegrefe correction to the recorded mISI for the consecutive electrical signals. The method may further include calculating a frequency-dependent mISI model based on the relationship between an expected mISI and stimulation frequencies.

[0133] In another embodiment, a system for measuring cardiac muscle function may include tests conducted in the presence of a compound being tested as a cardiac medication and again using a computer having a processor and memory connected to a power source, wherein the memory stores software having computer implemented commands. The system utilizes an array of electrodes connected to the computer and cardiomyocyte samples on the electrodes. The software implements an electrical stimulation protocol in a method having steps of delivering an electrical signal at a baseline frequency from the power source to the electrodes to stimulate respective cardiomyocyte samples; measuring, with the computer, a repolarization time for the respective cardiomyocyte samples at the baseline frequency; using the repolarization time, calculating a functional refractory period (mISI) for the respective cardiomyocytes at the baseline frequency; iteratively increasing respectively applied frequencies of consecutive electrical signals stimulating the cardiomyocyte samples; tracking the mISI for the electrical signal at the baseline frequency; tracking the recorded mISI for the consecutive electrical signals at the respectively applied frequencies; and calculating a frequency-dependent mISI dataset for the baseline frequency; applying a Holzegrefe correction to the recorded mISI for the consecutive electrical signals; calculating a non-drug ^related frequency-dependent mISI dataset of corrected mISI values (mISIc) for the baseline frequency; and calculating a frequency-dependent response curve, modeling the mISI for respective concentrations of the compound present at the cardiomyocyte samples, for the baseline frequency. The Holzegrefe correction may use an equation:

[0134] mISI(f) = mISIc x (fref / f)ß, wherein mISI(f) is the mISI at a given stimulation frequency, frefis the reference stimulation frequency to which the mISI is corrected, f is a frequency variable at which the cardiomyocyte is stimulated, and ß is the individual species rate-correction coefficient. The method may continue by iteratively changing the baseline frequency and calculating a respective non-drug related, frequency- dependent mISI dataset for respective baseline frequencies. The method may continue by modeling the respective non-drug related, frequency-dependent mISI datasets as graphical curves for each baseline frequency, wherein the graphical curves comprise corrected mISI values as a function of the respectively increasing applied frequencies. The method may continue for each baseline frequency of the respective non-drug related, frequency-dependent mISI datasets, identifying a half maximal value of the corrected mISI and an associated half maximal frequency (f50) from the graphical curves for the respective non-drug related, frequency-dependent mISI datasets for each baseline frequency. By further comprising extrapolating the graphical curves, for the respective non-drug related, frequency-dependent mISI datasets, to a frequency value of zero (0) they system may include identifying the half maximum value (EC50_0) of the corrected mISI values.

[0135] Calculating a frequency dependent half maximum effective concentration EC50(f) for the compound to which the cardiomyocyte samples are exposed by a relationship may be expressed as:

[0136] EC50(f) = EC50_0 / (1 + (f / f50)h), wherein h is a Hill coefficient corresponding to a sigmoidal shape.

[0137] The method may further include calculating a frequency dependent mISI response to doses of the compound wherein the frequency dependent mISI response corresponds to another equation:

[0138] 1 / mISI(f) = (1 / mISI0(f)) / (1 + [A] / EC50(f)), wherein mISI0(f) corresponds to mISI at frequency (f) from the respective non-drug related, frequency- dependent mISI datasets and [A] corresponds to compound concentration.

[0139] The methods and systems of this disclosure may include compiling final frequency dependent response curves for doses of the compound at respective concentrations ^for each baseline frequency wherein the respectively increasing applied frequencies have corresponding mISI values.

[0140] In conclusion, this disclosure successfully demonstrated the design and validity of a stress test-on-a-chip using human iPSC-derived cardiomyocytes patterned on an MEA device. This system predicted human cardiotoxicity after compound treatment with cardiomyocytes under exercise-related stress through electrical stimulation and pacing, showing that this system serves as a novel platform for preclinical testing. The unique capabilities of this system offer the potential to detect arrhythmogenicity that may go undetected at regular resting heart rates or using other preclinical methods. Ultimately, this system has the potential to be utilized routinely for cardiotoxicity screening to help bring drugs to market faster and offer new techniques in personalized medicine

[0141] References

[0142] 1. Witchel, H.J., Drug^induced hERG block and long QT syndrome. Cardiovascular therapeutics, 2011.29(4): p.251-259.

[0143] 2. Priest, B., I.M. Bell and M. Garcia, Role of hERG potassium channel assays in drug development. Channels, 2008.2(2): p.87-93.

[0144] 3. Garrido, A., A. Lepailleur, S.M. Mignani, P. Dallemagne and C. Rochais, hERG toxicity assessment: Useful guidelines for drug design. European journal of medicinal chemistry, 2020.195: p.112290.

[0145] 4. Nachimuthu, S., M.D. Assar and J.M. Schussler, Drug-induced QT interval prolongation: mechanisms and clinical management. Therapeutic Advances in Drug Safety, 2012.3(5): p. 241-253.

[0146] 5. Tamargo, J., R. Caballero, R. Gómez, C. Valenzuela and E. Delpón, Pharmacology of cardiac potassium channels. Cardiovascular research, 2004.62(1): p.9-33.

[0147] 6. Nichols, C.G., G.K. Singh and D.K. Grange, KATP channels and cardiovascular disease: suddenly a syndrome. Circulation research, 2013.112(7): p.1059- 1072.

[0148] 7. Chaves, A., G. Zingaro, M. Yordy, K. Bustard, S. O'sullivan, A. Galijatovic-Idrizbegovic, H. Schuck, D. Christian, C. Hoe and R. Briscoe, A highly sensitive canine telemetry model for detection of QT interval prolongation: studies with moxifloxacin, haloperidol and MK-499. Journal of Pharmacological and Toxicological Methods, 2007. 56(2): p.103-114. ^

[0149] 8. Ambhore, A., S.-G. Teo, A.R.B. Omar and K.-K. Poh, ECG series. Importance of QT interval in clinical practice. Singapore medical journal, 2014.55(12): p. 607.

[0150] 9. Lestuzzi, C., D. Stolfo, A. De Paoli, A. Banzato, A. Buonadonna, E. Bidoli, L. Tartuferi, E. Viel, G. De Angelis and S. Lonardi, Cardiotoxicity from capecitabine chemotherapy: prospective study of incidence at rest and during physical exercise. The oncologist, 2022.27(2): p. e158-e167.

[0151] 10. Sung, J.H., Y.I. Wang, N. Narasimhan Sriram, M. Jackson, C. Long, J.J. Hickman and M. L. Shuler, Recent advances in body-on-a-chip systems. Analytical chemistry, 2018.91(1): p.330-351.

[0152] 11. de Mello, C.P.P., C. Carmona-Moran, C.W. McAleer, J. Perez, E.A. Coln, C.J. Long, C. Oleaga, A. Riu, R. Note, S. Teissier, J. Langer and J.J. Hickman, Microphysiological heart–liver body-on-a-chip system with a skin mimic for evaluating topical drug delivery. Lab on a Chip, 2020.20(4): p.749-759.

[0153] 12. Grant, A.O., Cardiac ion channels. Circulation: Arrhythmia and Electrophysiology, 2009.2(2): p.185-194.

[0154] 13. Gintant, G., E.P. Kaushik, T. Feaster, S. Stoelzle-Feix, Y. Kanda, T. Osada, G. Smith, K. Czysz, R. Kettenhofen and H.R. Lu, Repolarization studies using human stem cell-derived cardiomyocytes: Validation studies and best practice recommendations. Regulatory toxicology and pharmacology, 2020.117: p.104756.

[0155] 14. Zwartsen, A., T. de Korte, P. Nacken, D.W. de Lange, R.H. Westerink and L. Hondebrink, Cardiotoxicity screening of illicit drugs and new psychoactive substances (NPS) in human iPSC-derived cardiomyocytes using microelectrode array (MEA) recordings. Journal of molecular and cellular cardiology, 2019.136: p.102-112.

[0156] 15. Kitaguchi, T., Y. Moriyama, T. Taniguchi, S. Maeda, H. Ando, T. Uda, K. Otabe, M. Oguchi, S. Shimizu and H. Saito, CSAHi study: detection of drug-induced ion channel / receptor responses, QT prolongation, and arrhythmia using multi-electrode arrays in combination with human induced pluripotent stem cell-derived cardiomyocytes. Journal of pharmacological and toxicological methods, 2017.85: p.73-81.

[0157] 16. Hinata, Y., Y. Kagawa, H. Kubo, E. Kato, A. Baba, D. Sasaki, K. Matsuura, K. Sawada and T. Shimizu, Importance of beating rate control for the analysis of drug effects on contractility in human induced pluripotent stem cell-derived cardiomyocytes. Journal of Pharmacological and Toxicological Methods, 2022.118: p.107228. ^

[0158] 17. Zeng, H., B. Balasubramanian, A. Lagrutta and F. Sannajust, Response of human induced pluripotent stem cell-derived cardiomyocytes to several pharmacological agents when intrinsic syncytial pacing is overcome by acute external stimulation. Journal of Pharmacological and Toxicological Methods, 2018.91: p.18-26.

[0159] 18. Wei, F., M. Pourrier, D.G. Strauss, N. Stockbridge and L. Pang, Effects of Electrical Stimulation on hiPSC-CM Responses to Classic Ion Channel Blockers. Toxicol Sci, 2020.174(2): p.254-265.

[0160] 19. Meyer, T., K.-H. Boven, E. Günther and M. Fejtl, Micro-electrode arrays in cardiac safety pharmacology: a novel tool to study QT interval prolongation. Drug safety, 2004.27: p.763-772.

[0161] 20. Tadano, K., S. Miyagawa, M. Takeda, Y. Tsukamoto, K. Kazusa, K. Takamatsu, M. Akashi and Y. Sawa, Cardiotoxicity assessment using 3D vascularized cardiac tissue consisting of human iPSC-derived cardiomyocytes and fibroblasts. Molecular Therapy-Methods & Clinical Development, 2021. 22: p.338-349.

[0162] 21. Yamamoto, W., K. Asakura, H. Ando, T. Taniguchi, A. Ojima, T. Uda, T. Osada, S. Hayashi, C. Kasai and N. Miyamoto, Electrophysiological characteristics of human iPSC-derived cardiomyocytes for the assessment of drug-induced proarrhythmic potential. PloS one, 2016.11(12): p. e0167348.

[0163] 22. Tomai, F., F. Crea, A. Gaspardone, F. Versaci, R. De Paulis, A. Penta de Peppo, L. Chiariello and P.A. Gioffrè, Ischemic preconditioning during coronary angioplasty is prevented by glibenclamide, a selective ATP-sensitive K+ channel blocker. Circulation, 1994.90(2): p.700-705.

[0164] 23. Negroni, J., E. Lascano and H. Del Valle, Glibenclamide action on myocardial function and arrhythmia incidence in the healthy and diabetic heart. Cardiovascular & Hematological Agents in Medicinal Chemistry (Formerly Current Medicinal Chemistry-Cardiovascular & Hematological Agents), 2007.5(1): p.43-53.

[0165] 24. Seino, S. and T. Miki, Physiological and pathophysiological roles of ATP-sensitive K+ channels. Progress in biophysics and molecular biology, 2003.81(2): p. 133-176.

[0166] 25. Okai, Y., K. Matsune, K. Yamanaka, T. Matsui, E.P. Kaushik, K. Harada, H. Kohara, A. Miyawaki, H. Ozaki and M. Wagoner, Video-based assessment of drug-induced effects on contractile motion properties using human induced pluripotent stem cell-derived cardiomyocytes. Journal of Pharmacological and Toxicological Methods, 2020. 105: p.106893. ^

[0167] 26. Li, G.-R. and M.-Q. Dong, Pharmacology of cardiac potassium channels. Advances in pharmacology, 2010.59: p.93-134.

[0168] 27. Friedel, H.A. and R.N. Brogden, Pinacidil: a review of its pharmacodynamic and pharmacokinetic properties, and therapeutic potential in the treatment of hypertension. Drugs, 1990.39(6): p.929-967.

[0169] 28. Itzhaki, I., L. Maizels, I. Huber, L. Zwi-Dantsis, O. Caspi, A. Winterstern, O. Feldman, A. Gepstein, G. Arbel and H. Hammerman, Modelling the long QT syndrome with induced pluripotent stem cells. Nature, 2011.471(7337): p.225-229.

[0170] 29. Jahangir, A. and A. Terzic, KATP channel therapeutics at the bedside. Journal of molecular and cellular cardiology, 2005.39(1): p.99-112.

[0171] 30. Oleaga, C., C. Bernabini, A.S. Smith, B. Srinivasan, M. Jackson, W. McLamb, V. Platt, R. Bridges, Y. Cai and N. Santhanam, Multi-organ toxicity demonstration in a functional human in vitro system composed of four organs. Scientific reports, 2016.6(1): p.1-17.

[0172] 31. Rickards, A. and J. Norman, Relation between QT interval and heart rate. New design of physiologically adaptive cardiac pacemaker. Heart, 1981.45(1): p.56-61.

[0173] 32. Holzgrefe, H., G. Ferber, P. Champeroux, M. Gill, M. Honda, A. Greiter-Wilke, T. Baird, O. Meyer and M. Saulnier, Preclinical QT safety assessment: cross- species comparisons and human translation from an industry consortium. Journal of Pharmacological and Toxicological Methods, 2014.69(1): p.61-101.

[0174] 33. Fridericia, L., Die Systolendauer im Elektrokardiogramm bei normalen Menschen und bei Herzkranken. Acta Medica Scandinavica, 1921.54(1): p.17-50.

[0175] 34. Wei, F., M. Pourrier, D.G. Strauss, N. Stockbridge and L. Pang, Effects of electrical stimulation on hiPSC-CM responses to classic ion channel blockers. Toxicological Sciences, 2020.174(2): p.254-265.

[0176] 35. Zeng, H., J. Wang, H. Clouse, A. Lagrutta and F. Sannajust, Resolving the reversed rate effect of calcium channel blockers on human-induced pluripotent stem cell- derived cardiomyocytes and the impact on in vitro cardiac safety evaluation. Toxicological Sciences, 2019.167(2): p.573-580.

[0177] 36. Lu, H., M. Hortigon-Vinagre, V. Zamora, I. Kopljar, A. De Bondt, D. Gallacher and G. Smith, Application of optical action potentials in human induced pluripotent stem cells-derived cardiomyocytes to predict drug-induced cardiac arrhythmias. Journal of pharmacological and toxicological methods, 2017.87: p.53-67. ^

[0178] 37. Chakraborty, P., R.A. Rose, K. Nair, E. Downar and K. Nanthakumar, The rationale for repurposing funny current inhibition for management of ventricular arrhythmia. Heart Rhythm, 2021.18(1): p.130-137.

[0179] 38. Müller-Werdan, U., G. Stöckl and K. Werdan, Advances in the management of heart failure: the role of ivabradine. Vascular Health and Risk Management, 2016.12: p.453.

[0180] 39. Chen, C., G. Kaur, P.K. Mehta, D. Morrone, L.C. Godoy, S. Bangalore and M.S. Sidhu, Ivabradine in cardiovascular disease management revisited: a review. Cardiovascular Drugs and Therapy, 2021.35(5): p.1045-1056.^^

Claims

CLAIMS 1. A system for measuring cardiac muscle function, comprising: a computer comprising a processor and memory connected to a power source, wherein the memory stores software having computer implemented commands; an array of electrodes connected to the computer; cardiomyocyte samples on the electrodes, wherein the software implements an electrical stimulation protocol in a method comprising: delivering an electrical signal at a baseline frequency from the power source to the electrodes to stimulate respective cardiomyocyte samples; measuring, with the computer, a repolarization time for the respective cardiomyocyte samples at the baseline frequency; using the repolarization time, calculating a functional refractory period (mISI) for the respective cardiomyocytes at the baseline frequency; iteratively increasing respectively applied frequencies of consecutive electrical signals stimulating the cardiomyocyte samples; tracking the mISI for the electrical signal at the baseline frequency; tracking the recorded mISI for the consecutive electrical signals at the respectively applied frequencies; and calculating a frequency-dependent mISI dataset for the baseline frequency.

2. The system of Claim 1, wherein the consecutive electrical signals are electrical pulses added to the electrical signal at the baseline frequency.

3. The system of Claim 2, wherein the method continues by iteratively changing the baseline frequency and calculating a respective frequency-dependent mISI data set for respective baseline frequencies.

4. The system of Claim 1, further comprising stimulating the cardiomyocyte samples to a metabolic steady-state at the baseline frequency. ^5. The system of Claim 1, wherein the baseline frequency corresponds to a heartbeat frequency of a human in a resting state and the applied frequencies correspond to elevated heart rates of the human.

6. The system of Claim 1, wherein iteratively increasing respectively applied frequencies of consecutive electrical signals comprises adding a series of pulses of increasingly higher frequencies to the baseline frequency.

7. The system of Claim 1, wherein the recorded mISI for the consecutive electrical signals is statistically corrected relative to the baseline frequency.

8. The system of Claim 7, wherein correcting the recorded mISI comprises applying a Holzegrefe correction to the recorded mISI for the consecutive electrical signals.

9. The system of Claim 7, further comprising calculating a frequency-dependent mISI model based on the relationship between an expected mISI and stimulation frequencies.

10. A system for predicting cardiac muscle function in the presence of a compound being tested as a cardiac medication, comprising: a computer comprising a processor and memory connected to a power source, wherein the memory stores software having computer implemented commands; an array of electrodes connected to the computer; cardiomyocyte samples on the electrodes, wherein the software implements an electrical stimulation protocol in a method comprising: delivering an electrical signal at a baseline frequency from the power source to the electrodes to stimulate respective cardiomyocyte samples; measuring, with the computer, a repolarization time for the respective cardiomyocyte samples at the baseline frequency; using the repolarization time, calculating a functional refractory period (mISI) for the respective cardiomyocites at the baseline frequency; iteratively applying consecutive electrical signals having respectively increasing applied frequencies and stimulating the cardiomyocyte samples; tracking the mISI for the electrical signal at the baseline frequency; ^tracking a recorded mISI for the consecutive electrical signals at the respectively increasing applied frequencies; applying a Holzegrefe correction to the recorded mISI for the consecutive electrical signals; calculating a non-drug related frequency-dependent mISI dataset of corrected mISI values (mISIc) for the baseline frequency; and calculating a frequency-dependent response curve, modeling the mISI for respective concentrations of the compound present at the cardiomyocyte samples, for the baseline frequency.

11. The system of Claim 10, wherein the Holzegrefe correction comprises an equation: mISI(f) = mISIc x (fref / f)ß, wherein mISI(f) is the mISI at a given stimulation frequency, fref is the reference stimulation frequency to which the mISI is corrected, f is a frequency variable at which the cardiomyocyte is stimulated, and ß is the individual species rate-correction coefficient.

12. The system of Claim 10, wherein the method continues by iteratively changing the baseline frequency and calculating a respective non-drug related, frequency-dependent mISI dataset for respective baseline frequencies.

13. The system of Claim 12, further comprising modeling the respective non-drug related, frequency-dependent mISI datasets as graphical curves for each baseline frequency, wherein the graphical curves comprise corrected mISI values as a function of the respectively increasing applied frequencies.

14. The system of Claim 13, further comprising, for each baseline frequency of the respective non-drug related, frequency-dependent mISI datasets, identifying a half maximal value of the corrected mISI and an associated half maximal frequency (f50) from the graphical curves for the respective non-drug related, frequency-dependent mISI datasets for each baseline frequency.

15. The system of Claim 14, further comprising extrapolating the graphical curves, for the respective non-drug related, frequency-dependent mISI datasets, to a frequency value of zero (0) and identifying the half maximum value (EC50_0) of the corrected mISI values. ^16. The system of Claim 15, further comprising calculating a frequency dependent half maximum effective concentration EC50(f) for the compound to which the cardiomyocyte samples are exposed by a relationship expressed as: EC50(f) = EC50_0 / (1 + (f / f50)h), wherein h is a Hill coefficient corresponding to a sigmoidal shape.

17. The system of Claim 16, further comprising calculating a frequency dependent mISI response to doses of the compound wherein the frequency dependent mISI response corresponds to another equation: 1 / mISI(f) = (1 / mISI0(f)) / (1 + [A] / EC50(f)), wherein mISI0(f) corresponds to mISI at frequency (f) from the respective non-drug related, frequency-dependent mISI datasets and [A] corresponds to compound concentration.

18. The system of Claim 17, further comprising compiling final frequency dependent response curves for doses of the compound at respective concentrations for each baseline frequency wherein the respectively increasing applied frequencies have corresponding mISI values. ^

Images