SYSTEMS AND METHODS FOR SELF-INOCULATION IN SERIAL EXPANSION AND PRODUCTION PROCESSES

Abstract

A system and method for self-inoculating a bioreactor in a serial expansion process includes an expansion chamber for expanding an initial cell solution to a viable cell density; a bioreactor for inoculation with the expanded cell solution; a fluid communication path between the expansion chamber and the bioreactor; a pump for controlling fluid flow through the fluid communication path; a Raman spectrometer for generating Raman spectrum data; a multivariate model that provides predictions for processing variables in the expansion chamber; and a computer system for controlling the pump to perform bioreactor self-inoculation from the expansion chamber, through the fluid communication path, when the computer system determines from the Raman spectrum data that one or more predefined triggering events have occurred.

Description

SYSTEMS AND METHODS FOR SELF-INOCULATION IN SERIAL EXPANSION AND PRODUCTION PROCESSES LfrRfrnn / zznz / E / YiAi CROSS REFERENCE TO RELATED APPLICATIONS This application claims the benefit and priority of U.S. provisional patent application No. 62 / 925,940 filed on October 25, 2019, and where permitted, is incorporated in its entirety by reference hereto. FIELD OF INVENTION The inventions included herein comprise bioreactor systems and methods for monitoring and controlling a serial expansion process in a bioreactor system. Particular embodiments also include bioreactor systems comprising a spectrometer and methods using spectroscopy to monitor and control a serial expansion process. BACKGROUND OF THE INVENTION Therapeutic antibodies, and in particular monoclonal antibodies (mAbs), have become an important tool in modern medicine for the development of target proteins that can be used in the treatment of a wide variety of diseases, including cancer and autoimmune diseases. The target proteins of interest are produced by a cell line that is expanded from an initial cryopreserved cell solution through a serial expansion process in one or more stages until a predetermined viable cell density (VCD) is achieved. At that point, the expanded cell solution is introduced into a production bioreactor to inoculate a culture medium maintained there. After inoculation, the cell culture continues to grow in the bioreactor until a target protein is expressed in a desired quantity, after which the cell culture fluid can be harvested and the target protein isolated and purified. Traditional serial expansion processes involved multiple stages of cell growth and expansion, using enlargement vessels, between the initial cryopreserved cell solution and the final production bioreactor. In early processes, the initial cryopreserved cell solution could be expanded through several stages, which might include, for example, one or more shaking flasks, one or more shakers, one or more Wave bags, and one or more expansion chambers before reaching the predetermined cell volume (CV) for inoculation into a production bioreactor. More recently, more efficient serial expansion processes have been developed to achieve a predetermined CV in fewer steps. However, modern processes still require cell expansion using at least one chamber to reach a predetermined CV before inoculation of a culture medium into a final production bioreactor. In general, the final target protein concentration can be increased and batch-to-batch consistency can be reduced in a production bioreactor by using an inoculum with the same VCD (velocity of the cell). However, there is a range of inoculum VCD that leads to a target production bioreactor yield. For example, a very low inoculum VCD can cause undesirable lactate cell metabolism, and an excessively high inoculum VCD can result in lower cell growth in the production bioreactor because the cells exit the exponential growth phase. Undesirable lactate cell metabolism and reduced cell growth in the production bioreactor result in lower quantities of target proteins than could have been produced from those cells, thus equating to a loss of yield and an overall decrease in process efficiency. Therefore, it is desirable that this cell expansion reach a predetermined VCD (velocity of cell expansion) that leads to the desired lactate cell metabolism in the production bioreactor and sustains exponential cell growth, and that the production bioreactor be inoculated as soon as possible after reaching the predetermined VCD. It will be noted that the target VCD range will vary from cell line to cell line, depending on the properties of the different cell lines. However, another complication arises: cell expansion can also vary between individual production runs of a common cell line due to variations in the culture medium and other operating conditions. As a result, the time to inoculate a production bioreactor, after cell expansion in an upstream expansion chamber to a predetermined VCD, can be variable. Despite the many advances made to date in the technique, further improvements to the serial expansion processes are still needed to further advance the technique and improve overall performance. As a non-exhaustive example, the technique would benefit from improvements that facilitate the inoculation of a production bioreactor after cell expansion to a predetermined VCD. SUMMARY OF THE INVENTION The present invention relates to systems and methods for using Process Analytical Technology (PAT) tools and a PAT Information Manager to provide monitoring and control strategies to increase process consistency. In one aspect, the systems and methods according to the present invention reduce reliance on manual operations for obtaining and verifying offline samples to confirm target cell densities and to initiate the transfer of a cell culture between bioreactors, such as when inoculating a final production bioreactor. Raman spectroscopy is used in conjunction with software from LfrRfrnn / zznz / E / YiAi PAT data management, to enable continuous monitoring of cell growth and automatic transfer to a cell culture between two containers when a predefined trigger event is detected (e.g., when a target viable cell density is detected). The systems described herein operate to monitor a cell culture in an expansion chamber using a Raman spectrometer and to control the inoculation of a production bioreactor with an inoculum from the expansion chamber based on Raman spectral data. In some examples, the system control scheme includes inoculating a production bioreactor using an in-line pump based on a determination that a cell culture in an upstream expansion chamber (e.g., an upstream bioreactor of a relatively smaller volume) has reached a predetermined viable cell density (VCD). Such systems and methods can be used with cell cultures including mammary cells, for example, Chinese hamster ovary (CHO) cells, and the cell culture can be grown to produce proteins including antibodies, antigen-binding fragments of these, or fusion proteins. The systems herein also include one or more processors communicating with a computer-readable medium (e.g., physical, non-transient memory) that stores software code for the execution of one or more processors to receive data, including a VCD from the Raman spectrometer cell culture, and to perform an inoculation of a production bioreactor based on the Raman spectral data. The software code stored on the computer-readable medium can be further configured to use one or more multivariate models, such as a partial least squares regression model, to interpret the Raman spectral data. The software code can also be configured to control the system to perform one or more signal processing techniques on the spectral data, for example, a noise reduction technique. The systems disclosed herein operate to monitor and control a serial expansion process and may include an expansion chamber to receive a starter cell solution for expansion into a viable cell culture; a bioreactor in fluid communication with the expansion chamber to receive a viable cell culture; a pump to efficiently transfer a viable cell culture from the expansion chamber to the bioreactor via a fluid communication path between the expansion chamber and the bioreactor; a multivariate model to correlate Raman spectrometer data with one or more process variables of the cell expansion process within the expansion chamber using Raman spectrometry; the Raman spectrometer is adapted to generate Raman spectrometer data; and a computer system in serial communication with the Raman spectrometer to receive Raman spectrometer data and in signal communication with the bioreactor. LfrRfrnn / zznz / E / YiAi the pump to control the pump to transfer a viable cell culture from the expansion chamber to the bioreactor. The Ñaman spectrometer can be adapted to generate Ñaman spectral data, and a multivariate model is correlated with Ñaman spectral data to one or more process variables. The computer system can then be adapted to compare the process variable measurements to one or more predefined process setpoints to determine whether one or more process variable measurements have met a predefined trigger value. When the computer system determines that a process variable measurement in the Ñaman spectral data has met a predefined trigger value, the control system instructs the pump to perform an auto-transfer of a volume of cell culture from the expansion chamber to the bioreactor, thereby auto-inoculating the culture medium in the bioreactor with a cell culture from the expansion chamber. The computer system processes the ñaman spectral data from the ñaman spectrometer to generate a multivariate model of one or more process variables, which may include a partial least squares regression model. By comparing process variable measurements from the ñaman spectral data with one or more process fit points, the computer system can use process variable measurements from a plurality of predefined isolated regions of the ñaman spectral data, such as wavelength regions of 800–850 cm⁻¹; 1260–1470 cm⁻¹; 1650–1840 cm⁻¹; and / or 2825–3080 cm⁻¹. The systems herein can be used to self-inoculate a bioreactor by expanding a cell solution in the expansion chamber, generating ÑAMN spectrum data, using the multivariate model to predict one or more process variables in the cell expansion process in the expansion chamber, comparing, with the computer system, predictions of process variables from the ÑAMN spectrum data and predefined process set points; and triggering the pump to self-inoculate the bioreactor with a viable cell culture from the expansion chamber when the computer system determines that one or more predictions of process variables from the ÑAMN spectrum data satisfy a predefined trigger value. The systems currently available can process Ñaman spectrum data received from the Ñaman spectrometer to generate a multivariate model of one or more process variables; and then obtain predictions of process variables from the multivariate model to compare against stored, predefined trigger values. After completing a serial expansion process, the system can store the multivariate model for use in monitoring and controlling a subsequent serial expansion process. In a process of In subsequent serial expansion, the system may use one or more multivariate models from one or more previous serial expansion processes for comparison with one or more variable measurements in the subsequent serial expansion process. The system may use one or more variable models from one or more previous serial expansion processes to control the processing conditions in the expansion chamber and / or the bioreactor. Both the preceding general description and the detailed description below are for illustrative and explanatory purposes only and are intended to provide further explanation of the invention as claimed. The accompanying figures are included to provide further understanding of the invention; they are incorporated herein and form a part of this specification, illustrate embodiments of the invention, and together with the description serve to explain the principles of the invention. BRIEF DESCRIPTION OF THE DRAWINGS Additional elements and advantages of the invention can be verified from the following detailed description provided in relation to the figures described below: FIG. 1 shows an example of a system according to the present description. FIG. 2 shows an example of a computer architecture that can be used with the computer system of the system in FIG. 1; FIG. 3 shows an example of a method for using the system in FIG. 1 to self-inoculate a bioreactor; Figures 4a-4d show steps for collecting and processing spectrum data and generating a regression model from collected spectrum data, using the system in Figure 1; FIG. 5 shows an example of a regression model generated from Raman spectrum data using the system in FIG. 1; FIG. 6 shows data ranges from a spectrum model to be used to generate a regression model using the system in FIG. 1; FIG. 7 shows a comparative example of two regression models generated using different regions of Raman spectrum data using the system in FIG. 1; and FIG. 8 shows a weighted regression model of predicted process values ​​generated by the system in FIG. 1. DETAILED DESCRIPTION OF THE INVENTION The following description discusses the present invention with reference to the examples shown in the accompanying figures, but does not limit the invention to such examples. As used herein, the singular forms of un / a, uno / a, and el / la include their plural referents, unless the context clearly indicates otherwise. The use of any examples or illustrative vocabulary (e.g., such as) provided herein is intended merely to further explain the invention and not LfrRfrnn / zznz / E / YiAi does not place a limitation on the scope of the invention unless otherwise stated. Nothing expressed in the specification shall be construed as indicating that any unclaimed element is essential or otherwise fundamental to the implementation of the invention. Unless made clear in the context, terms such as “first,” “second,” “third,” etc., when used to describe multiple devices or elements, are used only to convey the relative actions, positions, and / or functions of the separate devices, and do not require a specific order for such devices or elements or any specific quantity of such devices or elements. The word “substantially,” as used herein with respect to any property or circumstance, refers to a degree of deviation that is sufficiently small not to appreciably detract from the identified property or circumstance. The exact degree of deviation permissible in a given circumstance will depend on the specific context, as will be understood by someone skilled in the art. The use of the terms “around” or “approximately” is intended to describe values ​​above and / or below a specified value or range, as understood by someone skilled in the art within the relevant context. In some cases, this may encompass values ​​within a range of approximately + / - 10%; in others, values ​​within a range of approximately + / - 5%; in still others, values ​​within a range of approximately + / - 2%; and in yet others, values ​​within a range of approximately + / - 1%. The applicable ranges in each case will be clear from the context, and no further limitations are implied. The expressions “includes” and / or “that includes”, when used in this descriptive memory, shall be understood to specify the presence of the characteristics, integers, steps, operations, elements and / or components expressed, but do not exclude the presence or addition of one or more additional characteristics, integers, stages, operations, elements, components and / or groups thereof, unless indicated herein or clearly contradicted by the context. References to value ranges herein, unless otherwise stated, serve as a shortcut to refer individually to each separate value that falls within the stated ranges, including reference values ​​for each range, each separate value within each range, and all intermediate ranges included by each range, where each is incorporated into the specification as if individually mentioned herein. All methods described herein may be performed with the individual steps executed in any suitable order. Unless otherwise stated herein, or clearly contradicted by the context, the methods may be performed in the precise order disclosed and without intermediate steps, with one or more additional steps interposed between the disclosed steps, or with the disclosed steps performed in any order other than the LfrRfrnn / zznz / E / YiAi exact order disclosed, with one or more steps performed simultaneously and with one or more disclosed steps omitted. The terms cell culture and cell culture media can be used interchangeably and include any solid, liquid, or semisolid medium designed to promote the growth and maintenance of microorganisms, cells, or cell lines. Components such as polypeptides, sugars, salts, nucleic acids, cellular debris, acids, bases, pH buffers, oxygen, nitrogen, viscosity-modulating agents, amino acids, growth factors, cytokines, vitamins, cofactors, and nutrients may be present within the cell culture medium. Examples include a mammalian cell culture process using mammalian cells or cell lines, such as a Chinese hamster ovary (CHO) cell line grown in a chemically defined basal medium. As used herein, the term “nutrient” can refer to any compound or substance that provides essential nutrients for the growth and survival of a crop. Examples of nutrients include, but are not limited to, simple sugars such as glucose, galactose, lactose, fructose, or maltose; amino acids; and vitamins, such as vitamin A, B vitamins, and vitamin E. As used herein, the term “signal communication” can refer to any way of communicating a signal between two or more devices, including, but not limited to, physical connections (e.g., directly connected signal paths) and non-physical connections (e.g., wireless signal paths). Unless otherwise stated, signal communication between two devices can be direct (e.g., a transmitter in a first device communicating directly with a receiver in a second device) or indirect (e.g., a transmitter in a first device and a receiver in a second device communicating with each other via an intermediate transceiver). In one example, as shown in FIG. 1, a system 10 is provided comprising an expansion chamber 110, a spectrometer 120, a pump 130, a production bioreactor 140, and a computer system 150. The expansion chamber 110 and the bioreactor 140 are in fluid communication with each other via a feed line 135, with the flow through the feed line 135 controlled by the pump 130. The spectrometer 120 has at least one probe 125 adapted to monitor a cell culture within the expansion chamber 110 and is in signal communication with the computer system 150. The computer system 150 is in signal communication with at least the spectrometer 120 and the pump 130, although it may be in signal communication with one or more, or each, of the expansion chambers 110, the bioreactor 140, and a network. In some examples, two or more of the spectrometers ñaman 120, the computer system 140 and the pump LfrRfrnn / zznz / E / YiAi 130 can be provided as a single, integrated device. The expansion chamber 110 and bioreactor 140 can be operated as batch, fed-batch, or continuous units. Their volumes range from approximately 2 L to approximately 10,000 L. For example, the expansion chamber 110 could be a 50 L stainless steel unit, and the bioreactor 140 could be a 250 L unit. Both the expansion chamber 110 and bioreactor 140 must maintain cell counts ranging from approximately 0.25 x 10⁶ cells / mL to approximately 100 x 10⁶ cells / mL. In one example, the 120 spectrometer is a nAM spectrometer that can monitor and collect data on any component of a cell culture that has a detectable nAM spectrum. The systems and methods described herein can be used to monitor any component of cell culture media, including components added to the cell culture, substances secreted by cells, and cellular components present after cell death. Cell culture media components that can be monitored by these systems and methods include, but are not limited to, nutrients such as amino acids and vitamins, lactate, cofactors, growth factors, cell growth rate, pH, oxygen, nitrogen, viable cell count, acids, bases, cytokines, antibodies, and metabolites. The 150 computer system can be implemented using one or more specially programmed general-purpose computer systems, such as embedded processors, systems-on-a-chip, personal computers, workstations, server systems, and minicomputers or central processing units, or in networked distributed computing environments. The 150 computer system may include one or more 1502A-1502N processors (CPUs), 1504 input / output circuitry, a 1506 network adapter, and 1508 memory. The 1502A-1502N CPUs execute program instructions to perform the functions of the systems and methods described herein. Typically, the 1502A-1502N CPUs are one or more microprocessors, such as an Intel Core® processor. The input / output circuits 1504 provide the ability to input data into the computer system 150 or to send data from it. For example, input / output circuits 1504 may include input devices such as keyboards, mice, touchpads, trackballs, scanners, analog-to-digital converters, etc.; output devices such as video adapters, monitors, printers, etc.; and input / output devices such as modems, etc. The network adapter 1506 interconnects the computer system 150 to a network 1510, which may be any public or private LAN or WAN, including, but not limited to, the Internet. Memory 1508 stores the program instructions that are executed by the CPU 1502A-1502N, and the data that it uses and processes to perform the functions of the L+R+nn / zznz / E / YiAi Computer system 150. Memory 1508 may include, for example, electronic memory devices such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), electronically programmable read-only memory (EEPROM), flash memory, etc., and electromechanical memory such as magnetic disk drives, tape drives, optical disk drives, etc., which may use an electronic integrated drive interface (IDE), or a variation or improvement thereof such as enhanced IDE (EIDE) or ultra-direct memory access (UDMA), or an interface based on a small computer system interface (SCSI), or a variation or improvement thereof such as fast SCSI, wide SCSI, fast and wide SCSI, etc., or serial advanced technology connection (SATA), or a variation or improvement thereof, or a Fibre Channel Discrete Access (FC-AL) interface. Memory 1508 may include driver routines 1512, driver data 1514, and operating system 1516. The driver routines may include software for performing the processing required to implement one or more drivers. The driver data may include data necessary for the driver routines to perform the processing. In one configuration, the driver routines may include multivariate software for performing multivariate analysis, such as a PLS regression model. In this configuration, the driver routines may include SIMCA (Sartorius Stedim Data Analytics AB, Umeå, Sweden) for performing PLS modeling. In another configuration, the driver routines may also include software for performing noise reduction on a dataset. In this configuration, the driver routines may include MATLAB Runtime (The Mathworks Inc., Natick, Mass.) for performing noise reduction filter models.Furthermore, the controller routines may include software, such as MATLAB Runtime, to operate an automatic control unit, for example, a proportional-integral-derivative (PID) controller. The software for operating the system should also be able to calculate the difference between a predefined setpoint and a measured process variable (for example, a measured nutrient concentration) and provide a prediction of when the predefined setpoint will be reached. When functionality to predict the time to reach a predefined setpoint is included, as with a PID controller, the computer system 150 is also in signal communication with the pump 130 so that the correct amount of inoculum can be pumped into the expansion chamber 110 and / or the bioreactor 140, as predefined. The system 10 can monitor and control the process variables in the expansion chamber 110 and the bioreactor 140, as shown in the figure.1, or in a plurality of expansion chambers and / or a plurality of bioreactors. FIG. 3 shows a flow diagram for a method 200 of performing a serial expansion process with system 10. After the introduction of a cryopreserved cell solution in a culture medium into the expansion chamber 110, the spectrometer LfrRfrnn / zznz / E / YiAi Raman 120 collects Raman spectral data (FIG. 4a) from the expanding cell culture in the expansion chamber 110 (step 201). Raman spectroscopy is a form of vibrational spectroscopy that provides information about molecular vibrations, which can be used for sample identification and quantification. The Raman 120 spectrometer collects Raman spectral data using a probe 125, which can be either a contact or non-contact probe. A non-contact probe 125 allows for in situ Raman analysis of the cell culture without the need for contact with or removal from the cell culture. In situ Raman analysis is advantageous because it is non-invasive and therefore reduces the risk of contaminating the cell culture, which can introduce unwanted influences on the cell culture and the resulting proteins. Raman spectral data are acquired at a regular frequency, ensuring that the spectral data are continuously updated. Spectral data can be collected every 10 to 120 minutes, every 15 to 60 minutes, or every 20 to 30 minutes. The appropriate sampling frequency can be determined on a case-by-case basis, for example, based on the specific cell line and / or processing conditions, as deemed necessary to ensure that the spectral data accurately reflect the current state of a given cell culture. Any commercially available Raman spectrometer can be used; non-exhaustive examples include the RamanRXN2 and RamanRXN4 spectrometers (Kaiser Optical Systems, Inc., Ann Arbor, Mich.). After collection, in step 202, the raw spectrum data are transmitted to the computer system 150, where they are preprocessed and the processed Raman data (FIG. 4b) are stored in memory 1508 in a dedicated location for later use (step 202). In some examples, processing the Raman spectrum data includes applying one or more special filters to correct for initial changes. For example, the raw spectrum data may be treated with a point attenuation technique or a normalization technique. Normalization may be necessary to correct for any variations in laser power and exposure time by the Raman spectrometer. In some examples, the raw spectrum data may be treated with point attenuation, such as 1. derived with point attenuation of 21 cm⁻¹, and normalization, such as standard normal variance (SNV) normalization. In parallel with the Raman spectral data collection (step 201), process variable data are also collected using an alternative “offline” method (step 203) and stored in memory 1508 (step 204). Offline process variable data can be collected using a suitable analytical method, for example, by manually obtaining a sample from the cell culture and analyzing it on a local analyzer, such as the BioProfile Flex® analyzer (Nova Biomedical Corporation, Massachusetts, USA). Verifiable offline process data are collected at LfrRfrnn / zznz / E / YiAi is collected at a lower frequency than Raman spectral data, for example, once every 24 hours, once every 12 hours, or once every 6 hours, and serves as the baseline for the Raman spectral data. When collected, the offline process variable data is also stored in the 150 computer system in a dedicated location. When collecting offline process variable data, the 150 computer system can also store information from PAT systems and / or data management systems (for example, laboratory data management system and / or continuous online process data). The computer system 150 uses the processed Raman data to generate a multivariate model that reports one or more cell culture process variables (step 205). When offline process variable data are available, the computer system 150 compares the Raman spectral data with the corresponding offline process variable data to correlate the peaks between the two datasets. The computer system uses data management and / or PAT data to correlate the offline process variable data with the corresponding Raman spectral data. Any type of multivariate software package, for example, SIMCA 13 (Sartorius Stedim Data Analytics AB, Umeå, Sweden), can be used to correlate the peaks between two sets of spectral data. The multivariate modeling performed by the computer system 150 may include, but is not limited to, partial least squares (PLS), principal component analysis (PCA), orthogonal partial least squares (OPLS), multivariate regression, canonical correlation, factor analysis, cluster analysis, graphical procedures, and similar methods. In the example shown in Figures 4c–4d, a PLS regression model is created by fitting available measurement values ​​obtained from Raman spectral data and offline process variable data (Figure 4c), and the model is optimized (step 206) by removing outliers to provide a linear predictive model (Figure 4d). Such a PLS regression model can be used to provide predicted process values, for example, predicted concentration values ​​for a particular variable to be monitored by the computer system 150 for control over system 10. Model optimization may involve applying additional signal processing techniques to the multivariate model and its predicted process values. For example, a noise reduction technique might be applied to the predicted process values ​​to perform data attenuation and / or signal rejection. Noise reduction techniques provide a filtered model. One such technique combines raw measurements with a model-based estimate of what the measurements should yield according to the model. The noise reduction technique can combine a current predicted process value with its uncertainties, which can be determined by the repeatability of the predicted process values ​​and the process conditions. LfrRfrnn / zznz / E / YiAi current. Once the next predicted process value is observed, the predicted process value estimate is updated using a weighted average, where more weight is given to estimates with greater certainty. By using an iterative approach, the final process values ​​can be updated based on the previous measurement and current process conditions. In this respect, the algorithm must be recursive and capable of running in real time to utilize the current predicted process value, the previous value, and the experimentally determined constants. The noise reduction technique improves the robustness of the measurements in the ñaman analysis and the predictions in the PLS. The computer system 150 includes an automatic control unit (ACU) 155 that operates, in a step 207, to evaluate modeled spectrum data to determine whether the pump 130 should be activated to transfer a volume of cell culture from the expansion chamber 110 to the bioreactor 140, to inoculate a culture medium in the bioreactor 140. The ACU 155 stores one or more predefined setpoint values ​​that can define a triggering event to execute an auto-inoculation.The ACU 155 can be any type of automatic controller capable of comparing filtered process values ​​with one or more predefined setpoint values ​​and automatically executing a predefined action after determining that one or more filtered process values ​​satisfy a condition of a corresponding setpoint value (e.g., is at or above a maximum setpoint value; at or below a minimum setpoint value; or similar). If the ACU 155 determines that the predefined conditions for inoculation have been met, then the ACU 155 drives pump 130 to create a fluid flow through fluid line 135, thereby self-inoculating the bioreactor 140 (step 208); otherwise, the process returns to data collection via an iterative loop (step 209). In one example, the ACU 155 stores a predefined setpoint value (also referred to herein as a “trigger value”) based on a target VCD for a cell culture that is currently undergoing serial expansion. This predefined setpoint value can be set to a target VCD so that desirable cell metabolism can occur in the production bioreactor while maintaining the cells in the exponential growth phase. For example, a VCD-based trigger value can be set to a value equal to a predetermined target VCD, a value that is -2.5% of the target VCD, a value that is -5% of the target VCD, a value that is -10% of the target VCD, and so on.In that example, if the ACU 155 determines that a measured VCD value is equal to or greater than a predefined VCD-based trigger value, then the ACU 155 treats that condition as a trigger event to actuate pump 130 to make a fluid flow through fluid line 135, so that a culture medium in bioreactor 140 is automatically inoculated with a cell culture volume of. LfrRfrnn / zznz / E / YiAi the expansion chamber 110. The ACU 155 can store any number of predefined trigger values, establishing conditions for any type of triggering event. For example, a first trigger value can be set based on a target VCD, and a second trigger value can be set based on a minimum lactate value. The VCD-based trigger value might represent a target VCD for inoculating the 140 bioreactor, as described above, while the lactate-based trigger value might identify a predetermined minimum lactate level that signals a change in the cell growth state. This lactate-based trigger value could be set to a value equal to a predetermined minimum lactate level, a value that is +2.5% of the minimum lactate level, a value that is +5% of the minimum lactate level, a value that is +10% of the minimum lactate level, and so on.In this example, the ACU 155 can be adapted to self-inoculate the bioreactor 140 after detecting any triggering event, so that the system 10 inoculates the bioreactor 140 once the predefined VCD trigger value is reached, although it can trigger self-inoculation at a lower VCD if a lactate level measurement is detected that is equal to or less than the predefined lactate-based trigger value, thereby ensuring the inoculation of the bioreactor 140 before a change in the cell growth state. As another example, the ACU 155 can operate with a first trigger value based on a target cell growth rate (CGR), a second trigger value based on any predetermined processing variable that indicates a change in the cell growth state, and a third trigger value based on a model-predicted CGR. A trigger value based on a model-predicted CGR can be set to the maximum model-predicted CGR considered acceptable for a given serial expansion process; a value that is -2.5% of the maximum cell growth rate; a value that is -5% of the maximum cell growth rate; a value that is -10% of the maximum cell growth rate; and so on.In this example, the ACU 155 can be adapted to auto-inoculate the bioreactor 140 after detecting any of the triggering events, such that the system inoculates the bioreactor 140 once the predefined VCD trigger value is reached. However, it can trigger inoculation at a lower VCD if a variable processing value is detected that satisfies a predetermined condition indicating a change in the cell growth state. Furthermore, early auto-inoculation is also triggered if a model-predicted VCD equal to or greater than a predefined model-predicted VCD trigger value is detected. In this way, if a cell culture begins to experience a higher model-predicted VCD, the system will automatically auto-inoculate. LfrRfrnn / zznz / E / YiAi model before the detection of a predefined VCD trigger value and without the detection of any other processing variable that provides a warning of a change in a cell growth state, then the system can trigger self-inoculation before achieving an unacceptable cell growth state. The ACU 155 can operate with any number of predefined trigger values, based on any number of different process variables, which may include, but are not limited to, any one or a combination of: one or more nutrients (such as amino acids and vitamins); lactate; cofactors; growth factors; cell growth rate; pH; oxygen; nitrogen; viable cell count; cell death count; acids; bases; cytokines; antibodies; and metabolites. The computer system 150 may also have controls that enable changes to the system, including the ACU 155, in real time from a platform interface. For example, there may be an interface that allows a user to select one or more trigger conditions based on a number of different processing variables (e.g., a trigger condition based on VCD; a trigger condition based on lactate; a trigger condition based on cell growth rate, etc.) to enter a desired value to use as a predefined setpoint value in a trigger condition (e.g., trigger values ​​based on a target VCD; a minimum lactate level; a maximum cell growth rate, etc.) and to adjust one or more preset trigger values.The ACU 155 must be able to respond to a change in one or more predefined trigger values ​​to adjust the conditions by which self-inoculation is triggered. In a first serial expansion process, Raman spectral data were used to generate a multivariate model based on process variable measurements in the ranges of 450–1800 cm⁻¹ and 2600–3100 cm⁻¹, while excluding measurements in the range of 1800 < x < 2600 cm⁻¹. In parallel with the Raman spectral measurements, offline spectral measurements were also taken using a BioProfile Flex® analyzer. Raman spectral measurements were taken every 15–60 minutes, while offline process variable measurements were taken approximately every 4 hours, 24 hours, and 72 hours after the introduction of the cryopreserved cell solution into expansion chamber 110. Figure 5 shows data from two spectral measurements, along with a target VCD (4.0 x 10⁶ cells / mL) for inoculation of bioreactor 140. As shown in FIG. 5, the two spectral datasets were relatively in agreement with each other at the 24-hour mark, corresponding to the second offline measurement, although they began to diverge around the 32-hour mark. While the two datasets converge again around the 72-hour mark, which LfrRfrnn / zznz / E / YiAi corresponds to the fourth offline measurement. A maximum deviation of approximately 2.0 x 10⁶ cells / mL was observed around the 48-hour mark, which corresponds to the third offline measurement. This deviation is significant. For example, when targeting an inoculation VCD of 4.0 x 10⁶ cells / mL, if ACU 155 relied on Raman spectral data, then autoinoculation of bioreactor 140 would occur at approximately 44 hours. However, based on the offline spectral data, autoinoculation at 44 hours would be premature since the VCD at that time would actually have been approximately 2.5 x 10⁶ cells / mL. In contrast, assuming that the offline spectrum data are accurate, the target VCD for inoculation would not be achieved until approximately 60 hours.Thus, the self-inoculation based on Raman spectrum data would have occurred in a suboptimal VCD, which could lead to a substantial deficit in the production process outcome. A deviation like the one in FIG. 5 can be problematic for accurate and reliable inoculation of a bioreactor. For example, if there is an error in the Raman spectra data, then autoinoculation might be performed too early, before an optimal VCD is actually reached. On the other hand, when offline process variable measurements are taken only once every few hours, manual inoculation could be performed too late, such as when manual inoculation is not performed at the 48-hour mark when the VCD is below the target (as in FIG. 5) and is instead performed at the 72-hour mark after the target VCD has been exceeded (FIG. 5). In an effort to improve the accuracy and reliability of Raman spectrum measurements, a cell addition study was performed using a CHO cell suspension at six different densities, as identified in Table I below: LfrRfrnn / zznz / E / YiAi Table I Target cell density (x106) Actual cell density (NovaFLEX®, x106) 1 1.03 4 3.85 7 7.68 10 9.39 13 13.09 16 15.65 Raman spectral measurements were taken in triplicate from six different cell densities, replicating the scan time used in the upstream Raman data collection. A variable influence on prediction plot (VIP) was used to identify the wavelength regions in the Raman spectral data that correlated most strongly with cell density measurements. From this study, the wavelength regions of 800–850 cm⁻¹, 1260–1470 cm⁻¹, 1650–1840 cm⁻¹, and 2825–3080 cm⁻¹ were identified as best reporting cell density (Fig. 6). Figure 7 shows the results of a comparative example between measurements obtained from two sets of spectral data. The first set is based on Ñaman spectral measurements taken according to conventional practices, and the second set is based on Ñaman spectral measurements taken according to innovative practices. Both data sets are plotted against offline spectral data. The conventional Ñaman spectral measurements were taken at wavelengths of 450–1800 cm⁻¹ and 2600–3100 cm⁻¹, while the innovative Ñaman spectral measurements were taken at wavelengths of 800–850 cm⁻¹, 1260–1470 cm⁻¹, 1650–1840 cm⁻¹, and 2825–3080 cm⁻¹. As can be seen, while both sets of data show some deviation from offline measurements, the measurements taken in accordance with the innovative practices of the present showed less variability between samples. In a further aspect of the present invention, in addition to the computer system 150 that retains a predicted process variable from the multivariate model, the computer system 150 can retain a time series of predicted process variables, either in the expansion chamber 110 or the bioreactor 140. This time series can be subjected to a noise reduction technique, which can be predictive or retrospective. By incorporating a noise reduction technique, the ACU 155 can be a locally weighted regression model, as in Cleveland, W.S., "Locally Weighted Regression and Smoothing Scatterplots." Journal of the American Statistical Association, Vol. 74, No. 368 (1979): 829-836, which is incorporated in full by this reference. For example, the weighting function can be a fifth-order polynomial that de-emphasizes the batch and emphasizes the recent data points.Prior knowledge informs the intuition that cell growth at N-1 is sigmoid, implying that the mid-growth region is approximately linear. Local regression models can estimate this linearity and extrapolate it to update a predicted inoculation time, and further calculate a time between measurements where the estimated process value will equal the trigger value. This approach reduces the variance in the prediction of a single ñaman measurement, as shown in FIG. 8. Although the present invention is described with reference to particular embodiments, a person skilled in the art will understand that the foregoing disclosure deals only with illustrative embodiments; that the scope of the invention is not limited to the disclosed embodiments; and that the scope of the invention may comprise additional embodiments with various modifications and LfrRfrnn / zznz / E / YiAi modifications relating to the examples disclosed herein without departing from the scope of the invention as defined in the claims and equivalents appended hereto. To the extent necessary for the understanding or completion of the disclosure hereof, all publications, patents, and patent applications referenced herein are expressly incorporated herein by reference to the same extent as if individually incorporated. No license, express or implied, is granted to any patent incorporated herein. The present invention is not limited to the exemplary embodiments illustrated herein, but is instead characterized by the appended claims.

Claims

1. A system for controlling a serial expansion process comprising: an expansion chamber for receiving an initial cell solution for expansion into a viable cell culture; a bioreactor in fluid communication with the expansion chamber for receiving a viable cell culture; a pump for transferring a viable cell culture from the expansion chamber to the bioreactor via a fluid communication path between the expansion chamber and the bioreactor; and a spectrometer having at least one probe for monitoring the cell expansion process within the expansion chamber using spectrometry.The ñaman spectrometer is adapted to generate ñaman spectral data; a multivariate model provides predictions of process variables based on ñaman spectral data; and a computer system communicates with the ñaman spectrometer to receive ñaman spectral data and with the pump to control the pump's operation for transferring a viable cell culture from the expansion chamber to the bioreactor, wherein the ñaman spectrometer can be adapted to generate ñaman spectral data and a multivariate model provides predictions of one or more process variables, and the computer system is adapted to compare measurements of the process variables to one or more predefined process setpoints to determine whether one or more process variable measurements have met a predefined trigger value, and wherein the computer system is adapted,After determining that a variable process measurement in the ñaman spectrum data has met a predefined trigger value, the pump is controlled to perform an auto-transfer of a cell culture volume from the expansion chamber to the bioreactor.

2. The system according to claim 1, wherein the computer system processes the ñaman spectrum data received from the ñaman spectrometer to generate a multivariate model of one or more process variables.

3. The system according to claim 2, wherein the computer system generates a partial least squares regression model.

4. The system according to claim 3, wherein the computer system is adapted, when comparing predictions of process variables from the multivariate model to one or more predefined process fitting points, to use measurements of process variables from a plurality of predefined isolated regions of Raman spectrum data.

5. The system according to claim 4, wherein the computer system uses measurements of Raman spectral data process variables in the wavelength regions of 800-850 cm-1; 1260-1470 cm-1; 1650-1840 cm-1; and 2825-3080 cm-1.

6. A method of self-inoculating a bioreactor using a system according to claim 1, comprising: expanding a cell solution in the expansion chamber; generating Raman spectral data, using the Raman spectrometer, to provide data to a multivariate model that predicts one or more process variables of cell expansion in the expansion chamber; the computer system that compares predictions of process variables from the multivariate model with predefined process setpoints in the computer system; the computer system that controls the pump to self-inoculate the bioreactor with a viable cell culture from the expansion chamber when the computer system determines that one or more process variable predictions from the multivariate model satisfy a predefined trigger value.

7. The method according to claim 6, wherein the predefined trigger value is a viable cell density value.

8. The method according to claim 7, wherein the predefined trigger value is set to a viable cell density value that is equal to or within the range of -10% of a predetermined target viable cell density.

9. The method according to claim 6, wherein the predefined trigger value is a lactate level value.

10. The method according to claim 9, wherein the predefined trigger value is set to a lactate level value that is equal to or within the range of +10% of a predetermined target viable cell density.

11. The method according to claim 6, wherein the predefined trigger value is a predefined VCD model.

12. The method according to claim 11, wherein the predefined trigger value is set to a VCD value that is equal to or within the LfrRfrnn / zznz / E / YiAi range of -10% of a predetermined maximum cell growth rate.

13. The method according to claim 6, wherein the computer system stores a first predefined trigger value based on a predetermined viable cell density; and stores a second predefined trigger value based on a predetermined processing variable other than the viable cell density; and the computer system is adapted to control the pump to self-inoculate the bioreactor with a viable cell culture from the expansion chamber when the computer system determines that a variable prediction of the multivariate model process satisfies either the first or the second predefined trigger value.

14. The method according to claim 6, wherein the second predetermined trigger value is a lactate level value.

15. The method according to claim 6, wherein the predefined trigger value is a VCD value of the intended model.

16. The method according to claim 6, wherein the computer system processes ñaman spectrum data received from the ñaman spectrometer to generate a multivariate model of one or more process variables; and obtains process variable measurements from the multivariate model to compare predefined trigger values.

17. The method according to claim 16, wherein the computer system generates a partial least squares regression model.

Images