Control of cell production in a bioreactor

Abstract

We propose a process implemented by a control system to control cell production in a bioreactor. The bioreactor comprises a cell culture vessel containing a culture medium and microcarriers on the surface of which the cells grow. The process includes dynamic parameter adjustment during production. The dynamically adjusted parameters include at least a culture medium temperature, a culture medium pH, a dissolved oxygen concentration in the culture medium, and the amount of microcarriers present in the vessel. The process improves the quality and quantity of biomass produced in the bioreactor. [Fig. 1]

Description

Title of the invention: Control of cell production in a bioreactor Technical field

[0001] This disclosure relates to a method implemented by a control system to control cell production in a bioreactor, a computer program to implement such a method, a control system for such a bioreactor, and a bioreactor incorporating such a control system. Technical background

[0002] One of the challenges of the major national bioproduction initiative defined by the government in 2021 is controlling the quantity and quality of biomass produced. However, the initial growing conditions between production batches are never identical. This can cause significant variability in the final product and lead to the rejection of products whose characteristics and quality no longer meet the expected specifications.

[0003] It is now widely agreed that the Quality-by-Testing approach is unsuitable for the rapid and efficient development of innovative pharmaceutical products. The emergence in the mid-2000s of the pharmaceutical QbD (Quality by Design) good practice (international guidelines ICH Q8-Q12) has made it possible to introduce predictive risk analysis as early as possible in the development chain (references Bastogne, 2017, Bastogne, 2022, Bastogne et al., 2022 and Bevers et al., 2022 from the list provided below). Other technological and digital advances, grouped under the term Process Analytical Technology, have provided the means on the ground to monitor certain critical variables in real time during product manufacturing.In addition to this, there are control strategies, also defined in the ICH Q8(R2) guideline. Based on measurements provided by analytical technologies, these strategies allow for adjustments to the process to ensure the product conforms to expected specifications. There are at least three levels of sophistication for these control strategies. The most advanced level allows for automated process control and naturally falls within the broader framework of the Industry 4.0 paradigm. For the automation community, there is nothing truly new on the horizon, except for the increasing availability of real-time data (spectra, images, and / or genomics). (online), which now allows for the implementation of more sophisticated controllers, such as those based on predictive control, to better manage quality and productivity objectives and correct certain variables during cultivation if necessary. The first studies, published under the name "Quality by Control," were applied to bioproduction between 2015 and 2020. However, these existing production control solutions in a bioreactor are not sufficient.

[0004] List of references: - Bastogne, T., 2022. iQbD: a TRL-indexed Quality-by-Design Paradigm for Medical Device Engineering. Transactions of the ASME, Journal of Medical Devices 16. - Bastogne, T., 2017. Quality-by-design of nanopharmaceuticals-a State of the art. Nanomedicine: Nanotechnology, Biology and Medicine 13, 2151-2157. - Bastogne, T., 1997. Identification des systèmes multivariables par les méthodes des sous-espaces. Application to a band entrance system (Thèse de Doctorat). Université Henri Poincaré, Nancy 1. - Bastogne, T., Caputo, F., Prina-Mello, A., Borgos, S., Barberi-Heyob, M., 2022. A State of the Art in Analytical Quality-by-Design and Perspectives in Characterization of Nano-enabled Médicinal Products. In review in Journal of Pharmaceutical and Biomédical Analysis. - Bevers, S., Kooijmans, SAA, Van de Velde, E., Evers, MJW, Seghers, S., Gitz-Francois, JJJM, van Kronenburg, NCH, Fens, MHAM, Mastrobattista, E., Hassler, L., Sork, H., Ahmed, KE, Andaloussi, SE, Breckpot, K., Bastogne, T., Schiffelers, R.M., De Koker, S., 2022. LNPs tuned for systemic immunization induce strong antitumor immunity by engaging splenic immunity. Molecular Therapy. - Camacho, E.F., Bordons, C., 2007. Nonlinear model prédictive control: An introductory review. Assessment and future directions of nonlinear model prédictive control 1-16. - Ferrari, C., Balandras, F., Guedon, E., Olmos, E., Chevalot, L, Marc, A., 2012. Limiting cell aggregation during mesenchymal stem cell expansion on microcarriers. Biotechnology progress 28, 780-787. - Guideline, I.H.T., others, 2005. Validation of analytical procedures: text and methodology. Q2 (RI) 1, 05. - ICH Management Committee, 2018. ICH Q14: Analytical Procedure Development and Révision of Q2(R1) Analytical Validation. (Final Concept Paper). International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use., ICH Secrétariat, Switzerland. - Lindberg, Y., 2014. A comparison between MPC and PID controllers for éducation and steam reformers. Master’s thesis. - Maciejowski, J.M., Huzmezan, M., 2007. Prédictive control, in: Robust Flight Control: A Design Challenge. Springer, pp. 125-134. - Nekanti, U., Mohanty, L., Venugopal, P., Balasubramanian, S., Totey, S., Ta, M., 2010. Optimization and scale-up of Wharton’s jelly-derived mesenchymal stem cells for clinical applications. Stem cell research 5, 244-254. - Rafiq, Q.A., Ruck, S., Hanga, M.P., Heathman, T.R., Coopman, K., Nienow, A.W., Williams, D.J., Hewitt, C.J., 2018. Qualitative and quantitative démonstration of bead-to-bead transfer with bone marrow-derived human mesenchymal stem cells on microcarriers: Utilising the phenomenon to improve culture performance. Biochemical engineering journal 135, 11-21. - Van Overschee, P., De Moor, B., 1996. Subspace identification for linear Systems - Theory, implémentation, applications. Kluwer Academie Publishers. - Voisin, C., Cauchois, G., Bensoussan, D., Huselstein, C., 2021. Effects of Cell Confluence on the Immunological and Migration Receptors of Wharton Jelly's Mesenchymal Stem Cells.

[0005] In this context, there is still a need to improve production in a bioreactor. Summary

[0006] A method implemented by a control system is proposed for this purpose to control cell production in a bioreactor. The bioreactor comprises a cell culture tank containing a culture medium and microcarriers on the surface of which the cells grow. The method includes, during production, a dynamic adjustment of parameters. The dynamically adjusted parameters include at least a temperature of the culture medium, a pH of the culture medium, a concentration of dissolved oxygen in the culture medium, and a quantity of microcarriers present in the tank.

[0007] The dynamic adjustment of the quantity of microcarriers may include a measurement of the cell confluence rate at the surface of the microcarriers in the tank, and, an addition of microcarriers in the tank so that the measured cell confluence rate follows a predetermined setpoint kinetics.

[0008] The addition of microcarriers into the tank can be carried out by a pump.

[0009] The measurement of the cell confluence rate can be carried out by an inline sensor.

[0010] Production can take place over a period divided into successive time intervals. The addition of microcarriers to the tank can include, for each time interval, a prediction of the evolution of the cell confluence rate in the tank. The addition of microcarriers to the tank can include, for each time interval, three steps. The first step can be a comparison of the predicted evolution with the predetermined setpoint kinetics. The second step can be a determination of the quantity of microcarriers to be added for the next time interval based on the result of the comparison. The third step can be an addition of the determined quantity of microcarriers to the tank.

[0011] The prediction can be based on a dynamic model comprising a state matrix, an input matrix, and an output matrix. The content of the matrices can be determined from training data.

[0012] Determining the quantity of microcarriers to be added may involve applying an optimization algorithm based on an objective criterion. The objective criterion may preferably have the following formula:

[0013] minj(Au) = min{E^ [j(r+-r(t+j) f + Au(Z + jl) ]2

[0014] in which:

[0015] -Am is the increment of the control signal for adding microcarriers to the tank minimizing a cost function J, taking into account the various constraints;

[0016] - y ( t + ) denotes the prediction of the cell confluence rate at time t + j in function of the measurements up to the moment;

[0017] denotes a reference signal at time t+j of the setpoint kinetics predetermined to continue;

[0018] - designates the beginning of a prediction horizon;

[0019] - N2 denotes the end of the prediction horizon;

[0020] - designates a calculation horizon for the control signal; and

[0021] - 2 is a predetermined weighting term.

[0022] The microcarriers can be balls and / or discs. The microcarriers can have a diameter between 50 micrometers and 1000 micrometers.

[0023] The bioreactor may include a culture medium agitator and at least one sensor for measuring the quantity of suspended particles in the vessel. The method may further include a dynamic adjustment of the culture medium agitator based on the quantity of suspended particles measured by at least one sensor.

[0024] A computer program is also proposed comprising instructions which, when the program is executed by a bioreactor control system, cause the latter to implement the process.

[0025] A control system for the bioreactor is also proposed. The control system can be, for example, a SCADA system. The system comprises a processor coupled to a memory. The memory stores the computer program. The processor is configured to execute the instructions of the computer program in order to control the bioreactor according to the process.

[0026] A bioreactor integrating the control system is also proposed. Brief description of the figures

[0027] Non-limiting examples will be described with reference to the following figures:

[0028] Fig. 1 shows a flowchart of an example of implementation of the process.

[0029] Figure [Fig. 2] illustrates an example of a bioreactor control system.

[0030] Fig. 3 shows an example of tracking the kinetics of a predetermined setpoint.

[0031] Figure 4 illustrates an example of a control system including regulation of the agitation.

[0032] Fig. 5 shows examples of acceptable value ranges in accordance with international guidelines ICH Q8(R2).

[0033] Fig. 6 shows an example of a block diagram of the determination of the quantity of microcarriers to be added for each time interval.

[0034] Fig. 7 shows an example of the dynamic model.

[0035] Figure 8 shows an example of minimization by the objective criterion.

[0036] Figure 9 shows an example of a signal for stimulating the feed rate of the pump in microcarriers.

[0037] Fig. 10 shows an example of calibration of confluence rate measurements.

[0038] Fig. 11 illustrates an example of an MPC regulator. Detailed description

[0039] With reference to the flowchart in [Fig. 1], a method implemented by a control system is proposed for controlling cell production in a bioreactor. The bioreactor comprises a cell culture tank containing a culture medium and microcarriers on the surface of which the cells grow. The method includes, during production, a dynamic adjustment S10 of parameters. The dynamically adjusted parameters include at least a temperature of the culture medium, a pH of the culture medium, a concentration of dissolved oxygen in the culture medium, and a quantity of microcarriers present in the tank.

[0040] The process improves the quality and quantity of biomass produced in the bioreactor.

[0041] Indeed, the dynamic adjustment of the quantity of microcarriers makes it possible to control the confluence rate on microcarriers, which is a critical point for improving quality and production yield. In particular, microcarriers are Cell culture media and the dynamic adjustment of cell quantity in the bioreactor help reduce cell degradation when cell confluence on microcarriers is too high. Cell confluence affects cell behavior and growth. When cell confluence on microcarriers becomes too high, the quality of production decreases and the quantity no longer increases (as the surface area of ​​all microcarriers in the chamber becomes colonized). It has been shown that confluence can be one of the main mechanisms inhibiting their expansion (Nekanti et al., 2010). Voisin et al. (2021) also demonstrated that confluence impacts the immunomodulatory phenotype and migration. Furthermore, once the entire surface area is reclaimed by the cells, a phenomenon called contact inhibition occurs, and the cells cease growing. Cell confluence on microcarriers also leads to cell and microcarrier aggregation. Moreover, Ferrari et al. (2012) showed that some cells can migrate from microcarriers to other microcarriers during confluence. Rafiq et al. (2018) demonstrated this particular bead-to-bead transfer property with human bone marrow mesenchymal stem cells, using different types of microcarriers in the same culture.Therefore, automatically controlling the addition of new microcarriers reduces the percentage of confluence and thus the cell damage associated with this phenomenon. Furthermore, by improving parameter settings, the process also reduces the risk of differing initial culture conditions between different production batches.

[0042] In particular, the process improves quality and production yields, especially in the context of developing innovative therapy drugs. This is the main motivation behind national calls for tenders such as the major challenge for biopharmaceuticals in 2020 and, more recently, the one related to innovations in biotherapies and bioproduction.

[0043] The process is implemented by the control system. This means that the steps (or almost all of the steps) of the process are executed by the control system, or any other similar system. Thus, the steps of the process are executed by the control system, possibly in a fully automatic or semi-automatic manner. In some examples, the triggering of at least some of the steps of the method can be effected by an interaction between the user and the control system.

[0044] A typical example of implementing the method consists of executing the method using a control system adapted for this purpose. The control system may include a processor coupled to memory and a graphical user interface (GUI). The memory may contain a computer program including instructions for executing the process. Memory can also store a database. Memory is hardware adapted to this type of storage, possibly comprising several distinct physical parts (for example, one for the program, and possibly one for the database).

[0045] Controlling cell production in the bioreactor means controlling cell production in the bioreactor by adjusting certain critical variables (i.e., parameters) of the process in real time, i.e., dynamically. By dynamic adjustment, it is understood that the process is executed during cell production in the bioreactor, and that the process adjusts the parameters during production (i.e., several times, at several different times during production). For example, production may take place over a period divided into successive time intervals, and, at each time interval, the process may include an adjustment of the parameters.The process may include, after each elapsed time interval, determining the parameter values ​​for the elapsed time interval, and then adjusting the parameters (i.e., the parameter values) for a subsequent time interval. The subsequent time interval may be the one immediately following the elapsed interval, or the second one (or any interval following the elapsed one). The process may include recording the determined value of each parameter (for example, in a database). The dynamically adjusted parameters are critical production parameters.

[0046] The production period and / or the division into time intervals can be predetermined. For example, the length of the production period and / or the time intervals considered can be defined by the user before the process is executed (for example, through user actions on the control system). The production period can be several hours, for example, to several days. The division into time intervals can be a division into intervals of one or more seconds, for example, one or more tens of minutes.

[0047] For one or more parameters (e.g., temperature, pH, and / or oxygen concentration), determining the parameter value may involve measuring (e.g., using a sensor) the parameter value in the tank during the time interval. For example, determining the temperature may involve measuring the temperature in the tank using a temperature sensor. Similarly, the pH may be measured using a pH sensor, and the oxygen concentration may be measured using an oxygen sensor. The determined parameter value may then correspond to an average of the measured parameter value over the time interval (e.g., when the measurement includes several measurement points over the interval). Alternatively, the value The determined parameter can be a value taken during the time interval (for example, at the beginning or in the middle of the interval). For the quantity of microcarriers, the value of this parameter can be calculated. For example, determining the value might involve calculating the quantity of microcarriers in the tank based on the quantity of microcarriers determined for a previous time interval (for example, the quantity initially present in the tank for the first time interval) and the quantity of microcarriers added since the end of that previous time interval.

[0048] The cells produced in the bioreactor can be adherent cells that grow on the surface of microcarriers. For example, the cells can be Vero cells or CHO cells. The cells can colonize the surface of the microcarriers present in the cell culture vessel (i.e., the microcarriers initially present in the vessel and those added during the execution of the process). The cells can grow (i.e., expand and multiply) by feeding on the culture medium present in the vessel.

[0049] The process includes a dynamic adjustment S10 of at least the parameters temperature, pH, dissolved oxygen concentration, and quantity of microcarriers in the tank. For example, the process may include, for each of these parameters, the execution of commands to increase or decrease the parameter value (these commands being executed at different times during production). In some examples, the process may include, in parallel, a dynamic adjustment of one or more other production parameters (other than temperature, pH, dissolved oxygen concentration, and the quantity of microcarriers in the tank).

[0050] In examples, the process can repeat the dynamic S10 adjustment for different production batches. For each production batch, the process can perform, during production, the dynamic S10 adjustment of at least the parameters temperature, pH, dissolved oxygen concentration, and quantity of microcarriers in the vessel in the same way. The process thus makes it possible to compensate for the effects due to uncontrollable variations in initial culture conditions between the different production batches, thereby improving the reproducibility of production in the bioreactor.

[0051] The dynamic adjustment S10 of the temperature, pH and dissolved oxygen concentration is now discussed.

[0052] The process can perform the dynamic adjustment S10 of the temperature, pH, and dissolved oxygen concentration in any way. For example, the process can dynamically adjust these parameters to follow a predetermined setpoint (e.g., a curve indicating a desired evolution for each parameter as a function of time) and / or to keep the value of these parameters within predetermined ranges (e.g., by standards or standards). Adjusting these parameters can be done by adding substances (e.g., air for oxygen concentration or carbon dioxide gas for pH) or energy to the tank (e.g., heat for temperature). The dynamic adjustment S10 can, at each time interval, and for each time interval, include a measurement of the parameter (temperature, pH, and dissolved oxygen concentration), and then, for each parameter, the addition of a quantity of the respective substance to the parameter so as to follow the predetermined setpoint value of the parameter.

[0053] The S10 adjustment of the quantity of microcarriers is now discussed in more detail.

[0054] The S10 adjustment of the microcarrier quantity may include an S20 measurement of the cell confluence rate at the surface of the microcarriers in the cuvette. The S20 measurement of the cell confluence rate may be performed by an online sensor (or probe). The online sensor may be a multi-angle sensor, which is configured to recover the coverage rate on the microcarriers. The online sensor may be a near-infrared spatial resolution spectroscopic probe. For example, the online sensor may be the SAM-Spec sensor manufactured by Indatech (Chauvin-Arnoux). Alternatively, the online sensor may be a permittivity measurement probe in a liquid medium (such as the Incyte Arc sensor manufactured by Hamilton). Alternatively, the measurement may be an offline measurement.For example, the measurement may include taking a sample from the tank (e.g., automatically by a sampling device located in the tank, or manually by an operator). After sampling, the measurement may include processing the sample to determine the cell confluence rate. The processing may include determining the absorbance and / or turbidity of the sample, and then calculating the cell confluence rate in the tank based on the determined absorbance and / or turbidity.

[0055] As previously described, production can take place over a period divided into successive time intervals. The addition S30 of microcarriers to the tank can include, for each time interval, a prediction S31 of the evolution of the cell confluence rate in the tank. The prediction can be a prediction of the evolution of the cell confluence rate in the tank at future times. The time interval over which the prediction is made can be a future time interval. For example, the prediction S31 can take place during a time interval (or time step) t, and the prediction S31 can predict the evolution of the cell confluence rate for the time interval (or time step) t+1, that is, the one that follows the time interval currently elapsed while the process executes the prediction S31 for interval t. The prediction can be based on the evolution of the Cell confluence rate before the time interval. For example, the prediction can be based on the cell confluence rate measured for previous time intervals. The prediction can also be based on measured values ​​of other parameters (temperature, pH, and / or dissolved oxygen concentration), for example, the values ​​measured for these other parameters during their dynamic adjustment.

[0056] In some examples, the prediction can be based on a dynamic model comprising a state matrix, an input matrix, and an output matrix. The contents of the matrices can be determined from training data. The dynamic model can be a dynamic model of the bioreactor describing it by a state representation. The contents of the matrices can be predetermined. The contents of the matrices may have been determined before the process was executed. The dynamic model can be a model belonging to the family of black-box models, that is, an empirical model whose matrix contents are determined from training data. The training data may have been collected during experiments performed on the bioreactor to be controlled. The dynamic model makes it possible to establish predictions about the cell confluence rate of the microcarriers in the vessel during the time interval.

[0057] The S31 prediction of the evolution of the cell confluence rate can be a prediction over the time interval. For example, the prediction may include several predicted values ​​of the cell confluence rate at different times within the time interval. Alternatively, the prediction may include a single predicted value of the cell confluence rate, this value being an average for the interval or at a specific time within the interval (e.g., in the middle or at the end of the interval). Alternatively, the prediction may include a predicted change from a measured and / or predicted value for the previous time interval (e.g., an increase or decrease in the cell confluence rate).

[0058] After the prediction, the addition S30 of microcarriers may include a comparison S32 of the predicted evolution with the predetermined setpoint kinetics. The predetermined setpoint kinetics may include an evolution of the setpoint cell confluence rate as a function of time. This evolution may be optimal in terms of the quality and quantity of biomass produced. The predetermined setpoint kinetics may be represented by a curve showing the evolution of the setpoint cell confluence rate as a function of time. The setpoint kinetics may have been predetermined based on prior trials. The curve representing it may be sigmoidal in shape, for example, with parameters: K a static gain and T a time constant. The comparison S32 may include calculating the difference between the predicted evolution for the time interval and the rate of cell confluence given by the setpoint evolution for this time interval, and for example also a calculation of a difference between the predicted evolution for one or more previous time intervals and, for each, the respective cell confluence rate given by the setpoint evolution.

[0059] The addition S30 may include a determination S33 of the quantity of microcarriers to be added for the next time interval based on the result of the comparison. The result of the comparison may include all the calculated differences. The determination S33 may include the application of an optimization algorithm. The optimization algorithm may be based on an objective criterion. The objective criterion may take as input a control signal for adding microcarriers to the tank. The optimization algorithm may include minimizing this objective criterion. The determined quantity of microcarriers to be added may minimize the result of the objective criterion. The minimization may include determining a control signal for adding microcarriers to the tank that minimizes the result of the objective criterion. The objective criterion may preferably have the following formula:

[0060] mmJ(Au) = mm^ -r(t + j) Au( / + j-1)]2|

[0061] in which: - A u is the increment of the control signal for adding microcarriers to the tank minimizing a criterion J, taking into account the various constraints; - y(t + denotes the prediction of the cell confluence rate at time t + j as a function of the measurements up to time z; - r(t+ j) denotes a reference signal at time t+j of the predetermined setpoint kinetics; - Nt denotes the beginning of a prediction horizon; - N2 designates the end of the prediction horizon; - Nu designates a calculation horizon for the control signal; and - A is a predetermined weighting term

[0062] The prediction horizon: N = A2 - 2Vj can be greater than or equal to 1 and / or less than or equal to 50. For example, N can be equal to 10. The weighting term can be predetermined by prior comparison trials. The weighting term can be greater than or equal to 0 and / or less than or equal to 10. For example, the weighting term can be equal to 1.

[0063] After the determination S33, the addition S30 may include the addition S34 of the determined quantity of microcarriers to the vessel. The process may perform steps S31 to S33 at the time interval t (using the notation introduced previously). The process may perform the addition S34 of the determined quantity of microcarriers during The time interval t+1 (the time interval following the time interval t during which the prediction is performed). The addition S34 of the determined quantity of microcarriers to the tank may include applying the control signal for adding microcarriers to the tank (i.e., minimizing the result of the objective function). The addition of microcarriers to the tank may be performed by a pump. The control signal may be a control signal for this pump. For example, the pump may include a valve, and the control signal may define an opening value for this valve (the pump valve being configured so that the opening value changes the flow rate of microcarriers added to the tank). The opening value may be applied during the time interval t+1.

[0064] The microcarriers can be of any shape. The microcarriers can be spheres and / or disks. For example, all the microcarriers in the tank (initially present or added) can be spheres. Alternatively, all the microcarriers in the tank can be disks. Alternatively still, the tank can comprise a mixture of spheres and disks. The microcarriers can have a diameter between 50 micrometers and 1000 micrometers.

[0065] In examples, the bioreactor may include an agitator. The agitator may be any device capable of agitating the contents of the vessel. For example, the agitator may be a propeller. Alternatively, the agitator may be a flow agitator or an external device such as a wave bioreactor. The bioreactor may also include at least one sensor for measuring the quantity of particles suspended in the vessel. This sensor may be configured to measure the absorbance and / or turbidity of the vessel container. For example, the bioreactor may include two inline sensors: an absorbance sensor for the vessel container and a turbidity sensor for the vessel container. The process may include a measurement of the quantity of particles (for example, in real time, at each interval t).The measurement may include a calculation of the cell confluence rate based on the result of at least one sensor present in the cuvette (for example, based on the measured absorbance and / or turbidity).

[0066] When the bioreactor includes an agitator and at least one sensor for measuring the quantity of particles suspended in the vessel, the process may further include a dynamic adjustment of the agitator based on the quantity of particles suspended measured by at least one sensor. The dynamic adjustment of the agitator may include, for each time interval, a measurement of the quantity of particles suspended at a time interval (or time step) t, and then, the determination of a control signal for the agitator to be applied for the following time interval t+1. After that, the dynamic adjustment may include the application of the determined control signal at the time interval t+1 (and Also at this time, a repetition of the measurement and signal determination for the following time interval (t+2) is performed. The determined control signal can be such that the amount of suspended particles remains within a given range of values. For example, the determined control signal can increase agitation when the amount of suspended particles is below the values ​​in the given range and decrease agitation when the amount of suspended particles is above the values ​​in the given range.

[0067] Dynamic agitation control contributes to improving the quality and quantity of biomass produced in the reactor. Indeed, the quantity of suspended particles is one of the parameters influencing the quality and quantity of biomass produced because it increases the culture surface area. Thus, it increases the bioreactor's efficiency while preventing excessive cell confluence that could lead to cell degradation. Adjustments to the quantity of microcarriers and the agitation speed are interdependent. As the number of microcarriers increases, the number of cells also increases, and due to this greater mass, it becomes necessary to increase the agitation speed to maintain overall homogenization above a critical value. If the agitation is too low, all the microcarriers and cells will gradually settle to the bottom of the bioreactor.Specifically, dynamic agitation control (in addition to adjusting the amount of microcarriers) helps reduce cell damage. Dynamic agitation control allows microcarriers to remain in suspension, preventing cell rupture (which can occur with excessive agitation).

[0068] In particular, the hydrodynamic effect, agitation parameters, and agitation speed influence cell damage. Specifically, hydrodynamic death mechanisms result from collisions of a cell-coated microcarrier with other beads, collisions with reactor components (primarily the propeller), and interaction with turbulent eddies. Cell death increases with agitation power and decreases with fluid viscosity. Automatically adjusting agitation in the bioreactor therefore reduces collision damage and more accurately replicates settings between different batches produced in the bioreactor.

[0069] With reference to Figures 2 to 11, examples of implementation of the process will now be described.

[0070] Figure 2 illustrates an example of a control system. The control system controls a pump 102 to automatically adjust, during the culture time, the quantity of microcarriers to be added to a bioreactor which includes a cell culture tank 101 containing cells, microcarriers and a medium of culture, but also a cell confluence rate sensor 103 which sends measurement data to the control system 104. This sensor can be a near-infrared spatial resolution spectroscopic probe, such as SAM-Spec manufactured by Indatech (Chauvin-Amoux), or a capacitance probe like Hamilton's Incyte-Arc. From the information from these sensors, the control unit 104 calculates the microcarrier feed rate so that the cell confluence measured at the surface of the microcarriers closely follows a predefined setpoint kinetic and does not exceed a predefined upper limit.

[0071] Figure 3 shows an example of monitoring the predetermined setpoint kinetics. Figure 3 shows the measured cell confluence 110 at the surface of the microcarriers and the predetermined setpoint kinetics 111. The process determines at each time step the quantity of microcarriers to add to the tank so that the cell confluence in the tank follows the predetermined setpoint kinetics 111.

[0072] To carry out the operation of controlling the time trajectory of the confluence rate of the microcarriers, the central control unit may include software which, depending on the differences between the measurement signal of the confluence rate and the desired reference trajectory of this same quantity, calculates the adjustments (variations) to be made on the supply rate of bare microcarriers, in order to reduce this difference to the maximum.

[0073] Figure 4 illustrates an example of a control system including regulation agitation. The multivariable bioreactor control system comprises a biomass culture tank 201, a temperature sensor 207, a pH probe 209, a dissolved oxygen sensor 208, a biomass analysis probe 213, an optical density sensor 214, a turbidity sensor 215, a sodium hydroxide (NaOH) feed pump 206A, mass flow regulators 206B, 205A, 205B for air, nitrogen, and CO2 supplies, a heating element 204, and a culture medium and microcarrier feed pump 203. The bioreactor contents are continuously mixed by an agitator 202. Temperature 210, dissolved oxygen 211, and pH 212 are controlled by three independent control loops acting on actuators 203, 204, 205, 206.These five critical process parameters – medium temperature, pH level, dissolved O2 concentration, agitation power and quantity of microcarriers – have combined effects on the quantity and quality of the biomass produced. A central control unit 214 simultaneously calculates the modifications to be made to the setpoint values ​​of the local controllers managing the temperature, pH level and dissolved oxygen concentration (dO2), as well as to the microcarrier feed rate and the agitator speed such that: .

[0074] - the quantity of biomass measured 220 by a probe inside the bioreactor always follows the same reference trajectory 221 during the culture time. The culture trajectory is imposed by the user responsible for the bioreactor;

[0075] - the entire set of microcarriers is kept in suspension without increasing unnecessarily increasing the agitation power, which would result in shocks and deterioration of the condition of the cells on the microcarriers;

[0076] - cellular confluence at the surface of microcarriers follows a kinetic predefined and that it does not exceed a predefined upper limit;

[0077] - the values ​​of the temperature and dissolved oxygen concentration variables, agitator rotation speed and pH level are always maintained within a range of acceptable values, called Control Operating Region 232, a sub-region of the "Design Space" 231 in accordance with international guidelines in Quality by Design (ICH Q8 to Q12) (see the ranges of acceptable values ​​in accordance with international guidelines shown in [Fig.5]).

[0078] The central control unit includes software which, based on the deviations between the measurement signals and the desired reference trajectories, simultaneously calculates the adjustments (variations) to be made to the setpoints of the local controllers, in order to reduce these deviations while maintaining the values ​​of the control variables within a working region.

[0079] The control software may incorporate a predictive control algorithm. The software may include instructions to determine the quantity of microcarriers to add at each time interval. Figure 6 shows an example of a block diagram illustrating the predictive control.

[0080] Compared to a conventional closed-loop control scheme, the setpoint r of the predictive control contains a quantity at different times in the future: t + 1, , t + N2, in order to form a horizon corresponding to the reference signal. This horizon of N2 future setpoint (or reference) values, denoted +, is compared at each time step (iteration) to the N2 future predicted values ​​jAt + i) of the output variable, which here corresponds to the average cell confluence rate of the microcarriers.

[0081] The difference between the desired values ​​Xf + i) and the predicted values ​​y{t + i) is then used by the optimization algorithm to generate a horizon of future control signal values ​​Nu, denoted , where only the first value corresponding to time t+1 is applied to the actuator, here the microcarrier feed pump. This control horizon is then recalculated at each time r to avoid discrepancies between the predictions and the observed reality. Nu is a control unit parameter that can be set by an automation engineer during the implementation and testing phases of the control unit on site.

[0082] An example of a minimization algorithm is now discussed. The bioreactor may include an MPC (Model (Based) Predictive Control) regulator which is constructed from 3 main elements: a prediction model, a target function and a control law.

[0083] The prediction model can be a dynamic model of the bioreactor described by a state-space representation. Figure 7 shows an example of the dynamic model. This dynamic model can consist of three matrices: a state matrix (A), an input matrix (B), and an output matrix (C). It can be a model belonging to the family of black-box models, that is, an empirical model whose matrix contents are determined from training data collected during experiments performed on the system to be controlled. This model makes it possible to establish predictions γ on the cell confluence rate of the microcarriers, predictions used in the MPC regulator.

[0084] The objective criterion may be of the form:

[0085] minj(Au) = min{£^ -rit + j) r + £*“X[ Au(Z + jl) ]2}

[0086] which describes a mathematical optimization problem where the objective is to find the best increment of the control signal A u that minimizes the cost function J, taking into account various constraints. This function consists of the following elements:

[0087] - y(t + jj / ) denotes the prediction of the output (cell confluence rate) at at time t+j, knowing all the measurements up to that time

[0088] -r(t+j) denotes the reference signal at time t + j.

[0089] - designates the beginning of the prediction horizon.

[0090] - N2 denotes the end of the prediction horizon.

[0091] - Nu designates the calculation horizon of the control signal.

[0092] - 2 is a weighting term used to choose between pursuit performance and energy cost expended on the input signal. Like 2VW, this may be a parameter to be adjusted by the automation engineer during the testing phases.

[0093] This objective function is composed of 2 terms:

[0094] - the first having the role of minimizing the quadratic error between the output y(t) and the reference r(t), as illustrated in the diagram in [Fig.8], for better trajectory tracking accuracy;

[0095] - the second having the role of minimizing the energy cost on the signal of command and allows obtaining a physically feasible control signal.

[0096]

[0097]

[0098] The control law can correspond to the optimal control value: the solution to the objective criterion, determined by solving the algorithm. optimization, allowing the calculation of the next value of the control signal, not^ei^i) + A lit)' Table 1 below shows the contents of an example MPC regulator optimization algorithm that minimizes a target function while taking into account a set of constraints. Choose a starting point xÿ from among the set of feasible solutions; Define W as the active set at x. aa For k - 0, 1, 2... do Solve the objective function to obtain &uk If = 0 (or close enough), then Calculate the Lagrange multipliers A, for the active set W, . AI ! If k > 0 for all i belonging to W then Point xk is the optimal solution. Return x, IK Otherwise Remove constraints with Lagrange multipliers in W negatives then do ! negatives then do ! W, and ï, x, i k+1 fc+l k. END i Otherwise i Calculate the optimal step size ak as well as the search direction p. i to au, Au, + a, p ik krk i If there are blocking constraints, then i Obtain by adding one of the blocking constraints to W^. Otherwise I i W <- VF i k+1 k END END END

[0099] Any optimization problem such as this one is defined using an objective function but also a set of constraints that define the set of feasible solutions, which corresponds to the set of xi to be tested in order to find the optimal solution to the problem. Let xt be a feasible point; a constraint xt is 0^Xkj. said to be active at point i / jn and inactive at point xk i / \ n. The set The active set Wk at point xk is the set of constraints active at the current point. For each point xk and its associated active set Wk, optimization is performed using the Lagrangian method. The optimal solution is found when all Lagrange multipliers are positive. Otherwise, this indicates that the objective function can still be optimized by deactivating the constraints corresponding to negative Lagrange multipliers. The algorithm then performs a step size ak Pk, where ak is the step length parameter, corresponding to the largest value in the range [0,1] for which all constraints are satisfied, and Pk the pitch direction parameter. The constraints / 1 for which the minimum is ® A kj The constraints reached are called blocking constraints: if = 1 and no new constraint is active at x*+i, then there is no blocking constraint at that iteration. If < 1, this means that the step along Pk has been blocked by a constraint external to Wk. In this case, a new active set is formed by adding one of the blocking constraints to Wk. The algorithm is then repeated for each iteration, adding constraints to the active set Wk until the optimal solution, which minimizes the objective function on its active set Wjt, is found.

[0100] The dynamic model used to estimate the predictions y of the output variable (cell confluence rate) can be a black-box state model determined from a supervised machine learning method based on subspace decomposition (for example, with the subspace decomposition algorithms described in Van Overschee and De Moor, 1996 and / or Bastogne, 1997). Commercial software tools exist for implementing these decomposition methods and allow a state model to be determined directly from the training data, notably, for example, in the System Identification toolbox of the Matlab software from Mathworks.

[0101] The determination method may be executed before the process is executed. Alternatively, the determination method may be included in the process, which may include, before production, an execution of the determination method to determine the state model that will be used, during production, to make predictions.

[0102] The determination method can be based on three steps.

[0103] The first step may include conducting experiments dedicated to acquiring training data. The experiments may be based on stimulating the input signal (the microcarrier feed rate) with a pulse signal of varying heights and widths, such as that shown in [Fig. 9]. These parameters may be manually adjustable. In the field, an automation engineer, assisted by a bioprocess engineer. During the experiments, the measurement signals from the confluence sensor can be recorded and samples can be taken to measure in the laboratory, using a reference technique, an average number of cells per microcarrier: y(t).

[0104] The second step may include calibration of the measurement signals. This step may include determining a calibration function fQ to reconstruct a y-f(m) estimate of the cell confluence rate. The good practices defined in ICH Q14 (ICH Management Committee, 2018) and Q2 (Guideline and others, 2005) guidelines may be used for this purpose. Figure 10 shows an example of such calibration of confluence rate measurements.

[0105] The third step may include identifying the state model. The third step may consist of applying subspace decomposition algorithms (e.g., such as those described in Van Overschee and De Moor, 1996 and Bastogne, 1997) to the signals (“( / ), y(t)) to estimate the state model matrices. following :

[0106] j üdt 4- 1. ) "A'dOi.&MFwfO (MG Orfc L GO)

[0107] where x(t) is an n-dimensional state vector corresponding to the model order. This order is also estimated by the identification algorithm. M(t) are independent, centered Gaussian random variables with constant variances, representing measurement and state noise. These random variables allow us to describe the uncertainty in the confluence rate predictions.

[0108] The dynamic model used to estimate the predictions of the output variable (cell confluence rate) can also be a white-box model, i.e. a model using biological laws of cell culture to describe the behavior of the bioreactor, such as that defined in the equations below, which considers a first nonlinear dynamic equation representing the cell concentration X(t) and a second nonlinear differential equation representing the substrate concentration S(t).

[0109] n ■ A (i) - ■ A (i) ""ï"' = ■ X(i) - • S(M

[0110] in which:

[0111] Ls is the substrate yield coefficient,

[0112] S in is the substrate concentration of the inlet stream,

[0113] F s is the flow rate of the inlet stream,

[0114] V(t) is the volume of the bioreactor, and

[0115] is the cell growth rate. This parameter is itself a function of process parameters such as temperature (T), pH, dissolved oxygen (DO2) in the bioreactor, but also some key nutrients in the culture medium such as the amount of glucose and glutamine.

[0116] The process has several advantages and consequences for the quality and yield of bioproduction. First, it allows control of cell colonization on the surface of the microcarriers and, more specifically, the phenomenon of cell confluence, which can lead to rapid degradation of the cells' biological state if it exceeds a certain level. Second, it allows the bioreactor to follow a predefined kinetic profile of average confluence during culture. Third, it improves the reproducibility of production between batches.

[0117] An example of adjusting production parameters in the bioreactor (temperature, pH level, dissolved oxygen concentration, substrate quantity, nutrient concentration, stirring speed, and microcarrier quantity) is now discussed. Parameter adjustment relies on predicting the evolution over time of critical quality attributes (output variables to be controlled) such as cell confluence, the number of live cells, and the population doubling level. This prediction is based on a dynamic model of the bioreactor, such as the state-space models described previously.

[0118] The bioreactor control can incorporate a Model Predictive Control (MPC) algorithm, which adjusts the values ​​of critical process parameters in real time to follow predefined trajectories of critical quality attributes. Figure 11 illustrates an example of such an MPC controller. This MPC optimization algorithm calculates the best fit for the process parameter values ​​based on the previously described bioreactor model prediction in order to minimize the error between the predicted kinetics, such as Cc(t), X(t), and PDL(t), and the desired trajectory of the same critical quality attributes: Cc(t), X(t), and PDL(t). This controller also minimizes energy and material costs.

[0119] The MPC controller contributes to improving biomass production in the bioreactor by enabling the following of predefined ideal trajectories for certain critical quality attributes such as cell confluence (Cc(t)), the number of live cells (X(t)), and the population doubling level (PDL(t)). This controller also allows for the consideration of value range constraints on the input variables (critical process parameters).

Claims

Demands

1. A method implemented by a control system (104) for controlling cell production in a bioreactor, the cells produced in the bioreactor not being human embryonic stem cells, the bioreactor comprising a cell culture tank (101) containing a culture medium and microcarriers on the surface of which the cells grow, the method comprising, during production, a dynamic adjustment (S10) of parameters including at least: - a temperature of the culture medium; - a pH of the culture medium; - a concentration of dissolved oxygen in the culture medium; and - a quantity of microcarriers present in the tank, wherein the dynamic adjustment (S10) of the quantity of microcarriers comprises: - a measurement (S20) of the cell confluence rate (110) on the surface of the microcarriers in the tank;and - an addition (S30) of microcarriers to the tank so that the measured cell confluence rate (110) follows a predetermined setpoint kinetic (111). in which production takes place over a period divided into successive time intervals, the addition (S30) of microcarriers to the tank comprising, for each time interval: - a prediction (S31) of an evolution of the cell confluence rate in the tank; - a comparison (S32) of the predicted evolution with the predetermined setpoint kinetic; - a determination (S33) of a quantity of microcarriers to be added for the next time interval as a function of the result of the comparison; and - an addition (S34) of the determined quantity of microcarriers to the tank.

2. Method according to claim 1, wherein the addition (S30) of microcarriers into the tank is carried out by a pump (102).

3. Method according to claim 1 or 2, wherein the measurement of the cell confluence rate is carried out by an inline probe (103).

4. A method according to any one of claims 1 to 3, wherein the prediction is based on: - a dynamic model comprising a state matrix, an input matrix and an output matrix, the contents of the matrices being determined from training data, or - a transparent-box type nonlinear state model using biological laws of cell culture to describe the behavior of the bioreactor.

5. A method according to any one of claims 1 to 4, wherein the determination of the quantity of microcarriers to be added comprises the application of an optimization algorithm based on a target criterion, the target criterion preferably having the following formula: imnj(Au) = min|EGvJjG+jk) -nt+j) r+L*”X[ A u(r+j -1 ) ]2} in which: - Au is the increment of the control signal for adding microcarriers to the tank minimizing a cost function J, taking into account the various constraints; - y(i + j^) denotes the prediction of the cell confluence rate at time t + j as a function of measurements up to time z; - r(t + j) denotes a reference signal at time t + j of the predetermined setpoint kinetics to be pursued; - iVj denotes the beginning of a prediction horizon; - N2 denotes the end of the prediction horizon; - Nu designates a calculation horizon for the control signal;and -2 is a predetermined weighting term.

6. A method according to any one of claims 1 to 5, wherein the microcarriers are balls and / or discs, the microcarriers having a diameter between 50 micrometers and 1000 micrometers.

7. A method according to any one of claims 1 to 6, wherein the bioreactor comprises a culture medium agitator and at least one sensor for measuring the quantity of particles suspended in the tank, the method further comprising: - a dynamic adjustment of the culture medium agitator as a function of the quantity of particles suspended measured by at least one sensor.

8. A computer program comprising instructions which, when the program is executed by a bioreactor control system, cause the latter to implement the process according to any one of claims 1 to 7.

9. A control system for a bioreactor such as a SCADA system, the system comprising a processor coupled to a memory, the memory having stored the computer program of claim 8, the processor being configured to execute the instructions of the computer program in order to control the bioreactor according to the method of any one of claims 1 to 7

10. 1 d / . Bioreactor incorporating a control system according to claim 9.

Images