Method for operating the clarification setup

Abstract

The present invention relates to a method for operating a clarification setup (1) of a bioprocessing facility (2), the clarification setup (1) comprising a fluidized bed centrifuge (3) and a pumping device (4), the fluidized bed centrifuge (3) comprising at least one centrifuge chamber (5) rotated about a geometric centrifuge axis (6), the bioprocessing facility (2) comprising an electronic process controller (7) for controlling the fluidized bed centrifuge (3) and the pumping device (4), the fluidized bed centrifuge (3) comprising a particle loading cycle (8) and a pumping device (9), the fluidized bed centrifuge (3) comprising at least one centrifuge chamber (5) rotated about a geometric centrifuge axis (6), the bioprocessing facility (2) comprising an electronic process controller (7) for controlling the fluidized bed centrifuge (3) and the pumping device (4), the fluidized bed centrifuge (3) comprising at least one centrifuge chamber (5) rotated about a geometric centrifuge axis (6), the bioprocessing facility (2) comprising at least one ... and / or a particle washing cycle (9) and a reverse operation for a particle discharge cycle (10), and if a particle loading cycle (8) is provided, the cell broth loaded into the centrifuge chamber (5) progresses to form a growing particle accumulation in the centrifuge chamber (5), and the clarification setup (1) includes a monitoring sensor device (11) with at least one optical sensor (12) for generating monitoring sensor data transmitted to an electronic process control device (7). It is proposed that in a monitoring routine (13), image-related data (14) representative of an optical image (15) of the centrifuge chamber contents (16) is generated by the monitoring sensor device (11) and a particle filling level (17) is calculated by the electronic process control device (7) from the image-related data (14) based on a computational model (18).

Description

[Technical field]

[0001] This application has been filed as a PCT international patent application dated January 11, 2023 in the name of Sartorius Stedim North America Inc., a U.S. corporation, as applicant in all countries, and Robert Soeldner, a German national, and Jonas Austerjost, a German national, as inventors and applicants in all countries. This application claims priority to EP Application No. 22151042.3, filed January 11, 2022, the contents of which are incorporated herein by reference in their entirety.

[0002] The present invention relates to a method for operating a clarification setup of a bioprocessing installation according to the general part of claim 1, a clarification setup according to claim 15 and the use of a centrifuge chamber for carrying out said method according to claim 16. [Background technology]

[0003] The term "bioprocess" currently refers to any kind of biotechnological process, especially biopharmaceutical processes. An example of such a bioprocess is the use of a bioreactor to cultivate microorganisms or mammalian cells under given conditions, where the cell broth is transferred from the bioreactor to a downstream process. The term "bioreactor" in this context means any manufactured device or system that supports a biological environment by allowing the monitoring and control of at least one parameter. The term "biological product" currently refers to any kind of compound produced in such a bioprocess. Examples of such biological products are proteins, especially antibodies, growth factors or hormones, metabolites or any other molecules, as well as cells or cell components, e.g. organelles.

[0004] The method for operating a clarification setup can be applied in various fields of biotechnology. High efficiency in this field has been driven by the increasing demand for biological products such as biopharmaceuticals. Efficiency in this sense is not only related to the cost-effectiveness of the components used but also to the controllability of the processes involved. The method relies on centrifuge chamber filling level determination to optimize the time efficiency as well as the controllability of the bioprocess.

[0005] A known method for operating a clarification setup (EP 2310486) comprises a fluidized bed centrifuge with a number of centrifuge chambers, each of which is assigned a chamber inlet and a chamber outlet. The clarification setup also comprises a pumping device assigned to the fluidized bed centrifuge and a liquid network for transporting the liquid, which is the cell broth to be clarified. Finally, the clarification setup comprises an electronic process control device for controlling at least the fluidized bed centrifuge and the pumping device.

[0006] In order to accurately determine the particle loading level of a centrifuge chamber in a fluidized bed centrifugation process, currently it is necessary to provide at least the fluid flow throughput as well as cell suspension properties such as cell diameter, cell concentration, etc., to calculate the respective particle loading level at a given time.

[0007] The term "particle" should be understood in the broad sense and means any particulate solid compound which may be of biological or non-biological origin, such as in particular cells, cell debris, proteins, (magnetic) microbeads, etc.

[0008] The term "particle filling level" should be understood in a broad sense. Particle filling level is a measure of the filling of the centrifuge chamber and can be reflected by the particle filling rate of the centrifuge chamber, a specific particle mass, volume, concentration, etc. Furthermore, the particle filling level can be expressed by a specific particle filling level at a given time or as a change in particle filling level over time, thereby also reflected in the particle filling rate.

[0009] However, this only serves as an approximation and if the fill level is underestimated in the fluidized bed centrifugation process, the system may be erroneously overfilled and cells may be introduced into the subsequent downstream process, resulting in inefficiencies in the fluidized bed centrifugation process, whereas if the fill level is overestimated, the fluidized bed centrifugation process will be inefficient due to a higher number of centrifugation cycles required to process a particular volume, such as a particular bioreactor volume.

[0010] Within the known methods, the calculation of the filling level is time-consuming and not even very accurate, which impairs controllability to some extent, so that the efficiency and reproducibility of the overall process can only be reached to a certain level.

[0011] Another known method, which is the starting point of the present invention ("Operation and Maintenance Manual kSep400", Sartorius Stedim Biotech), relies on the generation of monitoring sensor data from a camera sensor, which is manually analyzed by a user. Based on the user's experience, the monitoring sensor data are utilized to operate the clarification setup. Again, this method is time consuming and has limited reproducibility. Summary of the Invention [Problem to be solved by the invention]

[0012] It is therefore an object of the present invention to improve the known methods for operating a clarification set-up in such a way that the bioprocess efficiency and reproducibility are improved with as little effort as possible. [Means for solving the problem]

[0013] The above mentioned problem is solved in a method for operating a clarification setup as set out in the general part of claim 1 by the features of the characterizing part of claim 1.

[0014] Due to the suspension state of the particles in the centrifuge chamber, particle accumulation proceeds in an orderly and reproducible manner. To a large extent, this effect occurs independently of the medium being treated. In particular, it was found that this results in a visible phase boundary, which moves across the centrifuge chamber with the ongoing particle agglomeration, the position of the phase boundary representing the target filling level of the centrifuge chamber. It is particularly interesting that, in contrast to known non-fluidized bed centrifuges, the proposed invention has been realized to result in a visible phase boundary without subjecting the centrifuged cells to such high acceleration forces that could otherwise lead to cell damage and even cell death. Thus, the cells are separated here under milder conditions, leading to a higher viability, which is of crucial importance if the cells are to be reused in a subsequent bioprocess. Furthermore, the proposed invention allows a more accurate estimation of the number of cells in the centrifuge chamber. This is due to the fact that particle agglomeration under non-fluidized bed conditions is fundamentally different due to the much higher acceleration forces acting on the cells. For this reason, in the proposed solution, the cells of the particle agglomeration are not packed as densely as they would be under non-fluidized bed conditions. This allows for a more accurate calculation of particle loading levels, which is particularly advantageous for more accurate calculation of the required cell numbers required to inoculate a potential subsequent bioprocess. Furthermore, it has been found that the above-mentioned systematic particle aggregation allows for a systematic calculation of loading levels based on a computational model, which can also be performed in real time.

[0015] In particular, it is proposed that in the monitoring routine image-related data representative of an optical image of the centrifuge chamber contents is generated by a monitoring sensor device, and the particle loading level is calculated by an electronic process control device from the image-related data based on a computational model.

[0016] The achievement of the present invention is the automatic determination of the filling level by a camera, preferably in real time, accelerating the manufacturing process in a bioindependent, cost-effective and non-invasive manner. The real-time determination of the centrifuge chamber load reduces unnecessary waiting time, minimizes the number of centrifugation cycles required to process a bioreactor volume, and prevents particle breakthrough of valuable cells. The term "particle breakthrough" refers to the point in time during the fluidized bed centrifugation process when at least one centrifuge chamber is filled to capacity with particles, preferably cells, and continued loading of the centrifuge chamber leads to overfilling of said centrifuge chamber.

[0017] By using automatic filling level determination by image processing, an accurate signal for each chamber can be obtained. As an image-related approach, the determined filling level is independent of broth composition such as amount of viable cells, cell type and cell diameter, and can even reflect the presence of impurities or dead cells.

[0018] Claims 2 and 3 are directed to preferred embodiments of the computational models used in the monitoring routines. According to these embodiments, different optical properties or phase boundaries in the centrifuge chamber are used to solve the problem of determining the particle loading level in a biologically independent manner.

[0019] Claim 4 defines a particularly preferred embodiment regarding the translucent properties of the centrifuge chambers. Such a design makes it possible to determine the particle loading level in a non-invasive manner. In particular, due to the translucent properties, there is no need to provide windows or the like in each centrifuge chamber.

[0020] The preferred embodiments as defined in claims 5 and 6 refer to different design and installation options of the camera unit as at least one monitoring sensor of the monitoring sensor device, which allows increased flexibility not only in the choice of camera but also in terms of the respective installation.

[0021] Particularly preferred embodiments for the definition and calculation of the particle loading level are defined in claims 7 and 8. These specifications allow for an optimized and simplified determination of the particle loading level.

[0022] Claims 9 and 10 are directed to preferred embodiments relating to options for the calibration sensor device and the calibration routine. These features are particularly advantageous for calibrating the monitoring sensor device and thus improving the accuracy of the computational model as well as the determination of the absolute maximum particle loading level.

[0023] According to a particularly preferred embodiment of the first alternative of claim 11, at least one parameter of the clarification setup is adjusted by the electronic process control device based on the particle filling level, allowing for uniform filling of each centrifuge chamber, which allows for simultaneous emptying of all centrifuge chambers and thus a more efficient use of the centrifuge chamber capacity with less effort.

[0024] A preferred embodiment according to the second alternative of claim 11 refers to a decision point based on which the electronic process control device initiates a centrifugation cycle.

[0025] According to a preferred embodiment of claim 12, a nominal maximum particle filling level and / or a nominal zero particle filling level are stored in the electronic process control device, respectively defining an offset for preventing particle breakthrough as well as unnecessary waiting times with high efficiency.

[0026] A preferred embodiment according to claim 13 is directed to a downstream setup in the bioprocessing facility, and the electronic process control device adapts at least one parameter of the downstream setup in the adaptation routine based on a predefined adaptation strategy. This is particularly advantageous, since it allows the downstream method to have not only increased flexibility in the selection of the subsequent downstream unit, but also reduced waiting time. Thereby, the subsequent biological product purification can be individually customized according to the needs of a specific biological product, resulting in increased bioprocessing efficiency and reduced process costs with high flexibility.

[0027] Claim 14 is directed to a preferred embodiment in which the presence of contaminants is detected during the analytical routine. These specifications make it possible to obtain an additional level of information that helps to distinguish between live and dead cells, as well as contaminants, which is essential for determining, for example, the number of cells required for reuse in a subsequent bioprocess.

[0028] According to the teaching of the second independent claim according to claim 15, a clarification set-up for carrying out the proposed method is claimed per se. All explanations given for the proposed method are fully applicable to the proposed clarification set-up.

[0029] The teaching of the third independent claim as defined in claim 16 is directed to the use of the proposed centrifuge chamber for carrying out the proposed method. All the explanations given above are fully applicable to the proposed centrifuge chamber.

[0030] A fourth teaching, which may be stated as an independent claim, is directed to an electronic process control device, which is designed to implement the proposed method according to any one of the preceding claims. Again, all the explanations given above apply in their entirety to this proposed fourth teaching. The electronic process control device preferably includes a data processing system for implementing the proposed method.

[0031] A fifth teaching, which may also be stated as an independent claim, is directed to a computer program product for the proposed electronic process control device, which is configured to implement the proposed method, in particular the above-mentioned routines. Again, all explanations given for the proposed method are fully applicable to the proposed computer program product.

[0032] A sixth teaching, which may also be written as an independent claim, is directed to a computer-readable storage medium on which a computer program is stored. Here too, all the explanations given for the proposed method are fully applicable to the proposed readable storage medium.

[0033] In the following, one embodiment of the present invention will be described with reference to the drawings. [Brief description of the drawings]

[0034] [Figure 1] FIG. 1 is a schematic diagram of a preferred embodiment of the proposed bioprocessing facility in which the proposed method can be performed. [Diagram 2] FIG. 2 is a perspective view of the operating principle of the method proposed according to FIG. 1. [Diagram 3] FIG. 2 shows a flow chart illustrating the preferred core operating principle of the method proposed according to FIG. 1. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0035] The proposed method for operating the clarification setup 1 of a bioprocessing installation 2 is preferably assigned to the upstream and downstream processes of a bioprocess treating liquids, in particular cell broths for cell culture and / or bioproduction.

[0036] The term "liquid" should be understood in a broad sense: it includes not only pure liquids but also emulsions and suspensions, e.g. heterogeneous mixtures of at least two different liquids or heterogeneous mixtures of solid particles and liquids.

[0037] The term "cell broth" refers to a suspension of particles in a solvent, specifically cells and / or cell debris in a medium. It specifically describes the entire culture medium and the respective organisms cultured in the culture medium.

[0038] The term "upstream processing" includes all steps related to cell banking, inoculum (seed train) development, medium development, growth rate optimization and the cultivation process itself, as well as the corresponding in-process control. Cell harvesting can be considered as both part of upstream processing and part of downstream processing. The term "downstream processing" includes all steps related to the recovery and purification of biological products, especially biopharmaceuticals, from natural sources such as animal or plant tissues or cell broths, including recycling of recoverable components and proper treatment and disposal of waste products.

[0039] Generally, cell culture is currently used to produce biopharmaceuticals, especially proteins, such as human insulin, growth factors, hormones, vaccines, or antibodies, antibody derivatives, etc. The biological product may also be a non-biological drug, such as enzymes for food processing, enzymes for laundry detergent, biodegradable plastics, or biofuels. The focus of the present invention is the biopharmaceutical products secreted from cells into the supernatant, such as antibodies or exosomes. Additionally or alternatively, the product may be the cells themselves, especially mammalian cells, including stem cells, or immune cells, such as CAR-T cells for cancer treatment.

[0040] As shown in Figures 1-3, according to all embodiments, the proposed method for operating a clarification setup 1 of a bioprocessing installation 2 employs at least one fluidized bed centrifuge 3 for clarification of cell broth by centrifugation and a pumping device 4 assigned to the fluidized bed centrifuge 3. The pumping device 4 comprises at least one pump. The fluidized bed centrifuge 3 comprises at least one centrifuge chamber 5, preferably an even number of centrifuge chambers 5, more preferably just four centrifuge chambers 5, which are rotated around a geometric centrifuge axis 6, resulting in a fluidized bed during normal operation. Preferably, each centrifuge chamber 5 comprises at least one assigned pump for pumping liquid into and out of the chamber. The bioprocessing installation 2 further comprises an electronic process controller 7 for controlling at least the fluidized bed centrifuge 3 and the pumping device 4.

[0041] "Centrifugation" is the term used to describe the sedimentation of particles in an artificially created gravitational field, where large acceleration forces are used to achieve significant reductions in separation times.

[0042] Here, the centrifuge is designed as a fluidized bed centrifuge 3 for carrying out a continuous centrifugation process. A preferred set-up of a fluidized bed centrifuge 3 is described in EP 2 485 846 A1, the contents of which are incorporated herein by reference.

[0043] The fluidized bed centrifuge 3 includes a rotor to which at least one centrifuge chamber 5 is attached, which can be rotated about a centrifuge axis 6, preferably by an electric motor. The centrifuge rotation speed and pump speed are adjustable by an electronic process controller 7 for the purpose of establishing a fluidized bed of particles, such as cells or cell debris, in the fluidized bed centrifuge 3. A fluidized bed is achieved when the centrifugal force on the particles equals the force of the opposing fluid flow, resulting in a net force on the particles of zero.

[0044] According to Fig. 1, the cell broth is led to a fluidized bed centrifuge 3. The fluidized bed centrifuge 3 is operated in forward operation for a loading cycle 8 and / or a washing cycle 9, and in reverse operation for a particle discharge cycle 10.

[0045] "Forward operation" refers to one of the two possible fluid flow directions in a fluidized bed centrifuge and describes an operation that leads to the separation of liquid, such as culture medium and cells, from solid particles. This "forward operation" allows, on the one hand, washing the separated cells with a buffer or medium, preferably a culture medium, more preferably a concentrated medium, and / or, on the other hand, clarification of the cell broth. The goal here is to clarify the liquid supernatant from solid particles, such as cells, cell debris, etc., which are considered as biomass. The product obtained in this forward operation is the supernatant of the cell broth, which contains the desired bioproduct, such as a recombinant protein, in particular an antibody.

[0046] The term "enriched medium" describes a medium that contains higher concentrations of vitamins, growth factors, micronutrients, and carbon sources, nitrogen sources and / or amino acid concentrations, etc., and preferably allows each organism to grow at its maximum growth rate due to optimized nutrient concentrations. Growth factors and micronutrients are included in the medium for organisms that cannot produce all the necessary vitamins themselves. Inorganic nutrients, including trace elements such as iron, zinc, copper, manganese, molybdenum and cobalt, are typically included in unrefined carbon and nitrogen sources, but may have to be added if refined carbon and nitrogen sources are used.

[0047] The loading cycle refers to the cycle of loading the cell broth to be centrifuged into each centrifuge chamber 5 in the forward operation of the fluidized bed centrifuge 3. Thus, during the particle loading cycle 8, the cell broth loaded into the centrifuge chamber 5 progresses to form a growing particle accumulation in the centrifuge chamber 5.

[0048] A wash cycle refers to a cycle during which each centrifuge chamber 5 is washed with medium or buffer during forward operation of the fluidized bed centrifuge 3. This wash cycle preferably serves to supply fresh nutrients to the cells.

[0049] Alternatively, the fluidized bed centrifuge 3 can be operated in reverse operation. "Reverse operation" means the second of the two possible fluid flow directions in the fluidized bed centrifuge 3 and describes an operation that leads to the discharge of separated solid particles, preferably cells. The resulting product in reverse operation is cells in a cell broth.

[0050] The discharge cycle therefore refers to the cycle assigned to the fluidized bed centrifuge 3, during which, in reverse operation, solid particles, preferably cells, are discharged from the centrifuge chamber 5. This discharge cycle 10 serves inter alia for the reuse of cells in a subsequent bioprocess.

[0051] Furthermore, the clarification setup 1 includes a monitoring sensor device 11 with at least one, here preferably an optical sensor 12, generating monitoring sensor data which are transmitted to the electronic process control device 7. The optical sensors 12 are aimed towards the respective centrifuge chambers 5 as shown in FIG.

[0052] Of particular importance to the present invention is that during the monitoring routine 13, monitoring sensor data in the form of image related data 14 representative of optical images 15 of the centrifuge chamber contents 16 are generated by the monitoring sensor device 11. From these image related data 14, the electronic process control device 7 calculates the particle filling level 17 on the basis of a computational model 18.

[0053] The computational model 18 is to be understood as a rule system for calculating the filling level 17 based on the image-related data 14, as will be explained later. In a preferred embodiment, the computational model 18 can be exchanged between two different operation instances or even within one and the same operation instance. Furthermore, the computational model is highly adaptable to different bioprocess settings resulting in altered optical properties, such as those concerning the choice of culture medium used, which may result in different contrasts, different brightness, different densities, etc. According to another preferred embodiment, the computational model can be adapted to different cell types, including different particle sizes, different shapes, different concentrations, etc. This adaptability makes the proposed method very flexible.

[0054] The term "image related data" should be understood in a broad sense. It represents at least one image and in this sense may be a regular photographic representation. Furthermore, the term "image related data" may include monitoring sensor data of other sensors of the monitoring sensor device 11 relating to other properties of the liquid. Such properties may be, for example, flow weight, flow rate, cell type, carbon source concentration, nitrogen source concentration, amino acid concentration, pH, temperature, oxygen concentration, carbon dioxide concentration, electrical conductivity, pressure, DNA concentration, protein concentration or biomass concentration.

[0055] The monitoring routine 13 or other proposed routines are preferably initiated according to a common predefined strategy or an individual predefined strategy. For example, the monitoring routine 13 may be generally initiated when the loading cycle 8 is started. However, it may even be more effective to have the monitoring routine 13 initiated after, for example, a predefined amount of loading time, rather than before a certain amount of particle agglomeration has occurred.

[0056] In the embodiment according to Fig. 2, here preferably in the monitoring routine 13, the particle filling level 17 has been calculated based on different optical properties of the centrifuge chamber contents 16 according to a calculation model 18. These different optical properties are in particular different translucency, different colour and / or different brightness of the centrifuge chamber contents 16 inside and outside the particle accumulation. This may for example be different colours of solid particles, preferably cells, caused by properties specific for a certain particle type, preferably cell type, such as different turbidity, colour, brightness and / or density differences inside and outside the particle accumulation.

[0057] The term “particle accumulation” here refers to the mass of particles accumulated within the centrifuge chamber 5 .

[0058] In a preferred embodiment, in the monitoring routine 13, a phase boundary 19 between the particle accumulation and the particle-free remaining part of the centrifuge chamber contents 16 is detected in the optical image 15 according to a computational model 18. Such a phase boundary 19 may arise due to differences in composition between the particle accumulation and the particle-free remaining part of the centrifuge chamber contents 16, such as differences in density, differences in affinity, e.g. differences in hydrophilic or hydrophobic properties, differences in color, translucency and / or brightness. Then, based on the computational model 18, the particle filling level 17 is calculated from the position of the phase boundary 19. The phase boundary can be detected by image processing, e.g. line recognition algorithms. It can also be detected simply by determining two regions with different optical properties, such as contrast, brightness, etc., in the optical image 15 representing the centrifuge chamber contents 16. These different optical properties preferably represent particle accumulation (e.g. comprising lower brightness) on the one hand and particle-free remaining part (e.g. comprising higher brightness) of the centrifuge chamber contents 16 on the other hand.

[0059] According to another computational model 18, the filling level 17 is calculated by applying a correlation function to the image-related data 14, resulting in an indication of the size relationship between the area of ​​particle agglomerations and the area of ​​the remaining part of the centrifuge chamber contents 16 free of particles in each optical image 15.

[0060] In another preferred embodiment according to Fig. 2, the centrifuge chamber 5 itself is made of a translucent material. Preferably, the centrifuge chamber 5 is designed as a single-use part, preferably made of a translucent and biocompatible plastic or bioplastic material, like PE, PP, PS, PVC, PET, PUR, etc. Thus, the monitoring sensor device 11 is detecting an optical image 15 of the centrifuge chamber contents 16 through the translucent material of the centrifuge chamber 5. This is not only cost-effective, since it is not necessary to introduce a sight glass for the optical sensor 12 in each centrifuge chamber 5, but also very flexible, since different areas of the centrifuge chamber contents 16 can be viewed by the optical sensor 12 without changing the setup.

[0061] According to another preferred embodiment, as can be seen in Figures 1 and 2, the at least one optical sensor 12 of the monitoring sensor device 11 is a camera unit, preferably a 2D or 3D camera. Exemplary 2D cameras used are two-dimensional CCD array sensors used in video cameras and digital cameras and CMOS sensors used in smartphones and tablets. The optical sensor 12 is preferably located inside or outside the fluidized bed centrifuge 3, more preferably inside or outside the centrifuge rotor chamber.

[0062] The viewing direction of the optical sensor 12, here preferably a camera unit, is preferably essentially parallel to the geometrical centrifuge axis 6. However, it may also be inclined, which may be advantageous in terms of optimized use of the available space. The term "essentially parallel" here does not mean an ideal parallel line in the mathematical sense, but in the colloquial sense that the camera unit is arranged in a better part parallel to the geometrical centrifuge axis 6 (see FIG. 2).

[0063] In Fig. 2, two options for the light devices 20, 21 are displayed. In the first option, the light of the light device 21 is transmitted through the centrifuge chamber contents 16 to the monitoring sensor device 11. This improves the reproducibility of the measurements by the optical sensor 12, independent of any other ambient conditions. In this preferred alternative, the light device 21 is positioned essentially opposite the monitoring sensor device 11. The term "essentially opposite" here does not mean an ideal opposite arrangement in the mathematical sense, but an opposite arrangement in the colloquial sense, the light device 21 being, for the better part, positioned opposite the monitoring sensor device 11. This means that the cell broth loaded in the centrifuge chamber 5 is transmitted by the light of the light device 21, which supports unambiguous detection of particle agglutination areas.

[0064] In an alternative embodiment included in Fig. 2, light from a light device 20 is shone on the centrifuge chamber contents 16 and reflected to the monitoring sensor device 11. The reflection is preferably achieved by a reflective element such as a mirror, a reflector, etc. The light source of the light devices 20, 21 is preferably at least one LED, more preferably at least one strobe LED.

[0065] Preferably, the monitoring sensor device 11 and / or the light device 20, 21 are synchronized with the rotation of at least one centrifuge chamber 5. "Synchronization" refers to the coordination of bidirectional events in time for the system to operate in unison.

[0066] According to a particularly preferred embodiment, the stroboscope LED or array of LEDs and / or the camera unit are synchronized with the rotation of the at least one centrifuge chamber 5. In particular, the synchronized operation of the monitoring sensor device 11 ensures that any optical information not related to the centrifuge chamber 5 itself is faded out.

[0067] In another preferred embodiment according to Fig. 2, the particle loading level 17 represents the range between a particle-free state of the centrifuge chamber 5 and a state of maximum particle accumulation, which is the boundary line of particle breakthrough and is defined by the state followed by particle breakthrough as the loading cycle 8 proceeds. According to a particularly preferred embodiment, the maximum particle accumulation state corresponds to an absolute maximum particle loading level and / or the particle-free state of the centrifuge chamber 5 corresponds to an absolute zero particle loading level. The absolute zero particle loading level corresponds to a complete absence of particles, in particular cells, in the centrifuge chamber 5. The absolute maximum particle loading level corresponds to a maximum particle loading level just before particle breakthrough during the loading cycle 8.

[0068] According to another preferred embodiment shown in Fig. 2, the fluidized bed centrifuge 3 comprises a monitoring opening 22 in the monitoring panel through which the centrifuge chamber contents 16 can be monitored. The monitoring opening 22 is assigned and aligned to the optical sensor 12. The fluidized bed centrifuge 3 preferably comprises just one monitoring opening 22 in the monitoring panel in order to monitor the centrifuge chamber contents 16 of all centrifuge chambers 5. In this case, the monitoring panel with the monitoring opening 22 is preferably fixed while the centrifuge chambers 5 are rotated about the centrifuge axis 6.

[0069] Alternatively, according to a particularly preferred embodiment, the fluidized bed centrifuge 3 includes one monitoring opening 22 in the monitoring panel for each centrifuge chamber 5, each monitoring opening 22 being aligned with the assigned centrifuge chamber 5. In this preferred embodiment, each monitoring opening 22 in each monitoring panel moves with the respective centrifuge chamber 5 about the centrifuge axis 6.

[0070] Here, preferably in the monitoring routine 13 , a particle filling level 17 is calculated in accordance with a calculation model 18 from the brightness and / or color distribution over the monitoring opening 22 on the basis of the calculation model 18 .

[0071] It is particularly preferred that the monitoring opening 22 extends along a radial direction relative to the geometric centrifuge axis 6. Such a realization of the monitoring opening 22 reduces the image processing for the monitoring opening 22 and thus simplifies the image processing considerably. In particular, detection of the phase boundary 19 within the monitoring opening 22 is possible with little technical effort.

[0072] In another embodiment, according to Fig. 1, the clarification setup 1 comprises a calibration sensor device 23 with at least one calibration sensor 24 for generating calibration sensor data transmitted to the electronic process control device 7. Here, preferably, the calibration sensor device 23 detects the presence of particles in the liquid line 25 downstream of the fluidized bed centrifuge 3. Additionally or alternatively, the calibration sensor device 23 detects the flow rate of particles in the liquid line 25 downstream of the fluidized bed centrifuge 3. These detections are preferably performed during the loading cycle 8 and / or the washing cycle 9. The at least one calibration sensor 24 is therefore preferably designed as a particle sensor, further preferably as a cell detection sensor and / or a flow rate sensor.

[0073] Preferably, according to Fig. 1, in a calibration routine, the monitoring sensor data of the monitoring sensor device 11 is correlated with the calibrated sensor data of the calibrating sensor device 23 in order to improve the accuracy of the computational model 18. For example, the monitoring sensor device 11 detects a certain particle loading level 17, which is then correlated with the calibrated sensor data. If the calibrating sensor device 23 detects the absence of particles, preferably cells, the correlation gives the information that no particle breakthrough has occurred. Additionally or alternatively, an absolute maximum particle loading level is derived.

[0074] Additionally or alternatively, in a calibration routine, which may be the calibration routine described above or a different calibration routine, the loading cycle 8 proceeds up to and beyond the point of particle breakthrough. The particle fill level 17 calculated at the point of particle breakthrough is stored in the electronic process controller 7 as the absolute maximum particle fill level. In this scenario, if the calibration sensor device 23 detects the presence and / or flow rate of particles, preferably cells, this gives information that particle breakthrough has indeed occurred. This particular particle fill level 17 is then saved in the electronic process controller 7 as the absolute maximum particle fill level.

[0075] According to a particularly preferred embodiment, in the adjustment routine at least one parameter of the clarification setup 1 is adjusted by the electronic process control device 7 based on the particle filling level 17 of the centrifuge chamber 5 according to a predefined adjustment strategy. The at least one adjustment parameter of the clarification setup 1 is preferably the centrifugation speed of the rotor of the fluidized bed centrifuge 3 and / or the volumetric flow rate of at least one pump, preferably all pumps, of the pumping device 4 assigned to the centrifuge chamber 5.

[0076] 3, in another preferred embodiment, based on decision point 26, the electronic process controller 7 in an automation routine initiates a centrifugation cycle according to a predefined automation strategy based on the calculated particle loading level 17. The fluidized bed centrifuge 3 is operated in such a "centrifugation cycle" which includes loading, washing and draining cycles.

[0077] For example, if the monitoring sensor device 11 detects that the measured particle filling level 17 of the centrifuge chamber 5 corresponds to an absolute maximum particle filling level according to the above-mentioned predefined automation strategy based on the calculated particle filling level 17, the electronic process controller 7 initiates a cleaning cycle and / or a draining cycle to prevent particle breakthrough (FIG. 3). According to another embodiment, if the measured particle filling level 17 of the centrifuge chamber 5 corresponds to an absolute zero particle filling level, the electronic process controller 7 initiates a centrifuge loading cycle 8 to prevent unnecessary waiting times.

[0078] In another particularly preferred embodiment, according to Fig. 1, the bioprocessing installation 2 comprises a culture setup 27, in particular a bioreactor, with at least one upstream unit 28 for producing a biological product. In general, when using a continuous upstream process such as perfusion culture, the cell broth level of the upstream unit 28 is preferably static, so that a continuous liquid flow containing cells and / or product can be fed to the fluidized bed centrifuge 3. However, in particular when the continuous upstream process is stopped or when a discontinuous upstream process such as a batch or fed-batch process is used, the upstream unit 28 preferably comprises at least a fill level sensor 29 for generating fill level sensor data 30 of the upstream unit 28.

[0079] The fill level sensor 29 is configured to detect the cell broth fill level in the upstream unit 28. Thereby, the harvest of the upstream unit 28 can be automatically terminated by the electronic process controller 7 when the fill level falls below a predefined lower fill level threshold. The fill level sensor 29 can be any sensor for qualitatively and / or quantitatively determining the presence of cell broth, such as a capacitance sensor, an optical sensor, an air bubble sensor, etc.

[0080] According to Fig. 3, in the information retrieval 31, the electronic process controller 7 retrieves at least the biomass state 32 of the upstream unit 28 from the fill level sensor data 30 generated by the fill level sensor 29. The biomass state 32 thereby indicates whether or not there is still cell broth present in the upstream unit 28 to be treated. The electronic process controller 7 then starts the monitoring routine 13 or the cleaning cycle 9 and / or the discharge cycle 10 based on the derived biomass state 32.

[0081] Here, preferably, as seen in Fig. 3, in information retrieval 31, the electronic process control device 7 retrieves the biomass status 32 of the upstream unit 28 and determines whether the cell broth is fully processed or not yet processed based on decision point 33. If according to the retrieved information the cell broth is fully processed (see "Yes" in Fig. 3), the electronic process control device 7 starts the cleaning cycle 9 and / or the discharge cycle 10.

[0082] However, if the cell broth is not yet completely processed (see "No" in FIG. 3), the electronic process controller 7 proceeds to a monitoring routine 13.

[0083] Additionally or alternatively, according to a particularly preferred embodiment, the upstream unit 28 includes a biomass sensor for generating biomass sensor data of the upstream unit 28. The biomass sensor can be any sensor for qualitatively and / or quantitatively determining the presence of biomass in the upstream unit 28, such as a capacitive sensor, an optical sensor, etc. If the biomass state 32 retrieved in the information retrieval 31 by the electronic process controller 7 of the upstream unit 28 reaches a predefined biomass threshold, such as a cell density threshold, the electronic process controller 7 initiates a loading cycle 8. Thereby, the harvesting of the upstream unit 28 can be automatically initiated once a predefined lower biomass threshold is exceeded, which is particularly advantageous for fed-batch or perfusion processes, where the cells are preferably maintained in a growth state. This automated approach employing a fill level sensor 29 and / or a biomass sensor makes the proposed method very flexible.

[0084] Preferably, a nominal maximum particle loading level is stored in the electronic process controller 7. This nominal maximum particle loading level is preferably below the absolute maximum particle loading level by a predefined upper limit offset. This upper limit offset is preferably up to 25%, more preferably up to 10%, more preferably up to 5%, more preferably up to 1% of the total centrifuge chamber volume. At decision point 26, according to an automated strategy, if the nominal maximum particle loading level is approached during the loading cycle 8, the loading cycle 8 is terminated and a wash cycle 9 or a drain cycle 10 is preferably initiated by the electronic process controller 7.

[0085] Additionally or alternatively, a nominal zero particle filling level is stored in the electronic process controller 7. This nominal zero particle filling level is above the absolute zero particle filling level by a predefined lower offset. This lower offset is preferably up to 25%, more preferably up to 10%, more preferably up to 5%, more preferably up to 1% of the total centrifuge chamber volume. Based on decision point 26, and according to an automated strategy, when the nominal zero particle filling level is approached during the discharge cycle 10, the discharge cycle 10 is terminated and preferably a loading cycle 8 is initiated.

[0086] As shown by Fig. 1, the bioprocessing installation 2 comprises a downstream setup 34 with at least one downstream unit 35. This downstream setup 34 is preferably at least one of the group of a filtration setup, a viral inactivation setup, a chromatography setup and / or a viral filtration setup. The at least one downstream unit 35 is preferably at least one of the group of a microfiltration unit, an ultrafiltration unit, a capture chromatography unit, a viral inactivation unit, a diafiltration unit, an intermediate (purification) chromatography unit, a polishing chromatography unit, a viral filtration unit and / or a sterile filtration unit.

[0087] In the adaptation routine, the electronic process control device 7 adapts, based on a predefined adaptation strategy, at least one parameter of the downstream setup 34 according to the image-related data 14. According to a particularly preferred embodiment, and purely by way of example, this at least one individually adaptable parameter is preferably the selection of the type of chromatography column, the required column volume, the required flow rate, the selection of the required stationary phase of the chromatography column and / or the type of filter, the selection of the required filter pore size and / or the pH for viral inactivation, the duration of viral inactivation, etc.

[0088] Preferred types of chromatography columns are affinity chromatography, in particular Protein A affinity chromatography, ion exchange chromatography (IEX) such as anion exchange chromatography (AEX) or cation exchange chromatography (CEX), hydrophobic interaction chromatography (HIC), size exclusion chromatography (SEC), or any other type of chromatography. Moreover, these chromatography types can be operated in axial or radial flow. The at least one downstream unit 35 comprises at least one chromatography column, preferably a plurality of chromatography columns, set up for multicolumn simulated moving bed (SMB) chromatography.

[0089] According to another preferred embodiment, in the analysis routine, irregularities attributable to the presence of impurities, preferably biological contaminants, are detected on the basis of predefined irregularity features in the optical image 15 .

[0090] The irregularity features in the optical image 15 are coupled with characteristic turbidity within the liquid, which can be easily detected by the camera unit as optical contrast and / or particle agglomerations, particle clouds, changes in color and / or brightness, etc. Preferably, the attributed irregularity state is "contamination".

[0091] Preferably, at least one component of the clarification setup 1, in particular the fluidized bed centrifuge 3, the centrifuge chamber 5, the upstream unit 28 and / or the downstream setup 34, and more preferably all components of the bioprocessing equipment 2, are designed as disposable components.

[0092] According to another independent teaching, a clarification setup 1 of a bioprocessing installation 2 for carrying out the above-mentioned method is claimed as such, comprising a fluidized bed centrifuge 3 for clarification of a cell broth by centrifugation and a pumping device 4 assigned to the fluidized bed centrifuge 3. The fluidized bed centrifuge 3 comprises at least one centrifuge chamber 5 rotated about a geometric centrifuge axis 6. Furthermore, the bioprocessing installation 2 comprises an electronic process control device 7 for controlling at least the fluidized bed centrifuge 3 and the pumping device 4. The fluidized bed centrifuge 3 is operated forward for a loading cycle 8 and / or a particle washing cycle 9 and reverse for a particle discharge cycle 10. Here, preferably, during the loading cycle 8, the cell broth loaded in the centrifuge chamber 5 proceeds to form a growing particle accumulation in the centrifuge chamber 5. The clarification setup 1 comprises a monitoring sensor device 11 with at least one optical sensor 12 for generating monitoring sensor data transmitted to the electronic process control device 7. See all previous descriptions.

[0093] It is essential that in the monitoring routine 13, image-related data 14 representing an optical image 15 of the centrifuge chamber contents 16 is generated by the monitoring sensor device 11 and a particle loading level 17 is calculated from the image-related data 14 by the electronic process control device 7.

[0094] According to another independent teaching, the use of a centrifuge chamber 5 made of a translucent material for carrying out the above-mentioned method is claimed per se. See all previous descriptions.

[0095] An electronic process control device 7 of a bioprocessing installation 2 for carrying out the claimed method may also be claimed per se. Again, reference is made to all the above statements.

[0096] It is very important that the electronic process control device 7 is designed to implement the proposed method by controlling at least the fluidized bed centrifuge 3 and the pumping device 4.

[0097] Preferably, the electronic process controller 7 is designed to implement the proposed method by controlling the upstream unit 28, the clarification setup 1 with the fluidized bed centrifuge 3 and / or the downstream setup 34 with the downstream unit 35. The electronic process controller 7 can be realized as a central unit controlling all or at least most of the components of the bioprocess installation 2. The electronic process controller 7 can also be realized in a distributed structure comprising a number of distributed units. In some embodiments, the at least one electronic process controller 7 directs the opening and closing of one or more valves 35, the rotation speed of the rotor directly or via a motor and / or the flow direction and / or velocity of fluids and / or particles from the upstream unit 28, such as a bioreactor.

[0098] Such an electronic process control device 7 includes, for example, at least one digital control unit (DCU) and / or at least one multifermenter control system (MFCS), which includes a local processor unit and a local data storage device itself. The MFCS also provides a centralized process management system and sends requests to the digital control unit. Additionally or alternatively, such an electronic process control device 7 preferably includes a computer and / or a server and / or a smartphone or the like. Here, preferably, the electronic process control device 7 includes at least one microprocessor that is individually adjustable and / or programmable and / or in which software can be implemented. All the above explanations are fully applicable to this teaching.

[0099] Preferably, the electronic process control device 7 includes a data processing system for implementing the above-described methods, preferably including a local data storage device and a local processor unit.

[0100] Finally, independent teachings may be directed to a computer program product for an electronic process control device 7 and a computer readable storage medium on which the computer program product is stored, preferably non-volatilely.

Claims

1. A method for operating a clarification setup (1) of a bioprocessing facility (2), said clarification setup (1) comprising a fluidized bed centrifuge (3) for clarification of a cell broth by centrifugation and a pumping device (4) assigned to said fluidized bed centrifuge (3), said fluidized bed centrifuge (3) comprising at least one centrifuge chamber (5) rotated about a geometric centrifuge axis (6), said bioprocessing facility (2) comprising an electronic process control device (7) for controlling at least said fluidized bed centrifuge (3) and said pumping device (4), said fluidized bed centrifuge (3) being adapted to clarify a particle broth. a clarification setup (1) operated in a forward direction for a particle loading cycle (8) and / or a particle washing cycle (9) and a reverse direction for a particle discharge cycle (10), wherein if a particle loading cycle (8) is provided, during said particle loading cycle (8), the cell broth loaded into the centrifuge chamber (5) progresses to form a growing particle accumulation in the centrifuge chamber (5), and the clarification setup (1) comprises a monitoring sensor device (11) with at least one optical sensor (12) for generating monitoring sensor data that are transmitted to the electronic process control device (7), wherein in the monitoring routine (13), image-related data (14) representing an optical image (15) of the centrifuge chamber contents (16) are generated by the monitoring sensor device (11), and a particle filling level (17) is calculated by the electronic process control device (7) from the image-related data (14) based on a calculation model (18). method.

2. wherein in the monitoring routine (13), according to the calculation model (18), the particle filling level (17) is calculated based on different optical properties, in particular different translucencies, different colors and / or different brightnesses, of the centrifuge chamber contents (16) inside and outside the particle accumulation. The method of claim 1.

3. In the monitoring routine (13), a phase boundary (19) between the particle accumulation and the remaining part of the centrifuge chamber contents (16) that does not contain particles is detected in the optical image (15) according to the calculation model (18), and the particle filling level (17) is calculated from the position of the phase boundary (19) based on the calculation model (18). The method of claim 1.

4. the centrifuge chamber (5) itself is made of a translucent material, and the monitoring sensor device (11) detects the optical image (15) of the centrifuge chamber contents (16) through the translucent material of the centrifuge chamber (5). The method of claim 1.

5. the at least one optical sensor (12) of the monitoring sensor device (11) is a camera unit, preferably a 2D or 3D camera, and / or the viewing direction of the camera unit is essentially parallel or inclined to the geometrical centrifuge axis (6). The method of claim 1.

6. a light device (20, 21) is assigned to the monitoring sensor device (11) with light irradiated through the centrifuge chamber contents (16) onto the monitoring sensor device (11) or with light irradiated by the centrifuge chamber contents (16) and reflected back to the monitoring sensor device (11); and / or the monitoring sensor device (11) and / or the light device (20, 21) are synchronized with the rotation of the at least one centrifuge chamber (5), The method of claim 1.

7. the particle loading level (17) represents a range between a particle-free state of the centrifuge chamber (5) and a maximum particle loading state, the particle loading level (17) being a boundary line to particle breakthrough, the particle loading level (17) being followed by particle breakthrough as the loading cycle (8) progresses, preferably the maximum particle loading state corresponding to an absolute maximum particle loading level and / or the particle-free state of the centrifuge chamber (5) corresponding to an absolute zero particle loading level, The method of claim 1.

8. the fluidized bed centrifuge (3) comprises a monitoring opening (22) of a monitoring panel through which the centrifuge chamber contents (16) can be monitored, and in the monitoring routine (13), the particle filling level (17) is calculated from the brightness and / or color distribution on the monitoring opening (22) based on the calculation model (18), and preferably the monitoring opening (22) extends along a radial direction relative to the geometrical centrifuge axis (6). The method of claim 1.

9. the clarification setup (1) comprises a calibrating sensor device (23) with at least one calibrating sensor (24) for generating calibrated sensor data transmitted to the electronic process control device (7), the calibrating sensor device (23) preferably detecting the presence and / or flow rate of particles in a liquid line (25) downstream of the fluidized bed centrifuge (3) during the loading cycle (8) and / or the cleaning cycle (9), characterized in that, preferably, in a calibration routine, the image-related data (14) of the monitoring sensor device (11) is correlated with the calibrated sensor data of the calibrating sensor device (23) in order to improve the accuracy of the calculation model (18) and / or to derive an absolute maximum particle loading level, The method of claim 1.

10. wherein in a calibration routine, a loading cycle (8) is progressed up to and beyond the point of particle breakthrough, and the particle loading level (17) calculated at the point of particle breakthrough is stored in the electronic process control device (7) as an absolute maximum particle loading level. The method of claim 1.

11. wherein during an adjustment routine, at least one parameter of the clarification setup (1) is adjusted by the electronic process control device (7) based on the particle loading level (17) of the centrifuge chamber (5) according to a predefined adjustment strategy, and / or during an automation routine, the electronic process control device (7) initiates a centrifugation cycle according to a predefined automation strategy based on the calculated particle loading level (17), The method of claim 1.

12. a nominal maximum particle loading level is stored in the electronic process control device (7), the nominal maximum particle loading level preferably being below an absolute maximum particle loading level by a predefined upper offset, and in accordance with the automation strategy, when the nominal maximum particle loading level is approached during the loading cycle (8), the loading cycle (8) is terminated, preferably the cleaning cycle (9) and / or the discharge cycle (10) is initiated by the electronic process control device (7); and / or a nominal zero particle loading level is stored in the electronic process control device (7), the nominal zero particle loading level being above an absolute zero particle loading level by a predefined lower offset, and in accordance with the automation strategy, when the nominal zero particle loading level is approached during the discharge cycle (10), the discharge cycle (10) is terminated, preferably the loading cycle (8) is initiated by the electronic process control device (7). The method of claim 11.

13. the bioprocessing installation (2) comprises a downstream setup (34) with at least one downstream unit (35), and in an adaptation routine, the electronic process control device (7) adapts at least one parameter of the downstream setup (34) according to the image-related data (14) based on a predefined adaptation strategy, The method of claim 1.

14. characterised in that in the analysis routine irregularities attributable to the presence of impurities, preferably contaminating organisms, are detected on the basis of predefined irregularity features of said optical image (15). The method of claim 1.

15. A clarification setup (1) of a bioprocessing facility (2) for carrying out the method according to any one of claims 1 to 14, said clarification setup (1) comprising a fluidized bed centrifuge (3) for clarification of a cell broth by centrifugation and a pumping device (4) assigned to said fluidized bed centrifuge (3), said fluidized bed centrifuge (3) comprising at least one centrifuge chamber (5) rotated about a geometric centrifuge axis (6), said bioprocessing facility (2) comprising an electronic process control device (7) for controlling at least said fluidized bed centrifuge (3) and said pumping device (4), 3) is operated in a forward operation for a particle loading cycle (8) and / or a particle washing cycle (9) and in a reverse operation for a particle discharge cycle (10), wherein if a particle loading cycle (8) is provided, during said particle loading cycle (8), the cell broth loaded into said centrifuge chamber (5) progresses to form a growing particle accumulation in said centrifuge chamber (5), and said clarification setup (1) includes a monitoring sensor device (11) with at least one optical sensor (12) for generating monitoring sensor data that are transmitted to said electronic process control device (7), wherein in the monitoring routine (13), image-related data (14) representing an optical image (15) of the centrifuge chamber contents (16) are generated by the monitoring sensor device (11), and a particle filling level (17) is calculated by the electronic process control device (7) from the image-related data (14) based on a calculation model (18). Clarification Setup (1).

16. Use of a centrifuge chamber (5) made of a translucent material for carrying out the method according to any one of claims 1 to 14.

Images