A method for predicting filtration pressure and flow rate of biological preparations and its application

Abstract

The present invention discloses a prediction method for the filtration pressure and flow rate of a biological agent and its application. A simulation system is formed with a peristaltic pump, a filter and corresponding pipelines under laboratory conditions, and a filtration pressure prediction formula and a flow rate prediction formula are obtained based on the simulation system. Specifically, one or more different types of solutions are selected, and each solution includes samples of multiple viscosities, and each sample is tested separately with a simulation system, and the filtration pressure and flow rate of each sample under different pump speeds are recorded, and each group of flow rate, viscosity and pump speed data are subjected to polynomial regression fitting to obtain a flow rate prediction formula; each group of filtration pressure, viscosity and pump speed data are subjected to polynomial regression fitting to obtain a filtration pressure prediction formula, and then the flow rate prediction formula and the filtration pressure prediction formula are used to predict the flow rate and filtration pressure of a solution of a certain viscosity filtered by a peristaltic pump, a filter and a corresponding pipeline system used in the simulation system at different pump speeds. The prediction method of the present invention accurately predicts the flow rate and filtration pressure in actual production, can greatly reduce the research and development cost, and is convenient for application in the prediction of parameters in actual production.

Description

Technical Field

[0001] The present invention relates to the technical field of biopharmaceuticals, and in particular to a method for predicting the filtration pressure and flow rate of biological preparations and an application thereof. Background Art

[0002] Compared with traditional small molecule drugs, biologics have the advantages of high specificity, strong binding ability, and low off-target rate, and have shown great potential in the treatment of many diseases, especially in the treatment of cancer, autoimmune diseases, and genetic diseases.

[0003] Sterile filtration is a key step in production. It removes microorganisms from liquids by physical interception to ensure that the quality standards of sterile drugs are met, thereby ensuring the safety of the solution. During the filtration process of biological preparations, molecules will be subjected to high filtration shear forces. Excessive shear forces may cause damage or denaturation of the molecular structure, affecting its biological activity and function. Therefore, in the development stage of drug production filtration process, it is crucial to study the impact of filtration process on molecular quality and determine the appropriate filtration process parameters and ranges.

[0004] During the filtration process, the pore size of the filter membrane can be regarded as a channel for liquid flow. According to the Hagen-Poiseuille equation, the shear force on the molecules is positively correlated with the flow rate. In actual production, it is difficult to monitor the flow rate, so it is difficult to use the corresponding laboratory model to simulate the poor filtration flow rate conditions in actual production, and then evaluate the specific impact of filtration shear force on product quality in production. At the same time, filtration pressure is also a key process parameter in the sterilization filtration stage. Excessive filtration pressure may destroy the integrity of the filtration system, affect the sterility of the product, and thus lead to production failure. Therefore, how to estimate the filtration pressure and flow rate of the drug stock solution during the filtration process and evaluate its impact on product quality and production process has become an urgent problem to be solved.

[0005] Usually, under laboratory conditions, the same filtration system and peristaltic pump as in actual production can be used, and a balance can be used to monitor the weight of the drug stock solution filtered out within a certain period of time. The filtration flow rate of the drug stock solution can be calculated to support the evaluation of the impact of shear force during the filtration process on product quality. At the same time, a pressure detection device can be used to monitor the filtration pressure during the filtration process. However, this experiment requires the use of a large amount of drug stock solution, and many projects are difficult to support this evaluation experiment due to resource constraints. In addition, due to the high cost of drug stock solutions, this also greatly increases the company's R&D costs.

[0006] Therefore, establishing a prediction model to estimate the filtration flow rate and filtration pressure of drug stock solution under production conditions is an effective way to solve the above problems, while reducing R&D costs and reducing the risk of production failure. However, there are few literature reports on how to predict the filtration flow rate and filtration pressure of drug stock solution. Summary of the invention

[0007] The technical problem to be solved by the present invention is to provide a method for predicting the filtration pressure and flow rate of biological preparations and its application. By using drug stock solution or simulated solution, a filtration pressure and flow rate prediction model is constructed, which can effectively predict the filtration pressure and flow rate in the production process, and then support the evaluation of the impact on product quality, reduce R&D costs, and reduce the risk of production failure.

[0008] In order to solve the above technical problems, the first aspect of the present invention is to provide a method for predicting the filtration pressure and flow rate of a biological agent, which comprises the following steps:

[0009] S1. Select one or more different solutions, and each solution includes samples of multiple viscosities, and the viscosity of the samples ranges from 1 to 100 cp;

[0010] S2, testing the samples in step S1 respectively with a simulation system, adjusting the pump speed in the simulation system, and recording the corresponding filtration pressure at different pump speeds, as well as recording the mass of the filtered solution at different pump speeds at different time points, and then calculating the filtration flow rate at different pump speeds;

[0011] S3, performing polynomial regression fitting on each set of flow rate, viscosity and pump speed data to obtain a flow rate prediction formula; performing polynomial regression fitting on each set of filtration pressure, viscosity and pump speed data to obtain a filtration pressure prediction formula;

[0012] When fitting, the interaction order is at least third order;

[0013] The obtained flow rate prediction formula and filtration pressure prediction formula are as follows:

[0014] Flow rate = A1 + A2v + A3μ + (v-A4) 2 ×A5+(v-A4)×(μ-A6)×A7+(μ-A6) 2 ×A8+(v-A4) 3 ×A9+(v-A4) 2 ×(μ-A6)×A 10 +(v-A4)×(μ-A6) 2 ×A 11 +(μ-A6) 3 ×A 12 +F1(v,μ);

[0015] Filtration pressure = B1 + B2v + B3μ + (v-B4) 2 ×B5+(v-B4)×(μ-B6)×B7+(μ-B6) 2×B8+(v-B4) 3 ×B9+(v-B4) 2 ×(μ-B6)×B 10 +(v-B4)×(μ-B6) 2 ×B 11 +(μ-B6) 3 ×B 12 +F2(v,μ);

[0016] Get A1 to A respectively 12 , B1 to B 12 , and A in F1(v,μ) n and B in F2(v,μ) n ;

[0017] Where μ is the viscosity, v is the pump speed, and F1(v,μ) and F2(v,μ) are terms greater than the third order;

[0018] S4. Use the flow rate prediction formula and the filtration pressure prediction formula to predict the flow rate and filtration pressure of a solution of a certain viscosity at different pump speeds when filtering the solution using a peristaltic pump, a filter and a corresponding pipeline system used in a simulation system.

[0019] In the present invention, a combination of a commonly used peristaltic pump, a filter and a pipeline can be used to form a simulation system under laboratory conditions, and a prediction formula for flow rate and filtration pressure is obtained based on the simulation system. In the formula, viscosity (e.g., viscosity of a drug stock solution) and the pump speed of a peristaltic pump are independent variables. When the same peristaltic pump, filter and pipeline combination used in subsequent research and development and actual production are used, for a given drug stock solution, the flow rate and filtration pressure at different pump speeds can be directly obtained, and the verified predicted value is close to the measured value, and the predicted filtration pressure can be used as the basic basis for regulating the pump speed of the peristaltic pump in actual production, and the predicted flow rate can support the assessment of the impact of shear force on product quality; the present invention can also greatly reduce the use of drug stock solution and other material costs and labor costs of repeated experimental operations by predicting flow rate and filtration pressure, and is applicable to the filtration process of the actual drug stock solution, such as any commercially available filter model (membrane area), any model of peristaltic pump and drug stock solution of any viscosity range. It should also be noted that polynomial regression fitting can be achieved by commercial software. It should also be noted that F1(v,μ) and F2(v,μ) are terms greater than the third order, which means that if the third order is taken, F1(v,μ) and F2(v,μ) do not exist. If the fourth order is taken, F1(v,μ) is (v-A4) 4 ×A 13 +(v-A4) 3 ×(μ-A6)×A 14 +(v-A4) 2 ×(μ-A6) 2×A 15 +(v-A4)×(μ-A6) 3 ×A 16 +(μ-A6) 4 ×A 17 , when F2(v,μ) takes the fourth order and higher, the same applies.

[0020] In a specific embodiment, the simulation system constructed under laboratory conditions further includes a filtration pressure detection device, a filtrate collector, a weighing device, and a timer;

[0021] Among them, the liquid inlet of the peristaltic pump is connected to the solution sample containing device, the liquid outlet of the peristaltic pump is connected to the liquid inlet of the filter, the liquid outlet of the filter is connected to the filtrate collector, the filtration pressure detection device is connected in series between the peristaltic pump and the filter, and the filtrate collector is placed on the weighing device.

[0022] Preferably, the filters are one or two connected in series.

[0023] In this specific embodiment, the simulation system set up in the laboratory is based on the peristaltic pump, filter and corresponding pipeline used in actual production, so it can realize the simulation of actual production. The prediction formula of filtration pressure and flow rate obtained based on the simulation system is accurate and close to the result of actual production, which ensures the reliability of the prediction result as the basis for adjusting the pump speed in actual production; the filtration pressure detection device is used to detect the filtration pressure of the simulation system, for example, including a pressure sensor and a pressure value display device, these components can be realized by existing technology, and will not be repeated here; the weighing device is used to weigh the mass of the filtered filtrate in the filtrate collector; the timer is preferably used in conjunction with the weighing device to record the weight of the filtered filtrate in the filtrate collector at different time points, so as to facilitate the calculation of the flow rate; it should be noted that the flow rate of the present invention is the rate of change of the filtrate mass over time. In addition, it can be understood that the filter is one or two in series, more preferably two, because the two-in-series mode is currently the most common application mode.

[0024] In a specific embodiment, the solution is a drug stock solution or a simulated solution that is the same as that in actual production, and preferably the drug stock solution is selected from a mixture of one or more of protein drugs, vaccines, cell drugs, and gene drugs; preferably the simulated solution is selected from a bovine serum albumin solution and / or a polyethylene glycol solution, wherein the protein drugs include monoclonal antibodies, bispecific antibodies, multispecific antibodies, antibody-drug conjugates, and fusion proteins; polyethylene glycol preferably has samples of multiple molecular weights, for example, a polyethylene glycol molecular weight of 2000-20000.

[0025] Preferably, the sample range includes: 1-100cp; or

[0026] 1-50cp and 50-100cp; or

[0027] 1-20cp, 20-40cp, 40-60cp, 60-80cp and 80-100cp; or

[0028] 1-10cp, 10-20cp, 20-30cp, 30-40cp, 40-50cp, 50-60cp, 60-70cp, 70-80cp, 80-90cp and 90-100cp;

[0029] The pump speed in step S2 is 5-150 rpm.

[0030] In the present invention, in order to improve the accuracy of the formula prediction, the range of the solution is as wide as possible, and the density of the data is as high as possible within the range of viscosity and pump speed.

[0031] In a specific implementation, the value range of A1 is 2 to 10; the value range of A2 is 0.002 to 0.08; the value range of A3 is -0.3 to -0.01; the value range of A4 is 30 to 70; the value range of A5 is -0.002 to -0.0001; the value range of A6 is 10 to 20; the value range of A7 is -0.008 to -0.001; the value range of A8 is 0.004 to 0.04; the value range of A9 is 2×10 -8 ~2×10 -5 ; A 10 The value range is 0.00001~0.0002; A 11 The value range is 0.0001~0.0005; A 12 The value range is -0.003 to -0.00005.

[0032] The value range of B1 is 2 to 20; the value range of B2 is 0.02 to 0.4; the value range of B3 is 0.01 to 0.8; the value range of B4 is 30 to 70; the value range of B5 is -0.008 to -0.0001; the value range of B6 is 10 to 20; the value range of B7 is -0.02 to 0.01; the value range of B8 is -0.05 to -0.001; the value range of B9 is 0.00001 to 0.0002; 10 The value range of B is -0.0002~0.0005; 11 The value range of B is -0.0002~0.001; 12 The value range is -0.001 to 0.01.

[0033] In order to solve the above technical problems, the second aspect of the present invention is to provide an application of the aforementioned method for predicting the filtration pressure and flow rate of biological preparations in predicting the filtration pressure and flow rate in the actual production process. It can be understood that when the aforementioned method is applied to predict the flow rate and filtration pressure in actual production, the prediction is relatively accurate, which can greatly reduce the research and development costs, and the production needs can be met by adjusting based on the predicted values ​​without affecting the product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 It is a simulation system built under laboratory conditions of the present invention;

[0036] Among them: 1-containing device; 2-peristaltic pump; 3-filter; 4-filtration pressure detection device; 41-pressure sensor; 42-pressure value display device; 5-filtrate collector; 6-weighing device. DETAILED DESCRIPTION

[0037] The following will be combined with the accompanying drawings to clearly and completely describe the technical solutions in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] In the present invention, the simulation system constructed under laboratory conditions is as follows Figure 1 As shown, the experimental solution is placed in a containing device, the liquid inlet of the peristaltic pump is connected to the containing device through a pipeline, and two filters in series are located downstream of the peristaltic pump, wherein a filtration pressure detection device is installed between the peristaltic pump and the filter close to it, and the outlet of the filter far away from the peristaltic pump is connected to the filtrate collector through a pipeline; the peristaltic pump transports the solution in the containing device to the filtrate collector, and during the transportation process, the solution is filtered through the filter, and the pressure detection device at the front end (upstream) of the filter detects the filtration pressure; in addition, the filtrate collector is placed on a weighing device so that the weighing device can display and record weight changes at any time, wherein the timer is not shown, and can be built into the weighing device or an independent timer, the peristaltic pump is turned on, the air in the system is discharged, and the timing starts after the solution fills the filtration system (it should be noted that the timing start time has no effect on the present invention, and the present invention mainly uses the time interval when calculating the flow rate).

[0039] In the present invention, the working process of the simulation system is: prepare one or more solutions, each solution contains multiple viscosities, add the solution to be filtered into the containing device, start the peristaltic pump, exhaust the air in the system at a lower pump speed, make the solution fill the filtration system, and pause the peristaltic pump. Set the peristaltic pump speed, turn on the peristaltic pump and the timer, record the filtration pressure at the pump speed, and record the change in the weight of the filtrate at the same pump speed over time, so as to calculate the flow rate at the pump speed. Adjust the pump speed, repeat the above operation, and obtain the results of the filtration pressure and flow rate at different pump speeds. Change the sample (i.e., change the type and / or viscosity of the solution), and adjust a series of pump speeds to collect data. Enrich the data type to make the constants in the flow rate prediction formula and the filtration pressure prediction formula more accurate, and improve the accuracy of the formula prediction. Similarly, for any desired peristaltic pump, filter and corresponding pipeline combination, repeat the above operation to obtain the corresponding filtration pressure and flow rate prediction formula.

[0040] Embodiment 1

[0041] Sample: polyethylene glycol (PEG) solution with viscosities of 2.9 cp, 6.5 cp, 12.1 cp, 16.7 cp, and 21.9 cp;

[0042] Peristaltic pump: 530 peristaltic pump from Watson-Marlow;

[0043] Pump down tube: 6.4*2.4mm silicone tube;

[0044] Filter: Millipak 20 filter, filter membrane area 100 cm 2 ;

[0045] The test results are shown in Table 1:

[0046] Table 1

[0047]

[0048] Polynomial regression fitting was performed on the above multiple sets of data, and the interaction order was selected as 3rd order, and the flow rate and filtration pressure prediction formula was obtained as follows:

[0049] Flow rate = 2.6 + 0.0066 × v – 0.13 × μ - (v - 32.8) 2 ×0.00047-(v-32.8)×(μ-10.4)×0.0033+(μ-10.4) 2 ×0.0073+(v-32.8) 3 ×0.000012+(v-32.8) 2 ×(μ-10.4)×0.000054+(v-32.8)×(μ-10.4) 2×0.00024-(μ-10.4) 3 ×0.00018;

[0050] Filtration pressure = 19.5 + 0.17 × v + 0.24 × μ - (v - 33.3) 2 ×0.0048-(v-33.3)×(μ-10.3)×0.012-(μ-10.3) 2 ×0.087+(v-33.3) 3 ×0.00015+(v-33.3) 2 ×(μ-10.3)×0.00032+(v-33.3)×(μ-10.3) 2 ×0.00069+(μ-10.3) 3 ×0.0050;

[0051] Where μ is the viscosity and v is the pump speed.

[0052] Verification Example 1:

[0053] Taking polyethylene glycol (PEG) solution (viscosity of 7.5cP), BSA solution (viscosity of 2.9cP) and monoclonal antibody solution (viscosity of 2.8cP) as examples, select the same peristaltic pump, pump tube and filter as in Example 1, record the filtration pressure at the corresponding pump speed, and record the change of the weight of the filtered solution over time at the same pump speed, and calculate the flow rate at the pump speed. According to the flow rate prediction formula and the filtration pressure prediction formula in Example 1, the flow rate and filtration pressure at the corresponding pump speed are calculated to obtain the corresponding predicted values. The actual values ​​and predicted values ​​of the filtration flow rate and filtration pressure are shown in Table 2:

[0054] Table 2

[0055]

[0056] The data in the above table show that, compared with the predicted values ​​obtained based on the prediction formula in Example 1, the actual test values ​​are all within the range of ±20% of the predicted flow rate, the actual filtration pressure values ​​are all within the range of ±3psi of the predicted filtration pressure, and the prediction formula obtained based on the data of PEG solutions with different viscosities at different pump speeds is not only applicable to PEG solutions, but also to BSA solutions and monoclonal antibody solutions.

[0057] Embodiment 2

[0058] Sample: bovine serum albumin (BSA) solution, viscosities of 2.8cp, 6.8cp, 21.8cp, and 32cp;

[0059] Peristaltic pump: 530 peristaltic pump from Watson-Marlow;

[0060] Pump down tube: 6.4*2.4mm silicone tube;

[0061] Filter: Millipak 40 filter, filter membrane area 200 cm 2 ;

[0062] The test results are shown in Table 3:

[0063] Table 3

[0064]

[0065] Polynomial regression fitting was performed on the above multiple sets of data, and the interaction order was selected as 3rd order, and the flow rate and filtration pressure prediction formula was obtained as follows:

[0066] Flow rate = 3.4 + 0.0024 × v – 0.11 × μ - (v - 33.2) 2 ×0.00069-(v-33.2)×(μ-15.3)×0.0048+(μ-15.3) 2 ×0.0051+(v-33.2) 3 ×2.9×10 -8 +(v-33.2) 2 ×(μ-15.3)×0.000016+(v-33.2)×(μ-15.3) 2 ×0.00032-(μ-15.3) 3 ×0.000090;

[0067] Filtration pressure = 2.9 + 0.26 × v + 0.78 × μ - (v - 35) 2 ×0.0017-(v-35)×(μ-15.9)×0.0085-(μ-15.9) 2 ×0.025+(v-35) 3 ×0.000049-(v-35) 2 ×(μ-15.9)×0.00012-(v-35)×(μ-15.9) 2 ×0.000061-(μ-15.9) 3 ×0.00073;

[0068] Where μ is the viscosity and v is the pump speed.

[0069] Verification Example 2:

[0070] Taking BSA solution (viscosity of 21.8 cP) as an example, select the same peristaltic pump, pump tube and filter as in Example 2, record the filtration pressure at a pump speed of 25 rpm, and record the change in the weight of the filtered solution over time at the same pump speed, and calculate the flow rate at the pump speed. According to the flow rate prediction formula and the filtration pressure prediction formula in Example 2, the flow rate and filtration pressure of the BSA solution at a pump speed of 25 rpm are calculated to obtain the corresponding predicted values. The actual values ​​and predicted values ​​of the filtration flow rate and filtration pressure are shown in Table 4:

[0071] Table 4

[0072]

[0073] The data in the above table show that, compared with the predicted values ​​obtained based on the prediction formula of Example 2, the actual test values ​​are within the range of ±20% of the predicted flow rate, and the actual filtration pressure value is within the range of ±3psi of the predicted filtration pressure. Moreover, the prediction formula obtained based on the data of BSA solutions with different viscosities at different pump speeds can accurately predict the filtration pressure and flow rate of BSA solutions with different viscosities at different pump speeds.

[0074] Embodiment 3

[0075] Sample: polyethylene glycol (PEG) solution with viscosities of 6.5 cp, 12.1 cp, 16.7 cp, and 21.9 cp;

[0076] Peristaltic pump: 530 peristaltic pump from Watson-Marlow;

[0077] Pump down tube: 9.6*2.4mm silicone tube;

[0078] Filter: Millipak 100 filter, filter membrane area 500cm 2 ;

[0079] The test results are shown in Table 5:

[0080] Table 5

[0081]

[0082] Polynomial regression fitting was performed on the above multiple sets of data, and the interaction order was selected as 3rd order, and the flow rate and filtration pressure prediction formula was obtained as follows:

[0083] Flow rate = 7.8 + 0.026 × v – 0.28 × μ - (v - 37.4) 2 ×0.0018-(v-37.4)×(μ-13.6)×0.0076+(μ-13.6) 2 ×0.0084+(v-37.4) 3×0.000027+(v-37.4) 2 ×(μ-13.6)×0.00011+(v-37.4)×(μ-13.6) 2 ×0.00043-(μ-13.5) 3 ×0.00057;

[0084] Filtration pressure = 16.7 + 0.051 × v + 0.26 × μ - (v - 38.3) 2 ×0.0076-(v-38.3)×(μ-13.9)×0.0099-(μ-13.9) 2 ×0.035+(v-38.3) 3 ×0.00016+(v-38.3) 2 ×(μ-13.9)×0.00011+(v-38.3)×(μ-13.9) 2 ×0.00054+(μ-13.9) 3 ×0.000049;

[0085] Verification Example 3:

[0086] Taking PEG solution (viscosity of 7.5 cP) as an example, select the same peristaltic pump, pump down tube and filter as in Example 3, record the filtration pressure at 20rpm, 40rpm and 80rpm pump speeds, and record the weight change of the filtered solution over time at the same pump speed, and calculate the flow rate at the pump speed. According to the flow rate prediction formula and filtration pressure prediction formula in Example 3, the flow rate and filtration pressure of the PEG solution at 20rpm, 40rpm and 80rpm pump speeds are calculated to obtain the corresponding predicted values. The actual values ​​and predicted values ​​of the filtration flow rate and filtration pressure are shown in Table 6:

[0087] Table 6

[0088]

[0089] The data in the above table show that, compared with the predicted values ​​obtained based on the prediction formula in Example 3, the actual test values ​​are all within the range of ±20% of the predicted flow rate, the actual filtration pressure values ​​are all within the range of ±3psi of the predicted filtration pressure, and the prediction formula obtained based on the data of PEG solutions with different viscosities at different pump speeds can accurately predict the filtration pressure and flow rate of PEG solutions with different viscosities at different pump speeds.

[0090] Embodiment 4

[0091] Sample: polyethylene glycol (PEG) solution with viscosities of 5.6 cp, 9.5 cp, 18.5 cp, and 32.2 cp;

[0092] Peristaltic pump: Watson-Marlow 630 peristaltic pump;

[0093] Pump down tube: 6.4*3.2mm silicone tube;

[0094] Filter: Millipak 200 filter, filter membrane area 1000cm 2 ;

[0095] The test results are shown in Table 7:

[0096] Table 7

[0097]

[0098]

[0099] Polynomial regression fitting was performed on the above multiple sets of data, and the interaction order was selected as 3rd order, and the flow rate and filtration pressure prediction formula was obtained as follows:

[0100] Flow rate = 6.7 + 0.059 × v – 0.22 × μ - (v - 62.6) 2 ×0.00048-(v-62.6)×(μ-16.5)×0.0047+(μ-16.5) 2 ×0.0079+(v-62.6) 3 ×2.9×10 -6 +(v-62.6) 2 ×(μ-16.5)×0.000010+(v-62.6)×(μ-16.5) 2 ×0.00017-(μ-16.5) 3 ×0.00018;

[0101] Filtration pressure = 2.7 + 0.18 × v + 0.47 × μ - (v - 62.6) 2 ×0.0014+(v-62.6)×(μ-16.5)×0.0013-(μ-16.5) 2 ×0.022+(v-62.6) 3 ×0.000011-(v-62.6) 2 ×(μ-16.5)×0.000038-(v-62.6)×(μ-16.5) 2 ×0.00014+(μ-16.5) 3 ×0.00094;

[0102] Verification Example 4:

[0103] Taking PEG solution (viscosity of 5.6cp, 9.5cp, 18.5cp, 32.2cp) as an example, select the same peristaltic pump, pump down tube and filter as in Example 4, record the filtration pressure at 15rpm, 50rpm, 30rpm and 80rpm pump speeds, and record the weight change of the filtered solution over time at the same pump speed, and calculate the flow rate at the pump speed. According to the flow rate prediction formula and the filtration pressure prediction formula in Example 5, the flow rate and filtration pressure of the PEG solution at different pump speeds are calculated to obtain the corresponding predicted values. The actual values ​​and predicted values ​​of the filtration flow rate and filtration pressure are shown in Table 8:

[0104] Table 8

[0105]

[0106] The data in the above table show that, compared with the predicted values ​​obtained based on the prediction formula in Example 4, the actual test values ​​are all within the range of ±20% of the predicted flow rate, the actual filtration pressure values ​​are all within the range of ±3psi of the predicted filtration pressure, and the prediction formula obtained based on the data of PEG solutions with different viscosities at different pump speeds can accurately predict the filtration pressure and flow rate of PEG solutions with different viscosities at different pump speeds.

[0107] Embodiment 5

[0108] Sample: Polyethylene glycol (PEG) solution, viscosities are 5.6cp, 9.5cp, 18.5cp, 32.2cp; bovine serum albumin (BSA) solution, viscosities are 2.8cp, 6.8cp, 21.8cp, 32cp

[0109] Peristaltic pump: 630 peristaltic pump from Watson-Marlow;

[0110] Pump down tube: 9.6*3.2mm silicone tube;

[0111] Filter: Millipak 200 filter, filter membrane area 1000cm 2 ;

[0112] The test results are shown in Table 9:

[0113]

[0114] Polynomial regression fitting was performed on the above multiple sets of data, and the interaction order was selected as 3rd order, and the flow rate and filtration pressure prediction formula was obtained as follows:

[0115] Flow rate = 8.3 + 0.029 × v – 0.22 × μ - (v - 61.5) 2×0.00049-(v-61.5)×(μ-16.6)×0.0061+(μ-16.6) 2 ×0.028+(v-61.5) 3 ×5.7×10 -6 +(v-61.5) 2 ×(μ-16.6)×0.000034+(v-61.5)×(μ-16.6) 2 ×0.00032-(μ-16.6) 3 ×0.0017;

[0116] Filtration pressure = 12.7 + 0.13 × v + 0.084 × μ - (v - 61.5) 2 ×0.0010-(v-61.5)×(μ-16.6)×0.0011-(μ-16.6) 2 ×0.014+(v-61.5) 3 ×0.000013+(v-61.5) 2 ×(μ-16.6)×0.000018-(v-61.5)×(μ-16.6) 2 ×0.000047+(μ-16.6) 3 ×0.0011;

[0117] Verification Example 5:

[0118] Taking PEG solution (viscosity of 9.5cp and 18.5cp) and BSA solution (viscosity of 2.8cp, 21.8cp, 32cp) as examples, select the same peristaltic pump, pump downtube and filter as in Example 5, record the filtration pressure at the corresponding pump speed, and record the change of the weight of the filtered solution over time at the same pump speed, and calculate the flow rate at the pump speed. According to the flow rate prediction formula and the filtration pressure prediction formula in Example 5, the flow rate and filtration pressure of the solution at different pump speeds are calculated to obtain the corresponding predicted values. The actual values ​​and predicted values ​​of the filtration flow rate and filtration pressure are shown in Table 10:

[0119] Table 10

[0120]

[0121] The data in the above table show that, compared with the predicted values ​​obtained based on the prediction formula in Example 5, the actual test values ​​are all within the range of ±20% of the predicted flow rate, the actual filtration pressure values ​​are all within the range of ±3psi of the predicted filtration pressure, and the prediction formula obtained based on the data of PEG solutions and BSA solutions with different viscosities at different pump speeds can accurately predict the filtration pressure and flow rate of PEG solutions and BSA solutions with different viscosities at different pump speeds.

[0122] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for predicting filtration pressure and flow rate of biological preparations, characterized in that: Under laboratory conditions, a simulation system is built with a peristaltic pump, a filter and corresponding pipelines, and the filtration pressure prediction formula and the flow rate prediction formula are obtained based on the simulation system, including the following steps: S1. Select one or more different solutions, and each solution includes samples of multiple viscosities, and the viscosity of the samples ranges from 1 to 100 cp; S2, testing the samples in step S1 respectively with a simulation system, adjusting the pump speed in the simulation system, and recording the corresponding filtration pressure at different pump speeds, as well as recording the mass of the filtered solution at different pump speeds at different time points, and then calculating the filtration flow rate at different pump speeds; S3, performing polynomial regression fitting on each set of flow rate, viscosity and pump speed data to obtain a flow rate prediction formula; performing polynomial regression fitting on each set of filtration pressure, viscosity and pump speed data to obtain a filtration pressure prediction formula; When fitting, the interaction order is at least third order; The obtained flow rate prediction formula and filtration pressure prediction formula are as follows: Flow velocity = A1 + A2v + A3μ + (v - A4) 2 ×A5 + (v - A4)×(μ - A6)×A7 + (μ - A6) 2 ×A8 + (v - A4) 3 ×A9 + (v - A4) 2 ×(μ - A6)×A 10 +(v - A4)×(μ - A6) 2 ×A 11 +(μ - A6) 3 ×A 12 + F1(v, μ); Filtration pressure = B1 + B2v + B3μ + (v-B4) 2 ×B5+(v-B4)×(μ-B6)×B7+(μ-B6) 2 ×B8+(v-B4) 3 ×B9+(v-B4) 2 ×(μ-B6)×B 10 +(v-B4)×(μ-B6) 2 ×B 11 +(μ-B6) 3 ×B 12 +F2(v,μ); Get A1 to A respectively 12 , B1 to B 12 , and A in F1(v,μ) n and B in F2(v,μ) n ; Where μ is the viscosity, v is the pump speed, and F1(v,μ) and F2(v,μ) are terms greater than the third order; S4. Use the flow rate prediction formula and the filtration pressure prediction formula to predict the flow rate and filtration pressure of a solution of a certain viscosity at different pump speeds when filtering the solution using a peristaltic pump, a filter and a corresponding pipeline system used in a simulation system.

2. The method according to claim 1, characterized in that The simulation system built under laboratory conditions also includes a filtration pressure detection device, a filtrate collector, a weighing device, and a timer; Among them, the liquid inlet of the peristaltic pump is connected to the solution sample containing device, the liquid outlet of the peristaltic pump is connected to the liquid inlet of the filter, the liquid outlet of the filter is connected to the filtrate collector, the filtration pressure detection device is connected in series between the peristaltic pump and the filter, and the filtrate collector is placed on the weighing device.

3. The method according to claim 2, characterized in that The filters are one or two in series.

4. The method according to claim 1, characterized in that In step S1, the solution is the same drug stock solution or simulated solution as that in actual production.

5. The method according to claim 4, characterized in that The drug stock solution is selected from a mixture of one or more of protein drugs, vaccines, cell drugs, and gene drugs; The simulated liquid is selected from bovine serum albumin solution and / or polyethylene glycol solution.

6. The method according to claim 1, characterized in that In step S1, the viscosity of the sample ranges from 1 to 100 cp; or 1-50cp and 50-100cp; or 1-20cp, 20-40cp, 40-60cp, 60-80cp and 80-100cp; or 1-10cp, 10-20cp, 20-30cp, 30-40cp, 40-50cp, 50-60cp, 60-70cp, 70-80cp, 80-90cp and 90-100cp; The pump speed in step S2 is 5-150 rpm.

7. The method according to claim 1, characterized in that In step S3, The value range of A1 is 2 to 10; the value range of A2 is 0.002 to 0.08; the value range of A3 is -0.3 to -0.01; the value range of A4 is 30 to 70; the value range of A5 is -0.002 to -0.0001; the value range of A6 is 10 to 20; the value range of A7 is -0.008 to -0.001; the value range of A8 is 0.004 to 0.04; the value range of A9 is 2×10 -8 ~2×10 -5 ; A 10 The value range is 0.00001~0.0002; A 11 The value range is 0.0001~0.0005; A 12 The value range is -0.003 to -0.00005. The value range of B1 is 2 to 20; the value range of B2 is 0.02 to 0.4; the value range of B3 is 0.01 to 0.8; the value range of B4 is 30 to 70; the value range of B5 is -0.008 to -0.0001; the value range of B6 is 10 to 20; the value range of B7 is -0.02 to 0.01; the value range of B8 is -0.05 to -0.001; the value range of B9 is 0.00001 to 0.0002; 10 The value range of B is -0.0002~0.0005; 11 The value range of B is -0.0002~0.001; 12 The value range is -0.001 to 0.

01.

8. The method for predicting solution filtration pressure and flow rate according to any one of claims 1 to 7, and its application in predicting filtration pressure and flow rate in actual production process.

Images