WO2018042458A1

WO2018042458A1 - Perfusion cell culture process

Info

Publication number: WO2018042458A1
Application number: PCT/IN2017/050373
Authority: WO
Inventors: Suman BANDYOPADHYAY; Anand PRABHAKAR; Srikanth RAPALA
Original assignee: Reddy's Laboratories Limited
Priority date: 2016-09-02
Filing date: 2017-09-01
Publication date: 2018-03-08

Abstract

The invention describes a feeding strategy for a cell culture process for increasing titer/vv of the therapeutic antibodies without impacting product's qualitative characteristics.

Description

PERFUSION CELL CULTURE PROCESS

FIELD OF THE INVENTION

The present invention relates to cell culture methods. In particular, the invention discloses a feeding strategy for a perfusion cell culture process to produce monoclonal antibodies, particularly therapeutic monoclonal antibodies.

BACK GROUND OF THE INVENTION

The discovery of monoclonal antibodies (Mabs) has ushered in an era of targeted therapeutics. For optimal therapeutic efficacy, Mabs are required in large doses and for a longer duration of time. This necessitates production of increased quantity of Mabs. There can be broadly two strategies for quantitative increase in Mabs production; 1) expanding manufacturing capacity, 2) improvement of cell culture processes. Expanding the manufacturing capacity incurs large investment and demands dependable future product pipeline to ensure return on investment. On the other hand, improvement of cell culture processes requires exploration of multitude of levers ranging from culture media components and additives to process parameters. Adding to this complexity is the interdependence of the aforementioned levers, and the need to achieve requisite quantity without compromising product quality.

With challenges unique to each strategy, a combination of the above two strategies offers maximal utilization of the production capacity to produce desirable quality and quantity of the Mabs consequently reducing the manufacturing cost of goods (COGS). Fed batch methods represent the most widely used choice for large scale production of Mabs owing to its operational simplicity and familiarity. However, continuous buildup of toxic by-products in fed-batch method severely limits cell culture longevity. Furthermore, despite recent developments and multiple levers available for optimization, the productivity in the fed-batch methods is an impediment.

Perfusion system is another well-known production method which offers to overcome the challenges presented by fed-batch bioreactor. Perfusion system maintains the cell longevity by replenishing the culture with fresh nutrient media alongside removing the inhibitory/waste by-products. The principal aspect of perfusion system, is the added requirement of a filtration device. However, fouling of the filter membrane limited the use of perfusion system and hence led to the evolution of the Alternating Tangential Flow (ATF) system, which owing to its alternating tangential flow through the filter device prevents the problem of membrane fouling. Despite this, continuous harvesting of proteins in perfusion system led to operational complexity. Recently, perfusion systems are incorporated with a filter membrane having a molecular weight cut-off (MWCO) smaller than the product of interest (e.g. protein or therapeutic Mabs) that which concentrate the product inside the bioreactor and enables a single time harvest. Such systems with a specific filtration membrane circumvent the need for continuous downstream processing.

However, despite several advancements in perfusion systems, a study by Cacciuttolo have determined fed-batch culture to be cheaper than perfusion culture; as the cost of large volumes of media requirement and harvest tanks in perfusion outweigh the benefits of higher productivity by the perfusion system (Cacciuttolo M. 2007.

Perfusion or fed-batch? A matter of perspective. In: Butler M, editor. Cell culture and upstream processing. London, UK: Taylor Francis Group, p 173-184).

Furthermore, scale up of perfusion process presents significant challenges in terms of reduced volumetric productivity, scale up and operability (WiUian C. Yang, et al.,

Journal of Biotechnology 217 (2016) 1-11).

Besides scale up and operability, it is imperative in any cell culture process to maintain the quality attributes of therapeutic proteins, such as glycosylation and variants content in case of Mabs, within defined acceptance limits to ensure efficacy and safety of therapeutic proteins.

Continuous improvements happen in the cell culture process / production methods. However, few perfusion based cell culture method/s exist that has achieved increased (volumetric) productivity without a compromise on product quality. Most importantly, such method should also be enabling for large-scale production of the product, particularly for therapeutic Mabs.

SUMMARY OF THE INVENTION To address the aforementioned problem, the present invention provides a simple yet effective feeding strategy for a perfusion cell culture process that achieves increased volumetric productivity i.e. titer/vv, without impacting the qualitative profile of therapeutic Mabs.

The said feeding strategy comprises at least two distinct phases; a phase of nutrient media addition and a phase of removal of spent media, which are decoupled from each other. Alternatively, the feeding strategy comprises at least three phases; viz., a phase of nutrient media addition, hold period and a phase of removal of spent media.

The feeding strategy can be successfully implemented at large scale / commercial scale production of therapeutic Mabs, in-turn addressing the issue of scale -up /operability.

BRIEF DESCRIPTION OF DRAWINGS

Figure 1: Representation of two phase feeding strategy (A) and three phase feeding strategy (B).The feeding strategy as represented is repeated every day, for example from day 4 onwards and thereafter, till the day of harvest, x and y in part A of figure represent duration of each phase and may be equal or not.

Figure 2: Represents % viability of the cells in Example 1(·) and Example 2A (o).

Figure 3: Represents titer/total bioreactor volume used for Example 1 and Example 2A.

Figure 4: Represents titer (in mg/ml.), % high mannose (HM) & % afucosylated glycans (AF) in an anti-CD20 antibody composition so produced in Example 1 and Example 2A. Figure 5: Represents % galactosylated glycans (GAG) in an anti-CD20 antibody composition so produced in Example 1 and Example 2A.

Figure 6: Represents % viability of the cells in Examples 2B and 2C.

Figure 7: Represents viable cell density(xl0⁶ cell /ml.) in Examples 2B and 2C.

Figure 8: Represents titer (mg/ml.) in Examples 2B and 2C.

Figure 9: Represents % of total afucosylated (TAF) glycans in an anti-CD20 antibody composition as produced in Examples 2B and 2C.

Figure 10: Represents % of galactosylated glycans (Gal) in an anti-CD20 antibody composition as produced in Examples 2B and 2C.

DETAILED DESCRIPTION OF THE INVENTION

Definitions

The term "glycoprotein" refers to protein or polypeptide having at least one glycan moiety wherein glycan refers to a monosaccharide or polysaccharide moiety. Thus, any polypeptide attached to a saccharide moiety is termed as glycoprotein.

The term "glycoform" or "glycovariant" have been used interchangeably herein, and refers to various oligosaccharide entities or moieties linked in their entirety to the Asparagine 297 (as per Kabat numbering) of the human Fc region of the glycoprotein in question, co translationaHy or post translationally within a host cell. The glycan moieties added during protein glycosylation include M3, M4, M5-8, M3NAG etc. Examples of such glycans and their structures are listed in Table 1. However, Table 1 may in no way be considered to limit the scope of this invention to these glycans.

The "glycoform composition" or "distribution" as used herein pertains to the quantity or percentage of different glycoforms present in a glycoprotein. The term "anti-CD20 antibody composition" refers to quantity or percentage of different glycoforms present in an anti-CD20 antibody preparation as obtained by performing steps as elucidated in following examples.

"Afucosylated glycans" described here, consists of grycan moiety wherein fucose is not linked to the non-reducing end of N-acetlyglucosamine. Without limitation, examples of afucosylated glycans include M3NAG, GO, G1A, GIB etc.

"Fucosylated glycans" described here, consists of glycan moiety wherein fucose which is linked to the non-reducing end of N-acetlyglucosamine. Without limitation, examples of fucosylated glycans include M3NAGF, GIAF, GIBF, G2F, G2SF, G2S2F etc. Various methods described in the art such as Wuhrer et. al., Ruhaak L.R., and Geoffrey et. al. can be used for assessing glycovariants present in a glycoprotein composition (Wuhrer M. et al., Journal of Chromatography B, 2005, Vol.825, Issue 2, pages 124- 133, Ruhaak L.R., Anal Bioanal Chem, 2010, Vol. 397:3457 3481, Geoffrey, R. G. et. al. Analytical Biochemistry 1996, Vol. 240, pages 210-226). The term "osmolality" as used herein is defined as a measure of the osmoles of solute per kilogram of solvent (osmol/kg) and may include ionized or non-ionized molecules. The osmolality may change during the cell culture process for e.g. by addition of feed, salts, additives or metabolites.

The term "Viable cell count (VCC)" refers to the no. of viable cell per unit volume at a particular time point in cell culture. The unit usually used for its measurement is million viable cells/ ml. The "VCC profile" represents graphically the VCC over the entire duration of cell culture.

The term "% Viability" refers to percentage of viable cells out of total number of cells including the viable and non-viable cells at a particular time point in cell culture. The "% viability profile" represents graphically the % viability over the entire duration of cell culture. The term "spent media" refer to media which has been depleted of nutrients, dehydrated and may be accumulated with toxic metabolites or by- products.

Perfusion process / system / scheme / mode refers to a cell culture process that involves addition of fresh nutrient media while at the same time spent media containing metabolites/ toxic by-products (e.g. lactate, ammonium) is removed.

The term "decoupling" or "decoupled" are used interchangeably and refers to a perfusion scheme wherein there is addition of nutrient media and removal of spent media, however steps or phases of addition and removal do not happen at the same time or simultaneously. In other words, the addition of nutrient media and removal of spent media are two distinct phases and one phase is initiated only after completion of the other phase.

The term "volumetric productivity" or "titer/w" or "titer/rv" refers to amount of protein i.e. titer, obtained per unit volume of the culture media. The per unit culture media can range from per ml to per reactor volume.

"Cycle time" in ATF is defined as the time taken for the diaphragm in the ATF module to completely inflate (Pressure Cycle) and deflate (Exhaust Cycle) during which the cell culture broth in the reactor gets displaced back and forth between the reactor vessel and the Hollow Fiber Module.

The term "nutrient media" refer to a liquid or gel designed to support the growth and proliferation of the recombinant cells producing polypeptide of interest. A typical nutrient media is composed of numerous components such as an appropriate source of energy, complement of amino acids, vitamins, inorganic salts, glucose, growth factors, hormones etc. The term "basal media" refers to the media in which the cells are seeded/inoculated to initiate the cell culture process. The term "feed media" refers to the media added to replenish the depleted, or to provide additional, nutrient components to the cells during cell culture process.

Detailed description of the embodiments

The present invention discloses a feeding strategy for a cell culture process for production of therapeutic Mabs.

An embodiment of the invention comprises a feeding strategy for a perfusion cell culture process for production of therapeutic Mabs.

Another embodiment of the invention comprises a feeding strategy, for a perfusion process for production of therapeutic Mabs, wherein the strategy involves decoupling of nutrient media addition and spent media removal.

Yet another embodiment of the invention comprises a feeding strategy, for a perfusion cell culture process for the production of therapeutic Mabs, wherein the feeding strategy comprises two distinct phases: a phase of nutrient media addition and a phase of removal of spent media and wherein the two phases are decoupled. Further embodiment of the invention comprises a feeding strategy, for a perfusion cell culture process for the production of therapeutic Mabs, wherein the feeding strategy comprises three distinct phases; a phase of nutrient media addition, a phase of hold period and a phase of removal of spent media. Preferably, the phase of hold period is in between the phases of media addition and removal of spent media. In any of the above mentioned embodiments, the said feeding strategy, may be initiated from start of the culture i.e. day 1 and repeated each day until harvest.

In any of the above mentioned embodiments, the said feeding strategy, may be initiated from day 4 and repeated each day until harvest. In any of the above mentioned embodiments, in the said feeding strategy, different phases may be of different duration relative to each other.

In any of the above mentioned embodiments, in the said feeding strategy, the duration of phases of media addition and hold period individually and even in combination, may be shorter in duration as compared to the phase of spent media removal.

In any of the above mentioned embodiments, in the said feeding strategy, the three distinct phases may be of duration as: a phase of fresh nutrient media addition carried 20 for 1 hour; followed by 5 hours of phase of hold period; followed by a phase of removal of spent media carried over for a duration of 18 hours; which is total of 24 hours cumulative.

In any of the above mentioned embodiments, in the said perfusion cell culture process, an alternating tangential flow (ATF) filtration system may be employed. In particular, the molecular weight cut-off (MWCO) of the filter used in ATF filtration is lower than the molecular weight of the monoclonal antibody which is to be concentrated inside the bioreactor. Preferably, the MWCO is 50 kDa.

In any of the above mentioned embodiments, for maintaining operable volume range of the bioreactor, a volume of nutrient media equivalent to volume of the media which will be perfused on day from when feeding strategy is initiated is removed.

In any of the above mentioned embodiments, the said feeding strategy results in increased titer/vv, without impacting the qualitative profile of therapeutic Mabs so produced. In particular, the % glycan species in the therapeutic Mabs so produced remains unchanged.

In any of the above mentioned embodiments, volume of the feed that is perfused each day may vary and is dependent on the cell density & viability which in turn is a characteristic of cell type, cell culture media/feed used. The cell culture media that are useful in the application include but are not limited to, the commercially available products PF-CHOTM (HyCloneTM), PowerCHOTM (Lonza), Zap-CHOTM (Invitria), CD CHO (Thermo Fisher Scientific), CD OptiCHOTM (Thermo Fisher Scientific) and CHO-S-SFMII (Thermo Fisher Scientific), ProCHOTM (Lonza), CDM4CHOTM (HycloneTM), DMEM (Invitrogen), DMEM/F12 (Invitrogen), Hams F10 (Sigma®), Minimal Essential Media (Sigma®), and RPMI -1640 (Sigma®) and IS CHO-CD G10.3TM (Irvine scientific®).

The cell culture feed that are useful in the invention include but are not limited to, the commercially available products Cell Boost TM 2 (Thermo Scientific 5 Hyclone TM, Catalogue no SH 30596.03), Cell Boost TM 4 (Thermo Scientific HyClone TM, Catalog no. SH30928), PF-CHOTM (HyCloneTM, Catalog no. SH30333.3).

Selection and optimization of cell culture media/ feed/ supplements and other parameters such as temperature, pH, osmolality and temperature and pH shifts etc. are well known in the art and general guidance is available in reference such as Freshney, R. I. Culture of animal cells (a manual of basic techniques which is incorporated herein by reference.

In another embodiment of the invention, the cells are mammalian cells. The mammalian cells may be Chinese Hamster Ovary (CHO) cells or mouse myeloma cell lines such as Sp2/0-Agl4 cells or NS0 cells. In one embodiment the bioreactor is a production bioreactor. In particular, the production bioreactor has capacity of at least 2.3 L. More particular, the production bioreactor is a large-scale production reactor with a capacity of at least 210 L.

In another embodiment of the invention the perfusion process is used for one or more stages of seed generation, seed expansion and for production bioreactor. The process according to the invention is used for the production of a recombinant protein, in particular monoclonal antibodies or their fragments thereof. More particularly the aforementioned monoclonal antibodies or their fragments thereof have therapeutic utility. Preferably the therapeutic antibodies are selected from a group of a human antibody, a humanized antibody, a chimeric antibody or fragments of such antibody thereof. Example of monoclonal antibodies are anti-CD20 antibody, anti- HER2 antibody, anti-VEGF antibody, anti-TNF alpha antibody etc.

Certain specific aspects and embodiments of the invention are more fully described by reference to the following examples, being provided only for purposes of illustration. These examples should not be construed as limiting the scope of the invention in any manner. EXAMPLES Cell line and Media:

A Chinese Hamster Ovary cell expressing anti-CD20 antibody was investigated. Proprietary basal media (IX POWER CHO 2 (Lonza, Catalog no. BE15-771)) and feed (CB4 (Hyclone, Catalog No.:SH30928.03) w/o glucose and w/o glutamine) were used in experiments.

Seed Generation

A seed expansion train consisting of no. of spinner/shake flasks and bioreactor was employed. The seed was generated in IX POWER CHO 2 media.

Example 1

Control culture process

The production bioreactor of 210 L integrated with XCell™ ATF System was assessed. The MWCO of filter membrane was 50kDa. The basal media was inoculated with the seeding density of 1.5xl0⁶ cells/ml The temperature was controlled at 37oC and pH at 7.05. Feed media was added as bolus on each of day 2, 3 and 4 in varying amounts of v/v: Day 2: 3%v/v, Day3: 4%v/v and Day4: 3%v/v. In this example, the feeding strategy employed from day 4 is nutrient addition and simultaneous removal of equal reactor volume of spent media. The total volume of media utilized in this scheme is 10 reactor volumes. A reactor volume is sum of volume of basal media and the total volume of media used in the feeding strategy.

Figure 2, 3, 4 and 5 represent % viability, titer/total bioreactor volume used, titer (in mg/ml.), % high mannose glycans & % afucosylated glycans of batch carried out as per example 1 and anti-CD20 antibody composition produced therefrom.

Example 2

Culture process

A total of three production bioreactors of 2.3 L and one of 210 L integrated with XCeU™ ATF System were assessed. The MWCO of filter membrane was 50kDa. The basal media was inoculated with the seeding density of 1.5xl0⁶ cells/ml. The temperature was controlled at 37°C and pH at 7.05. Feed media was added as bolus on each of day 2, 3 and 4 in varying amounts of v/v: Day 2: 3% v/v, Day3: 4% v/v and Day4: 3% v/v. To maintain the working volume in the production bioreactor a phase of the phase of spent media removal is initiated in the last 18 hours of day 3.

From day 4 onwards, the feeding strategy was initiated: media addition for about 1 hour followed by hold period for 5 hours and thereafter spent media removal for 18 hours. On day 6 the temperature was shifted from 37°C to 35°C, however the feeding strategy with three distinct phases was maintained. The feeding strategy (of 24 h duration) as aforementioned is repeated every day and is continued until harvest.

The representation of feeding strategy is showed for day 3-5 in figure 1. This strategy is representative until harvest for the production bioreactors. Example 2A

Table 1 represents the perfusion scheme for a 2.3 L production bioreactor. The total volume of media utilized in this scheme is 3 reactor volumes. The ATF flow rate was 0.6 LPM. Figure 2, 3, 4 and 5 represent % viability, titer/total bioreactor volume used, titer (in mg/ml.), % high mannose glycans & % afucosylated glycans of batch carried out as per example 2A and anti-CD20 antibody composition produced therefrom.

Titer per bioreactor volume and glycosylation characteristics of anti-CD20 antibody produced using process of example 2A.

Example 2 B

As can be noted from figure 1, there was decline in % viability from day 8 onwards in Example 2A as compared to Example 1. Thus, to improve in cell sustenance media exchange volume was increased to 0.3 RV on Day 5 (instead of 0.2 RV) in subsequent 2.3 L batches as represented in Table 2.

The optimization of exchange volume per day such as above can be based on cell type, cell density, culture media etc. and can be carried by person skilled in cell culture techniques.

The total volume of volume of media added and volume of spent media removed for each day and the perfusion rate for the same are given in Table 2. The ATF flow rate 5 was 0.6 LPM. The total volume of media utilized in this scheme is 3.16 reactor volumes.

Figure 6, 7, 8, 9 and 10 represent % viability, viable cell density(xl0⁶ cell /ml.), titer (in mg/ml.), % of total afucosylated & % of galactosylated glycans of two production reactor (2.3L) carried out as per example 2B and anti-CD20 antibody composition produced therefrom.

Example 2C

A no. of process parameters characteristic to perfusion mode were optimized, in particular for large scale, i.e. 210L production bioreactor. The parameters evaluated were minimizing the differences in flux, maintaining constant flow per fiber, maintaining constant ATF to filtration ratio, maintaining constant cycle time. The parameter proceeded for scale -up batches was minimizing the differences in flux & maintaining constant flow per fiber. Furthermore, agitation and aeration (e.g. scaled up linearly) were optimized for 200L production bioreactor. The agitation rate was fixed to be 110 LPM and aeration for air and oxygen was in range of 2.50LPM-8.50LPM and 0-4.0 LPM. The ATF flow rate was 54 LPM.

Figure 6, 7, 8, 9 and 10 represent % viability, viable cell density(xl0⁶ cell /ml.), titer (in mg/ml.), % of total afucosylated & % of galactosylated glycans of two production reactor (210L) carried out as per example 2C and anti-CD20 antibody composition produced therefrom.

Table 1

Day Perfusion Rate

0 NA

1 NA

2 NA

3 NA

4 0.2

5 0.2

6 0.3

7 0.2

8 0.2

9 0.1

10 0.1

11 0.1

12 0.1

Table 2

Day Perfusion Rate

0 NA

1 NA

2 NA

3 NA

4 0.2

5 0.3

6 0.3

7 0.2

8 0.2

9 0.1

10 0.1

11 0.1

12 0.1

13 NA

Claims

CLAIMS: We claim:

1. A feeding strategy for perfusion cell culture process, wherein, the strategy comprises two distinct phases: a phase of nutrient media addition and a phase of spent media removal, wherein the phases are decoupled from each other.

2. The feeding strategy of claim 1, wherein the cumulative duration of the two phases is 24 hours.

3. The feeding strategy of claim 1, wherein the phases are of different duration relative to each other.

4. The feeding strategy of claim 1, comprises an additional phase of hold period.

5. The feeding strategy of claim 4, wherein the duration of phases of nutrient media addition and hold period individually, and even in combination, are shorter in duration as compared to the phase of spent media removal.

6. The feeding strategy of claim 4, wherein the phase of nutrient media addition is of 1 hour; followed by 5 hours of phase of hold period; followed by a phase of removal of spent media carried over for a duration of 18 hours.

7. The feeding strategy of claim 1 or 4, is initiated from the day 1 onwards until day of harvest.

8. The feeding strategy of claim 1 or 4, is initiated from day 4 onwards until day of harvest.

9. The feeding strategy of claim 1 or 4, is used for production of anti-CD20 antibody.

10. The feeding strategy of claim 1 or 4, results in increased titer/vv of the produced Mabs without impacting the qualitative profile of the Mabs so produced.