Sunghoon Junga, Breanna Borysb, Tania Sob, Tiffany Dangb, Erin Robertsb, Michael Kallosb, Robert Kirianc, Josephine Lembongc, Taby Ahsanc, Jon Rowleyc, Yas Hashimuraa

a PBS Biotech Inc., 1183 Calle Suerte, Camarillo, CA, 93012, USAb Pharmaceutical Production Research Facility (PPRF), Department of Chemical and Petroleum Engineering, University of Calgary, 2500 University Drive NW, Calgary, Alberta, Canada T2N 1N4

c RoosterBio Inc., 5295 Westview Dr, Suite 275, Frederick, MD, 21703, USA

BackgroundA Reliable and Scalable Single-use Bioreactor System is Needed for Large-Scale Manufacturing of Cell Therapy ProductsØ Stirred-type mixing can result in a high degree of hydrodynamic shear stress and

an uneven gradient of energy dissipation, negatively affecting cell yield anddifferentiation efficiency, quality, and potency

Vertical-Wheel Technology Enables Superior, Scalable Expansion andDifferentiation Performance for Numerous Types of Cell Therapy Products

Ø Vertical-Wheel bioreactors provide the ideal conditions of homogeneous energydissipation distribution, complete particle suspension with minimal power input,and low hydrodynamic shear stress for various types of cell therapy products

Ø For processes involving cells grown on microcarriers such as MSCs, Vertical-Wheel bioreactors provide superior performance of cell expansion as well asrapid, in-vessel cell dissociation and harvesting

Ø For processes involving cells grown as aggregates such as ESCs and iPSCs,Vertical-Wheel bioreactors provide unmatched control of cell aggregate size andmorphology, leading to superior performance of cell expansion (whilemaintaining pluripotency) and subsequent differentiation efficiency

Ø Vertical-Wheel bioreactors have demonstrated unparalleled scalability across afull range of vessel sizes, enabling the scale up of cell therapy manufacturing tomeet commercial production needs

Superior Scalability and Robustness of Cell Therapy Manufacturing Using Vertical-Wheel Bioreactors

Experimental Results cont.

Conclusion

The Authors would like to thank RoosterBio (Fig. 3), University of Calgary (Fig. 4), Semma Therapeutics (Fig. 5), and University of Lisbon (Fig. 6) for allowing PBS

Biotech to show their experimental data

WWW.PBSBIOTECH.COM INNOVATIVE BIOPROCESS SOLUTIONS

Directed Differentiation Of Human iPSCs

To Cerebellar Organoids In Vertical-Wheel Bioreactor

INNOVATIVE BIOPROCESS SOLUTIONS WWW.PBSBIOTECH.COM INNOVATIVE BIOPROCESS SOLUTIONS

o Courtesy of Professor Joaquim Cabral and Ms. Teresa Silva

(Days)

Cell Line 1

Cell Line 2

Ø After 35 days of generation, iPSC-derived organoids were efficientlymaturated to GABAergic and Glutamatergic neurons (not shown) in PBS-0.1(scale bar, 100 µm)

Fig 6. Directed Differentiation of Human iPSCs to Cerebellar Organoids in PBS-0.1 Vertical-Wheel Bioreactor

Fig 5. Size and Morphology Comparison of ESC Aggregates Grown in PBS-0.5 and PBS-3 (Vertical-Wheel) vs. Spinner (Stirred)

Spinner (500mL)

PBS-0.5 (500mL)

PBS-3 (3L)

0

50

100

150

200

250

300

24 hr 72 hr

Cel

l Agg

rega

te D

iam

eter

(um

)

Hours in Culture

PSB 3MAGPBS MiniPBS-3 PBS-mini PBS-0.5

WCB PBS-0.5 PBS-3 PBS-15 PBS-80

Fig 2. A Representative, Large-Scale Cell Manufacturing Process Scheme Using the Family of Vertical-Wheel Bioreactors

Successful applications of Vertical-Wheel bioreactors Include:

Ø Rapid growth of freshly thawed human cells that were inoculated directly intobioreactors containing microcarriers

Ø Serial passaging of cells into progressively larger Vertical-Wheel bioreactorswhile maintaining high levels of cell growth and pluripotency

Ø Highly effective in-vessel dissociation of cells from microcarriers bymanipulating agitation speed during enzymatic dissociation

Ø Rapid, efficient medium exchange and cell harvest through the use of a built-in, adjustable height dip tube (PBS-3) or harvest valve (PBS-15 & PBS-80)

Fig 3. Comparable hMSC Growth Profile Across Scales and Robust Process Across Donors

Ø (A) Small scale 0.1L culture process (n=3) was successfully scaled up to 3Ldevelopment scale (n=3), 15L pilot scale (n=3), and 50L production scale (n=2)in Vertical-Wheel bioreactors

Ø (B) The culture process works for multiple donors and can be robustly scaled to50L production scale

A B

0.E+00

1.E+05

2.E+05

3.E+05

0 2 4 6

Viab

leCells/

mL

Time(Days)PBS40rpm PBS60rpm PBS80rpmNDS40rpm NDS60rpm NDS80rpm

Day6FoldExpansion

PBS NDS

40rpm 16.7± 1.1 4.9± 2.8

60rpm 11.2± 0.9 3.3± 0.3

80rpm 7.7± 0.5 6.3± 2.7

Ø Single cell inoculation with seeding density of 20,000 cells/mL led to successfulcell expansion in small-scale Vertical-Wheel bioreactors (PBS-0.1) compared totraditional horizontal-blade mixing (NDS)

Fig 4a. Comparison of iPSC Aggregate Expansion with Different Agitation Rates in PBS-0.1 vs. Spinner (NDS)

10

PBS

NDS

PBS

NDS

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

PercentofPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

Day1 Day3 Day5

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

40rpm(D5) 60rpm(D5) 80rpm(D5)

200um

200um

200um

200um

200um

200um 10

PBS

NDS

PBS

NDS

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

PercentofPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

Day1 Day3 Day5

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

40rpm(D5) 60rpm(D5) 80rpm(D5)

200um

200um

200um

200um

200um

200um 10

PBS

NDS

PBS

NDS

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

PercentofPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

Day1 Day3 Day5

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

40rpm(D5) 60rpm(D5) 80rpm(D5)

200um

200um

200um

200um

200um

200um

PBS(Vertical-Wheel)

NDS(Horizontal-

Blade)

Fig 4b. Comparison of iPSC Aggregate Size and Morphology with Different Agitation Rates in PBS-0.1 vs. Spinner

Ø After five days of cell expansion, there was a marked difference in the size andmorphology of cell aggregates grown in PBS-0.1 compared to horizontal-bladebioreactors, even with different agitation rates

Ø ESC aggregates grown in spinner flasks show a wide range of sizes andmorphologies, especially after 72 hours

Ø In comparison, ESC aggregates grown in Vertical-Wheel bioreactorsmaintain uniform size and morphology not only after 72 hours but alsoduring scale up from 500mL to 3L working volume

Ø Complete cell suspension and efficientfluid mixing can be rapidly achieved withlow power input

Ø Cells experience a microenvironmentwith low hydrodynamic shear stress andhomogenous distribution of energydissipation

Ø Optimal size and morphology of cellaggregates can be tightly controlled byadjusting the agitation rate

Fig 1. Vertical-Wheel Schematic

Experimental Results

Ø Much tighter distribution of cell aggregate sizes in PBS-0.1 compared tohorizontal-blade bioreactors, especially after five days

Ø Inverse correlation of agitation rate with cell aggregate diameter: higher rateresults in smaller average diameter and vice-versa

Fig 4c. Comparison of iPSC Aggregate Size Distribution with Different Agitation Rates in PBS-0.1 vs. Spinner

10

PBS

NDS

PBS

NDS

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

PercentofPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

Day1 Day3 Day5

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

0

20

40

60

80

50 250 450 650 850

Percento

fPop

ulation

Aggregate Diameter (µm)

40rpm(D5) 60rpm(D5) 80rpm(D5)

200um

200um

200um

200um

200um

200um

Aggregate Diameter (µm)

PBS(Vertical-Wheel)

NDS(Horizontal-

Blade)

0.E+00

1.E+05

2.E+05

3.E+05

0 2 4 6

Viab

leCells/

mL

Time(Days)PBS40rpm PBS60rpm PBS80rpmNDS40rpm NDS60rpm NDS80rpm

Day6FoldExpansion

PBS NDS

40rpm 16.7± 1.1 4.9± 2.8

60rpm 11.2± 0.9 3.3± 0.3

80rpm 7.7± 0.5 6.3± 2.7

0.E+00

1.E+05

2.E+05

3.E+05

0 2 4 6

Viab

leCells/

mL

Time(Days)PBS40rpm PBS60rpm PBS80rpmNDS40rpm NDS60rpm NDS80rpm

Day6FoldExpansion

PBS NDS

40rpm 16.7± 1.1 4.9± 2.8

60rpm 11.2± 0.9 3.3± 0.3

80rpm 7.7± 0.5 6.3± 2.7

0.E+00

1.E+05

2.E+05

3.E+05

0 2 4 6

Viab

leCells/

mL

Time(Days)PBS40rpm PBS60rpm PBS80rpmNDS40rpm NDS60rpm NDS80rpm

Day6FoldExpansion

PBS NDS

40rpm 16.7± 1.1 4.9± 2.8

60rpm 11.2± 0.9 3.3± 0.3

80rpm 7.7± 0.5 6.3± 2.7

0.E+00

1.E+05

2.E+05

3.E+05

0 2 4 6

Viab

leCells/

mL

Time(Days)PBS40rpm PBS60rpm PBS80rpmNDS40rpm NDS60rpm NDS80rpm

Day6FoldExpansion

PBS NDS

40rpm 16.7± 1.1 4.9± 2.8

60rpm 11.2± 0.9 3.3± 0.3

80rpm 7.7± 0.5 6.3± 2.7

0.E+00

1.E+05

2.E+05

3.E+05

0 2 4 6

Viab

leCells/

mL

Time(Days)PBS40rpm PBS60rpm PBS80rpmNDS40rpm NDS60rpm NDS80rpm

Day6FoldExpansion

PBS NDS

40rpm 16.7± 1.1 4.9± 2.8

60rpm 11.2± 0.9 3.3± 0.3

80rpm 7.7± 0.5 6.3± 2.7

0.E+00

1.E+05

2.E+05

3.E+05

0 2 4 6

Viab

leCells/

mL

Time(Days)PBS40rpm PBS60rpm PBS80rpmNDS40rpm NDS60rpm NDS80rpm

Day6FoldExpansion

PBS NDS

40rpm 16.7± 1.1 4.9± 2.8

60rpm 11.2± 0.9 3.3± 0.3

80rpm 7.7± 0.5 6.3± 2.7

24 hr 72 hr

SCALING A XENO-FREE FED-BATCH MICROCARRIER SUSPENSION BIOREACTOR SYSTEM FROM DEVELOPMENT TO PRODUCTION SCALE FOR MANUFACTURING XF hMSCs

Josephine Lembong1, Robert Kirian1, David Wang1, Joseph Takacs1, Ang-Chen Tsai2, Kenny Cruz2, Francisco Rosello2, Kayley Cox2, Yas Hashimura2, Jon Rowley1, Sunghoon Jung2, Taby Ahsan1

1 RoosterBio, Inc. 5295 Westview Drive, Suite 275, Frederick, MD 217032 PBS Biotech, Inc. 1183 Calle Suerte, Camarillo, CA 93012

ABSTRACT

• There have been > 900 clinical trials investigating the use of humanmesenchymal stem/stromal cells (hMSCs) for regenerative medicine,driving the need for an economical biomanufacturing paradigm.

• A scalable process and production technology platform that cangenerate billions to trillions of cells per manufacturing lot is needed tomeet the clinical demand.

• Suspension bioreactors show great promise in reaching commercially-viable working volumes. However, scalability of cell production remainsa challenge.

•

• This study developed a scalable xeno-free (XF) hMSC bioreactorprocess that maintains the final cell population doubling level (PDL)within the recommended range of 16-20 to ensure product quality.

• The ability to scale XF hBM-MSC expansion in a low shear, single-use,vertical-wheel suspension bioreactor was evaluated at small scale(0.1L), development scale (3L), pilot scale (15L), and production scale(50L), using high volume XF cell banks, an optimized XF fed-batchmedia system, and XF microcarriers in a scalable bioreactor system.

• Cell yields of >0.5M cells/ml were achieved in all bioreactor scaleswithin 5 days of culture with no media exchange. Comparable nutrientand waste profiles, cell growth curves, and pH were observed in allscales.

• Cells from all bioreactor scales maintain the hMSC critical qualityattributes of osteogenic, adipogenic, and chondrogenic differentiationpotential, as well as functional attributes of angiogenic cytokine (FGF,HGF, IL-8, TIMP-1, TIMP-2, and VEGF) secretion and inducibleimmunomodulatory potential (as measured by functional IDO activity),which are comparable to 2D control cells of similar PDL.

• Our expansion of XF hMSCs in a scalable bioreactor culture platformcan provide significant time and cost savings for translationalresearchers and product developers in the regenerative medicine,tissue engineering, and cell therapy fields.

• hMSCs seeded into bioreactor on day 0 are fed with RoosterReplenishTM-MSC-XF on day 3, and are ready for harvest on day 4 or 5 of culture.

ESTABLISHED hMSC BIOREACTOR EXPANSION PROCESS

FED-BATCH PROCESS OUTPERFORMS 1/2 MEDIA EXCHANGE & BATCH CULTURE, AND

INCREASES MEDIA PRODUCTIVITY IN BIOREACTOR

• Fed-Batch process shows a distinct advantage on final cell yield and mediaproductivity over ½ media exchange and Batch process. (‘*’ indicates statisticalsignificance between Fed-Batch and Batch systems at Day 5 and Day 6(p<0.05)). Comparison study was performed in scaled down 0.1L bioreactors.

COMPARABLE hMSC GROWTH PROFILE ACROSS SCALES & ROBUST PROCESS ACROSS DONORS

• hMSC proliferation were monitored throughout culture as observed by the formation of cell-microcarrier agglomerates during expansion.

• Distribution of cells on microcarriers throughout the bioreactor culture was comparable across scales,demonstrating a scalable seeding strategy. Blue dashed lines indicate end of bioreactor culture.

• The small scale 0.1L bioreactor culture process (n=3) was successfully scaled to the 3L developmentscale (n=3), 15L pilot scale (n=3), and 50L production scale (n=3), showing similar growth profiles.

• The scalable bioreactor process is robust and works for multiple donors.

COMPARABLE NUTRIENT / WASTE / pH PROFILE

• Concentration of glucose, glutamine, lactate, and ammonia were comparable in all bioreactorscales (0.1L, 3L, 15L, 50L), and are maintained at desired levels throughout culture to support cellexpansion. Glutamine is present in excess in media, and breaks down into ammonia withincreased storage duration.

hMSC MANUFACTURING PLATFORM EVOLUTION

• hMSC manufacturing platforms have evolved as the demand for cells increases.• Suspension bioreactor production of hMSCs is a scalable manufacturing platform that can

provide the necessary lot sizes of billion to trillions of cells while reducing the time, labor,and cost of goods (COGs) for regenerative medicine applications.

MAINTAINED CELL CRITICAL QUALITY ATTRIBUTES AND FUNCTIONALITY ACROSS BIOREACTOR SCALES

COMPARABLE FINAL HARVEST YIELDS

• Cell detachment from microcarriers was accomplished through agitation of the bioreactorwheel in the presence of an enzymatic cell dissociation reagent.

• Efficient removal of microcarriers from the bulk cell solution was accomplished by passingthrough a filtration device. ǂ Parameters for downstream processes in the 50L scale have notyet been optimized.

Pre-Harvest

Post-Harvest

Post Microcarrier Separation

Bioreactor Scale Cell Density at Harvest Total Cell Yield0.1L (n = 3) 615,000 cells/ml 61.5 Million cells3L (n = 3) 504,000 cells/ml 1.5 Billion cells15L (n = 3) 664,000 cells/ml 10 Billion cells50L (n = 3) 635,000 cells/ml 31 Billion cells

• XF hMSCs expanded in the bioreactor maintained their tri-lineage differentiationpotential, expansion potential, and cell surface marker expression identity.

• Cell functional attributes are maintained as shown by comparable inducibleindoleamine 2,3-dioxygenase (IDO) activity when stimulated with interferon-gamma(IFNᵧ), and angiogenic cytokine secretion profile.

0

0.1

0.2

0.3

0.4

0 1 2 3 4 5 6Cel

l Den

sity

(M c

ells

/ml)

Cultivation time (day)

RoosterReplenishTM-MSC-XFOR ½ Media Exchange

0

50

100

150

200

250

300

350

400

Med

ia P

rodu

ctiv

ity

(M c

ells

/lite

r med

ia)

**

Seed hMSCs on Microcarriers

Add Feed to Culture

Harvest and Vial hMSCs

Day 0 Day 3 Day 5

COMPARABLE CELL EXPANSION ON MICROCARRIERS

A B

SUMMARY

▪ A robust, scalable XF hBM-MSC culture process was developed in avertical-wheel suspension bioreactor up to production scale (50L).• Consistent yields of >0.5M cells/ml is achieved in all scales• hMSC critical quality attributes and functionality are maintained

▪ The demonstrated scalability of hBM-MSC culture allows significanttime and cost savings as a plug-and-play standardized system fortranslational researchers and hMSC-based product developers.

2D

3Lbioreactor

15Lbioreactor

50Lbioreactor

Figure adapted from Rowley & Montgomery, BioProcess International 2018

OSTEOGENESIS ADIPOGENESIS CHONDROGENESIScontrol diff control diff control diff