Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
ADVANCING THE ROBUST MANUFACTURE OF T-CELL THERAPIES THROUGH THE
APPLICATION OF STIRRED TANK BIOREACTORS
Alex Klarer1,Matthew Marsh1, Shaun Ranade1, David Smith1
1Hitachi Chemical Advanced Therapeutics Solutions, Allendale, NJ
Introduction
With the rise of T-Cell immunotherapies cell therapy developers have
encountered the difficulty of producing consistent, quality products.
Approved therapies Kymria™ and Yescarta™ as well as late phase
clinical products BB2121 and UCART19 have been critiqued for
potential obstacles to patient access due to high sticker prices and
variable manufacturing outcomes. Manufacturing processes need to
be developed to allow for increased control over each unit step to
improve the uniformity of the cell journey over the life of the product.
Bioreactor culture systems provide a pathway for a higher level of
process control over cell expansion.
With the high cost of developing and manufacturing cell therapies,
manufacturing techniques that allow for rapid translation to the clinic
and reduce operating costs need to be investigated. Stirred-tank
bioreactors have been heavily adopted by the biologics industry and
have well defined characteristics that facilitate the scale-up from small
scale development work to full scale clinical and commercial
manufacture. Established modelling of the fluid dynamics present
within these bioreactors may also reduce the reagent costs of cell
culture with more efficient mass transfer.
Conclusion
T-Cell immunotherapies have proven efficacy as viable medications for
previously untreatable diseases. However, manufacturing options for these
transformative therapies threaten patient access due to inconsistent
manufacturing, long processing times, and high cost. Stirred-tank bioreactors
present an option for the expansion of primary human T-Cells that has
previously not been pursued due to concerns about shear stress. Small scale
studies have shown that impeller driven mixing in stirred-tank bioreactors does
not increase cell death or limit expansion potential. Translating the learnings
from those small scale studies to the autologous manufacturing scale has
shown that the BIOne stirred-tank bioreactor produces a consistent culture
environment and high cell yields. The BIOne significantly outperformed rocking
bioreactor systems in both total yield and production efficiency. These
improvements could impact patient access by reducing the manufacturing
timeline and reducing the cost from both reagents and labor. The high yield of
the BIOne was only possible by integrating it with the Lovo spinning membrane
system to allow for media exchange; nutrient and metabolite analysis
throughout the culture reveals that increasing the exchange rates would further
prevent lactate accumulation and potentially further increase yields.
0.0
5.0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
mg/
dL
Day
Lactate Concentration Over Time
ambr - Low Perfusion
ambr - High Perfusion
Batch Fed
Static
Figure 2. The change in key metabolite
concentrations over the course of the thirteen
day culture (N=3).
0.0
2.0
4.0
6.0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
mg/
dL
Day
Glucose Concentration Over Timeambr - Low Perfusionambr - High PerfusionBatch FedStatic
Establishing Key Parameters of Stirred-Tank Bioreactor Cultures
Autologous Scale Expansion of Primary T-Cells
0.0E+00
2.0E+07
4.0E+07
6.0E+07
8.0E+07
1.0E+08
0 2 4 6 8 10 12
Cel
ls
Day
ambr - Low Perfusion
ambr - High Perfusion
Batch Fed
Static
0.0E+00
2.0E+07
4.0E+07
6.0E+07
8.0E+07
1.0E+08
0 2 4 6 8 10 12
Cel
ls
Day
ambr - Low Perfusion
ambr - High Perfusion
Batch Fed
Static
Figure 1. Growth curves from two donors are compared independently to
their control cultures (N=3). The largest sample to sample increase in
growth occurs between day five and seven. Bioreactor cultures performed
markedly better in two of the three donors, but more development is
necessary to improve consistency across various donor types.
The results of the small
scale study showed that the
shear stress of a stirred
tank bioreactor does not
inhibit cell proliferation and
may provide additional
efficiencies in mass transfer.
The experimental conditions
outperformed the static
controls and were able to
achieve cell densities
comparable to continuously
perfused systems using a
rudimentary mock batch-fed
process. The results
indicate that high
concentrations of lactate
prevented further expansion
of the cultures.
References
➢ Marenghi, Angela, et al. “Perfusion’s Role in Maintenance of High-Density T-Cell Cultures.”BioProcess International, 13 Jan. 2015, www.bioprocessintl.com.
➢ Nienow, Alvin W., et al. “The Physical Characterisation of a Microscale Parallel Bioreactor Platform with an Industrial CHO Cell Line Expressing an IgG4.” Biochemical Engineering Journal, vol. 76, 2013, pp. 25–36., doi:10.1016/j.bej.2013.04.011..
➢ Rameez, Shahid, et al. “High-Throughput Miniaturized Bioreactors for Cell Culture Process Development: Reproducibility, Scalability, and Control.” Biotechnology Progress, vol. 30, no. 3, May 2014, pp. 718–727.,
doi:10.1002/btpr.1874.
➢ Molleryd, Carin, et al. “Scaling up Clinical T Cell Expansion in a Xuriâ„¢ Cell Expansion System.” Cytotherapy, vol. 17, no. 6, 2015, doi:10.1016/j.jcyt.2015.03.379.
➢ Li, F., et al. “A Systematic Approach for Scale-Down Model Development and Characterization of Commercial Cell Culture Processes.” Biotechnology Progress, vol. 22, no. 3, Feb. 2006, pp. 696–703., doi:10.1021/bp0504041.
Medium Exchange Using Spinning Membrane Filtration
Figure 7. Nutrient and metabolite analysis shows that media exchange
rates were not sufficient to prevent lactate accumulation and cell death
late in the culture. Lactate concentrations and LDH, a measure of cell
lysis, have a linear coefficient of determination of 0.72.
y = 2000e-0.236x
0
500
1000
1500
2000
2500
0 5 10 15 20 25 30Vo
lum
e o
f Sp
ent
Med
ia [
mL]
Hour
Figure 6. Assuming the system functions as a true continuous stirred
tank reactor, spent media removal is modelled as an exponential decay
function to determine the true media exchange rate at a set perfusion
rate.
Configuration
T-Cell Isolation and Activation: Lymphocytes were isolated from apheresis collections by elutriation and incubated
with CD3/CD28 Dynabeads™ at room temperature with light agitation. The cell-bead conjugates were incubated at
37°C and 5% CO2 in static Vuelife culture bags for 72 hours to fully activate. The cells were partitioned into the
various culture conditions after the activation period.
Culture Media: TexMACS media (Miltenyi) was supplemented with 5% human AB serum, 0.1% Pluronic F-68
(Thermo-Fisher), 166 ppm Antifoam-C (Sigma Aldrich), and 30 U/mL IL-2 (R&D Systems). Media exchange, when
implemented, was set to 25% for cell concentrations >1.5x106 cells/mL, 50% for cell concentrations >2.0x106
cells/mL, 75% for cell concentrations >3.0x106 cells/mL, and 100% for cell concentrations >4.0x106 cells/mL.
Culture Devices: The Distek BIOne bioreactor was used as the stirred-tank culture vessel. Media exchange was
performed on the BIOne through integration with the Lovo (Fresenius Kabi) to remove media through the spinning
membrane and replacing removed media directly into bioreactor. Rocking motion cultures used the Xuri Cell
Expansion System (GE Healthcare) and exchanged media through the internal media filter. Cultures were controlled
to 50% dissolved oxygen and a pH of 7.15 with N2, CO2, and AIR via sparging in the BIOone and overlay in the Xuri.
Figure 3. Activated T-Cells expanded in Distek BIOne
bioreactors showed the capacity for significant expansion
across four experiments and two donors. In all cases the
cultures yielded greater than 15 x 109 cells prior to eight
days of total culture time.* The four experiments show
consistent growth profiles within donors which indicates that
the BIOne is able to create a culture environment that is
highly reproducible lot to lot.
*BIOne Run 2 was terminated after day five due to a
malfunction of the perfusion system.
-1.00
0.00
1.00
2.00
3.00
4.00
5.00
6.00
0 1 2 3 4 5 6 7 8 9
Pop
ula
tio
n D
ou
blin
gs
Days in Culture
BIOne Run 1
BIOne Run 2
BIOne Run 3
BIOne Run 4
-1.00
0.00
1.00
2.00
3.00
4.00
5.00
6.00
0 1 2 3 4 5 6 7 8 9
Pop
ula
tio
n D
ou
blin
gs
Days in Culture
BIOne Run 3
BIOne Run 4
Rocking Bioreactor
0.0E+00
5.0E+05
1.0E+06
1.5E+06
2.0E+06
2.5E+06
3.0E+06
3.5E+06
0 1 2 3 4 5 6 7 8
Cel
ls p
er M
illili
ter
of
Med
ia C
on
sum
ed
Days in Culture
BIOne
Rocking Bioreactor
0
0.5
1
1.5
2
2.5
3
0 2 4 6 8
Lact
ate
Co
nce
ntr
atio
n [
g/L]
Days in Culture
BIOne Run 3
BIOne Run 4
Rocking Bioreactor
0
20
40
60
80
100
120
140
160
180
200
0 2 4 6 8
LDH
Co
nen
trat
ion
[U
/L]
Days in Culture
BIOne Run 3
BIOne Run 4
Rocking Bioreactor
Figure 4. Comparing a single donor across the
BIOne stirred tank system and the rocking
bioreactor system indicates that the BIOne enables
increased cell expansion toward the later stage of
the cell culture. The average increase in expansion
was 1.22 population doublings. This difference in
expansion shown in this experiment could
correspond to a 24-48 hour reduction in total
culture length depending on the target yield.
Culture is typically the limiting factor in production
speed and reducing the length of time required to
achieve a therapeutic yield will result in faster
product release.
Figure 5. In addition to higher total yields, the
BIOne was shown to expand cells more efficiently
than the rocking bioreactor. 2.99 x 106 cells were
produced per milliliter of media consumed in the
stirred-tank bioreactor which is a 58% increase
over the 1.88 x 106 cells produced per milliliter in
the rocking bioreactor. Accounting for the high cost
of media components in standard T-Cell media,
more efficient cell expansion would help reduce the
cost of the resulting therapy.