Unlocking Potential: A Comparative Study of CAR T Cell Expansion in Stirred Tank vs Static Bioreactors

Biomanufacturers use bioreactors to expand genetically modified immune cells during cell therapy development due to their closed design and the potential to reduce contamination. Largely, biomanufacturers depend on static bioreactors or bioreactors that agitate expanding cells either using a rocking motion or a stirring action.

In a previous application note, we showed that expansion using automated stirred-tank bioreactors improves unedited T cell expansion over static bioreactors. In this study, we sought to learn the effects of expanding genetically edited T cells in a stirred tank bioreactor, in comparison to expansion in a static bioreactor.

Summary of methods

Frozen leukopak bags containing white blood cells (WBC) were thawed. T cells were isolated from leukopaks using the Gibco(TM) CTS(TM) DynaCellect(TM) Magnetic Separation System and the Gibco CTS Dynabeads(TM) CD3/CD8 beads. Following isolation, the T cells were activated in static GREX® 100 bioreactor containing CTS OpTmizer(TM) Pro Serum Free Medium (SFM) supplemented with 100U/mL of IL-2.

On the third day, the magnetic CTS Dynabeads CD3/CD28 beads were separated from the T cells using the CTS DynaCellect system.

Following the debeading process, electroporation buffer exchange was carried out on the Gibco CTS Rotea Counterflow Centrifugation System. The payload (CTS TrueCut Cas9 Protein, TrueGuide synthetic gRNA, and custom CD19 CAR DNA) and the activated T cells were then transferred to the CTS Xenon Electroporation System via a welded PVC tubing (connected to the line H on the Gibco CTS Rotea Single-Use Kit. The CTS Xenon Multishot Electroporation Cartridge was used in the electroporation of T cells.

Electroporated cells were then expanded either in a BioBLU Single-Use Stirred Tank bioreactor vessel or in a static GREX® 6-well bioreactor vessel. Post-electroporation analyses were performed at days 3 and 10.

All data is representative of at least three independent experiments.

Results

Figure 1: Cell viability post-electroporation.

While edited T cells expanded in the stirred tank vessels saw an initial dip at day 3 post-electroporation, viability increased to over 90% by day 10. On the other hand, while viability remained high, for CAR T cells expanded in static GREX vessels, this number dropped between the day of electroporation and day 10 post-electroporation (figure 1). This progressive overall decrease in viability, did not result in a decrease in expansion for CAR T cells expanded in GREX vessels however (figure 2B).

By day 10 post-electroporation, while the percent KO/KI did not differ between CAR T cells expanded in stirred tank vessels compared to static vessels (figure 2A), there was a four-fold increase in the total number of CAR T cells that were expanded in stirred tank vessels over static bioreactor vessels (figure 2B).

Figure 2: Gene editing efficiency and total number of edited cells at days 3 and 10 post electroporation.

An average of 30% of edited T cells expressed anti-CD19 CAR on day 10 following expansion in the stirred tank bioreactor. This represented an average of 2 billion edited cells. For GREX vessels, while the average of T cells expressing anti-CD19 CAR by day 10 was approximately 20%, this represented 500 million total edited T cells. The four-fold increase in the number of total edited T cells expanded in stirred tank bioreactor vessels suggests that agitation enhances edited T cell expansion.

Figure 3: Cell phenotypes of edited cells at day 10 post-electroporation.

Despite the clear difference in expansion, the proportion of cell types did not differ greatly between stirred tank-expanded cells and static GREX-expanded cells at day 10 post-electroporation. B cell contamination was highest in non-electroporated controls expanded in GREX bioreactor vessels.

Figure 4: Total number of effector memory and naïve central memory T cells on day 10 post-electroporation.

While it did not reach statistical significance, there was a slightly higher naïve central memory T cell count compared to effector memory T cells with stirred tank-expanded CAR T cells as compared to the GREX.

Figure 5: Killing assay on days 3 and 10 post-electroporation.

Finally, CAR T cells expanded in stirred tank vessels killed Nalm6 target cells more effectively than GREX-expanded cells both at day 3 and day 10 post-electroporation.

Overall conclusions:

A previous study reported an improvement in expansion when unedited T cells were cultured in stirred tank bioreactors. Our goal here was to learn how expanding edited/CAR T cells in stirred tank bioreactors would compare with expansion in a static G-REX bioreactor vessel.

The data shows that increased agitation during the expansion phase of the CAR T cell manufacturing process leads to an improved expansion of edited T cells. Having a significantly increased number of edited T cells is useful for patients receiving CAR T cell infusion as part of their treatment plan as it could enhance the proliferative capacity of the CAR T cells.

We also observed that agitation did not adversely affect the viability of edited cells or affect the proportion of T cells post-electroporation. The proportion of naïve central memory T cells was slightly improved when the edited cells were expanded in stirred tank bioreactors. The ability of CAR T cells to engraft into a patient following infusion is related to their state of differentiation. Less-differentiated naïve/central memory cells show the greatest potency in published pre-clinical studies (1-4).

Furthermore, CAR T cells expanded in stirred tank bioreactor vessels exhibited improved killing of target cells at three days post-electroporation, and to a lower extent, ten days post-electroporation. This data suggests a higher potency for stirred tank-expanded CAR T cells to target and attack cancerous cells.

References:

1. Berger C, et al. Adoptive transfer of effector CD8⁺T cells derived from central memory cells establishes persistent T cell memory in primates.  Clin. Investig. 2008;118:294–305. doi: 10.1172/JCI32103.
2. Graef P, et al. Serial transfer of single-cell-derived immunocompetence reveals stemness of CD8⁺central memory T cells. Immunity. 2014;41:116–126. doi: 10.1016/j.immuni.2014.05.018.
3. Hinrichs CS, et al. Adoptively transferred effector cells derived from naive rather than central memory CD8⁺T cells mediate superior antitumor immunity.  Natl Acad. Sci. USA. 2009;106:17469–17474. doi: 10.1073/pnas.0907448106.
4. Hinrichs CS, et al. Human effector CD8⁺T cells derived from naive rather than memory subsets possess superior traits for adoptive immunotherapy.  2011;117:808–814. doi: 10.1182/blood-2010-05-286286.