Which serum-free medium and cytokine combination is used for in vitro expansion of CAR-T cells?

When doing BCMA CAR-T, we remove IL-2 and replace it with IL-7 10 ng/mL + IL-15 5 ng/mL, adding from the day of activation, and from D3 onward, half-volume medium exchange with cytokine supplementation every 2 days. This combination helps maintain CD8 and Tscm phenotypes, but it varies between different donors. Some protocols use low-dose IL-2 (10-20 IU/mL) plus IL-7/IL-15 to avoid insufficient early expansion caused by completely withdrawing IL-2. To maintain a stem cell-like phenotype, IL-15 should not be lower than 5 ng/mL, and IL-7 can be 10-12.5 ng/mL. After D7, if proliferation is slow, IL-15 can be supplemented once more to 10 ng/mL, but do not frequently increase the concentration, as this tends to increase terminal differentiation. For cytokines, it is recommended to purchase clinical-grade carrier-free protein, aliquot and freeze after dissolution, and avoid repeated freeze-thaw cycles.

Slow late-stage expansion with X-VIVO 15 plus IL-2 is very common; it is not specifically optimized for CAR-T expansion. After switching to TexMACS GMP or ImmunoCult-XF T Cell Expansion, the D10 expansion fold can be more than 30% higher, and CD8 maintenance is slightly better, but the price is higher. Among domestic options, we have evaluated T cell serum-free media from Youkang and OPM, with a lead time of 3-5 days and a price about 40% lower than imported products, but their ability to support high-density expansion needs to be tested in-house. It is recommended to use the same donor and the same magnetic bead activation, and separately measure cell density, glucose/lactate, and pH from D3-D12 to compare expansion fold and viability. If 2-5% autologous plasma or serum replacement is added to the medium, viability after D7 is usually more stable, but introducing serum requires attention to lot-to-lot variability and clinical compliance.

CD8 ratio decline and slow proliferation should be assessed separately: it may be caused by excessive activation strength, high bead-to-cell ratio, or too high IL-2 dose, leading to CD4-favored expansion and exhaustion. We have performed DOE. When the bead-to-T cell ratio was reduced from 3:1 to 1:1, activation time was shortened from 3 days to 2 days, and IL-7 + IL-15 were then used, the CD8 ratio at D10 increased by 10-15 percentage points, and the Tscm ratio was also higher. It is recommended to monitor CD4/CD8, CD45RA/CD62L, CCR7, and PD-1/LAG-3/TIM-3 by flow cytometry, testing once each at D5 and D9. If the PD-1/LAG-3 double-positive ratio exceeds 30%, it indicates that the expansion protocol may be skewed toward exhaustion, and it is necessary to reduce IL-2 and remove beads earlier. After switching to a different lot of domestic medium, it is also recommended to perform small-scale validation with 3 donors; do not draw conclusions based on only one donor.

<p>Slower proliferation on Day 7 may not be due to the medium itself, but rather excessive density and accumulation of metabolic waste. We generally maintain a density of 0.5-2×10^6/mL, start counting every 2-3 days from D3, and split and replenish medium when the density exceeds 2×10^6, rather than waiting until D5 to change it. After D5, it is recommended to perform a half-volume medium change, retaining a portion of the conditioned medium to reduce interference with cytokines caused by centrifugation. If glucose is at 20 mM, proliferation will drop significantly; you can supplement with GlutaMAX, sodium pyruvate, or simply replace with fresh medium. When changing medium, supplement cytokines to the final concentration, not by simply adding linearly based on the volume of medium replenished. Additionally, the CD8 proportion drops after D7; you can check whether the magnetic beads have been removed. Prolonged magnetic bead stimulation tends to favor CD4 expansion, and removing the beads at D3-D5 usually improves this.</p>