As engineered T-cell therapies expand into new indications, researchers are re-evaluating how early manufacturing decisions shape the biology of the final product. A new closed, consolidated workflow, published in Cell and built on a Terumo BCT platform, is designed to bring activation, transduction, and expansion into a single environment — a shift that could clarify how culture conditions influence T-cell phenotype during discovery and preclinical optimization.
“Since continuous feeding of nutrients, cytokines, and gas exchanges are controlled and changeable throughout the entire process, along with the capability for daily sampling for tests, we can easily change parameters to allow growth of specific T-cell subsets, including T-cell memory phenotypes,” Richard C. Koya, immunologist at the University of Chicago School of Medicine, told DDN. “These are correlated with long persistency in vivo and higher anti-cancer effects.”
The new approach lands at a time when CAR-T manufacturing science is changing. Reviews of the field note that labor-intensive, multi-step workflows remain a core limitation for early-phase work and continue to drive cost and variability. By running the critical steps of T-cell generation in a functionally closed system, Koya’s team aims to reduce hands-on manipulation while preserving the ability to test how experimental variables — such as activation duration, cytokine exposure, or vector design — influence exhaustion markers, metabolic state, and downstream potency.
Reproducibility and decision-making earlier in the pipeline
For discovery groups evaluating new TCR constructs or co-stimulatory domains, Koya said the standardized, programmable nature of the device offers a more consistent experimental setting.
“As the process is fully controllable with a pre-defined automated/programmable instrument, reproducibility and quality control of T-cell generation are highly increased,” he noted. This consistency, he said, enables earlier mechanistic comparisons under matched conditions — a capability that could help teams screen designs more efficiently before committing to more resource-intensive preclinical work.
Koya also highlighted how upstream quality improvements can ripple into earlier go/no-go decisions. “Optimized and validated methods with higher upstream cell quality or transduction efficiency would definitely facilitate earlier implementation of new TCR-T immunotherapies,” he told DDN. “They would decrease the initial entry barrier into the field for screening and assessment of new targets for therapies.”
Balancing technical progress with biological unknowns
Even as automation helps standardize inputs, fundamental questions remain about what features actually drive durable responses in vivo. “There is still much research needed to define the actual cellular characteristics in the infusion product that will really induce effective anti-tumor responses in real patients,” Koya said. He noted that T-cell metabolic state is increasingly viewed as an important determinant of function, but its role in long-term efficacy is not fully understood.
He also pointed to persistent, non-biological barriers: regulatory timelines, complex chain-of-custody requirements, and multi-site logistics, all of which can slow translation regardless of manufacturing setup. These constraints have been widely cited as limiting factors in the broader CAR-T manufacturing landscape.
Implications for drug discovery
The consolidated workflow is not presented as a replacement for clinical-scale manufacturing but rather as a tool to strengthen early discovery and preclinical evaluation. By tightening control over culture variables and reducing manual variation, the platform gives researchers clearer readouts on how design choices — from vector configuration to cytokine supplementation — directly influence phenotype and functional potency.
For drug discovery programs, this could shorten iterative design cycles, support more reliable screening of emerging TCR or CAR constructs, and generate early-stage data that translate more effectively into later development. As automation and closed systems become more common, they may also help academic teams run Phase 1 studies with greater consistency while reducing dependence on scarce cleanroom resources.
In Koya’s view, the long-term impact will depend on how well the field connects controlled in-device biology with real-world performance. For now, the goal is to build platforms that allow researchers to ask better mechanistic questions earlier — and to do so with fewer constraints than traditional T-cell manufacturing methods allow.
