Dual Knockout CAR-T Cells: A New Breakthrough in Solid Tumor Therapy

While Chimeric Antigen Receptor (CAR) T-cell therapy has achieved remarkable success in hematologic malignancies, this "living drug" has repeatedly hit a brick wall when confronting solid tumors. Worldwide, tens of millions of people are diagnosed each year with solid malignancies such as lung, gastric, and pancreatic cancers. Pancreatic cancer, for instance, has long stagnated at a 5-year survival rate of around 10%, while mortality rates for ovarian cancer remain grimly high. Within the dense microenvironment of these solid tumors, CAR-T cells face a dual dilemma: they fail to infiltrate, and even when they do manage to enter, they are rapidly rendered dysfunctional.

Recently, a study published online in Nature, titled "In vivo genome-wide CRISPR screens of human T cells in solid tumours," offered a promising solution to this challenge. Scientists from the Gladstone Institutes and the University of California, San Francisco (UCSF) harnessed the world’s first platform capable of conducting genome-wide CRISPR screening in live animal models to pinpoint two critical genes that restrict the antitumor activity of CAR-T cells. Knocking out both genes simultaneously empowered edited T cells to demonstrate potent tumor-clearing capabilities across multiple mouse models of solid tumors.

Figure 1. Genome-wide in vivo CRISPR screen identifies GPCR signalling pathways regulating human T cell abundance in solid tumours. (Liu Q, et al., 2026)

The researchers emphasized that this breakthrough overcomes a long-standing hurdle in solid tumor research. Previously, scientists primarily engineered and evaluated CAR-T cells in cell culture dishes. However, real solid tumor microenvironments are extraordinarily complex—replete with immunosuppressive signals, nutrient deprivation, and severe hypoxia—making petri dishes poor proxies for in vivo human biology.

The primary barrier to executing high-throughput, genome-wide screens within solid tumors in vivo has been the low recovery yield of tumor-infiltrating T cells. Yields typically top out in the hundreds of thousands—far below the statistical threshold required for genome-wide coverage—causing real biological signals to get drowned out by noise.

To overcome this, the research team devised an ingenious "bait" strategy: they engineered tumor cells to express a signaling molecule that acts as a powerful chemoattractant for T cells, effectively mounting a "lighthouse" on the tumor. This subtle modification left the fundamental biology of the tumor intact while driving a dramatic surge in infiltrating T cells—boosting recovery yields from hundreds of thousands to millions of T cells per tumor, thereby sweeping away the main barrier to in vivo genome-wide screening.

Equipped with an abundance of cells, the team conducted two parallel, large-scale CRISPR screens to systematically knock out nearly 20,000 genes one by one:

1. The Infiltration Screen: Focused on identifying gene knockouts that enabled T cells to efficiently stream into the tumor core.
2. The Functionality Screen: Evaluated T cells already inside the tumor to determine which gene knockouts allowed them to continuously secrete interferon-gamma (IFN-γ), a critical anticancer signaling molecule.

Out of a vast ocean of genomic data, two distinct targets stood out:

• P2RY8 (The Spatial Brake): This gene encodes a cell-surface receptor that, when activated by specific ligands in the tumor microenvironment, transmits a "stop moving" signal to the T cell. Knocking out P2RY8 is the functional equivalent of removing the parking brake from the T cell. Experiments revealed that P2RY8-deficient T cells flooded into tumors in quantities far exceeding wild-type T cells, as if liberated from spatial restrictions.
• GNAS (The Functional Hub): The second and most exhilarating discovery involved the GNAS gene, which encodes the Gαs protein—a central signaling hub that integrates multiple immunosuppressive cascades within the cell. The tumor microenvironment harbors diverse immunosuppressive molecules, such as adenosine and prostaglandin E2 (PGE2), which act like different keys inserting into the same lock: Gαs. Knocking out GNAS rendered T cells completely deaf to these multi-pronged "stand down" signals, allowing them to sustain high levels of IFN-γ secretion and maintain full cytotoxic momentum inside the tumor.

These two genes execute a clear division of labor: P2RY8 governs whether T cells can enter, whereas GNAS dictates whether they remain functional once inside.

Synergistic Efficacy and Exceptional In Vivo Safety

The researchers then executed a bold combinatorial approach, knocking out both genes simultaneously. The results were striking: in a mouse model of lung cancer, even when administered at a low dose of CAR-T cells, two-thirds of the mice treated with dual-edited CAR-T cells achieved complete tumor clearance at the experimental endpoint. In contrast, none of the mice receiving unedited CAR-T cells reached a similar outcome. Significant tumor regression was likewise observed across multiple other solid tumor models, including melanoma, pancreatic cancer, gastroesophageal adenocarcinoma, and uterine sarcoma.

Crucially, the team validated the biosafety profile of these dual-edited cells. In in vitro cultures, the gene-modified T cells behaved identically to standard T cells. Furthermore, in mice tracked for over six months post-tumor clearance, no autoimmune-like toxicities or adverse events were observed. This "tumor-localized activation" profile drastically mitigates the risk of off-target toxicity, providing a valuable layer of safety for future clinical translation.

A Modular Framework for the Future of Cell Therapy

The researchers highlighted that these double-knockout cells remain indistinguishable from unedited cells outside the tumor, exerting their enhanced potency only upon entering the tumor microenvironment—a ideal behavior for precision cellular therapeutics. The open, versatile nature of this screening platform promises to yield many more discoveries; the two genes reported here represent just the beginning.

From a technological standpoint, the study’s greatest contribution lies in constructing a scalable, universal in vivo screening framework. This platform empowers scientists to systematically discover therapeutic gene targets under conditions that closely mimic real clinical landscapes, exposing gene functions that were previously masked in culture dishes. By combining the knockouts of P2RY8 and GNAS, the study provides direct proof that "relieving migratory inhibition" and "resisting functional suppression" act synergistically to boost cell therapy efficacy.

While a bridge must still be built between mouse models and human clinical applications—requiring rigorous, large-scale toxicology and efficacy studies—this research unquestionably illuminates the path forward. It proves for the first time that systematic, genome-wide screens are achievable within the extreme environment of a live tumor. As scientists use genetic scissors to dismantle the natural constraints placed on T cells by evolution, a scientifically grounded beacon of hope emerges for patients with solid tumors worldwide.