TP53-Mutant AML Cells Weaponize TGF-β1 to Neutralize Immunotherapy

TP53 mutations occur in 10–15% of newly diagnosed AML and up to 30% of therapy-related or relapsed cases, placing patients in the adverse-risk category with median overall survival under seven months. Even allogeneic stem cell transplantation fails to meaningfully improve outcomes, with two-year OS around 9–19% and relapse rates exceeding 60%. As bispecific T-cell engager (BiTE) molecules and CAR-T therapies enter AML clinical trials, understanding how TP53 status shapes the immune microenvironment is urgent.

The authors used three AML cell lines (MV4-11, MOLM-13, OCI-AML3) engineered with shRNA-mediated TP53 knockdown alongside matched wild-type controls, plus primary AML patient samples stratified by TP53 deletion status. T cells from healthy donors or AML patients were co-cultured with these targets at a 1:6 effector-to-target ratio using AMG 330, a CD3×CD33 BiTE molecule, or a control T-cell engager. Cytotoxicity, proliferation, and cytokine output were measured by flow cytometry over three to five days. Transwell insert experiments physically separated T cells from AML cells to isolate secretome effects. Supernatants were analyzed by quantitative proteomics, and T cells recovered from co-cultures underwent bulk RNA sequencing.

TP53 knockdown cells were significantly more resistant to AMG 330-mediated killing than wild-type counterparts across all three cell lines and in primary TP53-deleted AML samples. T-cell proliferation and secretion of proinflammatory cytokines (including IFN-γ and TNF-α) were markedly reduced in co-cultures with TP53-deficient cells. Critically, Transwell assays demonstrated that physical separation did not rescue T-cell function, implicating soluble factors in the immunosuppressive effect. Proteomic profiling of conditioned media identified TGF-β1 as significantly elevated in TP53-knockdown co-cultures. RNA sequencing of T cells exposed to TP53-deficient AML revealed transcriptional signatures consistent with cellular senescence and cell-cycle arrest, characterized by upregulation of cyclin-dependent kinase inhibitors and senescence-associated gene programs.

These findings establish a mechanistic link between TP53 loss-of-function, TGF-β1 hypersecretion, and T-cell exhaustion/senescence that collectively blunts BiTE efficacy. The immunosuppressive secretome of TP53-deficient AML thus represents both a mechanistic explanation for clinical resistance and a therapeutic target. Combining TGF-β pathway inhibitors with T-cell engagers, or using strategies to prevent T-cell senescence, emerge as rational approaches for this high-risk AML subset.

Caveats include the use of shRNA-based knockdown rather than true CRISPR-engineered TP53-null models, potential off-target shRNA effects, and limited primary patient sample numbers. In vivo validation and clinical correlative studies in BiTE-treated TP53-mutant AML patients are needed to confirm translational relevance.