包括 TLR2 和 TLR4 表达在内的早期免疫特征与 CD19 CAR-T 细胞治疗后完全缓解相关

Background/Objectives: CD19-directed chimeric antigen receptor T (CAR-T) cell therapy induces profound immune remodeling. Nonetheless, biomarkers predicting complete remission (CR) remain poorly defined.

We characterized longitudinal cytokine and immune-cell dynamics after CAR-T infusion and identified early immunological features associated with CR. Methods: Longitudinal immune profiling was performed in 18 patients with non-Hodgkin lymphoma, including 14 with relapsed/refractory diffuse large B-cell lymphoma treated with anti-CD19 CAR-T cells. Peripheral blood was collected at the baseline and days 7, 14, 21, 28, and 60 post-infusion. Multiparameter flow cytometry quantified lymphoid and myeloid subsets and Toll-like receptor (TLR)2 and TLR4 expression. Serum cytokines were measured by multiplex assays. Machine-learning-based feature selection identified variables associated with CR. Results: Two inflammatory waves were observed.

The first, at day 7, featured elevated IL-6, IL-10, IFN- , IFN- , and TNF- , accompanied by increased CD4 + T cells, HLA-DR high classical monocytes, and non-classical monocytes. The second, at days 21-28, showed increased IL-5, IL-6, IL-12, IFN- , and GM-CSF, with expansion of CD4 + and CD8 + T cells, regulatory T cells, NK-T cells, and non-classical monocytes.

TLR2 expression was significantly upregulated at day 7 on T-cell subsets and on classical and intermediate monocytes. An exploratory feature-selection analysis identified baseline and day-7 TLR2 and TLR4 expression on lymphoid and myeloid cells, early IFN- levels, and monocyte frequencies as variables associated with CR. Conclusions: Together, these data show that anti-CD19 CAR-T therapy induces two coordinated waves of cytokine release and immune-cell activation.

Moreover, the findings suggest that early modulation of innate immune features, particularly TLR2 expression, is associated with complete remission, although these biomarker relationships remain exploratory and require validation in larger cohorts.