Chemokine Receptor CXCR5 Emerges as Tumor-Intrinsic Switch Controlling Chemo-Immunotherapy Response in Triple-Negative Breast Cancer

Triple-negative breast cancer, the most aggressive subtype of breast cancer, has long resisted targeted therapy, and single-agent immune-checkpoint inhibitors produce response rates below twenty percent in this disease. A new study now reports that the chemokine receptor CXCR5, acting within tumor cells themselves, functions as a decisive molecular switch that determines how well triple-negative breast cancer responds to the combination of paclitaxel chemotherapy and PD-1 blockade. Using human tumor specimens, a syngeneic mouse model, and genetic gain- and loss-of-function experiments, the researchers found that tumor-intrinsic CXCR5 coordinates two processes at once: the recruitment of cytotoxic CD8-positive T cells into the tumor and the activation of intrinsic apoptotic signaling through the p38 MAPK pathway. The work, published as a mechanistic preclinical and retrospective clinical analysis, positions CXCR5 as both a candidate predictive biomarker and a potential therapeutic lever for chemo-immunotherapy in this hard-to-treat cancer.

The clinical backdrop is sobering. In the metastatic setting, triple-negative breast cancer carries a median overall survival of only twelve to eighteen months, and although checkpoint inhibitors have transformed many solid tumors, they rarely work alone in this subtype. Current phase-III trials therefore pair PD-1 blockade with paclitaxel, an approach that has raised pathological complete response rates from roughly fifteen percent to thirty-five percent. What remained unclear was why some patients benefit while others do not. The researchers began with a clue from their own center: bulk RNA sequencing of post-treatment biopsies showed that CXCL13, the sole known ligand for CXCR5, was the most upregulated gene in patients who achieved a partial response to tislelizumab, a PD-1 antibody, combined with paclitaxel, compared with non-responders. Immunofluorescence then revealed CXCL13-positive CD8-positive T cells co-localizing with CXCR5-expressing tumor cells in regions rich in tertiary lymphoid structures.

To establish whether the combination therapy genuinely improves outcomes, the team first analyzed twenty patients with metastatic triple-negative breast cancer, ten treated with paclitaxel alone and ten with paclitaxel plus a PD-1 inhibitor. Overall survival was significantly longer in the combination group, with a log-rank p-value of 0.003, and progression-free survival also favored the combination at p = 0.001. In parallel, the researchers implanted 4T1 murine breast cancer cells into immunocompetent mice and treated them with vehicle, paclitaxel, tislelizumab, or the combination over four cycles. Control tumors grew progressively and paclitaxel alone slowed expansion, but the combination produced the strongest suppression of tumor volume and weight, preserved splenic architecture that paclitaxel alone had disrupted, and markedly extended survival in Kaplan-Meier analysis. The mouse model thus reproduced the clinical signal and provided a platform for mechanistic dissection.

Immunofluorescence profiling of day-28 tumors showed what the combination was doing to the immune microenvironment. Paclitaxel alone modestly increased intratumoral PD-1-positive cells, and adding tislelizumab elevated this infiltration further. CD4-positive helper T cell density followed the same trend, but the most striking change appeared in cytotoxic lymphocytes: sparse granzyme-B-positive CD8-positive cells in control and single-agent arms gave way to a pronounced accumulation of these effectors under combination therapy. Western blotting of tumor lysates revealed a parallel shift in apoptotic signaling. The anti-apoptotic protein BCL2 declined, while the pro-apoptotic proteins BAX and cleaved caspase-3 rose markedly, and CXCR5 itself was upregulated in the combination group. Together, these data indicated that PD-1 blockade plus paclitaxel remodels the tumor toward a cytotoxic, pro-apoptotic state in which CXCR5 is prominently induced.

To test whether CXCR5 was merely a passive marker or an active effector, the team engineered 4T1 cells to overexpress Flag-tagged CXCR5 or to silence endogenous Cxcr5, then implanted the modified cells into mice. The results were decisive. Knocking down Cxcr5 blunted the therapeutic benefit of the paclitaxel-tislelizumab combination, whereas overexpression sensitized tumors to the treatment, producing the most pronounced growth inhibition among all experimental arms. Mice bearing CXCR5-depleted tumors died earlier, while those with CXCR5-overexpressing tumors survived markedly longer under combination therapy. Spleen measurements reinforced the immune dimension of the effect: knockdown tumors were associated with evident splenomegaly, whereas CXCR5-overexpressing hosts had smaller, paler spleens and the lowest mean splenic mass. The authors conclude that CXCR5 is a cell-intrinsic determinant governing both tumor growth restraint and responsiveness to chemo-immunotherapy.

The immune profiling of these engineered tumors told a consistent story. PD-1-positive T cell density was highest in CXCR5-overexpressing lesions and lowest in knockdown tumors, with CD4-positive infiltration following an identical pattern, and granzyme-B-positive CD8-positive cytotoxic cells markedly enriched in the overexpression group. In human cells, the team turned to MDA-MB-231, a human triple-negative breast cancer line, engineered to silence or overexpress CXCR5 and co-cultured with activated T cells derived from peripheral blood mononuclear cells. Crystal violet assays showed that CXCR5-overexpressing cells had significantly lower survival when exposed to T cells or to paclitaxel plus tislelizumab, while silenced cells resisted killing. Transwell experiments added a chemotactic mechanism: culture supernatants from CXCR5-overexpressing cells significantly increased CD8-positive T cell migration compared with control supernatants, indicating that the receptor promotes T cell recruitment from the tumor side of the axis.

A key question was whether CXCR5 simply makes cells more sensitive to paclitaxel. It does not. Cell viability assays after paclitaxel treatment showed no significant difference between CXCR5-overexpressing and control cells, meaning the receptor does not change intrinsic chemosensitivity. What it does change is immunogenic cell death. Paclitaxel-induced release of HMGB1, a damage-associated molecular pattern, was markedly enhanced in CXCR5-overexpressing cells and reduced when CXCR5 was silenced, and ATP levels, another crucial damage signal, were significantly elevated with overexpression and diminished with silencing. These findings connect CXCR5 to the mechanism by which chemotherapy primes antitumor immunity: the receptor amplifies the release of molecular flags that alert the immune system to dying tumor cells, rather than lowering the threshold for cell death by the drug itself.

The downstream signaling pathway was mapped with pharmacological inhibitors. Transwell assays showed that the enhanced CD8-positive T cell migration induced by paclitaxel plus tislelizumab was abolished by VX5, a CXCL13 inhibitor, confirming dependence on the CXCL13-CXCR5 axis. Recombinant CXCL13 protein added to the drug combination further enhanced CXCR5 expression, reduced BCL2, and increased BAX, accompanied by activation of p38 signaling. When CXCR5-overexpressing cells were treated with the AKT inhibitor LY294002 or the JNK/p38 inhibitor SP600125, only the JNK/p38 inhibitor attenuated the CXCR5-mediated downregulation of BCL2 and upregulation of BAX, identifying p38 as the critical downstream effector of CXCR5-driven apoptotic priming. The authors note that JNK and p38 can directly phosphorylate BAX at threonine 167, a modification required for its mitochondrial translocation, providing a plausible structural basis for the observed effect.

On the clinical side, transcriptomic comparison of pre-treatment tumors from six partial responders and six non-responders within the combination cohort identified fifteen differentially expressed immune-related genes. Responders showed higher expression of interferon-gamma signaling molecules including IFNG and CXCL13, the cytolytic effector GZMK, antigen presentation machinery components such as B2M, PSMB9, and BATF, and PD-L1 itself. Analysis of TCGA data through the KM plotter further showed that higher expression of both CXCL13 and CXCR5 was significantly associated with prolonged overall survival in triple-negative breast cancer patients, supporting their potential as favorable prognostic biomarkers. These results align with evidence from melanoma, lung cancer, and bladder cancer, where CXCR5-positive CD8-positive T cells within tertiary lymphoid structures correlate with checkpoint inhibitor response.

The authors are candid about the limitations. The clinical sample was small and retrospective, the study lacked direct CXCL13 level measurements and blocking experiments in vivo, and immunocompetent mouse models may not fully recapitulate the human tumor microenvironment. The chemokine network is redundant, with axes such as CXCR3-CXCL9/10 potentially compensating for CXCR5 loss, and other anti-apoptotic proteins like MCL1 and BCL-XL may sustain survival signals even when BCL2 falls. Rescue experiments with BCL2 overexpression were not performed to confirm that the sensitization truly depends on the BCL2-BAX axis. Nevertheless, the study proposes a concrete translational path: transient CXCR5 agonism, perhaps through engineered CXCL13-Fc fusion proteins or tumor-selective CXCR5-encoding mRNA-lipoplexes delivered neoadjuvantly, could convert immunologically cold triple-negative tumors into T-cell-inflamed ones, while circulating CXCR5-positive T cell dynamics might serve as a minimally invasive biomarker of early response or emerging resistance. For a disease with so few options, a single receptor linking T cell trafficking, immunogenic cell death, and apoptotic priming is a target worth pursuing rigorously.

Subject of Research: Tumor-intrinsic CXCR5 regulation of T cell infiltration and apoptosis in triple-negative breast cancer chemo-immunotherapy

Article Title: CXCR5 regulates CD8 + /GZMB + T cell infiltration and apoptotic signaling in breast cancer

Article References: Feng, W., Yan, J., Gong, X., Hu, J., Jin, M., Ma, C., & Li, F. (2026). CXCR5 regulates CD8+/GZMB+ T cell infiltration and apoptotic signaling in breast cancer. iScience, 29(11), Article 117874. https://doi.org/10.1016/j.isci.2026.117874

Image Credits: AI Generated

DOI: 10.1016/j.isci.2026.117874

Keywords: triple-negative breast cancer, CXCR5, CXCL13, PD-1 blockade, paclitaxel, tislelizumab, CD8 T cells, immunogenic cell death, p38 MAPK, apoptosis, tertiary lymphoid structures, biomarker