Wnt Inhibition Sensitizes PD-L1 Blockade Therapy by Overcoming Bone Marrow-Derived Myofibroblasts-Mediated Immune Resistance in Tumors

Tinglei Huang, Fuli Li, Xiaojiao Cheng, Jianzheng Wang, Wenhui Zhang, Baiwen Zhang, Yaoliang Tang, Qingli Li, Cong Zhou, Shuiping TuFrontiers in Immunology 2021 · DOI

Abstract

Cancer-associated fibroblasts (CAFs) are a recognized cause of tumor resistance to immune checkpoint blockade therapy, but the underlying mechanisms remain elusive. This study establishes a bone marrow-derived CAF (BMF)-rich tumor model to demonstrate that BMF-mixed tumor xenografts are resistant to anti-PD-L1 antibody immunotherapy compared to pure tumor xenografts. In vitro co-culture assays reveal that cancer cells induce BMFs to overexpress PD-L1 via the activation of the Wnt/β-catenin signaling pathway. Pharmacological inhibition of Wnt signaling or genetic knockout of PD-L1 in BMFs restores sensitivity to PD-L1 blockade. Combination therapy significantly potentiates therapeutic outcomes, suggesting that Wnt inhibition is a viable strategy to overcome BMF-mediated immunotherapy resistance.

Introduction

Immune checkpoint blockade (ICB) therapy, specifically targeting the PD-1/PD-L1 axis, has revolutionized oncology. However, fewer than 20% of patients respond effectively due to "immune resistance" driven by the tumor microenvironment (TME). Cancer-associated fibroblasts (CAFs) play a critical role in this resistance. This paper focuses on a specific subset—bone marrow-derived myofibroblasts (BMFs)—and investigates how they act as an immune shield for tumors, evading anti-PD-L1 therapy.

Methods and Experimental Design

The researchers employed a multi-disciplinary approach combining in vivo mouse models and in vitro cell signaling assays. BMF-rich tumor models were created by injecting mice with a mixture of tumor cells (lung carcinoma or colorectal cancer) and BMFs.

Experimental timeline for tumor inoculation and anti-PD-L1 treatment

To dissect cellular crosstalk, co-culture systems were established, and transcriptomic changes were analyzed via microarray. Genetic engineering (CRISPR/Cas9) was used to specifically knock out PD-L1 in BMFs, while pharmacological inhibitors (XAV-939 and Wnt-C59) were used to target the Wnt/β-catenin pathway.

Results

BMFs Drive Immunotherapy Resistance

The study confirmed that the presence of BMFs significantly impairs the efficacy of anti-PD-L1 therapy. While pure tumor xenografts responded well to treatment, BMF-mixed tumors showed negligible growth inhibition, confirming that BMFs act as an external immune shield.

Tumor volume data demonstrating that BMF-mixed tumors are resistant to anti-PD-L1 therapy compared to pure tumor models

The Role of the Wnt/β-catenin Pathway

Mechanistic analysis revealed that tumor cells activate the Wnt/β-catenin signaling pathway in BMFs, leading to a dramatic upregulation of PD-L1 on the fibroblast surface. Interestingly, PD-L1 expression did not increase significantly on the cancer cells themselves in these co-cultures, suggesting that BMFs serve as a primary "immune decoy." Knocking out PD-L1 specifically in BMFs increased the tumor inhibitory rate of anti-PD-L1 therapy from 17.8% to 55.1%.

Synergistic Effects of Wnt Inhibition

Combining the Wnt inhibitor XAV-939 with anti-PD-L1 therapy significantly potentiated the therapeutic outcome. This combination led to a surge in tumor-infiltrating CD8+ T cells (reaching 37%, up from ~23% in monotherapy) and increased IFN-γ production. Importantly, the treatment was well-tolerated, showing no significant toxicity in mouse models.

Conclusion

This research highlights the critical role of stromal PD-L1 in mediating immune resistance. By identifying the Wnt/β-catenin pathway as the driver of BMF-mediated protection, the study provides a clear therapeutic target. Combining Wnt inhibitors with standard ICB therapies offers a promising strategy to improve outcomes for patients with CAF-rich, immunotherapy-resistant tumors.

Summary updated

Abstract

Cancer-associated fibroblasts (CAFs) has been recognized as one cause of tumor resistance to immune checkpoint blockade therapy, but the underlying mechanisms still remain elusive. In the present study, a bone marrow-derived CAF (BMF) -rich tumor model is successfully established by subcutaneously mixed inoculation of BMFs and tumor cells into mice and the BMF-mixed tumor xenografts are demonstrated to be resistant to anti-PD-L1 antibody immunotherapy compared to the mere tumor xenografts. In vitro assays <i>via</i> the co-culture system of BMFs and tumor cells indicate that the co-cultured BMFs are induced to overexpress PD-L1, while there is no such a phenomenon in the co-cultured cancer cells. The further knock-out of PD-L1 in BMFs rescues the sensitivity of BMF-mixed tumor xenografts to PD-L1 blockade therapy. Mechanistically, <i>via</i> the microarray assay, we identify that the upregulation of PD-L1 in BMFs stimulated by cancer cells is medicated by the activation of the Wnt/β-catenin signaling pathway in BMFs. Moreover, the administration of Wnt/β-catenin signaling inhibitors, including XAV-939 and Wnt-C59, distinctly inhibits the upregulation of PD-L1 expression in the co-cultured BMFs. The further combination administration of XAV-939 significantly potentiates the therapeutic outcome of PD-L1 blockade therapy in BMF-mixed tumors. In summary, our study demonstrates that Wnt inhibition augments PD-L1 blockade efficacy by overcoming BMF-mediated immunotherapy resistance.
BibTeX
@inproceedings{huang2021wnt,
  author = {Tinglei Huang and Fuli Li and Xiaojiao Cheng and Jianzheng Wang and Wenhui Zhang and Baiwen Zhang and Yaoliang Tang and Qingli Li and Cong Zhou and Shuiping Tu},
  title = {{Wnt Inhibition Sensitizes PD-L1 Blockade Therapy by Overcoming Bone Marrow-Derived Myofibroblasts-Mediated Immune Resistance in Tumors}},
  booktitle = {Frontiers in Immunology},
  year = {2021},
  doi = {10.3389/fimmu.2021.619209},
}
Figure