Background Solid tumor microenvironments (TMEs), which are frequently characterized by dense fibroblastic infiltrates and matrix protein deposition, restrict tumor infiltration by immune effector cells such as T-cells and natural killer (NK) cells, limiting the efficacy of approaches such as CAR-T and CAR-NK cell therapies.^{1 2} Numerous proteases, such as matrix metalloproteinases, cathepsins and dipeptidyl peptidases are known to digest the extracellular matrix.³ We hypothesized that forced overexpression of such proteases would promote matrix protein degradation in the TME to facilitate effector cell invasion and anti-tumor effects. As an initial proof-of-concept we have modified NK cells to overexpress fibroblast activation protein (FAP). FAP is a type-II transmembrane serine protease that is primarily expressed in reactive stromal fibroblasts but can be expressed by NK cells.⁴ FAP promotes tissue remodeling and extracellular matrix digestion to facilitate cell migration.⁴ Here we report the impact of FAP overexpression on matrix invasion and anti-tumor effects of a human NK cell population.

Methods FAP overexpressing NK 92 cell (FAP OE) were generated and utilized to examine NK cell migration and invasion both in vitro and in vivo. FAP wild-type (WT) and CRISPR knockout (FAP KO) NK92 cells were used as controls.

Results FAP OE NK cells displayed enhanced migration along matrix. These cells spent less time in arrest, showed increased cell velocity, and increased cumulative distance traveled compared to wild type (WT) cells. FAP OE cells also displayed enhanced invasion through matrix (FAP OE: 28.2 ± 3.7% invasion vs WT: 11 ± 0.9% invasion, p<0.01), and into tumor spheroids (FAP OE: 9.8 ± 1.0 NK cells/cluster vs WT:4.7 ± 0.6 NK cells/cluster, p<0.001). In separate experiments, FAP KO cells displayed decreased invasion both along and through matrix (FAP KO: 5.5 ± 0.7 invasion vs WT: 12 ± 1.3% invasion, p<0.01). Importantly, forced overexpression of FAP promoted NK cell invasion into PANC1 tumor xenografts grown in NOD SCID mice following one or three IV infusions of either wild-type or FAP-OE NK92 cells (FAP OE: 1.1 ± 0.3 NK cells/mm³ vs WT: 0.3 ± 0.04, p<0.05). Therapy with FAP OE cells led to enhanced anti-tumor effects (four-fold reduction of tumor weights 4 weeks following treatment initiation, p<0.001).

Conclusions Forced overexpression of FAP, a proteolytic enzyme, can enhance the invasiveness and therapeutic effectiveness of intravenously infused NK cells into the pancreatic ductal adenocarcinoma TME. This specific example suggests a potential platform strategy to improve immune cell invasion into solid tumors.

Acknowledgements We would like to thank Dr. Kerry Campbell for providing NK cell lines, the pBMN plasmid, and continuous intellectual and technical support; Dr. Dean Lee for providing the NK feeder cells; the Georgetown Lombardi Comprehensive Cancer Center Tissue Culture and Biobanking, Flow Cytometry and Cell Sorting, Genomics and Epigenomics, Histopathology and Tissue, and Microscopy and Imaging Shared Resources. Graphical abstract and schematics were created with BioRender.com.

References

1. Xiao Z, Todd L, Huang L, Noguera-Ortega E, Lu Z, Huang L, Kopp M, Li Y, Pattada N, Zhong W, Guo W, Scholler J, Liousia M, Assenmacher CA, June CH, Albelda SM, Puré E. Desmoplastic stroma restricts T cell extravasation and mediates immune exclusion and immunosuppression in solid tumors. Nat Commun. 2023 Aug 22;14(1):5110.

2. Li T, Niu M, Zhang W, Qin S, Zhou J, Yi M. CAR-NK cells for cancer immunotherapy: recent advances and future directions. Front Immunol. 2024 Feb 9;15:1361194.

3. Edsparr K, Basse PH, Goldfarb RH, Albertsson P. Matrix metalloproteinases in cytotoxic lymphocytes impact on tumour infiltration and immunomodulation. Cancer Microenviron. 2011 Dec;4(3):351–60.

4. Fitzgerald AA, Weiner LM. The role of fibroblast activation protein in health and malignancy. Cancer Metastasis Rev. 2020 Sep;39(3):783–803.

Ethics Approval All NSG animal studies were reviewed and approved by the Georgetown University Institutional Animal Care and Use Committee (GU IACUC), protocol 2016-1181.