I3 - Natural Killer Cell Infiltration is Governed by Hydrogel Mechanics and HA Interactions in Solid Tumor Models

Introduction: : Adoptive transfer of natural killer (NK) cells holds promise as a less toxic and more broadly applicable alternative to T cell-based therapies, particularly for patients with limited treatment options. However, in the context of solid tumors, many administered NK cells fail to effectively infiltrate the tumor and reach cancer cells, limiting their therapeutic potential. Although the tumor extracellular matrix (ECM) is known to undergo extensive remodeling, the influence of these alterations on NK cell infiltration remains poorly understood. To address this gap, we developed a poly(ethylene glycol) (PEG)-based hydrogel system in which the biochemical and mechanical properties can be independently tuned. Here, we apply this platform to systematically investigate how two characteristic features of the tumor ECM, matrix stiffening and hyaluronic acid (HA) degradation, affect NK cell infiltration and tumor sampling.

Materials and

Methods: : PEG-based hydrogels containing integrin-binding and metalloprotease-degradable sites were formulated with varying stiffness by adjusting the PEG-diacrylate concentration. 1 mg/mL of methacrylated HA (MeHA) of low (LMW; 10 kDa) or high (HMW; 1000 kDa) molecular weight was incorporated via chemical crosslinking. Either A549 lung cancer cells or chemoattractant SDF1α were encapsulated in hydrogels, which were co-cultured with NK92 cells suspended in the supernatant. After incubation, hydrogels were fixed and stained. Confocal microscopy was used to quantify NK cell infiltration within the 3D matrix. To investigate HA-mediated signaling, the HA receptor RHAMM and CD44 were selectively blocked using antibodies.

Results, Conclusions, and Discussions:: We engineered three hydrogel formulations with Young’s moduli of 6, 15, and 22 kPa and incorporation of LMW or HMW MeHA did not alter the mechanical properties (Fig 1A and 1B). We then quantified NK92 cell infiltration towards a chemoattractant point source in hydrogels of varying stiffness and observed a biphasic response, with increased NK cell infiltration in hydrogels with mid-range stiffness (Fig 1C). This suggests that solid tumors may leverage mechanotransduction pathways to evade NK cell detection. Additionally, the inclusion of HMW MeHA promoted NK cell infiltration (Fig 1D), indicating that HMW MeHA may serve as a supportive substrate for NK cell migration within the ECM. To investigate this further, we blocked the HA receptors RHAMM and CD44, which are known to mediate cell interactions with HA. Blocking RHAMM significantly reduced NK cell infiltration into MeHA-containing hydrogels compared to controls, whereas blocking CD44 had no effect (Fig. 1E and F). Similar trends in NK cell migration in response to hydrogel stiffness and HA were observed when hydrogels contained lung cancer cells in lieu of the chemoattractant (Fig 1G and 1H). Our results demonstrate that ECM stiffness and HA composition influence NK cell infiltration in 3D. Specifically, the medium stiffness and the presence of HMW MeHA create favorable conditions for NK cell migration, suggesting that the ECM may not only act as a barrier to NK cell infiltration but can also serve as a guiding substrate depending on its mechanical and biochemical properties. Further, we demonstrate that RHAMM may play a critical role in NK cell migration within HA-rich matrices. The tunable hydrogel platform allows systematic analysis of ECM-immune interactions, revealing mechanisms that could inform the design of improved NK cell-based therapies. Ongoing work aims to elucidate the processes behind these interactions and link NK cell infiltration with cytotoxicity in vitro to better understand the functional outcomes in engineered tumor environments.

Acknowledgements and/or References (Optional)::