(P 74) MATRIX DENSITY INDUCES CHEMORESISTANCE WITHIN A TISSUE-ENGINEERED MODEL OF OSTEOSARCOMA

Objective: Despite efforts in developing molecular-targeted drugs and immunotherapies, treatment regimens and survival rates for osteosarcoma (OS) remain unchanged. The estimated 5-year survival rate is 60% in primary disease and a mere 20% in refractory and metastatic disease. OS generates a robust extracellular matrix (ECM) composed of mainly collagen (COL) type I as organic phase and hydroxyapatite (HA) as inorganic phase. The OS ECM is a three-dimensional acellular physical support structure that regulates cell proliferation, cellular signaling, and chemoresistance. The direct effect of the ECM on OS cell phenotype has largely been undefined. This work investigates the impact of tumor extracellular matrix density on malignant cell phenotype and response to chemotherapy.

Methods: The 143B OS cell line was embedded within a biomimetic bone matrix composed of COL and nano-HA, the biomineral component of the bone matrix. COL matrix density was varied between a final concentration of 1, 2 and 4 mg/mL, and nano-HA concentration was held constant at 5 mg/mL. Matrix microarchitecture and mechanical properties were analyzed through confocal reflectance imaging and shear rheology. Hydrogels with cells embedded throughout the matrix were cultured in unconstrained environments permitting cells to exhibit contractile forces to compact and remodel the matrix. Cell viability, cell morphology, and hydrogel dimension were imaged at 7 days of culture. To evaluate response to chemotherapy, cells were cultured within each matrix condition and then treated with doxorubicin for 24 hours.

Results: Matrix microscopy revealed disorganized COL fibrils characteristic of the native OS ECM. Reduced matrix density resulted in fewer COL fibrils and increased void spaces within the matrix. Nano-HA was uniformly distributed and incorporated within the COL fibers across all three COL concentrations (Fig. 1). OS exhibited contractile forces and induced higher gel compaction at lower density (1 mg/mL and 2 mg/mL) matrices compared to the 4 mg/mL density. These findings correlated with gel mechanical properties as storage moduli were approximately 40, 200, 800 Pa in the 1, 2, and 4 mg/mL conditions. Cells maintained high viability in all conditions at day 7 of culture (live cells = green, red cells = dead) (Fig. 2). Cells grown within 4 mg/mL COL+HA exhibited higher quantity of F-actin stress fibers compared to lower matrix densities (Fig. 3). There was a right shift in the half maximal inhibitor concentration (IC50) of cells cultured in 4 mg/mL COL+HA hydrogels compared to lower matrix densities (1 mg/mL and 2 mg/mL) indicating chemoresistance (Fig. 4).

Conclusion: This work elucidates the response of OS cells within a biomimetic matrix revealing that cells exhibit decreased chemosensitivity in higher COL densities. Future research will crosslink matrices and stiffen hydrogels while maintaining COL matrix density to understand whether our results are due to matrix density, matrix stiffness, or both. Additionally, we will explore the use of ECM-degrading agents to overcome chemoresistance and assess the potential of this approach with front-line chemotherapy to improve drug efficacy. Overall, this work demonstrates OS becomes chemoresistant in higher COL densities, highlighting the ECM as a novel drug target for OS treatment.