Novel PEEK fabrication using fused strand deposition reduces inflammation and enhances MSC differentiation promoting bone growth and implant osseointegration

Spine fusion devices fabricated from polyether ether ketone (PEEK) using traditional machining or with surface pores created by salt leaching result in a fibrous connective tissue interface. To overcome this limitation, we used fused strand deposition (FSD), which elevates fused filament fabrication (FFF) or fused deposition modeling (FDM) to generate a first-of-its-kind PEEK implant with an architecture designed to mimic trabecular bone. We examined the responses of human bone marrow stromal cells (MSCs) and macrophages isolated from the femurs of C57/Bl6 male mice to the additively manufactured porous PEEK (PP) and porous PEEK coated with HA (PP-HA) to determine if these modifications would improve cell response compared to solid PEEK (SP). PP and PP-HA constructs had similar 3D architectures but differed in hydrophilicity (PP-HA > PP). MSCs and macrophages were cultured on PP, PP-HA, and SP, and osteoblast differentiation and M1/M2 polarization were assessed. MSCs attached to the SP surface and the PP and PP-HA fibers, and synthesized osteoblast proteins in a surface-dependent manner. Notably, MSCs and macrophages produced VEGF mRNA and protein on PP-HA at levels higher than those on PP or SP. Macrophages grown on PP and PP-HA exhibited reduced expression of pro-inflammatory cytokines compared to cells on SP and increased levels of anti-inflammatory cytokines, but they did not exhibit a distinct M1 or M2 phenotype. These results show that additive manufacturing of a unique, fully porous PEEK implant using FSD, results in a surface that promotes MSC differentiation and decreases the pro-inflammatory response of macrophages to the surface, suggesting that PP and PP-HA implants will improve regeneration and eventually osseointegration in vivo. Importantly, the mechanisms involved are not the same as would be expected when testing Ti6Al4V substrates under the same experimental conditions and may have been obscured if the cells had been cultured using osteogenic media (OM). We correlated our in vitro findings with the clinical use of PP-HA implants in four patients, each of whom was treated with a different bone graft material; in all cases, fusion was achieved.