Relevant In Vitro Predictors of Human Acellular Dermal Matrix-Associated Inflammation and Capsule Formation in a Nonhuman Primate Subcutaneous Tissue Expander Model

Objective: Benchtop methods were evaluated for preclinical inflammation/capsule formation correlation following implantation of human acellular dermal matrices. Methods: Dermal matrices were compared with native dermis for structure (histology, scanning electron microscopy), collagen solubility (hydroxyproline), enzymatic susceptibility (collagenase), and thermal stability (differential scanning calorimetry). Results were compared with implantation outcomes in a primate tissue expander model. Results: Native dermis, electron beam–sterilized, and freeze-dried human acellular dermal matrices had equivalent morphology, acid-soluble collagen (60.5% ± 6.3%, 65.3% ± 3.2%, and 63.3% ± 2.4%, respectively), and collagenase resistance. Implant results showed minimal inflammation/matrix degradation, lack of capsule formation, insignificant elastic modulus change (57.65 ± 20.24 MPa out-of-package/44.84 ± 23.87 MPa in vivo), and low antibody induction (2- to 8-fold increase) for electron beam–sterilized matrix. Similar results for freeze-dried dermal matrix were previously observed. γ-Irradiated, γ-irradiated/freeze-dried, and ethanol-stored dermal matrices were statistically different from native dermis for acid-soluble collagen (82.4% ± 5.8%, 72.2% ± 6.2%, and 76.8% ± 5.0%, respectively) and collagenase digestion rate, indicating matrix damage. γ-Irradiated matrix-implanted animals demonstrated elevated inflammatory response, foreign body giant cells, capsule formation at the tissue expander junction, and robust matrix metalloproteinase-1 staining with significant elastic modulus decrease (37.43 ± 7.52 MPa out-of-package/19.58 ± 1.16 MPa in vivo). Antibody increase (32- to 128-fold) was observed 6 to 10 weeks following γ-irradiated matrix implantation. Ethanol-stored dermal matrix elicited an acute antibody response (4- to 128-fold increase, 2-4 weeks) and macrophage-concentrated synovial-like hyperplasia at the tissue expander junction, moderate matrix metalloproteinase-1 staining, and significant elastic modulus decrease (61.15 ± 9.12 MPa out-of-package/17.92 ± 4.02 MPa in vivo) by 10 weeks implantation. Conclusion: Demonstrated loss of collagen integrity in vitro may be predictive of inflammation/capsule formation in primate tissue expander models. These results may be further predictive of clinical observations.