PURPOSE: Breast implant–associated anaplastic large cell lymphoma (BIA-ALCL), a rare but potentially deadly complication of device-based breast reconstruction, has an incidence estimated to be as high as 1 in every 3,817 cases of textured device implantation and has been implicated in 17 deaths worldwide. Present hypotheses of BIA-ALCL pathogenesis propose that bacterial biofilms present on textured implants may lead to T-cell dysregulation in the setting of chronic inflammation or genetic susceptibility, but this theory remains lacking in experimental evidence partly due to the inadequacy of present in vivo and in vitro models. The purpose of this study is to utilize our high-fidelity ex vivo biomimetic, 3-dimensional breast model to study the effects of implant shells on patient-derived BIA-ALCL cells.
METHODS: Healthy patient-derived breast tissue was processed for its individual cellular constituents including adipocytes, organoids, and the stromal vascular fraction (which also includes immunologic cells). These constituents were then suspended within 50 µl of 0.3% type I collagen matrix along with patient-derived BIA-ALCL cells at a density of 200,000 cells/ml before being plated into 6 mm wells. As a control, BIA-ALCL cells were also suspended within type I collagen at the same seeding density and volume but without any breast components. Before plating, wells were lined with either textured, smooth, or no implant shells. These were 1 × 2 cm pieces of implant shell dissected from the whole implant, cleaned of any underlying residual silicone gel, and autoclaved before being placed into the wells with the superficial aspect of the shell facing into the well. Eight wells were plated per implant shell type: 4 biomimetic platform wells and 4 collagen-only controls. All groups started at an equal density of approximately 1,000 cells per 3-dimensional confocal snapshot. Wells were then imaged immediately and every other day using confocal microscopy before being processed using Imaris software to analyze cell proliferation over time.
RESULTS: BIA-ALCL cell proliferation was significantly more robust in the biomimetic platform relative to the collagen-only groups regardless of implant shell type. BIA-ALCL cells in both the textured and smooth shell biomimetic groups grew nearly 30% faster than those within biomimetic wells lacking implant shell, with statistical differences as early as day 2 following plating. By day 10, mean cell counts in the textured and smooth shell biomimetic groups were 4,021 ± 999 and 4,281 ± 633, respectively, compared with 2,399 ± 355 in the biomimetic group lacking implant shell (P = 0.015). There was no statistical difference in BIA-ALCL cell proliferation between the textured and smooth biomimetic groups or among any of the collagen-alone groups.
CONCLUSIONS: Using our unique tissue-engineered 3-dimensional ex vivo model of BIA-ALCL, we have demonstrated that BIA-ALCL cells thrive within the biomimetic platform when compared with collagen alone and that incorporation of smooth and textured implant shell leads to significantly increased BIA-ALCL cell proliferation when compared with no implant shell. These findings contribute to the implication of breast implant materials in the development of BIA-ALCL, and they demonstrate the promise of our platform for use in further investigation of BIA-ALCL pathogenesis and therapeutics.