Investigating the delivery of PD-L1-targeted immunoliposomes in a dynamic cervical cancer-on-a-chip model

The recent approval of pembrolizumab in recurrent or metastatic cervical cancer warrants further investigations into the usefulness of immunotherapies for more durable and less radical interventions. In this study, the targeting potential of anti-PD-L1-functionalized immunoliposomes was tested in a 3D in vitro cervical cancer-on-a-chip model. Immunolipsomes were synthesized and decorated externally with monovalent anti-PD-L1 Fab' fragments of commercially available atezolizumab. Cervical cancer cell lines with varying levels of PD-L1 expression were cultured as spheroids embedded in a collagen I matrix, and treated under flow of culture media. Flow cytometry and live-cell confocal imaging were used to measure the interactions and uptake of untargeted liposomes and immunoliposomes in this panel of cell lines. The immunoliposomes retained specific functionality regardless of protein corona formation in high serum environments. As such, spheroids expressing high levels of PD-L1 preferentially internalized immunoliposomes in a 3D environment with extracellular matrix present, while low PD-L1-expressing cell lines showed no preference for either formulation. Importantly, treatments performed in monolayer cultures (on plastic) showed no differences between immuno- and untargeted liposome uptake, including the way in which the endocytosed liposomes are trafficked subcellularly. This study demonstrates the importance of both active and passive accumulation strategies to achieve nanoparticle targeting. Immunoliposomes remain a promising platform for the development of targeted nanotherapies against cervical cancers. However, initial functional tests did not translate directly to biological performance and this should be kept in mind for future formulations. Furthermore, the in vitro model developed appeared useful for visualizing liposome uptake in a 3D, live tissue environment and represents a cost-effective and reproducible model for future studies.