Could 3D models of cancer enhance drug screening?

Abstract

Cancer is a multifaceted pathology, where cellular and acellular players interact to drive cancer progression and, in the worst-case, metastasis. The current methods to investigate the heterogeneous nature of cancer are inadequate, since they rely on 2D cell cultures and animal models. The cell line-based drug efficacy and toxicity assays are not able to predict the tumor response to anti-cancer agents and it is already widely discussed how molecular pathway are not recapitulated in vitro so called flat biology. On the other side, animal models often fail to detect the side-effects of drugs, mimic the metastatic progression or the interaction between cancer and immune system, due to biologic difference in human and animals. Moreover, ethical and regulatory issues limit animal experimentation. Every year pharma/biotech companies lose resources in drug discovery and testing processes that are successful only in 5% of the cases. There is an urgent need to validate accurate and predictive platforms in order to enhance drug-testing process taking into account the physiopathology of the tumor microenvironment. Three dimensional in vitro tumor models could enhance drug manufactures in developing effective drugs for cancer diseases. The 3D in vitro cancer models can improve the predictability of toxicity and drug sensitivity in cancer. Despite the demonstrated advantages of 3D in vitro disease systems when compared to 2D culture and animal models, they still do not reach the standardization required for preclinical trials. This review highlights in vitro models that may be used as preclinical models, accelerating the drug development process towards more precise and personalized standard of care for cancer patients. We describe the state-of-the art of 3D in vitro culture systems, with a focus on how these different approaches could be coupled in order to achieve a compromise between standardization and reliability in recapitulating tumor microenvironment and drug response.