Dottorato in TECNOLOGIE AVANZATE IN CHIRURGIA

Titolo della tesi: Modelli tridimensionali derivati da tumori maligni del colon retto per la sperimentazione di terapie personalizzate: approcci in vitro e in vivo

ABSTRACT Despite recent therapeutic advancements and an improved understanding of cancer biology, there are limiting factors that still complicate therapeutic decisions. The high variability of cellular drug responses between different patients (intertumor heterogeneity) and within different lesions of a single patient (intratumor heterogeneity) is widely recognised as a key cause of tumor progression and resistance to therapy. The development of three-dimensional (3D) in vitro cell culture systems has advanced our understanding of cellular processes, both in normal tissues and in cancer, by mirroring physiological characteristics and architectural features of in vivo tissues, therefore supporting personalised therapy decisions. The aim of this thesis is to further characterise and compare 3D models derived from colorectal cancer (CRC) resections such as spheroids and organoids, in order to identify the most accurate model that displays the characteristics needed for personalised medicine. To this end, a comparison between cultured spheroids and organoids derived from CRC samples has been conducted to establish the structural differences, markers expression, pharmacological sensibility and the presence of cancer stem cells (CSCs). Particular attention is given in evaluating the content of CSCs in both spheroids and organoids, considering that this parameter represents a fundamental element firmly connected to the response to anticancer therapies. Different experimental approaches such as immunofluorescence, Real-Time PCR, in vivo subcutaneous and orthotopic implantation are developed in this study with the purpose to obtain valid results in evaluating the most suitable 3D model to use in precision oncology. Our results suggest that the choice of 3D model for personalised medicine approaches has to consider both biologic characteristics as well as technical features. Our results also show that, although the spheroid model has a lower yield as compared to the organoid model, it is characterised by a higher concentration of CSCs. Therefore, spheroids should be the preferred model for experiments aimed at the identification of CSCs-targeted strategies. Differently, organoids are more suitable for routine pharmacological studies, as they retain the structure of the original colon cancer, they recapitulate genetic and phenotypic traits of the parental tumor and they allow a high success rate of production. Altogether, our studies further validate the usefulness of 3D cellular models for personalized medicine in colorectal cancer therapy and provide new evidences to support the appropriate model choice for targeted drug testing.