Development of a three-dimensional melanoma in vitro model

Melanoma is a type of skin cancer that occurs when melanocytes, melanin-producing cells located in the epidermis, start proliferating in an uncontrolled manner. It is the deadliest form of skin cancer, affecting a higher number of people each year. Despite remarkable efforts, current therapies fail mainly because tumours rapidly develop drug resistance. Research should focus on further understanding the mechanisms underpinning advanced-stage melanoma as well as identifying predictive and prognostic biomarkers in order to satisfy the need for better treatments and overcome drug resistance. In addition, the major challenge for future drug development also comprises finding alternative models for drug screening purposes. The use of animal models in research is highly controversial, with an ongoing debate on their ethical acceptability. Also, animal models are often poorly predictive of therapeutic outcomes due to the differences between animal and human physiological environments. For these reasons, the aim of this work was to develop a three-dimensional (3D) dermis equivalent harbouring melanoma spheroids to closely mimic the features of in vivo malignant melanoma. Firstly, the dermal architecture was reproduced by developing a hydrogel resembling the skin, namely its composition. The hydrogel was generated by crosslinking among collagen type I, hyaluronic acid (HA), four-arm PEG succinimidyl glutarate (4S-StarPEG) and fibrin, whose ratios were optimised until well-interconnected porosity and optimal rheological properties were obtained. The average pore size, assessed through Scanning Electron Microscopy (SEM) analysis, was 120 μm, suitable for supporting cell survival, and nutrients and oxygen diffusion. Loss and storage moduli, assessed through rheology, were in the range of 10-100 kPa, resembling skin viscoelastic properties. The tube inverting test showed a gel formation occurring in 180±30 s at 37°C. The entire optimization process was carried out with the aim of achieving a hydrogel suitable for future bioprinting applications. The dermal compartment was generated by embedding fibroblasts (FBs) into the hydrogel, and cell viability was evaluated by means of Live/Dead and Prestoblue assays. In this study, two different methods were used to fabricate melanoma cell spheroids, both employing the 451-Lu metastatic melanoma sub-line of WM164. The first approach used was the hanging drop method while the second one employed a 96-round bottom well plate. The latter turned out to be the best fabrication method in terms of ease of use and the number of spheroids obtainable in a short time. Various cell densities were tested and the spheroid diameters measured after 72 hours from the initial spot ranged from 300 to 1000 μm. At day 4, spheroids were incorporated into the hydrogel by pipetting and characterised for cell viability through the CellTiter-Glo assay, while the formation of a necrotic core was observed from Live/Dead stained images. The results demonstrate that this 3D in vitro melanoma model could provide a physiologically relevant platform for future drug screening applications. Additionally, it might offer insightful information on tumour-stroma interactions, focusing on how melanoma-fibroblast cross-talk influences tumour biology. The inclusion of perfused blood vessels within the skin equivalent would also be a further step, as angiogenesis is essential for tumour progression and metastasis.