3D cell culture techniques accurately represent how cells grow and how they are affected by diseases, including cancer. In such a view, 3D cultures offer a great opportunity for cancer treatment research, allowing to investigate molecular therapeutic targets and their pharmacological inhibition. Indeed, 3D cultures represent valuable simulators of cancer tissues, as they exhibit similar growth and treatment response patterns.
However, there are challenges to be addressed, as the reproducibility of 3D cultures depends on many variables, including the substrate in which cells grow, the proper phase of growth and cancer features.
The model applied in this study reduces the substrate-related variables, since a small bioreactor, integrating a 3D synthetic, inert, and biocompatible scaffold, was used to favor reproducible cell colonization and growth. Due to the high amount of cells, real-time monitoring of cells viability and cell number estimation with Real-Time Glo (Promega) (as reported in Candini et al. 2019) is not possible. However, the scope of this work does not require this specific type of evaluation. Histology, immunohistochemistry and NGS give evidence about cell presence, viability and behaviuor with the main aim to compare the 3D culture with its original tumor counterpart.
The growth phase can be controlled by stimulating, with appropriate growth factors, only the component of interest (epithelial if the growth of carcinoma is to be stimulated, mesenchymal if of sarcoma), generating an in vitro model where tumor progression acceleration may reveal the mechanisms underlying neoplastic transformation for a drug discovery approach.
Furthermore, the morphologic and immunophenotypic analogies between 3D growth and primary tumor, combined with the rapid establishment of primary cultures, reduced the quantity of culture medium and growth factors, being time- and cost-effective. Compared to 2D cultures and even to animal models, 3D cultures provide a more reliable tool in predicting how drug treatments will affect patients.[22]
Most of the in vitro knowledge regarding spontaneous tumors in small animals comes from the study of 2D cell cultures or of the tumor itself.[1] However, it is well known that the monolayer 2D growth is represented by neoplastic cells having unlimited access to the ingredients of the medium, including oxygen, nutrients and signal molecules, which does not mimic a natural neoplastic mass. An additional important drawback is due to the fact that cancer development and progression rely not only on the main cell population, but also on the interactions with the associated stroma, intended as the matrix and stromal cells. When neoplastic cells are placed back into an in vivo environment (i.e., an animal model) or are grown in a 3D system, proliferation and gene/protein expression are much closer to the ones occurring within the tumor than those grown in a 2D systems.[1]
In our study, it has been shown that morphology and immunophenotype of 3D cell cultures are able to adequately replicate the growth of canine primary carcinomas and soft tissue sarcomas, not only by the expression of the primary tumor differentiation markers, but also by re-creating a tumor microenvironment, characterized by stromal fibroblasts in all the epithelial tumor cultures and inflammatory cells (macrophages in one case).
While the expression of differentiation markers between primary tumors and 3D growths was overall concordant, in one sarcoma case actin was expressed by the primary tumor only. This discrepancy may be due to a de-differentiation in cell cultures rather than to the selective growth of cells negative for that marker, which represented half of the population in the primary tumor.
At the genetic level in sarcomas, no sequence alterations were found nether in 3D cell culture nor in primary tumor samples. Intriguingly, comparing the MDM2 amplicon coverage value with that of TP53, in one case we observed a preferential amplification in MDM2 amplicons, hypothesizing a possible MDM2 amplification, both in 3D cell culture and in primary tumor specimens from the same sample. Indeed, in the other two cases, the amplicons ratio (MDM2 coverage:TP53 coverage) was near 1, and also in these samples, the ratio was similar both in 3D cell culture and in primary tumor specimens.
As well as morphologic and phenotypic similarities, also genomic data are concordant between primary tumours and the corresponding 3D cell cultures, which can be therefore considered a suitable substrate for biobanking even after formalin-fixation and paraffin embedding.
In oncology, several diagnostic tools are available to evaluate the expression of molecules aimed at tailoring treatment, thereby providing a precision approach. However, many of these tests are based on the evaluation of fixed tissues where cells viability is lost, leading to the impossibility to investigate cell function during tumor growth and progression. Being the cell-to-cell and cell-to-extracellular matrix interaction a fundamental condition for more representative tumor studies, we investigated and demonstrated the reproducibility of morphologic, immunophenotypic and tumor microenvironment aspects in in vitro models using 3D cultures obtained from canine carcinomas and soft tissue sarcomas. Furthermore, in this in vitro model, a genetic similarity was demonstrated with parallel amplicon coverage in primary tumor and corresponding cell culture.