**1. Introduction**

20 Breast Cancer – Focusing Tumor Microenvironment, Stem Cells and Metastasis

Welch, D. (1997). Technical considerations for studying cancer metastasis in vivo. Clin Exp

The increased use of animals in fundamental and applied research due to the remarkable drug development in the 20th century has been an important matter of concern for people at large, but also for the scientific community. This led Russel and Burch to examine the decisions which could meliorate this situation, and they proposed, in 1959, the principle of the 3Rs (Reduce, Refine, and Replace) nowadays largely admitted as an ethical and incontrovertible principle (Russell & Bursch 1959). Alternatives to animal experiments (Scheme 1) then knew a fantastic boom with the permanent objective of a high scientific quality in order to prevent, treat and cure human illness.

Reaching the equilibrium between *in vitro* and *in vivo* models, observing the 3Rs rules, is very difficult. Effectively, *in vitro* systems allow an excellent control of all parameters of the experiments, and then, good quantifications. More the models are simple, more they are easy to handle, but more they also are dedifferentiated and keep away from the *in vivo* situation.

Scheme 1. *In vitro* systems as alternatives to the use of animals.

Within the framework of this book, the question becomes now: how the 3Rs could be the best way to phase out animal experiments when considering breast cancer? We try to bring some response elements in this chapter, emphasising the *in vitro* models the most useful and the most frequently used. But we also show that no model is perfect and sufficient by itself, and that pure *in vitro* models also need assistance of *in vivo* ones.

*In Vitro* Breast Cancer Models as Useful Tools in Therapeutics? 23

The relevance of cellular models is controversial since their over-simplicity implies difficulties in extrapolating results from the cell line to the tumour in humans and thus

Indeed, cell lines are homogeneous, theoretically consisting only of a single cell type (pure




The hundred of available cell lines do not cover all of the tumour features found in patients. Furthermore, the proportions of some characteristics are sometimes reversed, such as the ER and PgR status which is very different in cell lines, compared to that found in the patient population (Lacroix & Leclercq 2004). These dissimilarities can be explained by the fact that most lines are derived from pleural effusion and metastases containing cells which are already different from the original tumour and thus, more or less representative of this tumour. Indeed, the ER/PgR status sometimes differs between the metastasis and the original tumour from the same patient. Based on these observations, several teams have worked on the development of cell lines derived from a primary tumour (Amadori *et al.* 1993; Gazdar *et al.* 1998; Shen *et al.* 2009), which are much more representative of the *in vivo* cancerous tissues that lines derived from metastases, but which suffer from the same problems related to their relative homogeneity and instability in a long term use. Moreover, the establishment of cell lines from primary tumours remains a difficult achievement, failures mainly being the result of contamination by the

Besides these previous drawbacks, many criticisms have been made against BCCL because some of them have been proven not being from breast cancer origin. Indeed, some lines were contaminated by other cell types during their first years of use, then spread to other laboratories, and used on a large scale without further verifications of their true origin. Several cell lines were denounced as false, whereas it was not the case (Fogh *et al.* 1977; Nelson-Rees & Flandermeyer 1977). These contaminations have been subjects of controversial for a long time. However, studies have shown with certitude that two cell lines

The MCF-7-ADRr cell line was developed in 1986 by Batist. It is derived from the lineage of human mammary adenocarcinoma MCF-7 and was rendered resistant to adriamycin treatment after exposition to increased concentrations of this drug. The obtained resistant cell line was also resistant to other agents such as actinomycin D, vinblastine and vincristine. However in 1998, the lineage between MCF-7 and MCF-7-ADRr became controversial, as shown by DNA fingerprinting studies and genetic comparison, so that the true origin of the cell line was undetermined and the cell line was renamed NCI/ADR-RES. Liscovitch and Ravid, in 2007, have collected data showing that NCI/ADR-RES were carcinoma ovarian


raises the question of their representativeness.

and clonal) due to the way they are established:

most stable culture during subcultures.

These particularities reduce the similarity with the primary tumour.

than the one of the surviving cells.

systems active *in vivo.*


stroma surrounding the tumour. - Confusion with some cell lines

were not from their supposed origin.
