Abstract

Metastasis, a complex and multistep process, accounts for more than 90% of cancer fatalities. The specific features of malignant cells such as uncontrolled growth, altered metabolism and invasive behaviours constitute the hallmarks of cancer [1]. Recent studies have highlighted the importance of the tumour microenvironment on these signatures of cancer and on the onset of metastasis [2]. All these factors have been associated with numerous biological determinants resulting both from an accumulation of genes mutations and a malfunction of numerous regulatory signaling pathways. Owing to the heterogeneity in these factors across different cancers and patients, no common mechanistic pathway leading to metastasis induction and progression has yet been characterized [1,2,3]. In contrast to these tremendous variations of genotype and phenotype, most solid tumours display similar collective invasion behaviours [3]. The cohesive multicellular structures formed during invasion such as cell sheets, strands or clusters display stunning morphological similarities and identical mechanical behaviours across most cancer pathologies and patients [3,4]. This homogeneity in terms of morphology and migration behaviours suggest that the invasion stage of metastasis could be mainly controlled by physical interactions between cells and simple mechanical phenomena [4]. Using a newly introduced computational framework [5], we study the collective dynamics of a cancer cell population enclosed in a mechanically resistive tissue. We observe that simple physical rules are sufficient to account for some of the common morphologies of invading tumours, including regimes of individual and collective invasions [6]. Moreover, the introduction of a few fibroblast-like cells, with an enhanced ability to remodel the tumour microenvironment, leads to a strong increase in the invasiveness of the cancer cell population. Their invasion behaviour is here associated with the frequent occurrence of “cell fingers”, guided by fibroblastlike cells at their leading edge [6]. Overall our in silico experiments reproduce the phenomenology of invasion across the different cancer pathologies and provide new insights on the mechanisms controlling this complex phenomenon [6]. Based on simple biophysical hypotheses and generic cellular interactions, we believe that our approach will help to unfold the different biological contributions to metastasis and to disentangle the links between genes, environment and malignancy.