Genetic changes in tumour microenvironments.

Numerous in vitro and in vivo studies have established that carcinoma-associated fibroblasts differ phenotypically from fibroblasts associated with normal tissue but the mechanisms underlying these differences are unclear. Since carcinoma-associated fibroblasts can be propagated in vitro for extended periods and still maintain their cancer-promoting phenotype, some investigators have proposed that they might have acquired somatic genetic alterations analogous to those observed in malignant epithelium. Early molecular genetic studies appeared to validate this hypothesis by demonstrating remarkably high frequencies of clonal somatic genetic alterations in carcinoma-associated fibroblasts, including loss of heterozygosity, gene amplification, and point mutations in tumour suppressor genes such as TP53 and PTEN. The initial excitement of these paradigm-changing studies overshadowed concerns that there may have been a more mundane explanation for these observations. In addition to the fact that the data would necessarily invoke an unlikely scenario of the simultaneous generation of two symbiotic malignancies, subsequent molecular genetic studies found no evidence of frequent genomic aberrations. One striking common trait of those studies reporting frequent clonal somatic alterations in carcinoma-associated fibroblasts is the use of tissues and techniques which are well known to be highly prone to generating artefacts such as limiting and poor quality DNA followed by highly multiplexed PCR-based analyses. It is now clear that clonal somatic mutations are not the biological basis of the cancer-promoting attributes of carcinoma-associated fibroblasts.