An important question in cancer research is the causes and mechanisms for the development of multifocal carcinomas, especially for non-small cell lung carcinomas (as I recently blogged on). While a “field cancerization” effect has been postulated as a model for multifocal carcinomas for a long time, it remains enigmatic why cancers develop at one site and not another given that the entire “field” is presumably exposed to the same carcinogenic agent or stimulus. More specifically, it is also obscure what the role is of the peritumoral stromal changes observed around multifocal carcinomas. Are these changes a cause or consequence (or, more likely, both) of these tumors? Under normal circumstances, the epithelial tissue stromal architecture imposes growth restraints upon epithelial structures and exerts a tumor-suppressive effect on “transformed” neoplastic cells. A similar effect is observed in the metastatic process in which the vast majority of metastatic tumor cells fail to form tumors at distant metastatic sites. The conventional view has been that the tumor stromal microenvironment is reactive in nature; that is, the tumor stroma develops as a result of molecular cues from the tumor itself. A recent article in Cell (8 June 2012) by Bing Ho and colleagues presents data suggesting that the loss of mesenchymal Notch/CSL signaling causes stromal changes that are potent stimuli for the development of carcinomas (abstract).
Mice carrying a mesenchymal-specific deletion of CSL-RBP-Jk, a key effector in the Notch signaling pathway, develop spontaneous multifocal squamous cell carcinomas after dermal atrophy and chronic inflammation. The authors found that CSL-deficient dermal fibroblasts promote increased tumor cell proliferation through upregulation of c-Jun and c-Fos expression. This, in turn, was associated with higher levels of diffusible growth factors, inflammatory cytokines, and matrix-remodeling enzymes.
The authors then tested these findings with human skin biopsy specimens of multifocal actinic keratosis, a premalignant skin lesion for cutaneous squamous cell carcinoma strongly associated with sun exposure. Stromal fields adjacent to multifocal actinic keratosis showed decreased Notch/CSL signaling, albeit only with exposure to UVA and not UVB exposure, through DNA methylation and epigenetic silencing of Notch signaling in dermal fibroblasts. These findings imply that the phenomenon of field cancerization can be mediated through stromal alterations preceding the development of neoplastic epithelial clones—in other words, the stroma is precancerous due to mutagen exposure.
This model would be a fascinating one to pursue for future lung cancer research with both non-small cell and small cell lung cancers since these cancers are all also strongly associated with a known mutagenic exposure, are posited to develop from a field cancerization effect from this exposure, and are frequently associated with multifocal synchronous tumors. Unfortunately, there isn’t exactly an analogous premalignant “actinic keratosis” lesion for NSCLC, at least one that can be easily recognize and sample, although I think such atypical adenomatous hyperplasia foci may be easier to study in mouse models. Anyone interested in pursuing this, let me know! ; )