High-powered genetic sequencing techniques are increasingly being applied to a wide spectrum of human cancers in an attempt to bring some order and understanding to the seemingly chaotic world of neoplasia.
A recent open source article in Nature (Nature 2013;502:333-8
) looked at over 3000 examples of twelve common human tumors (glioblastoma, ENT squamous carcinoma, breast carcinoma, renal carcinoma, endometrial carcinoma, rectal carcinoma, pulmonary adenocarcinoma, pulmonary squamous carcinoma, bladder carcinoma, ovarian carcinoma, colonic carcinoma, and leukemia).
This study looked at not just DNA mutational data but also copy number variations, DNA methylation, micro RNA activity, cellular protein activity and clinical data. The purpose of the study was to find "core" cancer genes and carcinogenic pathways as distinct from "collateral genetic damage unrelated to pathogenesis; in essence to look for order in apparent chaos. 127 significantly mutated genes were identified across the 12 tumor types. Some were well known oncogenes and others are emerging as factors in oncogenesis. The average number of mutations in the significantly mutated genes varied across tumour types; most tumours had two to six, indicating that the number of driver mutations required during oncogenesis is relatively small. Mutations in transcriptional factors/regulators showed tissue specificity, whereas histone modifiers were often mutated across several cancer types.
The article is heavy reading and you may not wish to indulge, but there is an excellent accompanying editorial by Alan Ashworth (Nature 2013;502;306-7
) which outlines the power as well as the limitations of this high volume approach to cancer genetics. Numerous publications of this type have gone into the development of the "Cancer Genome Atlas." One of the most obvious goals of these studies is the identification of common and necessary genetic events in carcinogenesis that may lead to the identication of further therapeutic targets for drug development. Such drugs may well be active over multiple tumor types.
Is the day coming when no one will care about the "morphology" or even the anatomic origin of a tumor and will be only interested in its genetic abberations? If you had asked me that 10 years ago I would have laughed and said, "(expletive deleted) NO!" Now.... I'm not quite so sure. Probably not in my or your lifetime but NEVER is a very long time. Clearly pathologists need to embrace this new world and own a spot within it as we move into the molecular unknown of neoplasia.