In recent years, developments in next generation sequencinghave allowed us to study different cancers at levels unknown before. And, as our recent special collection on the genomics of cancer progression and heterogeneity, as well as numerous other studies show, one of the most profound and therapeutically inconvenient features of cancer is its heterogeneity.
It is often the case that metastatic tumor is not a good representation of the primary lesion and vice versa. What this means is that even if we are able to profile the primary tumor genetically, we may still be a long way from finding an appropriate drug to treat its metastases.
There has been some evidence, however, that this is not always the case. The group of David Solit from Memorial Sloan Kettering Cancer Center demonstrated a couple of years ago that mutations occurring in some of the most commonly altered genes in colorectal cancer – KRAS, NRAS, BRAF and several others – remain broadly unchanged in the metastases. What is more, changes, if there were any, were mostly associated with already undergoing therapy (so were not, strictly speaking, spontaneous mutations).
In a study published in Genome Biology last August David Solit, Michael Berger and colleagues widely extended these findings. By looking at a much wider panel of some 230 known cancer genes in 69 patients for whom primary and matched metastatic colorectal cancer samples were available, they showed that mutation profiles for KRAS, NRAS and BRAF genes were 100% concordant!
They additionally tested four more patients with whole genome sequencing, and these experiments confirmed that no additional site-specific alterations occurred.
As with lung cancer,the most well-described driver alterations in colorectal cancer are concordant between primary tumors and metastatic lesions.
Patrick Tan & colleagues
Genome Biol 2015, 16: 32
Last week Genome Biology published an article from Patrick Tan and colleagues at the National University of Singapore, who looked at an even larger panel of some 750 known cancer genes in 18 patients for whom they had matched samples of normal tissue, colorectal cancer primary lesions and liver-limited metastases.
They too observed very high mutation concordance levels. Importantly, they saw no major differences between the patients with different types of colorectal cancer, or between patients who did or did not receive the therapy.
These studies suggest that, at least in case of colorectal cancer, the tumor can be probed either from the primary or from the metastatic site during clinical testing. Additionally, the results of Patrick Tan and colleagues study confirm the linear progression model in this cancer type.