Are cancer causing mutations losing their "Driver" status? - a personal view

When I first started looking into dbSNP database in the year 2012, I was quite surprised to discover that the so called "Driver" mutations were catalogued in this human genetic variation database with a reference SNP cluster ID. dbSNP entries were meant to be a repository for germline variations. Whereas for somatic mutations which are detected in cancerous tissues, COSMIC database was the place to go looking.

To validate whether germline variants co-localize with somatic variants, I took a look into the TCGA datasets. Surprisingly, there were quite a lot of correlation among germline and somatic variants. Shown below is the plotting of all aggregated somatic variations versus aggregated germline variations found in the human genome. Clearly, a large majority of somatic variants are also found in the human genetic variation database. This suggests the spectrum of somatic and germline mutations are more or less identical for human tissues. In simple words, in some people healthy tissues contain genetic alterations that are only found in tumor tissues of cancer patients.

 

These findings can only lead to one conclusion that tumor-specific somatic mutations are also found in healthy tissues? But there was no consensus on how to deal with this information. 

Over the next couple of years with the wide-spread expansion in the usage of NGS from 2015 onwards, more and more labs encountered the same information that cancer-causing "Driver" mutations are also found in healthy tissues. This is reflected by many publications, that did manage to attract popular media and scientific news outlets.

 

 

The collection of tumor-specific somatic mutation in the COSMIC database are each assigned a unique COSMIC ID for all its entries (Forbes et al, Curr Protoc Hum Genet. 2009). For many of these COSMIC ID entries there are corresponding reference SNP ID found in human genetic variation database dbSNP.

For instance: 

BRAF p.V600E, COSMIC ID - COSV56056643, dbSNP ID - rs113488022

KRAS p.G12V, COSMIC ID - COSV55497419, dbSNP ID - rs121913529

Many benign skin naevi are composed of cells expressing BRAF p.V600E, yet the mutation does not induce neoplasia or malignancies (Wu et al, Am J Dermatopathol 2007). Similary, actively cycling gut cells are also known to contain cancer-causing somatic mutations (Tomasetti et al, Science 2017). Very likely they are just a part of shedding worn out cells. This suggests that, if I were to sequence callous tissues, I am very likely going to encounter many somatic driver mutations. I will wait until someone does this sequencing and posts a comment on this post.

We are probably going to see a paradigm shifting view on how cancer is caused in the following years, as more and more clarity emerges from human tissue sequencing studies. 

The question whether a somatic driver mutation per se is enough to cause cancer will become a debatable topic for human cancers. 

Why different individuals manifest differently to cancer-driving somatic mutations?

It is very likely that we are going to adopt a more coherent and inclusive definition for human cancers, one that considers tumor microenvironment as a major player in cancer etiology.