Driver mutation vs biomarker vs recurrent mutation

Question

[CY]

I have had concern for the relationship of these three terms for long time...

My understanding is that

1: For a specific cancer type, mutant 'A' could be its one of many driver genes. On the ohter hand, mutant 'A' must has high sensitivity and specificity to that cancer type to be its biomarker (Not sure if both sensitivity and specificity needed?). Seems like biomarker and driver has some overlap, like IDH1 as both driver gene and biomarker for glioma.

2: The discovery of driver mutation / biomarker is usually initiated from identifying recurrent mutations. Seems like this is for now the only way to discover driver mutation / biomarker although, in my opinion, driver mutation / biomarker does not has to be current.

3: There are many driver gene prediction tools. Anyone has recommendation of thich one to use?

I am sure I missed some good points about driver mutation / biomarker. Can anyone share some comments?

Answer 1

[Kevin Blighe]

Biomarker

A biomarker is absolutely anything of a biological origin that can be used for diagnosis and/or prognosis of any disease. It could be T- or B-cell counts, CRP levels, a genetic variant, toxicology data of urine, a metabolite in blood, DNA in blood, et cetera.

Driver mutation

A driver mutation is a somatic mutation in cancer for which evidence suggests its key role in cancer progression. The most well known genes that harbour driver mutations include TP53 (of course), ATM, PTEN, MYC, et cetera. Loss or gain of function of these genes can be equally bad and can drive multiple types of cancer. Other genes that also harbour driver mutations may act in more specific ways and be more associated with a specific type of cancer, or cancer of certain tissue, e.g., EGFR, ERBB2, BRCA1, et cetera.

Then again, you have other groups of genes that are involved in genome maintenance and chromosomal stability. Mutations in these may also be driver mutations because they result in genomic instability, which can drive tumour progression. Look up the 'CIN' signature genes.

Of course, driver mutations make for good biomarkers.

Recurrent mutations

These are just mutations that are frequent in a particular cancer type. They may not necessarily be driver mutations but are just frequently mutated due to certain cellular processes that are disrupted in one cancer over another, e..g, DNA repair or other mechanism. For example, PIK3CA is the most frequently mutated gene in breast cancer, whilst TP53 is the most frequently mutated gene across all cancers (carcinomas).

Some recurrent mutations may just be passenger mutations. They may still make for good biomarkers.

Kevin