Written by Liam Little, Science Writer
A new study, published in Science, has used whole genome sequencing (WGS) to explore mutational signatures across the largest reported cohort of multiple tumour samples. Novel mutational signatures were discovered relating to many different cancer types, helping to build a more complete picture of the cancer genome and unlocking new potential targets for therapeutic intervention. The concepts of common and rare mutational signatures are introduced, and the researchers present an algorithm to utilise these in the pursuit of personalised cancer treatments.
Exploring the cancer genome
The development of cancer is characterised by thousands of genetic aberrations and mutations. Understanding the mutational landscape can provide diagnostic insights and uncover the patterns of mutations developed during tumourigenesis. Modern WGS approaches now make it possible to explore the full landscape of both causative “driver” mutations and additional “passenger” mutations.
Within the study, researchers have performed mutational signature analysis of over 12,000 whole genome sequenced cancers collected for the 100,000 Genomes Project. These results were then validated against cohorts from the International Cancer Genome Consortium (ICGC) and the Hartwig Medical Foundation. This produced the largest cohort of WGS cancer samples to date, encompassing 19 different tumour types.
From common to rare mutational signatures
The increased cohort size provided an abundance of data, including a record of single-base substitutions (SBS) and double-base substitutions (DBS) in cancer genomes. The researchers were able to confirm many previously reported mutational signatures, but also discovered 40 SBS and 18 DBS signatures that had not previously been identified. The researchers attempted to characterise the new signatures they discovered.
The concepts of “common” and “rare” mutational signatures were introduced in the study. A small number of “common” signatures appear to be prevalent across each tumour type, independent of the size of the cohort. In contrast a larger number of “rare” signatures are seen in a specific organ and are dependent on the number of samples analysed.
Comparing unreported signatures with endogenous processes, pathways including base excision repair, mismatch repair and double strand break repair were highlighted. Environmental exposures can also be associated with mutational signatures, giving a more complete view of the cancer genome. Here, the researchers link mutations to exposures such as UV, platinum and tobacco.
Assessing mutational signatures in the future
Using their results, the researchers developed a signature-fitting algorithm, Fit Multi-Step (FitMS). By fitting common organ-specific signatures and then searching for rare signatures, FitMS can help to assess the presence of mutational signatures in future samples.
The authors also discuss the significance of common and rare signatures. “An increased number of samples may help discern common signatures that occur at low levels for specific tumour-types. Greater sample numbers may also help unveil signatures that occur at a low frequency in the population.”
