The mutational footprints of cancer therapies

Like Comment
Read the paper

Chemotherapies and radiotherapy revolutionized cancer treatment in the 20th century. Although new waves of very promising targeted therapies and immunotherapies have entered the toolbox of anti-cancer treatments in recent decades, chemotherapies and radiotherapy --alongside surgery or alone-- are still the workhorse of the treatment of primary tumors. Many chemotherapies and radiotherapy work by damaging the DNA of cells. Many tumor cells (but also fast-dividing healthy cells in the body) fail to repair the massive amount of damage created by these therapies and die. However, some cells (both tumoral and healthy) are able to withstand the load of lesions created, at the cost of acquiring mutations in their genomes.

Not unlike chemotherapies, other mutagenic agents (some of them external, like UV-light or tobacco) damage the DNA of our cells and leave behind hundreds or thousands of mutations imprinted in their genome. Each of these mutagens causes specific types of DNA lesions that give rise to particular mutation types, much like a signature, or footprint (Figure 1). Owing to a decade of sequencing primary cancer genomes, we have been able to unravel the mutational footprint of many of the mutagenic agents to which we are exposed in our daily lives [1-7]. The mutational footprints of most chemotherapies, on the other hand, are still unknown. This is because only cancer patients are exposed to them, and during a relatively short period of time. If the mutational footprint of chemotherapies is to be revealed, we reasoned, it must be through the analysis of the mutations in metastatic tumors of patients who received these drugs. Tracking the footprint of different widely-used chemotherapies would then allow us to measure the burden of mutations contributed to the patients’ tumors, an important step to understand their potential impact on healthy cells, and study their late side-effects. 

We were thus extremely fortunate that the genome of more than 3,500 such metastatic tumors --the largest collection to date-- had just been sequenced and made available for research by the Hartwig Medical Foundation ( Using two different approaches [1-4], we extracted the latent footprints of several mutational processes active throughout the lives of these patients. Some of them were already known, and corresponded to common mutational processes, such as aging, UV-light, tobacco, faulty DNA repair mechanisms, and others. However, we were able to show that some mutational signatures which have not been observed in primary tumors constituted the footprint of five chemotherapies (three platinum-based drugs, temozolomide and capecitabine) and radiotherapy. 

Using their mutational footprint, we could then measure that the contribution of mutations of the four most widely used chemotherapies (platinum-based drugs and capecitabine) to the metastatic tumors of these patients is comparable to that of the aging process. Given that the time of exposure to these chemotherapies is comparatively short, this means that over this period they actually contribute between a hundred and a thousand times more mutations than the endogenous aging process. Finally, we reasoned that the functional effects of the mutations contributed by different processes are probably mediated by their likelihood of leaving coding mutations --i.e., those more probably affecting the activity of proteins. Thus, we developed a mathematical approach to compute this likelihood for different mutational processes, including these four chemotherapies.

The discovery of the mutational footprints of these chemotherapies and the approaches devised in this study have opened up a door to the study of their functional effects on the healthy cells of patients. We hope that this door allows us to understand the late side-effects of these chemotherapies [8,9]. We also expect that it will allow us to develop a way to measure the mutational toxicity of these chemotherapies in real time --i.e., during the treatment. This way clinicians may be able to maximize the tu effect of the drugs while minimizing their mutational impact on healthy cells.


1. Alexandrov, L. B. et al. Signatures of mutational processes in human cancer. Nature 500, 415–21 (2013).

2. Alexandrov, L. et al. The Repertoire of Mutational Signatures in Human Cancer. bioRxiv 322859 (2018). doi:10.1101/322859

3. Kasar, S. et al. Whole-genome sequencing reveals activation-induced cytidine deaminase signatures during indolent chronic lymphocytic leukaemia evolution. Nat. Commun. 6, 8866 (2015).

4. Kim, J. et al. Somatic ERCC2 mutations are associated with a distinct genomic signature in urothelial tumors. Nat. Genet. 48, 600–606 (2016).

5. Nik-Zainal, S. et al. The genome as a record of environmental exposure. Mutagenesis 30, 763–770 (2015).

6. Helleday, T., Eshtad, S. & Nik-Zainal, S. Mechanisms underlying mutational signatures in human cancers. Nat. Rev. Genet. 15, 585–598 (2014).

7. Kucab, J. E. et al. A Compendium of Mutational Signatures of Environmental Agents. Cell 177, 821-836.e16 (2019).

8. Kopp, L. M., Gupta, P., Pelayo-Katsanis, L., Wittman, B. & Katsanis, E. Late Effects in Adult Survivors of Pediatric Cancer: A Guide for the Primary Care Physician. Am. J. Med. 125, 636–641 (2012).

9. Iyer, N. S., Balsamo, L. M., Bracken, M. B. & Kadan-Lottick, N. S. Chemotherapy-only treatment effects on long-term neurocognitive functioning in childhood ALL survivors: A review and meta-analysis. Blood 126, 346–353 (2015).

Abel Gonzalez-Perez

Research Associate, IRB Barcelona

No comments yet.