In recent years, biological sciences have witnessed a surge in the generation of data. This trend is set to continue, heralding an increased need for bioinformatics research. By 2018, sequencing of patient genomes will likely produce one quintillion bytes of data annually – that is a million times a million times a million bytes of data. Much of this data will derive from studies of patients with cancer. Indeed, already under The Cancer Genome Atlas (TCGA), comprehensive molecular analysis, including exome sequencing and expression profiling, has been performed on more than 10,000 patients.
Given this veritable flood of data, the bottleneck in cancer research is now how to analyse this data to provide insights into cancer biology and inform treatment.
The fruits of cancer bioinformatics research are in fact already making their mark on our understanding of cancer. To date, more than 1,000 studies have been published based on TCGA-collected data alone and these important studies, and others, provide a roadmap for the future of cancer bioinformatics.
Clinical relevance of the catalogue of drivers and mutational processes
A first step to understanding the mechanisms of tumour emergence, evolution, and drug resistance is to understand the driver events and mutational processes in cancer. Sequencing data has seen an explosion in tools to perform such analysis, shedding light on novel driver events, and the mutational process occurring during tumour evolution. However, these studies have also revealed that neither the catalogue of drivers nor our understanding of the processes that mould cancer genomes is close to complete. Moreover, in general, most studies have only focused on the coding part of the genome.
From a clinical perspective, a key question relates to the clinical relevance of these events and how they can guide clinical decision-making. A recent large-scale computational analysis found that although less than six percent of tumours are tractable by approved agents following clinical guidelines, up to 40.2% could benefit from different repurposing options, and over 70% could benefit when considering treatments currently under clinical investigation. As has been previously discussed, the scale of evidence for genomic events to guide precision medicine needs to be further established.
Cancer genome evolution
To fully understand cancer, we need not only to understand the events, but also when they occur during tumour evolution and how they may facilitate cancer drug resistance. Recent years have seen the development of a plethora of bioinformatics tools to dissect cancer genome evolution and intra-tumour heterogeneity, and these have been used to shed light on the complex clonal architecture of tumours, revealing clonal driver events present in the founding cancer cell, as well as later mutations in cancer genes present only in a subset of cancer cells. These studies highlight the need not only to consider whether a mutation is present or absent, but whether it is present in all or only a subset of tumour cells, which may have major implications for targeted there therapies.
A crucial next step will be to attempt to write the rulebook for cancer genome evolution and how cancer therapy influences tumour evolution.
One of the exciting breakthroughs in cancer research over the last years has come from work in immunology. The recognition of the importance of tumour-specific neoantigens that may act as targets for immune responses has increased interest in personalized vaccines and strategies to accentuate anti-tumour immune responses, including immune-checkpoint blockade, which seek to unleash a patient’s own T cells to destroy the tumour.
Recent work has begun to reveal the genomic underpinning of when immunotherapy succeeds or fails, and the next steps in extending this work will necessarily require a bioinformatic foundation. In particular, we need to improve the tools to identify neoantigens and also extend our understanding of how the cancer genome shapes and is shaped by anti-tumour immunity.
Taken together, this exciting work provides a platform for future bioinformatics research, both from a methods and an analysis perspective. As the flood of cancer genomic data continues, the need for bioinformatics is more pressing than ever.
Featured image credit: DNA by PublicDomainPictures. CC0 Public Domain via Pixabay.