Cancer heterogeneity and evolution – the review to end all reviews

Series on Tumor Evolution

“A particular successful guide to understanding and modeling cancer progression has been evolutionary theory, which has a long tradition in cancer research. Already 40 years ago, seminal work established an evolutionary view of cancer (Nowell 1976; Dexter et al. 1978; Fidler 1978), in which carcinogenesis is regarded as an evolutionary process driven by stepwise somatic mutations and clonal expansions,”

write the authors of an awesome new review paper titled Cancer evolution: mathematical models and computational inference. (Obviously, I am not biased at all. I would call it ‘awesome’ even if I wasn’t one of the authors – I swear!)

Writing the review article made me wonder how the long tradition of an evolutionary understanding of cancer plays out on PubMed. Using code by R-psychologist I plotted the following figure, which shows the number of PubMed hits for queries on ‘cancer evolution’ (red), ‘cancer heterogeneity’ (yellow), as well as reviews on these topics (green). You can find my complete analysis as an R markdown document on my webpage.

bla bla bla
Figure 1: The number of papers on pubmed on ‘cancer evolution’ (red), ‘cancer heterogeneity’ (yellow), as well as reviews on these topics (green). Code and exact PubMed queries in R markdown document on my webpage.

What is the smallest reviewable result?

The first paper on ‘cancer heterogeneity’ appeared in 1962, on ‘cancer evolution’ in 1965 and the first review appeared in 1975.

But look at how big the green curve of reviews is!

In half of the years, reviews made up >20% of papers in this research area, which means that for every 4 research papers there was a review paper.

That’s quite a lot.

My interpretation is simple: Cancer evolution is a conceptually stagnant field!

Conceptually stagnant means: All the concepts were out there for decades, but had to wait for the technologies to catch up with them. And once they did, the renewed vigor led to a telling, re-telling and re-re-telling of the same basic concepts.

Everything has surely been said by now, but maybe not yet by everybody.

Our own review focuses on mathematical models and methods for data analysis, which is still a bit of a niche – but for how long?

So, for you to be able to focus again on research and not having to read any more redundant summary papers, here is THE REVIEW TO END ALL REVIEWS – a list of facts about tumor evolutions that I would think the community broadly agrees on.

When citing papers in the following, I tried to use as many old papers as possible, to illustrate how well worn these concepts are.

The cancer evolution paradigm

Cancer is heterogeneous. This heterogeneity comes in at least 4 flavors:

  1. Inter patient: Many types of cancer are not one disease but many and molecular profiles can distinguish subtypes [22522925].
  2. Intra patient: Different sites in the same patient can have morphologically and genetically very different tumours.
  3. Intra site: Within each site, a tumour is a complex tissue full of interactions between the cell-autonomous compartment (the cancer cells) and a non-cell-autonomous compartments (immune cells, fibroblasts, blood cells, …)
  4. Inter cells: And finally, as you already know from the beginning of this series, the cancer cells themselves are genetically heterogeneous and a collection of different clonal populations. This means: There is not THE cancer genome. Each cancer has a collection of many different genomes.
bla bla
Figure 2: Four levels of tumor heterogeneity. (Yes, this is a nice figure. If you use it in your talks, put my name on the slide!)

Evolution leads to heterogeneity

Cancer genomes are formed by evolutionary processes within each patient.

Cancer evolution is a step-wise process of mutations and clonal expansions.

“It is proposed that most neoplasms arise from a single cell of origin, and tumor progression results from acquired genetic variability within the original clone allowing sequential selection of more aggressive sublines.” [959840]

Cancer genomes show traces of the evolutionary process in the form of mutations, which include single nucleotide variants (SNVs) and copy number aberrations (CNAs) and other genetic and epi-genetic changes.

“Cancer evolves dynamically as clonal expansions supersede one another driven by shifting selective pressures, mutational processes, and disrupted cancer genes.

These processes mark the genome, such that a cancer’s life history is encrypted in the somatic mutations present.” [22608083]

In the end, a cancer is a diverse mix of different cellular subpopulations.

“The list of characteristics by which subpopulations differ is also extensive: cellular morphology; tumor histology; karyotype and other cytogenetic markers; growth rate; cell products; receptors; enzymes; immunological characteristics; metastatic ability; and sensitivity to therapeutic agents. In addition, since sister cells usually remain contiguous in solid tumors, sublines tend to be localized regionally or zoned.” [6372991]

Drivers and passengers

Mutations are either drivers or passengers:

  1. Drivers are the selected few mutations that give a cell a fitness/growth advantage.
    • Driver mutations are necessary for tumor development and the obvious targets for precision therapy.
    • One of the most popular ways to identify drivers is to hunt for genes that are much more often mutated than expected by chance [23770567]
  2. Passengers are most of the mutations you find in a cancer genome.
    • Passengers hitchhike with drivers but don’t have a fitness advantage
    • They are however useful for many analyses:
      1. They are the basis for evolutionary inference (as discussed in the previous posts)
      2. they can form mutational signatures to infer which mutational processes were perturbed in the cancers history [23945592]

Different types of cancer can be driven by different types of mutations. The two main classes of cancer are dominated by either SNVs or CNAs, but not both. [24071851]

Different sub-clones can carry different sets of drivers. Else things would be boring, wouldn’t they …

Drivers are context specific. Which mutation is a driver and which is a passenger depends on the selective pressures (=drugs) acting on a cell and can vary over time for the same cell.

Diagnosis and therapy

Genetic heterogeneity can confound diagnosis if a biopsy doesn’t reveal the full spectrum of mutations in a patient.

“In practice such molecular analyses are often performed on single small biopsy specimens taken from large lesions. Interpretation of the results must take into account the question of whether a tumour specimen is truly representative of the entire tumour.” [8777166]

Genetic heterogeneity enables cancers to become resistant to therapy, because even if treatment wipes out most of the cancer cells, there might be a small population of cells (a minor sub-clone) that carries a mutation that makes it resistant. And until treatment changes this minor population will repopulate the tumor.

“Hence, each patient’s cancer may require individual specific therapy, and even this may be thwarted by emergence of a genetically variant subline resistant to the treatment.

More research should be directed toward understanding and controlling the evolutionary process in tumors before it reaches the late stage usually seen in clinical cancer.” [959840]

Not only a disease of the genome …

Different sub-clones can have different metastatic potential.

Seed and soil: the cancer cell needs to fit into the target tissue to metastasize there.

“The development of a metastasis is dependent on an interplay between host factors and intrinsic characteristics of malignant tumor cells. The process of metastasis is highly selective, and the metastatic lesion represents the end point of many destructive events that only a few cells can survive.” [354778]

And finally, our normal tissue is a defense against cancer and the cellular microenvironment plays a major role in shaping tumor development (a fact that genomics-y people are often happy to forget):

“If the microenvironment were not dominant, each cell would have its own way and the result would be either a uniform lump of similar fate or absolute chaos.

Early examples of the dominance of the microenvironment on the processes that unleash cancer were gleaned from the study of the functional consequences of exposure to carcinogenic chemicals.” [21383745]

This is the end

Phew … now everything has been said by me too!

It feels good to have all that stuff out of my system.

Please don’t summarize it again.


Acknowledgements: Thanks to Edith Ross and Geoff Macintyre for feedback on this post.


3 thoughts on “Cancer heterogeneity and evolution – the review to end all reviews

You gotta talk to me!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s