Lessons And Methods From Evolutionary Biology For Tumor Biology

Question

[user1202664]

A recent paper has confirmed just how great intratumor heterogeneity can be. This has implications for prognosis and cancer management.

However, extracting multiple samples from a single cancer case and assaying these samples is - given current technology - not practicable in terms of cost or timescale. These constraints often mean that microarray analysis or sequencing is confined to a snapshot of multiple cells from a single sample for each cancer case.

Are there any suggestions about how to characterize the molecular heterogeneity and genomic instability of cancers given these constraints? Specifically, I am thinking about how methods such as molecular phylogenetics can be applied to this situation. In many ways, the problems facing an evolutionary biologist are similar to those faced by tumor biologists: they must use a limited number of snapshots to infer the details of a heterogenous and dynamic process.

Answer 1

[Nathan Harmston]

So although its not using sequence data (its using methylation data) -

The Stem Cell Population of the Human Colon Crypt: Analysis via Methylation Patterns - http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1808490/

So by assuming that what you sample is the result of a branching process, you can begin to infer information about the parameters of that process and it relates to the underlying biology - number of stem cells, etc. I think this is a neat approach to looking at this, given a model and data trying to infer the details of the underlying process that lead to the observed tumour.

Maybe not what you were looking for, but I think it might be interesting for you nonetheless.

[user1202664]

This seems interesting and I'm going to look further into this.Only downside for me is that it will be difficult to go back to samples and do methylation assays.

Answer 2

[Zhidkov]

Actually Navin N., used the phylogenetic approach in characterization of tumor cells in breast cancer:[?] 1) Navin N, 2011[?] 2) Navin N, 2010 [?]

Ilia

[user1202664]

Navin's approach is interesting and possibly ideal. However, it is an intratumor approach that relies on multiple samples from a single tumor. I guess I was wondering how you could do it with a single sample from each tumor.