Cancer and Evolution

The same evolutionary science that makes sense of life's diversity also plays a critical role in cancer research.
This is a companion discussion topic for the original entry at
1 Like

Thanks for the opportunity to explain the connection between cancer and evolution.

I’m happy answer questions here.

Great article. As we learn that cancer cells arise and disappear, it is a little unnerving to realize our immune systems may well be a major force in selecting which cancers grow and thrive, as well as what keeps them in check. Lots of fertile ground for work.

Great article, Joshua. I noticed the other day that Peter Nowell just died. Dr. Nowell wrote a seminal paper in 1976, which was one of the first to present the cancer process in evolutionary terms.

I dont mention it often here, but I worked in cancer research (focusing on chemical and physical carcinogenesis) for the first 20 years of my research career. Among the model systems I worked on was radiation induced skin tumors in rats. We found evidence for the kind of molecular evolution in tumors that you mentioned in serial biopsies of growing tumors. The following is from the paper in Cancer Research (Garte SJ. Burns FJ. Ashkenazi-Kimmel T. Felber M. Sawey MJ. Amplification of the c-myc oncogene during progression of radiation-induced rat skin tumors. Cancer Research. 50(10):3073-7, 1990)

“Altogether, 11 tumors were biopsied 2 to 5 times each. In most cases, c-myc gene copy
number increased in the later biopsies, compared to earlier ones.”

In other words, we could actually see an increase in the copy number (by amplification) of an important oncogene called c-myc, as tumors grew in size and malignancy. At the time, I did think of this (as was by then a standard model) as very fast cellular evolution, but I wasnt very interested in evolution per se. My mistake. More recently Susan Rosenberg has also linked the behavior of DNA repair processes that lead to an enhanced mutation rate to the evolutionary significance of stress directed mutations.

I would love to hear more about your own theoretical work in this area. Best.

1 Like

Look how HeLa cell lines have changed over the generations…

Indeed. Actually most tumor cell lines continue to evolve in culture. Among these new phenotypes is the ability to grow on plastic, which is very difficult to get normal cells to do (speaking from experience). Then as time goes on, they acquire many more novel phenotypes, all of which are related to a very high mutation frequency, and extremely strong selection pressure. That is how tumor cells eventually become invasive, malignant, and drug resistant.

I thought that this was a great article! Thanks for posting it.

I always enjoy the insights of the Rev. John Polkinghorne, since he gives us a good perspective on how we deal with cancer. In his St.Edmunds Lecture (2002) Polkinghorne said, ‘The same cellular processes that have driven the fruitful history of evolution through genetic mutation, must necessarily allow other cells to mutate and become malignant. The anguishing fact that there is cancer in creation is not gratuitous, something that a more compassionate or competent Creator could easily have remedied.’

Then on Polkinghorne’s web site, somebody asked, “Could God not have guided evolution so that we have bodies which attempt to kill off cells which mutate and become malignant?”

And Polkinghorne/s former student, (Nicholas Beale) replied, “Well He has /we do - there is a wonderful and elaborate immune system whose mysteries we are only just beginning to fathom, (see eg - but the immune system is not infalliable, and we develop cancer and other diseases when the immune response is insufficient to prevent the disease. I think the fundamental reason the immune system is not infalliable is that it is built with stochastic processes and also that if the immune responses are too strong then (a) you get auto-immune diseases and (b) the energy used is excessive. John’s fundamental point - that it is the same molecular processes - of course remains.”

1 Like

So you’re shooting a moving target!

Our friend @Paul_Nelson pointed out this article to me. It provides a counterpoint to my article, arguing that evolution does not work well to explain cancer.

Of course, there are differences between the evolution of cancer, and the evolution of the species. I haven’t argued that they are identical, but that there is a great deal in common. One substantive difference not included in my article is that cancer does not reproduce sexually, which substantially changes the population genetics analysis. There are more differences too. This article points out some of them.

I think my point still stands though. Evolutionary theory is very useful in understanding cancer, whether or not we agree with the full evolutionary story.

Curious your thoughts. I am happy to respond to specific points you find in this article.

It is really amazing how much was figured out during this time. It was in the early 1970s that the evolutionary nature of cancer became clear. Before then it was really an open question. With that in mind, the fact that Armitage-Doll was able to calculate the number of mutations required to produce cancer in 1954 is truly amazing:

@beaglelady thanks for the addition. That was really a great addition.

Thanks for posting this link. I had not heard of this paper, possibly for good reason. I found it very unconvincing as an argument against evolutionary processes in cancer. The most charitable interpretation I can make is that the author suggests that somatic cancer cells do not behave exactly like populations of organisms when it comes to certain features of Darwinian mechanisms. This is self evident. He admits that natural selection and adaptation are present, but argues against a cumulative evolution in tumors to explain certain malignant adaptations like resistance to therapy. He says as part of his argument, that in many cases, the mutations that allow for drug resistance were already present in the normal cells. I find this to be a very weak argument. The same can be said for evolutionary processes. As Benkirk has often pointed out in this forum, most selection occurs due to pre existing mutations. Furthermore, the absolute necessity for cumulative evolutionary pathways is doubtful. Examples of rapid evolution are now commonly known.

Of course, the main point is that nobody argues that cancer actually IS Darwinian evolution. And the ways in which the author shows differences are either not convincing or trivial. Despite Paul Nelson’s embrace of this paper, which he probably believes deals a blow to the value of evolutionary thinking in cancer biology, I doubt very much that this paper would have any impact on the standard paradigm of viewing carcinogenesis and cancer treatment as largely connected to the Darwinian concepts of natural selection, adaptation and genomic instability.


Good news. It looks like ENV might be posting a response soon. Looking forward to it, and I’ll post a link here when it is up.

1 Like

Evolution is Not a Scientific concept - i.e. not verifiable as it lacks a clear, testable definition. If and only when this will ever be resolved, we will be able to compare Cancer to Evolution and evaluate a possible match.

Until then, a simple observation is that Cancer kills itself by killing the host (if we even want to go as far as assigning a self to Cancer). Transmission in a few cases is due to the incidental and preexisting behavior of the host which has nothing to do with the Cancer itself. If Cancer had modified the behavior of the host, then the resemblance might be somewhat more believable.

1 Like

@NonlinOrg welcome to the forums!

Good news, the modern understanding of evolution actually includes a fully developed and coherent mathematical theory. So we can compare cancer and evolution to see if it is a match. And it is.

To be clear here, the “behavior” I am referring to is that dogs are sexually promiscuous and Tasmanian devils like to bite each other. That is it. Nothing really dramatic here, and it is just a way of pointing out that the evolution of parasitic tumors is dependent on the environment which is shaped by the behavior of its hosts.

Evolution often (usually?) depends on incidental and preexisting environmental conditions. So this is one way that these examples actually match evolution. A great example where we see good concordance between evolution and cancer, so thanks for bringing it up!

Any how, we would love to hear more about you. What is your position on origins? Are you a young or old earther? Are you a Christian, like me, that believes that God created everything? Do you hold to Ken Ham-like de novo creation or the progressive creation of Hugh Ross? Who are the thinkers here that you most agree with?

Before I get into extended debates with people, I find it best to understand our baseline positions and identify some common ground. Let’s start with that.

I imagine this blog is yours? There are some well written posts there. Looking forward to seeing you write on there about cancer and evolution =).

And once again, welcome!


So I found this blog post of @NonlinOrg (I think its you) entitled “Evolution is Creation.”

I disagree with some of the terminology (but that is a technical point for later) but I have to say it is really beautifully written. And I do agree with a lot. I thought I would excerpt the key passage for me:

As an instrument of creation, a version of Evolution may be valid. One has to simply consider how a Creator would use the mechanics of Evolution and Abiogenesis as Creation tools. The automobile is clearly a Creation, yet it is shaped by the same Evolutionary principles observed by Darwin: each model could cover the Earth were it not for competing designs; a model is reproduced with variability; divergence of character helps models spread into all available environments; the better features are retained and passed to the next generation, while natural selection ensures the extinction of the uncompetitive designs; totally different “species” like the passenger car and the interstate truck display convergent evolution; etc. Unassisted growth and reproduction are of course missing because the automobile is only a human Creation. In turn, Abiogenesis is very much in line with Abrahamic Religions as mentioned in several faith writings.

You are describing here a “valid” “version of Evolution”, if it is an “instrument of Creation”. This is exactly the BioLogos position, and mine as well. As they write in their belief statement:

We believe that the diversity and interrelation of all life on earth are best explained by the God-ordained process of evolution with common descent. Thus, evolution is not in opposition to God, but a means by which God providentially achieves his purposes. Therefore, we reject ideologies that claim that evolution is a purposeless process or that evolution replaces God.
The Work of BioLogos - BioLogos

Any how, whether or not you personally adopt a position in agreement with our theistic evolution, your articulate defense of our position does give us some real common ground. That is a great starting point.

1 Like


Sir, I don’t believe you have a fair approach to developing “verifiable” definitions.

For example: a TESTABLE hypothesis is:

“If Natural Selection is what determines which gene pools can thrive and expand into available environmental niches, then when the Dinosaur-Killing arrived, we should expect to see an expansion of the surviving animals into niches previously occupied by Dinosaurs.”

If we look into the fossil record we see that where there had once been marine dinosaurs, that ate up most any fleshy object of significant size, we should find that mammalian forms emerge to fill the aquatic niches formerly held by reptiles.

And this is EXACTLY what the fossil record shows.

There are literally HUNDREDS of hypotheses like this, that are predicted and then found to be correctly predicted.

So if you want to be treated with credibility in these discussions, you are going to have to acknowledge what thousands of researchers have already discovered: evolution can be tested in hundreds of ways… .it is VERY testable.

1 Like

59 posts were split to a new topic: Are there problems with the evolutionary scenario?

This topic was automatically closed 6 days after the last reply. New replies are no longer allowed.