I Shared my Flesh with Thinking Cancer.

Some of you may remember that line.  It’s from “The Things“, that unabashed piece of fan-fic I wrote a while back (and which seems to be getting way more love than I was ever expecting). A few of you may even remember the research that inspired it — I  blogged about it a few years back, rambled on about somatic evolution and intracellular competition.  I mused about evolution in a world where Darwinian processes never really went away even at the intraorganismal level, where even within the organism it was Every Cell for Itself. I called it Cancer, For the Greater Good, and I wondered if some far-off world that played by those rules might have given rise to the shapeshifter from Carpenter’s movie.

Maybe not so far-off, as it turns out. If there’s anything to this paper by Davies & Lineweaver, I might as well have been talking about Earth — because if these guys are right, we are all cancer.

I don’t mean this proximally. I mean it in the sense that all birds are dinosaurs — because according to Davies & Lineweaver, cancer (more precisely, “tumor-like neoplasms”) is the common ancestor of all animal life. Every malignant lump on your breast, every metastatic colony proliferating through your marrow, is just a rebooted revisitation of your grandmother a million times removed.

Yo' Mama

The basic idea’s petty straightforward. Natural selection reaches into every corner of the biosphere, you see; and a billion years ago that meant every cell for itself because unicellular life was the only game in town. A mere six hundred million years back, though, all that had changed. Metazoans were everywhere — cells grouped into colonies with specialized subsystems called tissues and organs —and somehow, within those colonies, the whole beat-the-competition thing had fallen out of favor. Cells worked together, now; hell, red blood cells even gave up their nuclei for the good of the organism, which really puts the kibosh on any future solo career. I think it had something to do with inclusive fitness.

In between, presumably, there was something halfway between Cuba and the US, some intermediate form between everyone for themselves and everyone for the state. Some kind of loose affiliation of cells which valued their individual freedom, but were not above at least some level of cooperation. Modern-day sponges might be a pretty good example: some cellular specialization, a bit of the ol’ helping hand between cells, but nothing so altruistic as an actual tissue. Call it “Metazoa 1.0”. Davies and Lineweaver do.

According to D&L, that old 1.0 operating system is still sleeping down there in our genetic code; it’s just been turned off by the more recent regulatory genes of Metazoa 2.0. It hasn’t been eradicated outright, because a lot of those ancient genes are still useful (“…the genes responsible for the cellular cooperation necessary for multicellularity are also the genes that malfunction in cancer cells.”) It’s just been — tamed, is as good a word as any. Tamed, and deactivated.

Except when something happens to one of those bits of regulatory code that keep it comatose. When some base pair flips this way instead of that, Metazoa 1.0 wakes up, its ancestral toolkit intact, ready to party like it’s One Billion Years B.P.

You might be surprised at how extensive that ancient toolkit is; I know I was. I always thought of cancer as the uncontrolled growth of undifferentiated cells, jes’ regular folks except for the fact that something had broken their Off switch. But cancer has tricks up its sleeve; cancer’s got moves. The fact that it can grow its own blood vessels to keep its deepest cells fed and watered is just the beginning. Quoting directly from the paper:

Hallmarks of cancer include the silencing of tumor suppressor genes, switching off apoptosis and anoikis (programmed death when cells detach from the extracellular matrix), switching off senescence by manufacturing enzymes to repair eroding telomeres (Hanahan and Weinberg 2000), evading the immune system by removing surface receptors, dramatically changing the viscoelastic properties of cells to facilitate motility, invasion and colonization (Butcher et al 2009), secreting corrosive enzymes to dissolve through organ membranes, thus permitting the cells to enter the blood and lymphatic circulatory systems and spread around the body, thriving in hypoxic conditions by switching off the normal oxidation—phosphorylation metabolism of healthy cells and using the glycolytic cycle instead (the so-called Warburg effect; see Warburg (1956)), tolerating the resulting low pH conditions far better than healthy cells, shielding themselves from the ‘alien cell’ alarm signals from organs they invade, manufacturing their own mitogenic signals and growth factors to make them independent of chemical replication signals (Hanahan and Weinberg 2000), altering the physical and chemical properties of the extracellular matrix and other host tissues to optimize tumor growth and survival, and accelerating genetic instability to evolve immunity in changing conditions, while rapidly adapting the cytoskeleton dynamics to enable mitosis to operate across a range of karyotypes, including fullblown aneuploidy.

The usual explanation for all these tricks is that every new iteration of cancer reinvents the wheel; after all, it’s cancer, right? It has a high mutation rate, it replicates fast. Put those two things together and it can evolve new traits way faster than baseline cells. But wait a second, Davies & Lineweaver ask:   aren’t 99% of all mutations supposed to be deleterious? (Actually, no.  Last I heard, we’d decided the majority of mutations were actually neutral. ) They also point out that advanced cancer cells have grossly deformed nuclei, chaotic chromatin restructuring, any number of structural changes that would spell certain death in normal cells — and yet these fuckers just don’t die. That’s kind of what makes them such fuckers in the first place.

I’m no expert on this stuff. But we seem to have a choice here between “Internal Darwinism” (each new cancer reinvents the wheel, just happening to stumble upon a similar sequence of beneficial mutations) and “Rebooted Atavism” (all those traits are part of a dormant ancestral toolkit in every cell, and the only difference between “cancerous” and “normal” is that something broke the switch that put it to sleep). If anyone still cares about parsimony, I know which option my semilayperson’s brain is edging toward.

Of course, it’s easy to be led astray when you’re not an expert in the field. Maybe Davies & Lineweaver misrepresented the facts (at the very least, they presented them in the light most favorable to Metazoa 1.0). The paper itself doesn’t present any new findings, it merely presents a new way of looking at old ones. Expert reaction seems to be mixed, from hmmm, could be something here to Cool metaphor, but don’t push it.

The authors have, at least, put forth a number of testable predictions of the Atavism Model: that “The complexity of cancer in a species should reflect the number of cell types in that species”, for example, and that “the progression of cancer should reflect reverse phylogenetic history”. (They also make a couple of predictions that could prove highly significant in the treatment of cancer.) I await more expert opinion on whether those predictions, even if borne out, would actually support the model (seems to me that a greater number of cell types might just provide more room for mutations of any kind, for example).

But the great thing about being a science fiction writer is that I don’t really have to wait if I don’t want to. Here is an idea, peer-reviewed and legitimately published, thrown into discourse: We are all descended from Cancer. We are borne of the Holy Tumor. Isn’t that a thought. Doesn’t that get your mind going: to the imagination of ancient habitats, somewhere on this planet or within it. To isolated refugia, cut off from the rest of the world when stromatolites were still young, where 2.0 never happened and the cancerous Metazoan prototype was free to chart its own evolutionary course through a billion years.

Maybe those shapeshifters have been here all along.

Maybe all we have to do is roll away the stone.


Image credit: eMedicine 2009.

This entry was posted on Thursday, March 17th, 2011 at 5:39 pm and is filed under evolution, extraterrestrial life, just putting it out there.... You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

27 Responses to “I Shared my Flesh with Thinking Cancer.”

  1. David

    The link to “The Things” is broken (but easily fixed).

  2. Nestor


    How about harnessing it? I’d quite like to be a shapeshifter, me.

  3. Dirk Flinthart

    You’re right. There’s a wealth of cool waiting to be exploited there.

  4. Daniel

    Makes my family history of succumbing to multiple myeloma downright poetic. You should read up on a series called Parasite Eve, all about mitochondria run amok.

  5. David

    I feel I should mention that “The Things” is a *great* story. Love it.

    Hmm. Now I feel I should also mention that I went back and read it *before* you changed the link, and that I’m not such a helpless worm that I couldn’t figure out what it was supposed to be.

  6. socialnotworking: We are all descended from cancer.

    […] https://www.rifters.com/crawl/?p=1896   Contact/Feedback · Colophon · Labs […]

  7. Ben

    Can’t read the paper on this phone – do they mention canine transmissible venereal tumour? Because there’s a 6,000 year old cell line that can jump from host to host, suppressing immune defences well enough to prosper in alien flesh. It has to be the poster child for Metazoa 1.0.

    It even has the Thing’s penchant for hiding in dogs. But it’s more of a body-snatcher than a shape shifter.

  8. SpeakerToManagers

    Hmmm … suppose one some planet there were two different and incompatible versions of Metazoa 2.0 that developed independently at about the same time. Say they were separated (maybe on continents that were not connected for a long period of time, maybe because of slower continental drift than we’re used to) and diverged from each other over hundreds of millions of years. Now they are connected again, and it turns out that one of them is capable of asserting itself over Metazoa 1.0. So it can turn the other version into itself if it finds a cancer in it. What kind of competition might develop? And what kinds of chimeras of the two 2.0 versions might occur?

  9. Daniel

    Ahaha, that owns. I’ve always been bugged by particular tissue groups growing characteristic tumours with characteristic mutations, translocations, etc, and personally suspected some kind of epigenetic role, where a particular pattern of gene expression makes that particular cell type susceptible to some particular mutations. What a beautiful model!

    My main issue is that they’re already drawing a sharp line between multis and unis. After all, alot of unicellulars already use co-operative strategies, like bacterias using quorum sensing, biofilms, crowding out environment-crashing organisms like C. difficile, etc.

  10. Mirik

    Sounds like The Flood (with infection forms, spreadpods and cancerous mutants) creature/apocalyptic threat, which I suppose is a The Thing rip-off in itself, from the Halo series and gives that game some unsuspected and unintended depth.

    It’s also slightly creepy, sir. Well done! 🙂

  11. scilicious

    Interesting premise, I wonder if that puts a new spin on teratoma, too?

    I’ll have to read the paper later, no time now, but I’m hardly an expert myself. However, it does make sense to certain extents. However, how much what cancer does, what its moves are as you put is, could also be seen as a…well, response to evolutionary pressure? (Thinking mostly removal of receptors for leading the immune system astray.)

    Will ponder this a little at a later point.

  12. ScottC

    I sense a remake of Inherit the Wind coming up, with “tumor-like neoplasms” taking the role of ideological boogeyman from chimpanzees. Who’s our modern Spencer Tracy analog? Hanks?

    Thanks for this, Peter. I’m tucking this little tidbit away for Easter, just to irritate the Intelligent Design faction of my family.

  13. Vellum

    “In between, presumably, there was something halfway between Cuba and the US, some intermediate form between everyone for themselves and everyone for the state. Some kind of loose affiliation of cells which valued their individual freedom, but were not above at least some level of cooperation.”

    So, Canada, then?

  14. AngusM

    For some reason, I keep thinking of Agent Smith’s line from “The Matrix”: “Human beings are a disease, a cancer of this planet. You are a plague, and we … are the cure.” But that’s a digression.

    On one level, the problem with cancer is that it’s a dead end: it kills its host, and then it’s Game Over. Not much scope for refinement by natural selection there.

    But the interesting thing is that, as you say, it’s got moves. The examples that you give make it sound as if it’s not just dropping back to an earlier state and using the tricks it learned a billion years ago. It’s co-opting the tools of more advanced organisms to enable it to thrive in an environment that is surely very different from the one it started out in. Take the blood supply trick: when Metazoa 1.0 was the hip new kid on the block, you didn’t have veins and red blood cells that could be tweaked to its advantage.

    It sounds to me like there may be a middle ground between Internal Darwinism and Rebooted Atavism. Cancer isn’t starting from zero each time: it has a toolkit that it can make use of, but that toolkit isn’t specialized to a particular organism or organ; it is something rather lower-level. The cancer cells that can’t exploit their environment (parts of living creatures) simply wither and die; the ones where a favorable mutation or configuration allows them to deploy their toolkit in a helpful way will thrive. The precise form of the cancer will be shaped by epiphenomena – the characteristics of the organs where it has to make its living – which may explain why we have recognizable types of cancer.

    Of course, it’s all for nothing in the end: cancer can’t help doing what cancer does, but on the macro level, it always ends the same way for cancer (pace Henrietta Lacks). It can’t improve the toolkit. Until the day that cancer learns to exchange genetic material between hosts …

  15. gawp

    I scanned the paper, will read in more detail later; very nice. Love that kind of Big Idea.

    I saw a presentation by Stuart Kauffman at a Systems Biology conference a few years ago; he presented some ideas on cancer that mesh rather nicely with those in the paper.

    After hours of stale PowerPoint talks, he stood at the podium and freestyled with no slides or notes for an hour. Gripping stuff. At the end he could have led the room into revolution.

    The basic premise (hope I’m getting this right) was that gene networks are state machines; cancer occurs when the gene network falls into patterns (local minima) due to epigenetic effects or mutations that he terms “unused cell types” that involve rapid proliferation. There are likely other states like this, but if they’re not proliferating they aren’t likely to cause problems. This model suggests that it should be possible to smack cancer cells back into a non-damaging state by perturbing their gene networks appropriately. He was looking for a grant to do a bunch of high-throughput screening of the effect of thousands of compounds on different cancer types.

    I don’t think he posited *why* these cancerous states are there; in a network of 20,000+ genes interacting there are bound to be all sorts of maladaptive states you can fall into.

    If the ancestral architecture of the networks was originally for such a state as suggested in the Davies and Lineweaver paper, that would be a nice explanation, and may suggest treatments.

  16. rm3154


    > Until the day that cancer learns to exchange genetic material between hosts …

    Devil facial tumour disease [http://en.wikipedia.org/wiki/Devil_facial_tumour_disease]

    Devil facial tumour disease (DFTD) is an aggressive non-viral transmissible parasitic cancer—which likely originated in Schwann cells—that affects Tasmanian Devils.[1][2][3][4] The first “official case” was described in 1996, in Australia. In the subsequent decade the disease ravaged Tasmania’s wild devils, with estimates of decline ranging from 20% to as much as a 50% of the devil population, across over 65% of the state.[5][6] Affected high-density populations suffer up to 100% mortality 12-18 months.[7] The disease has mainly been concentrated in Tasmania’s eastern half. Visible signs of DFTD begin with lesions and lumps around the mouth. These develop into cancerous tumours that may spread from the face to the entire body. The tumours interfere with feeding, and the affected animal may starve to death. At present the population has dwindled 70% since 1996. Numbers as of 2010 show an 80% rate of infection throughout the population. Six females have been found with a partial immunity. Breeding in captivity has begun to try and save the population.[8]

  17. Mike Tevee

    Interesting hypothesis.

    Why do the more lurid images of “The Andromeda Strain” keep
    flying through my head?

    On the topic of Method to its Madness (smart cancers), or challenges to Darwin, John Cairns at Harvard had some evidence that advantageous mutations were more frequent than others, flipping the idea of Random Mutation on its head. They told him his reputation was on the line, but he would not back down and kept getting the same result. Turns out there was a subtle flaw in his experiment, but you gotta admire his courage.

  18. Nix

    (I hope you support blockquotes or this is going to look disgusting)

    You said:

    But wait a second, Davies & Lineweaver ask: aren’t 99% of all mutations supposed to be deleterious? (Actually, no. Last I heard, we’d decided the majority of mutations were actually neutral. )

    Davies and Lineweaver were talking about mutations with phenotypic effect, obviously. Synonymous mutations, mutations not affecting protein shape nor active site, and so on, are excluded by implication. And the majority of such mutations are deleterious.

    They also point out that advanced cancer cells have grossly deformed nuclei, chaotic chromatin restructuring, any number of structural changes that would spell certain death in normal cells — and yet these fuckers just don’t die. That’s kind of what makes them such fuckers in the first place.

    Well, yes, exactly, because the genes that would have killed them off are broken, as are the genes that would signal that the cell was misbehaving so that it could be phagocytosed.

    Where I have a big problem with this hypothesis is that it’s evolutionary nonsense. Essentially all a tumour’s way-cool lethal bags of tricks are derived from things other cells in the body do (many are digestive tricks or foetal tricks: after all, foetuses are *supposed* to trigger angiogenesis and grow like mad, and they have a lot of genes that get switched off right after that: they’re meant to stay off, but if they don’t… hell, one way tumours get so genetically shagged is that after they’ve broken the genes that repair DNA and keep chromosomes in shape, alus and other retrotransposons go crazy and start replicating like mad across the genome, landing in other genes and breaking them… and, sometimes, the SINE and LINE-derived machinery which B cells use to churn out such a huge number of antibodies kicks in and starts doing its business on the entire genome. Super-accelerated mutation here we come! Combine that with constant replication, and, y’know, it doesn’t matter if most of the mutations are deleterious: only a tiny fraction need to work for cancer to spread. That’s selection for you.)

    The idea of ‘reverting to the way things used to be’ is about as flawed as the stuff in Greg Bear’s _Darwin’s Radio_ about nifty new tricks bubbling under the surface then erupting out as cool! new! effective! traits! — namely, that things which are not expressed in the phenotype drift to uselessness in a very short period. Sure, a lot of ancestral toolkit is still useful — but if that ancient stuff goes wild when regulatory regions are broken, that doesn’t mean our ancestors were cancer: it just means that a toolkit which happens to be ancient goes mad when regulatory regions are broken. Regulatory regions evolve fast: in ancient days, there were probably *different* regulators. Nothing will survive if it is never expressed.

  19. Keith Z-G

    Hmm, perhaps it’s a stretch, but I wonder if this might fit together somehow with http://www.newscientist.com/article/dn20265-biologys-dark-matter-hints-at-fourth-domain-of-life.html ? Which, by the way, is an article that I couldn’t read through without mentally exclaiming “ßehemoth!” every few sentences.

  20. Krazy Kimchi Fan

    Nestor?! What are you doing here?! Why did you let that lazy comic artist of yours run away?

    Oh, and Watts, you made me very worried for a moment. I was thinking you’ve got cancer right before I started reading the post.

    “… just a rebooted revisitation of your grandmother a million times removed.”
    that line reminded me of Fat which I think you’ll enjoy reading:

  21. Nestor

    If you love them, you have to let them go. If they come back, then you can shackle them to a drawing table again. Which Krazy Kimchi fan are you? There were at least a dozen, back in the day.

    All this talk reminds me of a short story I read once, where the cemetery adjacent to a cancer ward was full of strange meaty plantlike growths, the cancers from the buried patients that had adapted and survived beyond their hosts…

  22. David A

    Peter, I trust you are familiar with slime moulds? Surely they are the perfect “Cananda” in between single-cellular and collaborative multi-cellular. The are single cellular organisms when food is abundant, but come together as a single organism when it is scarce, some cells even sacrifice themselves to form stalks for spores in emergencies. Awesome stuff.

    Lots of coverage on Ars lately, including the marvelous modelling of Tokyo’s subway system (http://arstechnica.com/science/news/2010/01/amoeboid-designs-complex-transportation-network-eats-oats.ars) and the revelation that they bring their own bacteria farms when migrating to new pastures (http://arstechnica.com/science/news/2011/01/slime-mold-macdonald-farms-its-bacterial-meals.ars)

  23. Krazy Kimchi Fan

    I used to go by the nickname “cow_2001”.

  24. Tim

    Hmmmm. Looks chewy. Perhaps it would benefit from honey-mustard sauce.

  25. Nestor

    Oh yeah, Cow_2001, I remember you. We still hang around on the site, mostly using it as a link dump and to talk about cartoons, feel free to drop by.

    (Apologies to our host for using his comment thread as a chat)

  26. Tertiary Matt

    Just read the story…

    I like to thing that the bits the Traveler lost when the crash happened ended up becoming Shoggoths. Because, well… why not?

    Hell of an interesting paper, I gotta say.

  27. Brycemeister

    Much more fascinating to my mind is-what kind of an environment would be necessary for a shapeshifting lifeform? What kind of bizarre environmental pressure would possibly cause the choice of nearly instant flexibility? And how would that shapeshifting be achieved?

    I’m not a biologist, but the stability of lifeforms seems to be because that’s the best way to preserve genetic information, and because outside of weather conditions, the environment itself is relatively stable over the long run. As well, something that had to evolve to meet, say, wildly changing environments might not even get a start, because of said shifting environments-and as well, species extinction might result if the environment changes too rapidly.

    The easy solution, of course, is the ol’ advanced civilization. Maybe the Thing was some extraterrestrials’ idea of a nifty biological probe (like Pathfinder, only alive and yuckier) or possibly some final solution to some war-the frankenstein monster thing, and they made it too well, and it ate the planet.

    I’ve thought about that-the Thing seems to absorb genetic material and biomass. So maybe it’s home planet might be covered in many feet of writhing, constantly shifting whatsits…eesh, that’s a nasty image…