It does seem extremely elegant relative to the very tiny amount of resources a single cancer cell has to work with.
I'm not sure why there was so much talk about 'nanomachines' some years ago when there are literally millions of pre-existing examples that are so vastly more sophisticated.
> Oh, I agree, imagine a "grey goo" apocalypse, where uncontrolled and hyper-aggressive nanotechnology spreads everywhere in a unceasing orgy of consumption and replication. Mutating according to its own unknowable logic, it eventually it covers the land in an inhospitable blanket of massive crawling megastructures and inscrutable malign hive-minds!
This line of thinking about humans is exceptionally noxious when it gives a pass to AI or nanobots or anything other than human that might exhibit the real behavior that humans are hyperbolically accused of. That being said, the same logic applies: Evolutionary advantages lead to outcomes that don't get a fuck about what you or I want. So far, we have no examples of monocultural life gaining a permanent advantage over whatever evolves next, at least not on a geological scale. But my response to this is always some variation of: If you hate people so much, why do you have kids?
I never know that for sure, but it's a safe guess. It's become a bit of a hilarious trope here in Portland. The people who are always complaining about the rest of society's wanton disregard for the environment are the ones who aren't riding the bus. Their kids are the reason they're so worried about climate change, which is caused by overpopulation, you see. Nothing racist or Malthusian. They're shuttling their kids to soccer practice in a Subaru, how much more eco-conscious could they get? And when dad puts the kids down, he gets online and complains about how everyone else's consumption is ruining the world.
When you talk to other people who don't have kids, they always make sure to let you know they don't.
IMO, “Single cancer cell” isn’t really a the unit. Solid tumors act more like a mini organ with different types of cells having roles: driving blood supply, immune suppression, or enabling and executing metastasis are examples.
Cancer isn’t sentient, but life finds a way.
Fu*k cancer.
I am not an oncologist but I thought that metastasis was more related to a tumor growing too close to a major blood highway. If some of those cells that like to grow out of control get into that system they end up getting lodged all over the place. It seems less like "enabling and executing" and more like it just grows and grows until it leaks out. The ability for tumor cells to suppress the immune system however is fascinating. It's almost like an alien.
Cancers actually drive blood vessel growth into the tumors to feed themselves. I’m sure they leverage that as part of dissemination.
Different primary tumors (eg Prostate, breast, lung) have preferred sets of metastasis sites that are not always explained by where they get stuck due to getting into the blood supply. Even getting into the blood supply requires functions of degrading extra cellular matrix, evading immune surveillance, getting out of blood vessels and “setting up camp” at the news destination, and those requirements can be cancer and/or met site preferential.
Much of the immune suppression mechanisms are actually regulated systems normal cells use to prevent runaway inflammation. Cancer finds that function and highjacks it to “hide”. Conversely (to further illustrate complexity), some cancers use aspects inflammation as ”fuel” to drive tumor growth especially early in the cancer.
I do agree- it is all truly fascinating.with that said: Fu*k cancer.
What I’ve never understood is how all this complex cooperative behavior emerges out of a relatively small number of mutations. I get that these are all repurposing of existing mechanisms, but still the overall coordination seems so much more purposeful than something that would happen with a few initial mutations and then some further selection. Particularly things like metastasis, which sounds like a many-stage process that would be hard to spontaneously evolve.
Yeah, it kind of makes you think that maybe some sort of cancer blob is the default state of cells and that everything that goes into making a collection of cells into a particular species is what's holding cancer back.
> but still the overall coordination seems so much more purposeful than something that would happen with a few initial mutations and then some further selection.
Isn't this just selection bias on a grand scale? You're looking at a specimen where the parts happened to end up arranged in a way that allows it to exhibit these properties that seem like coordination, while not paying much attention to the quintillions of specimens that did not.
It is more than just a few mutations. A good number of the common earlyish driver mutations are in the systems that detect and repair damage to DNA, which then accelerates mutation accumulation. (At some cost to stability of that lineage, but overall there are obliviously formulas that work)
And I’d also point out that these functions that are “sought after” by cancer are not all de novo, some things are complex functions that encoded for use by other normal cells at a other times and places in development of the organism. Some key mutations that enable the use of those functions by the cancer are crucial moments cancer development. Also many key mutations are loss of functions, which are often easier to accumulate. A example of genetic Losses of function is losing “contact inhibition” where normal cells stop growing when they get neighbors. Cancer cells loose this inhibition and just continue growing. This alone would require a suite of mutations.
Another comment in the thread talks about cancer evolution and the good of the organism not making sense. Cancer doesn’t care about the organism. The organism has evolved many layers of checks and balances that can work for decades, but throw in time and poisons like smoking, UV damage, and there will be loopholes found. Cancer is persistent and has the entire genome of functions to turn on and off in various ways. Seriously Fu*k cancer.
Human body and immune cells and cancer cells all are complex.
It is clear that for blind force , unconscious nature, and aimless chance could in no way interfere in such percipient, delicate art, and conscious subtle wisdom, and perfect providential balance; they could not be their work. It is impossible.
It is also utterly impossible that living creatures made themselves for then each of their particles would have to possess comprehensive knowledge and wisdom like a god, to be able to know , see and make all the parts of their bodies and form it , indeed , it would have to know ,see , and make everything in the world connected to it.
"Whoever created the mosquito's eye ,created the sun.
Whoever ordered the flea' stomach, ordered the solar system."
from Quran's light
That would imply that from a biological stand point it should have had a self serving purpose. But it kills the host. So what gives? Where is the logic?
It has the same purpose as us, just that while our "world" is the Earth (or whatever planet our species will conquer before extinction), its "world" is merely the human body it inhabits.
The death of a cancer's host body is like our equivalent of our star dying, and I guess our equivalent of interplanetary colonization would be a hypothetical cancer's ability to jump between hosts.
Neither humans nor any other known life-forms can meaningfully influence stellar evolution regards either the fate or timing of the Sun's end.
Cancer can and does bring about the death of its host.
(One might argue that both humans and other life forms have the potential and/or demonstrated capability to transform their own ecosystem beyond its viability to continue to sustain and support them, with both the present anthropogenic greenhouse-gas-based global warming and the Great Oxygenation Event and subsequent Snowball Earth serving as examples.)
That said, I understand your analogy.
There are virally- and clonally-transmitted or otherwise contagious cancers:
Cancer colonies self-sustain and self-replicate, causing local success, then kill the host without escaping it, causing colony failure and do not successfully become a new species of independent pathogen. Local optimization causes disaster at a larger scale, causing the system to collapse.
Nothing intelligent is happening. Zero-foresight self-amplifying systems that will inevitably crash out will simply amplify themselves to that point, then cease. Applications to areas outside of oncology are an exercise for the reader.
This is why I think cancer is more like a lifeform than a chaotic pile of cells. Matt Levine demonstrated that applying certain electric field to a growing embryo alters its morphology without changing its dna, and since cancer is always in the rapid growth phase, it would be interesting to do a similar experiment on it.
There is a device the US FDA approved for brain tumors (Optune by Novocure)that uses alternating electrical fields. It surprised the field when it got approved as its mechanism wasn't conventional or well understood. There were also some confounding factors in the study design. That said, it's in a bunch of other tumor types now in more studies.
There are many papers - development of immunotherapy for melanoma has been a big topic for the last few years, with some fantastic improvements but more to do.
As it happens I can spot that the linked paper is already out of date as I know at least one is the trials has been cancelled, because I was on it.
Immune evasion in cancer is really interesting and particularly important because immune therapy has really emerged as a groundbreaking therapy. But, while immunotherapy excels in blood cancers, it still struggles in solid tumors. Part of the reason for this is that a solid group of cancer cells can much more easily change their microenvironment to exclude immune cells.
So this paper explores one of the ways they do that. Cancer cells turn off a protein that sits on the surface of normal human cells that immune cells regularly bind to. Binding to this protein, CD58, causes changes in the immune cell. What these changes are critically depends on the other signals that the immune cell senses on the surface of the cell. If it senses another (somewhat famous) protein, PD-L1, the immune cell self destructs, as PD-L1’s presence typically means that a cell is normal and healthy. But, cancer cells frequently evolve to over generate PD-L1, thereby creating an micro-environment filled with self destruct signals. Further, the other protein CD58 actually increases the physical force of immune cell binding, so when cancer cells turn off CD58 they actually get more slippery.
This paper explores an interesting connection between the two proteins. There’s a third protein which binds to both of them and increases the amount of time they're active on the cell’s surface. Honestly I haven’t had time to fully digest this bit, but it seems like the regulatory protein preferentially binds to CD58, which is good because it promotes immune cell adhesion and recognition of the tumor cell. But! When CD58 is lost, it then binds to PDL1 and promotes immune cell evasion.
Why nature chose this as a regulatory mechanism is kind of baffling, but exactly the kind of thing that makes cell biology interesting
Warning: no background in science, dumb question inc
It seems like the cancer cells are pretty good at evolving to get around our immune system. Is there a reason our immune system doesn’t evolve to outcompete the thing that is killing it?
I should note this is somewhat outside my normal range, so actual experts may disagree.
There may be a few reasons. First, as another commenter has noted many cancers manifest late in life, after typical reproductive age. This is decent evidence that our immune system is in fact quite effective at killing cancer already. A rough estimate is that a human body experiences 10^16 cell divisions in a lifetime. Each division is a mutation and therefore cancer opportunity; not to mention the constant barrage of DNA damaging chemicals, radiation, and viruses. The numbers for precancerous mutations repaired (or cells destroyed) over a lifetime would be totally staggering.
The flipside of this is that a death from cancer after having kids doesn't reduce the prevalence of your genes in future generations by too much. You already have progeny after all. Humans are social animals and so there's maybe some benefit to having grandma or grandpa around. All that to say that (from an evolutionary perspective) there's not a huge, obvious benefit to having a super good immune system later in life.
There's also probably a balance between having an aggressive immune system and auto-immune disease. An over-eager immune system could attack normal cells as well as cancer cells.
Yes, also the parent generation eventually dying off confers survival advantages to the younger generation: fewer old folks consuming resources and providing little value in return.
Death isn't just a critical part of the individual life -- rather, it seems key to evolution itself, biological or otherwise. clearing the slate/setting a timeout on any given system seems to be a good way to evict the accumulation of fixed points/cycles and other non-global-maxima that accumulate.
I'm completely talking out of my ass here, but I suspect systems will tend towards rent seeking/"cheating"/maintaining a status quo/really, just "overfitting" to the environment by making the ecosystem around them dependent on them.
Death ensures resiliency and progress. If we ever actually achieve some form of (pseudo-)immortality, i'd hope that we introduce an "afterlife" which removes the post-mortal from "living" society/ecosystems after s certain point. at most, living society should be able to inspect & affect the post-living system (afterlife), but the post living system should be airgapped from being able to (at least, directly) affect the living one at all. that said, i'd be wary about indirect affects. to die is to not influence. i'm still influenced by the memory/simulacrum of my grand mother who raised me until middle school. thats some degree of affect on my ecosystem, but it isnt what i would call direct.
i suppose there need not be a singular timeout or ecosystem a system is depolyed to, so long as it converges to zero. Heck, not even monotonically nor with a consistent cohort of systems.
> "afterlife" which removes the post-mortal from "living" society/ecosystems after s certain point. at most, living society should be able to inspect & affect the post-living system (afterlife), but the post living system should be airgapped from being able to (at least, directly) affect the living one at all
> at most, living society should be able to inspect & affect the post-living system (afterlife), but the post living system should be airgapped from being able to (at least, directly) affect the living one at all.
I feel like this hypothetical scenario is contradictory (and, supposing there was engineering behind both this life and an afterlife, is an argument for keeping the two isolated). Data is code, so data collected from the afterlife is potential for "remote execution from the afterlife" (not that humans are computers, but for example traditionalists who were able to directly study the long-dead would act differently than if they had to piece tradition together from scraps of documents - particularly, it would freeze traditions/memes rather than letting them evolve) and therefore death losing its role as the cessation of influence.
> I'm completely talking out of my ass here, but I suspect systems will tend towards rent seeking/"cheating"/maintaining a status quo/really, just "overfitting" to the environment by making the ecosystem around them dependent on them.
This is also why institutions should have a lifetime. Corporations, government entities, all the things that don't have a natural life length really ought to.
> fewer old folks consuming resources and providing little value in return.
As a parent living by my in-laws, I can assure you that, far from providing "little value", raising children without support from grandparents would be severely more difficult. Personally, I see generations trying to be independent of one another as a large factor in making modern childrearing increasing difficult.
And this is aside from the knowledge, and labor that would have been so vital in former times. The old woman spinning and knitting is a historical trope, but when you have to hand make all of your textiles, she becomes quite useful indeed.
Sure, but eventually your in-laws will become unable to care for children, maybe even unable to care for themselves. At that point, from a strictly utilitarian standpoint, they become a burden (I'm not making an ethical claim here, because ethics is not part of how nature/natural selection works).
> There's also probably a balance between having an aggressive immune system and auto-immune disease.
This is the key. I’ve had an immunologist explain to me (over a bottle of wine, so my recollection is hazy) that our immune system is actually very very aggressive already and tuned within a few percentage points of being problematic.
Some autoimmune diseases appear to be due pathological differences in the intestinal microbiome, typically when its flora diversity is drastically reduced. The latter is more common in first world countries, and especially among highly sanitised environments, such as overuse (unnecessary use) of antimicrobial agents and consumption of too much food which has been processed such that all microbes have been destroyed. Allergies are also a class of immune disorder, more prevalent among those who have little gut flora diversity. Unfortunately, while high diversity reduces the likelihood of developing allergies, increasing intestinal flora diversity does not seem to reduce existing allergies.
Peto's paradox https://en.wikipedia.org/wiki/Peto%27s_paradox is interesting and relevant to your question. To summarize, larger species have evolved better defences against cancer, because they have more cells, so would otherwise experience far more cancer. There's some evidence that these defences reduce fertility, so aren't selected for in smaller species.
Also, the evolution of cancer cells to get around an immune system happens on a completely different timescale to the selective pressures for the human immune system to avoid cancer (days vs decades).
Yes, and there are a few exceptions in the form of small animals which rarely get cancer such as the naked mole rat. This is an area of active research to see if their cancer defenses can be used to develop drugs for humans.
There is a tricky balance between an effective immune system and one that's too gungho and you get auto-immune problems.
Frankly it's amazing it works at all - not only does the immune system have to be able to react to stuff it's never seen before - it needs to be able to distinguish between self and non-self - and self is pretty damn complex.
So to make an analogy - cancers like a bunch of terrorists in a busy shopping centre and the immune system is the government response. Only one side has to worry about bystanders.
As a noob, this makes me wonder what happens to a body without PD-L1. Does that make the immune cells themselves exhibit cancerous growth? Does it lead to auto-immune issues?
A glance at Wikipedia confirms that some pathogens are aware of PD-L1 and like to mess with proteins that regulate the immune system. Epstein-Barr Virus (EBV), the most common human virus, the herpes that 90% of people have had and a common cause of cancer, apparently has a protein that induces expression of PD-L1. Some other viruses in mice have been observed to do it.
PD-L1 already exists in human cells, for cancers leverage it, they just need to overexpress it, which can be a minor genetic change.
Pathogens, like bacteria, don't have PD-L1 and it's a very large transmembrane protein. Rather than a small genetic change to over express it, pathogens would need to generate the entire protein by scratch. Possible, but highly unlikely.
Retroviruses don't have cellular membranes. They could over-express PD-L1 in a cell it infects, but they don't need the cell to live, they want it to lyse and release more viruses (though I guess you could argue it wants it to survive immune attack long enough to do so).
Bacteria could potentially do it, but also bacteria cell walls are very different than human. I'm not sure an expressed transmembrane protein could actually be embedded in a bacterial cell wall and still remain functional.
I'm hung up on the title's use of the word elegant, which ordinarily has a positive connotation, to describe a very sad and deadly process. When something is elegant, it means we recognize something beautiful in its form, such as an elegant gown or an elegant solution to a vexing problem.
I would rather see a word such as grotesque or monstrous used, or a phrase such as brutally efficient. That indicates a similar type of imposing magnitude, but also a lack or distortion of beauty.
Elegance is less of a positive descriptor and more of a way to describe precision of craft in this connotation. Where the evolutionary pattern of malignant melanoma is very deft at adapting to its situation.
> Ugly things can be good, and beautiful things can be bad.
This is indeed very common in the real world, and often talked about in one way or another on HN. So much so that the parent almost certainly has seen multiple examples after nearly 15000 comments and 13 years on HN.
How do cancers have sophisticated mechanisms for evading immune responses when they don’t have any adaptive pressure? The cancer dies with the victim. It seems like the equivalent of mashing on the keyboard and getting a valid program.
I’m just another guy throwing ideas, maybe the cancer should be compared to a virus or bacteria, where their life cycle is much shorter than humans, the objective is maximum growth, then it dies as any other organism
They have immense adaptive pressure at the cellular level. The evolutionary process plays out within each patient. Aberrant cells that don’t have some sort of evasion mechanism will be identified and destroyed by the immune system.
I've heard a hypothesis that cancer behavior is not primarily evolved in the patient, but instead it is an ancient mechanism from the time of transition from single cell to multi cell organism.
A human cell can never survive on its own. But there might have been a time when cells could have benefited from "pulling the escape hatch" and taking their chances instead of continuing cooperation within a damaged early multicellular organism.
They do have adaptive pressure. 100s of millions of years of single cell evolutionary instincts live within our cells and the desire to survive. Being a multicellular organism is a relatively new learned behavior, and a human cell returning to that old mindset is basically cancer.
The only problem is that these individually minded (cancer) cells have every ability of your healthy cells and have basically stopped caring about the greater whole and care about themselves. Then they evolve at a micro level for their survival to fend off chemo, immune system, radiation etc. All it takes is one adapted/surviving cell to come back strong.
The sophisticated mechanisms for evasion exist because they have all the methods of evading your immune system that healthy multicellular organisms need to function and they multiply and increase their mutation rate to try new methods to survive and thrive.
I view cancer cells as single cells to understand their behavior with the adaptations of all the healthy cells returning to their “baser instincts”.
Source: Caretaker of a cancer patient and former cell bio major
Greater level of irony is that we say things like that while steadfastly refusing to look in the mirror. Civilization as it stands isn't the most sustainable either.
https://www.youtube.com/watch?v=uoJwt9l-XhQ
reply