Hacker News new | past | comments | ask | show | jobs | submit login
Elephants Rarely Get Cancer, Now We Know Why (2015) (discovermagazine.com)
153 points by Osiris30 3 days ago | hide | past | web | favorite | 85 comments

Didn't early humans also rarely get cancer?

My understanding is that cancer is one of those things you die from if you don't die at child birth, get eaten by predators, or get the flu, polio, malaria, the plague, etc.

That is, the reason cancer is such a big deal in modern times is because other causes of death have been controlled.

I don't know much about elephants but I imagine cancer is still low on their list of things to worry about (humans are probably a lot higher up) If they started living a lot longer, it would probably show up more.

People under 40-50 also rarely get cancer. Although I'm sure the human environment has produced some more causes of it that make those rates higher than they used to be.

From what I understand, there's only a limited number of times you can perform cellular division without getting erroneous DNA copies (cancerous cells).

Like humans, elephants start their life with one cell, which is then replicated over and over again until the end of their lives. But somehow, elephants manage to reach body sizes 100 times bigger than humans (meaning many more cellular divisions) without having cancerous cells.

I think that's the basic observation that triggers this kind of research.

Disclaimer: I am no expert at all.

Our gamete cells have been undergoing an unlimited number of cellular divisions since... well, since life started. What's so special about somatic cells that makes them cancerous after so many divisions?

> What's so special about somatic cells that makes them cancerous after so many divisions?

Nothing, actually. I'd rephrase the parent comment.

There is in fact a fixed (somewhat small) probability that replication ends up in a mutation. But looking at the continuous replication going on in our cells as a Bernoulli process, you can find a number N such that there is a probability of Q (say 99%) that a mutation happened before the Nth replication.

In animals, the gamete cells are selected at an early time in the embryo, and the don't divide as much times as the other cells. (Most os the cells of the testicles / ovary are just normal cells that provide support for the germinal cells.) Reducing the number of divisions reduce the number of mutations between generations. https://en.wikipedia.org/wiki/Germ_cell

Also when a cell becomes cancerous it usually has a broken version of many of the checks to avoid uncontrolled division and to ensure that the cell cooperates with the neighbor cells. This will probably cause problems in the formation of the embryo or fetus. Or kill the baby before reproduction age. So cancerous germ cells are self eliminated from the gene pool.

It's actually that they aren't special. With gametes the body cares about making sure things work out well, and you only need the one embryonic stem cell to be issue free (discarding any with mutations) to create a new healthy human. With the somatic cells though you just need one (oversimplifying massively here of course) cell to become cancerous to kill the person, and the DNA repair mechanisms become more likely to fail throughout life

Someone else basically said it but to put it more plainly: broken gametes produce no-fertilization, no-implantation, miscarriages, and non-viable or infertile offspring. So only the good ones get through.

Prehistoric human lifetime is very hard to estimate, but I have heard reports that while infant mortality was high, actual lifetimes could still reach 70+ years

It's true that life expectancy used to be very low due to very high infant mortality, but even after that it wasn't easy going:

> Based on Neolithic and Bronze Age data, the total life expectancy at 15 would not exceed 34 years. Based on the data from modern hunter-gatherer populations, it is estimated that at 15, life expectancy was an additional 39 years (total 54), with a 0.60 probability of reaching 15.


That's in the times when agriculture became a thing. It meant that people lived in a very crowded way where the limited tracts of very fertile land was, settled and with very poor sanitation. Infections were rampant, and famines in poor crop years, common. Hunter-gatherers had it much better - except there were a lot fewer of them.

I don't get this argument. Sure, some did reach old age but we are way more interested in what percentage of them it was

Assume half of babies die in childbirth or by age 3. Assume another half die before they reach 25. Even assume that another half die before age 60. That means there are still tribal elders that can pass on information and educate the small children they take care of while parents are out getting food. It's a huge deal for a species based on cultural learning.

> That means there are still tribal elders that can pass on information [...] It's a huge deal for a species based on cultural learning.

True, it might be easy to claim that there is no way to select for longevity because longevity happens after the birth of the offspring. Having elder generations around to improve the odds of survival for the descendants should make longevity a possible factor in selection.

FWIW, the human birth process is not viable without cultural accumulation. Baby heads are too damn big and kill women if they don't have a midwife to help them through. But it's a cycle and the midwife is only helpful herself because she has an oversized brain to pass along cultural knowledge.

Elephants have roughly as long a lifespan as humans and there are parts of the world where their populations thrive, even in the wild. I don't think dying young is a cause in this case.

Still, we have pediatric cancers that are not caused by lifestyle. I wonder what's the ratio for small elephants. Are the also more protected ?

But how often would those be encountered in an average city? Because that's about the size of the total elephant population and if a handful of individuals got a rare cancer, chances that biologists would learn of it are still slim.

I believe I've read this verbiage around cancer in pets, such as domestic cats living far longer than "intended".

Pet rats ("fancy rats") were largely separated fron the wild gene pool about 150 years ago. They can live to three years or more, and are quite prone to benign tumours.

(source: wife used to be a fancy-rat breeder)

This is a MUCH better explanation: https://www.ncbi.nlm.nih.gov/pubmed/25459141

Killing off the misbehaving cells is an obvious solution but this must also shorten the organisms lifespan via organ failure.

You only get so many divisions from the zygote before a cell line must be killed off, put into senescence, or it starts malfunctioning due to accumulated mutations (~60). A good solution produces as many differentiated functional cells as possible while minimizing divisions from the zygote.

Really, this article shows a fundamental misunderstanding of how our bodies work.

So just waving away the insane technology leaps required, if we had a digital copy of our DNA we could theoretically create pristine new stem cells and inject them into our bodies as needed to replace aging cells. You might make such a master copy at birth, or more likely in such a world, you're conceived from the master copy which your parents carefully edited before deciding to have you.

Fundamentally we could live a really long time in the far future, which is maybe closer to us today than we are to 0 A.D.

The "master copy" would be at conception. Even after a few divisions it is unlikely any of the cells carry the exact same DNA.

And just because you made it from zygote to birth does not mean your master copy was "pristine".

But yeah, if you could somehow replace your tissue stem cells in situ with ones that are closer to the original state it would generally allow your body to continue functioning for much longer. I mean some of those mutations could be beneficial to your particular environment too, but in general... Yeah.

You don’t have any single non-degraded copy but you do have billions of degraded copies. Assuming errors are non-correlated, one ought to be able to reconstruct the original sequence given enough of them.

The idea would be to use some far-future technology to enable a kind of DNA repair that cells themselves can’t do, using non-local information from other cells in the organism.

Most mutations are probably large scale chromosomal missegregation, etc where huge chunks get moved around or duplicated/deleted rather than just replacing/removing/adding a single nucleotide. From what I have read something like that happens about once every 10 divisions per cell line.

So I think that could be more difficult than you expect.

I think the same statistical methods apply to cancel the noise.

I doubt it, it isn't like the average is going to correspond to the actual original sequence.

It's worth pointing out that YC startup foreverlabs.com does freeze your stem cells for precisely that reason.

Interesting! Not a bad idea given there will likely be more stem cell based treatments available in the future. The cost benefit analysis seems to make sense to me.

Pulled from the comments (which are remarkably smart):

"Upregulated p53 activity in humans is implicated in Huntington's disease. Which makes sense, since the symptoms involve early death of brain cells. So this isn't a panacea in itself. But as we learn more of the gene regulatory networks involved, we may be able to develop something much more fine-tuned that has the benefits without the costs." - Suzanne Sadedin

"Mouse cells given extra copies of the p53 gene seemed to develop some cancer resistance, he said."

I wonder if they could try putting it in to tasmanian devils, which have suffered enormous cancer rates recently.

The cancers you're thinking of in Tasmanian Devils are the result of a transmissible (!) tumor. The cancerous cells aren't the result of mutations within the animal's body, but instead are introduced into it from another infected animal. As such, genetic therapies are unlikely to have an effect.

As such, genetic therapies are unlikely to have an effect.

Can you back that up somehow? That doesn't sound at all right to me.

Edit: For the upvote, downvote brigade -- with zero reply so far, but this comment has gone to zero and back to one umpteen times already -- I'm genuinely curious. This isn't some kind of gotcha question.

I have a genetic disorder. I have umpteen relatives who have had cancer. I actually have a serious vested interest in better understanding how this stuff works.

Short answer: Transmissible tumors are fucking weird.

They're less like normal cancers, and more like parasites made of cancerous cells. Moreover, they're not even typical cancer cells -- they're cancer cells which have been dividing (and continuing to mutate) since the disease first came into existence, to the extent that their genetic material has become wildly divergent from that of the host species.

Typical genetic therapies for cancer focus on the hope that enough of the normal apoptotic pathways still exist that the cancer cells can be "convinced" to recognize themselves as cancerous and undergo cell death. In the case of transmissible tumors, though, it's likely that these pathways have been entirely destroyed through selective pressures on the cancer.

Is it just me, or does that description sound unnervingly close to nature having created a Grey Goo scenario?

Cells are basically grey goo. It is also not uncommon for cancer cells to become independent of the host. See HeLa cells for a human example. These are cancer cells taken from a woman who died in 1951, but which are still living and dividing today (in some way forming a new species): https://en.wikipedia.org/wiki/HeLa

Nature is the grey goo scenario. This used to be a very nice rock in space and then life got it all slimy.

Okay, that makes more sense. Parasites can be an absolute bitch to treat.

as others have said, genetic therapies for these kinds of transmissible cancers are not so effective since the cells have already acquired the cancer phenotype. The usual regulatory mechanisms which prevent unchecked proliferation occur in each individual cell, and have already been circumvented by the time cells become cancerous.

I thought I would add that what makes the tasmanian case interesting is that though the body is generally pretty good about detecting and removing foreign cells (including viruses and bacteria), somehow these contagious cancers elude this detection and are allowed to proliferate [1]. It is likely that if the tasmanian devil's immune system were able to detect the intruder cancer cells as coming from another individual, it would eradicate them with ruthless efficiency. Why these cells are able to skirt the host immune system though is a different question.

[1] https://www.ncbi.nlm.nih.gov/pubmed/28695294

We're talking about genetic therapies that would make cells less prone to mutating into cancerous cells. This would have no effect on cells that are already cancerous, introduced from an outside source.

I read the article. It says this gene actively promotes apoptosis -- cell death -- in defective cells.

I mean, I guess it would depend on how you are introducing the gene. How many cells it is getting into.

But if it causes defective cells to suicide, why would it matter where the defect is coming from? I don't think we really have a good handle on what causes cancer. I suspect more cancer is due to some infectious agent than is generally believed and we know genetic variations have significant impact on the immune system and its ability to function at all.

What the parent means is that in the Tasmanian devils' case it's the cancerous cells of another animal which are transmitted, not some kind of virus or similar that introduces genetic damage in the host. Since the improved/strengthened cell death mechanism introduced by the therapy would only affect your own cells (as the introduced cells wouldn't have the altered genes), there would be no change to the cancerousness of the transmitted cells.

>I mean, I guess it would depend on how you are introducing the gene. How many cells it is getting into.

It's cancer. You're trivializing the ultimate source of the problem. If we could ensure universal targeting of cancer cells, we wouldn't need a technically elaborate method of cellular death, like gene modification.

You might as well suggest solving world hunger by just feeding everyone.

You're trivializing the ultimate source of the problem.

No, I'm not. But it's occurred to me that I'm imagining this as treatment for Tasmanian devils that already have tumors. Maybe other people are envisioning it as a preventative measure.

And maybe that's part of the problem in trying to discuss it.

It is by definition a preventative measure. Otherwise, it's no different then any other cancer treatment in that it has the same fundamental weakness working against it: natural selection.

If the cells knew when they were supposed to die, then it wouldn't be cancer.

If the cure is telling the cells when they're supposed to die, then the cells that don't listen survive, ergo the cancer survives.

Which, in the sense of (non-preventitive) treatment, makes it a problem of universal targeting.

See reply by Sniffoy

Could you go in to more detail on why genetic therapies wouldn't work on a transmissible tumor? Aren't those tumors still affecting the genes in the cells and causing them to grow out of control?

No, the tumor is the cells growing out of control, not some separate thing that acts on the genes of nearby cells. Once cells have mutated into a tumor, it's a bit late to make them less prone to doing so.

The article mentions possibly treating/curing cancer: "He believes there could be a drug that could mimic the effect of p53, or a way to deliver these genes to people at risk of developing cancer, or who already have it to treat or cure the disease."

Method of cellular death doesn't really matter, at this level of discussion of a hypothetical cure. Targeting does. And this strikes me as killing a mosquito with a cannon. If you can ensure universal targeting, there are much easier ways.

just looked then to see if naked mole rats have upregulated p53 too as they are resistant to cancer and it seems to be regulated by same mechanism


I live in Tasmania :)

That’s a great idea.

Also, I advocate to make endangered animals legal to keep as pets.

Look how successful cats and dogs are.

You'd have to domesticate them. And then it's a philosophical question of if they're even Tasmanian Devils anymore.

After all, the success of dogs didn't do anything to prevent the near extinction of the wolf.

There are many pets that aren't domesticated like fish, snakes, tarantulas.

That's not to say that Tasmanian devils are docile enough to be that kind of pet, though.

True, but...

Part of what makes the cancer so transmissible with Tasmanian devils is their predisposition for biting each other, especially on the face, well past the point of drawing blood.

I mean, they're not called devils because they're red with horns.

They made a devil reserve in australia, in a similar climate. Going well I think.

A Tasmanian devil though??!

Most medium sized, and above, dogs could bite your hand off no problem.

Ya’d need to keep them in an enclosure. They breed in captivity fine.[1]

I wonder if they could be bred to become house pets. Could take a while.

1. http://theconversation.com/tasmanian-devils-reared-in-captiv...

They are infamously aggressive. There is ongoing debate about the wisdom of keeping extremely aggressive dogs as pets, so the dog comparison isn't really a good rebuttal.

> He hopes to have a clinical trial within the next three to five years.

It's been just over 4 years since that article. I wonder what has happened since.

> The work has already given him a new tool when he talks with his patients.

> “When I have a patient in front of me diagnosed with the syndrome, they will almost certainly get cancer,” he said. “But in that moment I’m able to tell them elephants don’t get cancer and we are working with the zoo and the circus to learn from elephants so one day you never have to get cancer.”

This seems a bit unethical to raise hope for a cure that statistically has a very low chance of passing clinical trials

Next one due in 2021

Do other large mammals have similar cancer resistance?

How many cells does a blue whale have?

Able to answer this since I just read an article on this in the Economist (https://www.economist.com/science-and-technology/2019/06/29/...): It seems whale genes have evolved to supress cancer. These include ATR (https://en.wikipedia.org/wiki/Ataxia_telangiectasia_and_Rad3...) that senses DNA damade, AMER1 (https://ghr.nlm.nih.gov/gene/AMER1), which stifles cell growth, and RECK (https://en.wikipedia.org/wiki/RECK), which suppresses metastatis.

These genes are also found in humans but the hope is that studying genes from animals like elephants and whales we can find ones that we don't have.

I always thought it might be because with a larger volume of blood, their adaptive immune system could contain more permutations of antibodies, allowing them to detect more cancers.

Just a hypothesis I thought of:

A lot of cancer cells are developed because of constant background radiation, if you get older you just were exposed to more of it.

Elephants have a much higher volume to be affected by it, so just from an evolutionary perspective they had to develop a better resistance. The ones who got cancer early couldn't reproduce as well.

Does that make sense? Or does the article say exactly that?

Why not have a read and find out?

Validating your hypotheses? How quaint!

Well it just says many cells should mean a lot of cancer.

We are not even sure if we need a certain amount of background radiation.

Could we use CRISPR to copy this gene more times in humans then?

Yea, we probably don’t want to do that. Over-expression has a tendency to compete with other regulatory genes, enabling the reverse of what we’d aim to achieve.

[0]http://m.cshperspectives.cshlp.org/content/2/2/a001107.full [1]https://www.ncbi.nlm.nih.gov/m/pubmed/18812169/

p53: Regular or super?

Increased p53 expression under the endogenous promoter protects “super p53” mice from tumorigenesis without the undesirable effects of premature aging.


>Garcia-Cao and collaborators report the generation of novel transgenic mice, called “super p53.” The super p53 mice express wild-type endogenous p53, and also carry one or two extra copies of a normal p53 gene, inserted as transgenes in the form of large genomic fragments of 130 to 175 kilobases. Because the additional p53 gene is expressed from its own promoter, the transgenically expressed p53 seems to be regulated in the same fashion as endogenous p53.

Do humans happen to get more cancer than other animals? There’s this weird idea, I forget the source, that a less efficient cell garbage collection mutation allowed increased brain development, connections, etc... but at the cost of increased cancer, depression, ego, etc...

Because they aren't stuffing thier necks with sugar, pounds of meat, and dairy? Perhaps the reason they aren't ruining our climate either.

It apparently is because Elephants are better at cleaning up cancer cells. Presumably, because they have more copies of the “P53 gene”.

No. Because they have more copies of a gene for killing cancer cells. The article makes no mention of dietary reasons.

Well, it's either "pounds of meat and dairy" or "sugar", you need one or the other because humans can't feed on sunlight yet.

If you don’t eat, you won’t die of cancer clearly, starvation yes, but definitely not cancer.

Literally no.

Because they can't open cigarette packages?

An old video [1] in response to this old article may indicate another reason why elephants dont get cancer, and provide easier way to replicate those results than gene therapy or splicing. Hint: go vegan.

[1] https://nutritionfacts.org/video/how-not-to-die-from-cancer/...

Do human vegans get less cancer? (Honest question.)

There are studies on this, however they are controversial since diet is not easy to control amongst study subjects. Also, there are vested interests keeping public opinion divided on this matter.

One study from 2014: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3565018/ This compared non-vegetarians with multiple types of vegetarian diets.

> Conclusion: Vegetarian diets seem to confer protection against cancer.

> Impact: Vegan diet seems to confer lower risk for overall and female-specific cancer compared to other dietary patterns. The lacto-ovo-vegetarian diets seem to confer protection from cancers of the gastrointestinal tract.

A more recent perspective on the topic of diet as a cancer risk: https://www.tandfonline.com/doi/full/10.1080/01635581.2017.1...

> Abstract: The role that nutrition plays in cancer development and treatment has received considerable attention in recent decades, but it still engenders considerable controversy. Within the cancer research and especially the clinical community, for example, nutritional factors are considered to play, at best, a secondary role. The role of nutrition in cancer development was noted by authorities as far back as the early 1800s, generally under the theory that cancer is "constitutional" in its origin, implying a complex, multifactorial, multistage etiology. Opponents of this idea insisted, rather vigorously, that cancer is a local unifactorial disease, best treated through surgery, with little attention paid to the etiology and possible prevention of cancer. This "local" theory, developed during the late 1700s and early 1800s, gradually included, in the late 1800s and early 1900s, chemotherapy and radiotherapy as treatment modalities, which now remain, along with surgery, as the basis of present-day cancer treatment. This highly reductionist paradigm left in its wake unfortunate consequences for the present day, which is the subject of this perspective.

(I am citing the second publication only in defense of my original comment, for future reference. In retrospect, when discussing health benefits of diet, I should have used the term whole food plant based diet instead of vegan to avoid the baggage that term carries.)

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact