Hugo Hacker News

How elephants avoid cancer (2015)

cosbgn 2021-08-17 11:05:50 +0000 UTC [ - ]

One theory is that they actually get so many cancers that they fight each other and never reach the stage where they would become harmful for the Elephant. A good video about this was made by Kurzgesagt – https://youtube.com/watch?v=1AElONvi9WQ

jjk166 2021-08-17 13:43:29 +0000 UTC [ - ]

Hypertumors aren't really a good explanation for Peto's Paradox. Yes, larger animals would need larger tumors to kill them, and larger tumors are more likely to develop hypertumors, but even if cancer didn't kill a large animal we'd still expect to see all those "small" tumors that haven't yet developed hypertumors.

Further, Peto's paradox doesn't just apply to large organisms: small creatures get higher cancer rates. You'd expect there to be some threshold below which the odds of a hypertumor developing are incredibly low, and all organisms that can be killed by tumors smaller than this threshold would have a cancer incidence rate proportional to body mass. At the same time there should be some threshold where a hypertumor is basically guaranteed, above which cancer deaths would simply stop regardless of body mass. Instead, cancer rate is independent of body mass across the spectrum.

Then, the lack of correlation between cancer rates and body mass only applies between species. Within a single species, animals with more body mass do have higher rates of cancer and vice versa. This is directly contradictory to what we'd expect if cancer suppression was a result of increased body mass.

Finally, closely related to Peto's Paradox is the fact that longer lived species have lower per-cell division cancer rates than shorter lived species. Animals with different lifespans but of the same bodymass can be killed by tumors of the same size, so you need some explanation other than hypertumors (which occur as a function of the size of the tumor) to explain this observation.

Evolutionary adaptations that limit cancer are both a simpler and better explanation for these observations. Every animal is going to have some threshold in its life by which it is likely to have passed on its genes, and another threshold beyond which it is unlikely to better pass down its genes. Any gene that kills you before the first threshold is going to be strongly selected against, any gene that keeps you alive after the second threshold isn't going to have much selective pressure supporting it. A mouse in the wild can successfully pass on its genes in 3 months and is unlikely to live past 18 months due to injury and predation. While mice under ideal conditions in captivity have lived for up to 6 years, a gene that helps mice suppress cancer when they're 5 years old isn't doing wild mice any favors. On the other end of the spectrum, blue whales take a minimum of 16 years to reach genetic breakeven, and realistically given their distribution it could easily take much longer. A mutation that causes a 5 year old whale to get fatal cancer will never get passed on to offspring. We should logically expect that any species will continue to evolve cancer defense mechanisms up until the point at which cancer is no longer the limiting factor for passing on their genes, at which point evolution stops selecting for further improvements.

crubier 2021-08-18 03:15:39 +0000 UTC [ - ]

This is a very strong point. It puts some rationalization behind my intuition that hyper tumors are not a correct explanation, it emphasizes the strength of the gene explanation and this consistent with OP’s article.

MiddleEndian 2021-08-17 17:58:09 +0000 UTC [ - ]

bradrn 2021-08-17 11:32:43 +0000 UTC [ - ]

Also discussed by Scott Alexander in some detail, near the end of a mostly unrelated post: https://slatestarcodex.com/2014/06/14/living-by-the-sword/

ginko 2021-08-17 10:43:37 +0000 UTC [ - ]

Who added the question mark to the title and why? At least make it "How do Elephants Avoid Cancer?" so it's a grammatically correct sentence.

OJFord 2021-08-17 10:59:07 +0000 UTC [ - ]

I'm more irked, as usual, by HN's Automatic Click-Bait Case-ifying.

You can, and I do when submitting, remove it by editing the title, but of course most people don't and why should they. Seems to go against the guideline to, generally speaking, leave the title as it is in the original article, too.

codetrotter 2021-08-17 11:07:19 +0000 UTC [ - ]

> Automatic Click-Bait Case-ifying

You mean title case?

https://en.wikipedia.org/wiki/Title_case

> Title case or headline case is a style of capitalization used for rendering the titles of published works or works of art in English.

It’s got nothing to do with click-bait. It’s part of English grammar.

messe 2021-08-17 11:46:07 +0000 UTC [ - ]

It's nothing to do with grammar, it's a purely stylistic choice.

djhn 2021-08-17 12:28:53 +0000 UTC [ - ]

Oh but what is grammar if not a bunch of stylistic choices that over time have settled?

I hate title case, but unfortunately it seems to have become The Way We Write Titles.

hdjjhhvvhga 2021-08-17 12:52:11 +0000 UTC [ - ]

There are two things here. Title capitalization is definitely a part of English orthography. Any book you take, no matter what English-speaking country the author is from, will always use title case, unless it wants to make a point (avant-garde etc.).

However, there are differences as to what deserves to be capitalized. Just the titles of books, films, etc? What about the titles of articles, are they in any way different, or maybe worse than the titles of books etc.? I guess this need for consistency pervades the American usage, codified in the CMoS and other style guides, and hence applied universally.

thatcat 2021-08-18 00:31:21 +0000 UTC [ - ]

It's based on the AP style guide used in US print journalism.

playpause 2021-08-17 12:48:09 +0000 UTC [ - ]

That might just be an American perspective. It is not standard elsewhere. I just checked all the UK and Canadian newspaper sites I can think of, and not one of them uses Stupid Title Case.

OJFord 2021-08-18 11:59:40 +0000 UTC [ - ]

Ah I think you're on to something there - I almost said (and this would explain why) that Bloomberg is the only publication I read that I notice/can think of that does this.

It just has click-bait/spam associations to me since that's where I generally see it: I don't like the look of the title, so I look at the URL, if it's 'amazingwownewsdaily.com' or something I move on; if it's 'bloomberg.com' or something recognisable (-ly good) I think 'oh ok, ugh, but ok' and open it.

OJFord 2021-08-17 14:07:23 +0000 UTC [ - ]

Exactly, a stylistic choice frequently made by click-bait type sites like Buzzfeed, and not made by Nature in the submission.

hdjjhhvvhga 2021-08-17 12:43:42 +0000 UTC [ - ]

Capitalization is part of orthography, not grammar.

stronglikedan 2021-08-17 19:56:26 +0000 UTC [ - ]

Pedantically, yes. Ubiquitously, no.

2021-08-17 12:40:52 +0000 UTC [ - ]

OJFord 2021-08-17 14:11:51 +0000 UTC [ - ]

On the contrary it has nothing to do with grammar, and I relate it to click-bait because it's so frequently done by such articles by BuzzFeed and the like; with much more prevalence than among things I actually want to read.

It is typically a signal that I don't want to read something, which makes it especially confusing on HN, which is generally a list of things I do want to read.

delibes 2021-08-17 10:56:39 +0000 UTC [ - ]

Also needs 2015, and has been posted before https://hn.algolia.com/?q=elephants+cancer

lordnacho 2021-08-17 16:39:42 +0000 UTC [ - ]

I wondered about this the other day and didn't get much of an answer:

Are there first-hand research studies where we take a large sample of some animal (whales / mole rats / whatever) and check the animals for cancer?

Apart from the thought experiment of Peto's paradox, has someone actually taken the time and effort to document this? I'd be interested in reading some papers.

dsign 2021-08-17 12:38:16 +0000 UTC [ - ]

If this "brute approach"[1] of having multiple copies of a gene manages to prevent cancer, imagine what some (human) intelligent design can achieve.

[1] Gene duplication is common in nature, and it's one of the main mechanisms by which new genes with different functions evolve.

inciampati 2021-08-17 12:59:35 +0000 UTC [ - ]

I'm imagining that our intelligence will lead us to an elegant solution that initially appears to work but later proves to have catastrophic unintended side effects. Maybe just copying genes is enough? That brute force approach might be smarter than it appears, given how incredibly common it is in nature. A key thing is that the copies can repair themselves off each other. The more copies that are available, the more robust the system is to error. That's part of why humans have their ribosomal DNA in super high copy number in five psuedo-homologous regions on the short arms of our acrocentric chromosomes. The genome keeps lots of copies of the most important stuff. For humans, cancer wasn't so important, until recently. So we might need to accelerate things and increase our tumor suppressor copy number. It works for people with trisomy-21.

idnefju 2021-08-17 14:03:10 +0000 UTC [ - ]

Genetic engineering will be our only hope precisely because modern medicine is disregarding our evolution. Lots of bad mutations are being passed down because we can treat them, which of course allows people to live a good life.

lebuffon 2021-08-17 13:48:03 +0000 UTC [ - ]

Doctor: I've got good news and bad news. The good news is we have genetically modified you so that you will never get cancer.

Patient: What's the bad news?

Doctor: In about 8 months you'll be able to unzip your pants with your nose.

:-)

dsign 2021-08-17 14:25:52 +0000 UTC [ - ]

> In about 8 months you'll be able to unzip your pants with your nose.

That's not the end of the world :)

_jal 2021-08-17 13:45:34 +0000 UTC [ - ]

> imagine what some (human) intelligent design can achieve

We don't have to imagine. Everyone here is familiar with Clippy.

More seriously, I think this is a category error, rather than just hubris. I don't think human intelligence is a substitute for evolutionary grinding.

dsign 2021-08-17 14:24:58 +0000 UTC [ - ]

> We don't have to imagine. Everyone here is familiar with Clippy.

I loved Clippy

pharaohgeek 2021-08-17 13:10:55 +0000 UTC [ - ]

In humans, a defect in the TP53 gene that leads to genetic predisposition to cancer is called Li-Fraumeni Syndrome. It explains a lot of people who have a strong family history of cancer or those who get it early in childhood.

dang 2021-08-17 20:34:11 +0000 UTC [ - ]

One past thread:

How Elephants Avoid Cancer (2015) - https://news.ycombinator.com/item?id=14698886 - July 2017 (50 comments)

dajohnson89 2021-08-17 15:36:18 +0000 UTC [ - ]

Is there a big list of things a human can do, to reduce/minimize their risk of getting cancer? "dont smoke" is obvious but a lot of chemicals/behaviors are less well understood in terms of cancer risk.

vharuck 2021-08-17 16:10:37 +0000 UTC [ - ]

https://www.cancer.org/healthy.html

- No smoking, or chewing tobacco

- Don't drink excessively

- Stay at a good weight

- Use sunblock

- Eat healthy (notably, reduce consumption of red meats)

- Get your recommended cancer screenings (mammograms, colonoscopies, prostate exams, etc)

Other tips for specific populations:

- If you are or were a heavy smoker, talk with your doctor about a lung cancer screening

- Vaccinate children against HPV

- Some experts recommend Hepatitis C screening for Baby Boomers

123jay7 2021-08-17 17:30:56 +0000 UTC [ - ]

because they are happy all the time.