Hacker News new | past | comments | ask | show | jobs | submit login
How cancer was created by evolution (bbc.com)
132 points by lookupmobile on June 5, 2016 | hide | past | favorite | 88 comments

The most horrifying story I know of is canine transmissible venereal cancer. It used to be a dog, but was more evolutionarily successful evolving into a tumor.


The Tasmanian Devil facial cancer is another horrific transmissible cancer:


"It used to be a dog" is a truly provocative phrase, not in a bad way. Fascinating.

Good thing it seems not too lethal?

It seems fairly debilitating in late stages. The Tasmanian devil's one seems to be threatening the species, which is rather serious.

Yeah I mean as horrifying as "std cancer" is, "species-threatening cancer" is objectively worse.

I wish we spent more time studying why elephants [1] and bowhead whales [2] don't get cancer more often than they do. If we knew exactly why their cells are so resistant to cancer we would have some ideas of what we need to do to make ours the same.

1. http://www.ncbi.nlm.nih.gov/pubmed/26447685

2. http://www.ncbi.nlm.nih.gov/pubmed/25565328

We do know. A gene called p53 detects anomalies during DNA replication and can "arrest" the growth or replication phase, or force apoptosis (cell death).

1. http://www.the-scientist.com/?articles.view/articleNo/44215/... 2. https://en.wikipedia.org/wiki/P53

No we don't know. We know very little about the reasons for the relative absence of cancer in these organisms. The absence of cancer is not solely due to p53 differences, p53 just happens to be one of the few cancer preventing genes studied to date in these animals.

yeah but P53 redundancies are a pretty big deal in answering that question. We have one P53 gene and if it gets damaged the cell gets to run amok. Those other animals have like 12 p53 genes, and don't get cancer.

Correlation, meet causation.

What is the valid counterpoint?

The evidence that p53 duplication is an important mechanism is very interesting, but we know p53 is not the sole answer. A good example of this is both humans and mice have one copy of the p53 gene yet humans are vastly more resistant to developing cancer than mice.

Could that be because humans are more complex from a biochemical standpoint than mice?

It is because we have been selected for a much longer lifespan than mice. Basically during human evolution there has been a selection process for greater resistance to developing cancer - if we got cancer at the rate mice get it almost nobody would make it to be a teenager.

Humans are by animal standards very long lived and naturally quite resistant to developing cancer. We have many more anti-cancer biochemical and genetic systems (antioncogenes) that work much better than the same systems in mice.

Please, your statement is so ambiguous that it makes almost no sense.

1. Which mouse? An inbred laboratory possibly immunodeficient strain, a genetically engineered mouse to GET cancer at a high rate, or maybe a wild type free-range mouse?

2. Which cancer? Do you realize that it is almost impossible to get realistic models of prostate cancer or that mice simply live to short to get colorectal cancer - even when the relevant mutations are introduced?

Sorry if I was being ambiguous. All mice are less resistant to cancer than humans, even wild type mice. Mice in nature rarely live long enough to get cancer, but if you put them up in a mouse retirement home (lab), a large percentage of wild mice will get spontaneous cancers.

Yes mouse cancer in not a great model for human cancer and we probably should not be using mice - about the only good thing they have going for them as a model is they are small and cheap.

Personally I would like to see us using pets (dogs and cats) much more as their cancers are a much better match to human cancers in both how they develop and how they progress. That we are not taking better advantage of this resource to develop new treatments is a tragedy.

I suspect one reason could be that mice have much faster metabolisms than humans. Also, typically the speed of metabolism is typically inversely correlated with size, although there definitely are some outliers too.

Mice have far few cells than humans so they should be much less likely to get cancer (all things being equal). Peto’s paradox is all about this effect [1].

1. https://en.wikipedia.org/wiki/Peto%27s_paradox

But those cells burn at a lot higher rate.

And wow, the "related articles" were actually useful for once. http://www.bbc.com/earth/story/20151031-the-animal-that-does... mentions duplicating the gene in mice: 2 copies seems to be the sweet spot.

Yes but not a billion time more. The relative resistance of human cells to cancer compared to mice is more than just about metabolism.

part of "intelligently" informed "lifestyle choices" might be what these big mammals have what we don't.

Foremost the number of times a gene is found does not directly relate to its expression, and mutant proteins are often significantly more reactive than their counterparts. Also having more genes means the risk of getting a damaged gene goes up.

Most importantly people for the last 20 years have tried to genetically integrate working p53 into cancer cells, and it has little effect on most types of cancer.

Why is that important, I would imagine that ship has sailed by the time the cell was deciding to not die

was the hypothesis that introducing a gene would start apoptosis?

Some cancers (testicular if I recall), are the result of a p53 mutation. Crucially, fixing the mutated gene did not fix the cancer. The community I work with has a lot of skepticism about claims that simply increasing the quantity of tumor suppressor gene will fix cancer.

Most likely having more genes means that there was a higher chance of receiving a viable beneficial mutation somewhere early in the animals evolutionary history, although it does little for the individual animal in the present day.

Any tests to insert p53 (and others) into cells to see what can go wrong ?

The level of expression of p53 is quite strongly regulated in human cells and over expression (due to an increase in the copy number) will be lethal. With crispr/CAS9 it should be possible to do some really interesting experiments in mice to see if increasing the copy number (in a control way) reduces the rate of cancer in mice.

Its a factor. Not the only one by far.

In our ongoing Small RNA - mRNA interaction study, we are also seeing p53 and a few other genes heavily being the regulators or are being regulated disproportionately between Cancerous vs. normal tissues.

Slate star codex kind of speculates [1] that it might have something to do with scale, ie -- in order to keep growing in a creature that large, the cancer cells need to coordinate among themselves and specialize to a degree that is not needed at smaller scale. In other words, before the cancer can grow large enough to threaten the host, "the cancer gets cancer."

[1] http://slatestarcodex.com/2014/06/14/living-by-the-sword/

Speculation by a psychiatrist of course, and mostly in service to a metaphor about some totally other topic, but I wonder if there's something to the theory.

There is a lot of speculation in this area - see Peto's paradox [1] - what we are missing is lots of good data.

1. https://en.m.wikipedia.org/wiki/Peto%27s_paradox

And naked mole rats right?

Naked mole rats don't live as long as humans. What we really need to know is how to make a human cell almost never get cancer - looking at large, long lived animals in different lineages should give us a lot of clues of the way to go.

It is much more profitable to treat cancer than to cure it. So I don't really believe it's about science.

And because I can already see the down votes, let me ask you down-voters a question. With what thoughts do you imagine a CEO of a company like Phizer or GlaxoSmithKline wakes up? Is it how to cure more people, or is it how to make more money to keep shareholders happy ? And see how there is a conflict of interests, the sicker the people are the more they are willing to pay to get well and the more they need your products.

Slim chances.

I'm sorry to say that but most cancer research you hear of in media don't want to cure cancer. Its just a business. Otherwise Apple could throw 40 billions of $ in it and come up wit ha cure.

At least in USA, it become such a fat cow to milk that foundations grew everywhere out of nowhere... go to your local store and say "no thank you" when cashier ask you "would you like to donate to fight cancer" and see their look like you are a murderer or something. Its next to "charity" best way for crooks to get rich tax-free while sucking on unaware populus.

Can you imagine the amount of money someone would have thrown at them if they could cure cancer? I don't think greed is what's stopping people from cutting cancer.

There are plenty of cures for cancer. They just don't all work all of the time on all people and it's not necessarily permanent. It all depends on the type of cancer and what the cause is. You can get the cancer in any cell in your body and often you have more than one type at a time. Plus the treatment for preventing a recurrence can be just as unhealthy as the disease. That's why there can be so many variations in treatment and survival rates.

The biggest thing you are trying to stay ahead of is the cancer spreading to your lymph nodes which is where the anti-bodies and white blood cells that power your immune system are made. Crippling these makes you susceptible to other diseases like every day bacterial infections and viruses which then could end up killing you. It's certainly not because they don't want to cure cancer.

Most cancers (80%) can be prevented though, and 70% of them are associated with HPV, a virus that is transmitted through sex. This is why sleeping around is unhealthy. You could be spreading it to every one you sleep with orally as well as genitally. (just another reason marrying your first love and being monogamous is healthy behavior). HPV has been associated with cervical cancer, breast cancer, throat cancer, prostate cancer, mouth cancer (noticing the pattern...).

And that's just one factor associated with cancer. Eating healthy (fruits an dveggies, lower red meat especially meat treated with chemicals) reduces other types of cancers associated with your digestive system like colon cancer (and is good for the heart and circulatory system which starts affecting you the closer you get to 50). Basically your arteries harden and your circulation gets poorer as you age and that results in other problems that creep up like high or low blood pressure that causes things like chronic headaches or numb feet, being overweight which can also lead to diabetes, etc.

If you take care of yourself when you are young you will have fewer problems as you age, but they are inevitable the older you get. But sometimes cancer is just a genetic mutation you inherited and you will get it anyway.

80% of cancers can be prevented and 70% are associated with HPV? Do you a source for that?

They're completely wrong, and are using it as a bullshit (excuse my language) platform to push their "moral" righteousness.

This simply proven:

  Based on data from 2006 to 2010, about 33,200 HPV-associated cancers occur in the United States each year: about 20,600 among females, and about 12,600 among males. Cervical cancer is the most common HPV-associated cancer among women, and oropharyngeal cancers (cancers of the back of the throat, including the base of the tongue and tonsils) are the most common among men.
That's from: http://www.cdc.gov/cancer/hpv/statistics/

Now let's quickly look at how many cancer cases we're looking at each year:

  In 2016, an estimated 1,685,210 new cases of cancer will be diagnosed in the United States and 595,690 people will die from the disease.
From: http://www.cancer.gov/about-cancer/understanding/statistics

So I don't know where they got their numbers from.

You might also be interested in: http://www.aicr.org/research/research_science_policy_report.... -- which has a pretty great rundown of preventable cancer incidences. Quite interesting, really.

These two paragraphs don't follow:

"Figures like this show that cancer is not only extremely pervasive, but also becoming more and more common. But why will so many people develop the disease at some point in their lives?"

"The fact that tumours are constantly changing their genetic makeup is one of the reasons why cancers are so hard to "kill".

The article's faulty reasoning seems to imply that cancer is more and more common because of evolution. The scientist's work is examining the evolution of a cancer's cells since its initial genesis in the patient. Unless cancer is transmissible between individuals, there is no on-going evolution in cancer that is making it more efficient across the pathology of multiple patients. Thus, this doesn't add much to our knowledge of why cancer is more and more common.

> "Figures like this show that cancer is not only extremely pervasive, but also becoming more and more common. But why will so many people develop the disease at some point in their lives?"

You're right, the article did a bit of trickery here conflating general human evolution, with the main point of the article about cancer's isolated evolution within the host. The line that attempts to support this follows:

"Large and complicated animals like humans [aka lots of 'evolution'] are vulnerable to cancer precisely because they are large and complicated".

This claim is not supported by the article, and not the main point - but it's also just part of the intro and doesn't take away much either, chalk it up to journalistic creativity.

> "The fact that tumours are constantly changing their genetic makeup is one of the reasons why cancers are so hard to "kill"

That's what tfa is all about. In summary:

- cancers reproduce/evolve within a host optimizing against the body's defenses as well as the treatments we provide

- our treatments are effective at first, but since the cancer keeps evolving, they are all eventually rendered useless

- if the evolution is the problem, what if we try stopping that?

And that's what they're studying:

- find the specific meds that stop a particular cancer by giving them to the patient and seeing if it kills it

- as soon as u find them, stop giving them, so that the cancer doesn't go into hyper evolution against them

- let the cancer grow fat and lazy suffocating itself with a high-proportion of easily killable cells

- re-apply the medicine in full force to kill the whole thing

The article addresses why it's more and more common. Specifically, it suggests our cells haven't evolved to live that long, so if you're older than 40, then at some point or another your cells will mutate - which may or may not get the attention of your immune system.

> The article's faulty reasoning seems to imply that cancer is more and more common because of evolution

No, it doesn't. It just says that there is more of it, mostly because we live longer and smoke cigarettes and such.

Has this idea of "more cancer" been controlled on

1. population generally getting older


2. more people getting diagnosed?

Also more people getting their life prolonged by cancer treatment?

some cancers are transmissible (HPV-induced, for instance, and those of the tasmanian devils, the list goes on and on, not kidding)

The cancer caused by HPV is not transmissible. The virus that causes it is.

Transmissible cancer is possible, but it's of almost negligible risk to humans [1, 2].

[1] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3228048/

[2] http://www.cdc.gov/media/releases/2015/p1104-parasite-tumors...

There are four known ones:


But we have started doing things like making stem-cell lines that have surface markers removed and are compatible with almost anyone..

Fortunately we tend not to angrily bite each other as a greeting.

They don't all require that:

> Canine transmissible venereal tumor (CTVT) is sexually transmitted cancer in dogs. It was experimentally transplanted between dogs in 1876 by M. A. Novinsky (1841–1914). A single malignant clone of CTVT cells has colonized dogs worldwide, representing the oldest known malignant cell line in continuous propagation.

Fun fact: an HPV vaccine that covers even more varieties of HPV was approved by the FDA a year ago.

> "The figures are even worse in the UK. According to Cancer Research UK, 54% of men and 48% of women will get cancer at some point in their lives."

I am not sure if this is a good or a bad thing. If you accept that cancer is simply the disease you get when you live long enough and don't die because of anything else before (a view even supported by the article), a higher percentage could also mean that all other things are in control now, and only cancer and heart attacks are left.

So we hear about how eating healthy and exercise and yada yada help your cancer and heart attack risks. How much would each go down if we were perfect?

My gut feeling is we could fix 75% of current heart attacks with weight control + exercise + whatever, but only say 25% of cancer... i.e. that heart attacks are very much in the group of things we know how to fix, whereas cancer is still very much in the group of things we don't know how to fix.

But curious what the real numbers are...

Well, I suspect many things regarding health are outside of anyone's control. Ie. if you're a living organism in 2016 on planet Earth, then you probably are negatively affected due to pollution caused by humans, let individuals hope they're lucky. There's lot of nasty stuff out there that we've made and let loose on purpose or by accident. You can't have an ideally healthy lifestyle when all up and down the ecological chain our handiwork is present (people like to make money with the least amount of effort on their part).

How prevalent is cancer amongst regular sport fans ?

> all other things are in control now, and only cancer and heart attacks are left. Really? What about parkinson etc?

One has to wonder why we age in the first place. No doubt we have evolved to age because a species that does not die off cannot evolve. The advantages of living longer eventually are outweighed by the pressure to evolve more rapidly. So if we see age as an evolved trait, might cancer be part of that endgame? Or at least, our growing susceptibility to cancer and other diseases as we age evolved specifically to kill off the old so as to make room for new generations.

This of course would have evolved long long ago, back before we were mammals. As only a handful of the most primitive animals are essentially immortal (jellyfish and the like) this would seem a very successful evolutionary trait.

Why the downvotes? There are other articles (on HN recently!) that postulate that aging is evolved.

Now yes cancer itself, as mutation caused, can't really be a trait of the host (I go as far as to consider cancer cells a separate parasitic species). And since many cancers are immortal, and the aging relates to a secession of cell division in some sense they are opposite extremes with normal development in the middle. But in being opposites they are related---end games / limitation for multicellularity.

This reminds me of this culture of cancer cells from a woman who died from cancer in the 50s. The cancer is still alive in laboratories today and there were even proposals to classify it as a new species:


ex-gf was working with "her" cells some 6-8 years ago while researching some rare type of anemia, back in Prague. Didn't know it's actually used worldwide.

one way to become immortal in some sense :)

That is the name of Rebecca Skloot's excellent book on the subject: The Immortal Life of Henrietta Lacks (https://en.wikipedia.org/wiki/The_Immortal_Life_of_Henrietta...)

From the article:

"In the US there has been a 25% decline in death rate in the last two decades. 'More than half of that decline is driven by cancer prevention activities," says Brawley.

"This points to the fact that some of the cancers that would previously have killed people had been prevented. Almost a third of death from cancer in the US has been attributed to cigarette smoking, for example. This makes tobacco "the single most preventable cause of death in the world", according to Cancer Research UK."

Never starting a smoking habit, or quitting a smoking habit if you have already started, is a great way to prevent a lot cancer risk. Being moderate (not excessive) in alcohol use helps a lot too.


> But a more telling reason for the rise is that humans, on average, live a lot longer than they used to. "If you live long enough you will get cancer," says Biankin.

I keep hearing this, but do we have data that shows that cancer rates have not changed within age groups?

Average lifespan is skewed by much higher infant mortality rates of the past: an average lifespan of 45 doesn't mean everybody was dropping dead at 45; it means more people were dying during their infancy and youth.

A 20 year old roman could expect to live to ~50:


In the US in 1900, a 20 year old could expect to live to ~62:


Today, a 20 year old can expect to make it to about 80.

So the gains aren't really limited to infant and child mortality. A substantial portion of the gain has certainly come from improvements there though.

But the gains at age may come from other areas and not from medicine. In this talk


the professor mentions as an aside (there is a transcript of the talk)

> And what is interesting, if you take out the childhood mortality, the Victorian person between 1850 and 1880 lived slightly longer, if he was a male, than you do today.

Females had the risk if child birth, an area where medicine did do a lot.

It shows that sanitation/hygiene and food seem to have done a lot for lower mortality outside the risk factors of being very young and/or giving birth.

And that was before antibiotics! Which is nice to know when we get another doom article about the end of antibiotics, because it seems that while sure individuals will suffer it by no means warrants predictions of doom for mankind.

While Swanton's work is great in revealing how difficult it is to contain a tumor and how evolution allows escape from therapy and in exposing genetic diversity in tumors, the idea that you can "chop at the root of the tree" is (in my opinion) bunkum. First of all, even if you target trunk mutations you can evolve resistance. Second, you can't just keep loading up drugs into patients; they can often barely tolerate one of these therapies (which have off-target effects on normal cells), combination therapies compound the problem.

All this talk about tobacco but nothing about alcohol, for which there is much stronger evidence of DNA damage? Does the BBC have a hidden conflict of interest?

Do you have a link supporting the DNA damage and alcohol relationship? That's interesting.

Lung cancers are particular in that they feature an incredible amount of somatic mutation. If I remember its at least 10x as much as the next high-mutation rate cancer. Probably more even. Smoking tobacco provides a melange of mutagens: complex aromatic DNA adducts that result from combustion, heavy oils, and radiation from potassium heavy fertilizer. These would be enough to cause problems alone, but to make things as bad as possible the active ingredient is a critical signaling molecule in the cell cycle, which after copious application is likely to induce the survival of cells unresponsive to it. These cells are less likely to play nice with others.

"Do you have a link supporting the DNA damage and alcohol relationship? That's interesting."

Not the op but alcoholics are likelier to develop e.g. pancreatic cancer. I believe the direct mechanism, though, is not explained. "About 7 out of 10 cases of chronic pancreatitis are due to long term heavy drinking":


In the US, the lifetime risk of developing cancer is 42% in men and 38% in women

Woah, this isn't far from saying living in the US isn't far from a coin toss for a death sentence.

Not if you don't smoke. Smoking accounts for ~30% of cancer deaths in the US. A substantial number of the rest are due to skin cancer (lots of white people descended from Northern Europeans living in sunny climates).

There are lots of types of cancers. Some of the most common types (skin and prostate) aren't really much of a detriment to life expectancy.

This explains it. Thanks.

Luckily, cancer isn't a death sentence anymore. It is really, really shitty. REALLY shitty. But it is not a certain death sentence.

What if cancer is the evolution in action. Evolution has to go through a huge number of permutations to get a viable macro organism. By trying to cure cancer, we are trying to withhold the natural selection in some form and preventing the next stage of evolution.

There is no such thing as `next stage of evolution'. Evolution doesn't have stages.

Also, we are only interested in curing cancer in humans (and perhaps at most livestock). Wild animals will still get cancer.

(Do plants and funghi etc get cancer?)

Apparently plants do develop tumors, but they don't metastasize because of cell walls that stop them from spreading. Also, like this article says ( http://www.popsci.com/article/science/ask-anything-do-plants... ), plants don't have vital irreplaceable organs, so tumors are less dangerous for them.

Just to nitpick, we're not only interested in curing cancer in humans. From elsewhere in this thread, a transmissible cancer that threatens an entire species, for which attempts have been made to find a cure or vaccine:


Humans evolution can be characterised into major stages from ape to homosapien.

No it can't. There were only many small adaptions, not stages. Stages also suggests that there is a next stage to evolve to which is not correct.

You are conflating 'evolution is directionless' with 'no stages'. Stages can only be seen when looking back at the evolutionary history.

For example:

Study reveals human body has gone through four stages of evolution (http://www.eurekalert.org/pub_releases/2015-08/bu-srh083115....)


Stages are a human abstraction that help us understand and make sense.

You can say that for just about everything

Cancer usually refers to a cell mutation that is harmful to the organism. I'm sure DNA mutations happen all the time in individuals, but in most cases they don't provide the utility to increase "fitness" relative to the environment which would increase the individuals quantity of reproduction/offspring, thus spreading the gene. If people were still able to reproduce at older ages, I'm sure we could evolve to be more resistant to cancer; since cancer will most likely affect you later in life, it has no impact to how much you reproduce.

The sad truth is above 80% of the food you can find in a U.S. supermarket is transgenic or otherwise called GMO. All sorts of carbohydrate sources - which comprise a huge part of the U.S. diet - from cereals, refined sugars and the base of the american diet: corn fructose. Vegetables and meat that's been grown with also GMO feed has been genetically modified and released to the public without enough testing. One of the consequences is cancer. Cancer grows because the quality of your food, the biotechnology embedded in it and american eating habits are incredibly unhealthy and unsuited for its lifestyle. Research the lobbying between FDA and Monsanto, research the correlation between meat consumption and risk of heart disease and cancer, research the recommended daily intake of fibre and compare it to the average american diet, see the correlation with cancer. There lies the answer. This article is a lie. Monopolies ALWAYS create complexity (Zero to One - Peter Thiel). Here are what your typical bullshitting articles say: - Cancer is a living entity, like a virus. - Cancer is way more complex than we thought! - Cancer keeps growing and it's unstoppable - We still don't know how cancer is created

Start here: https://nutritionfacts.org

I lived this on my own skin with my mum getting cancer. This article has been completely lobbied by the pharmaceutical industry. No doubt.

There is certainly no scientific consensus that GMO/transgenic food causes cancer; if anything the community believes it does not. Perhaps we currently have flaws in implementation (e.g., Roundup-ready crops could lead to increased human exposure to the herbicide), but why is GMO itself a problem? Adding genes from harmless cold water fish to harmless standard tomatoes in order to produce cold resistant tomatoes, for example, seems to be a safe procedure. I would love to be pointed to credible studies that contradict my way of thinking about this.

You are right about consensus. But I invite you to investigate how big the cancer industry is in worth/year and historically how it all has been created, and who owns it. Also to study the history of the American food industrialisation. Also to investigate who owns and influences the top media sources in the world. To be clear the point I was trying to make wasn't "GMO is the only cause of cancer" or "Genetical modifications on foods produce cancer". My point is "Current GMO products consumption is one of the contributors to cancer" - I'm not stating how, I'm stating that it has an effect. I know far more about the effects of meat and dairy than I do about GMO. But somewhere to get started would be here: http://nutritionfacts.org/video/are-gmos-safe-the-case-of-ro... (The page has links to all the scientific sources cited)

This is pure quackery with no scientific basis.

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