Hacker News new | comments | ask | show | jobs | submit login
Cancer Progress: Much More Than You Wanted To Know (2018) (slatestarcodex.com)
219 points by sethbannon 13 days ago | hide | past | web | favorite | 69 comments

This author writes so well. The introduction, in just 169 words, tells you exactly what the article is about. I wish more articles had such good introductions.

Rather than starting with some personal anecdote contrived for some sort of empathy, as is the common style these days!

Thank you.

Or something like:


“I’ve always hated coffee” said Joe Bloggs, sipping his chai latte as he sits somewhat dishevelled in a grey overcoat that insulates him against the wintry London weather. The Hackney coffeehouse he chose for this interview specialises in Peruvian blends, an apt reflection of his latest film set in South America.

... (insert interview)

... Joe, for one, will be hoping reviews for his next film is more of a blend than a roast.


I've never realized how much I hate that meandering and how difficult it's becoming to go through those overlong pieces until I saw it verbalized in this thread.

I know the kind of article you mean. I always exit them quickly. If that style helps in SEO, surety it is offset by people not wanting to read that kind of crap. Funny post, though.

I really don't think people do it for SEO reasons. People that do that for SEO tend to write text that is on-topic but explain simple things in drawn out ways using lots of related keywords.

I think people that write articles that start with long anecdotes (which are usually linked back to in the end) are trying to emulate the writing style you'd usually see in newspapers and magazine going back decades. It doesn't work well on the web I feel when there's an abundance of information where you're trying to filter through what's useful to read and what isn't. It's different when you're e.g. reading a magazine and have limited articles to pick from.

> It doesn't work well on the web

It doesn't work in magazines and newspapers either, but certain writerly circles (mostly American ones) consist of telling each other that this kind of fluff is a good idea.

You will find this kind of writing in Time magazine, but not in The Economist. Which is one of many reasons why TE is a better paper.

I call that 'NPR style' and it's literary cancer.

Literary, but not literally, thank heavens.

Amateur writers often mix up journalistic with literary writing. Both has its place but when people can't make up their mind they end up producing a chimera. Their writing doesn't follow an inverted triangle principle while at the same time it fails to build an arc of tension. The lede is buried somewhere in the middle.

Pervasive SEO advice is also to blame. "Google loves long-form" is the common mantra. Just add a personal anecdote at the beginning to reach the target word count.

I don't blame them on the one hand, because truly neutral and impersonal journalism (think Reuters) look soul-destroying to write - if you're the creative kind. But people working for real news agencies (not the ones dependent on fluff) have to stick to facts and impersonal, unopinionated stories.

This is one of my biggest pet peeves. Clickbait title with an unrelated story in the beginning.

This is also how most non-fiction books begin.

Agreed. Worth checking his other posts as well, he has some fantastic articles on a wide variety of subjects.

This is where the tradition of an 'abstract' at the beginning of an academic paper comes from. Our blogging world has grown up enough to rediscover the ways of the old!

This alone makes me want to click on that article. None of that "these first couple paragraphs are filler until we actually start talking about the topic, and even then you're still lost until about 2/3rds of the way through."

Also the conclusion answers the question posed in the introduction without beating about the bush. If you only read the introduction and the conclusion and you still have learned something. The lede in plain sight - good job.

> I wish more articles had such good introductions

Check out the Financial Times. Major motivation behind my subscription is their to-the-point writing style.

When I read this in August it was fascinating. Since then an immediate family member was diagnosed with cancer, and it all feels so much more real and painful :(

Hang in there ... I've ridden that particular train[1][2]. You have to be strong, positive for the family member. Know you are not alone, there have been many people who've gone through this, and as I discovered, it helps to talk to them.

I am happy to report that as of 29-Dec, we are at 7 years post op. On the 1 year anniversary of when we got the diagnosis, I saw an xkcd strip[3]. That gave me some hope of an eventual return to normalcy. A few years later, I saw one of the followons[4].

Little things like that help. They really do.

[1] https://scalability.org/2011/11/ot-it-really-focuses-your-at...

[2] https://scalability.org/2012/01/ot-what-is-and-what-should-n...

[3] https://xkcd.com/1141/

[4] https://xkcd.com/1928/

I'm always confused at such comments, considering how variable and different cancer can be (which is discussed in the article by the way), and how vague people seem to like to be about it ; I wish more precise denominations would be the norm.

Because when I read such encouragements I imagine them being to say, someone with a pancreatic cancer (who is short term screwed considering the odds), from someone with maybe a breast cancer or prostate cancer (rather common and fairly beatable nowadays). And that feels very odd.

The message of encouragement is to the family, "you are not alone, and you should reach out to others to help." Regardless if it is easily treatable, or an aggressive terminal cancer.

When we went through the process, there was a huge support system for my wife. Basically everyone was telling us positive messages. Some of them were wrong, but this was intended to keep her spirits and hope up. It turns out that you heal better when you have hope.

Meanwhile, I had to calm the fears of our then 12 year old daughter, help her to understand what was occuring. Simultaneously, I had to do this for my wife. And her family, and my family. All while dealing with my own worries.

It's not just the patient you have to take care of. It's everyone.

This is why knowing that others have done this, and survived, is so very important. Knowing that you can talk to them, voice your fears that you dare not do in front of your loved ones ... you need this when you are going through this.

Cancer sucks, no matter which one you deal with. I keep getting back to xkcd, as Randall and his wife went through it. Part of his "you are not alone, and it helps you to talk about it" are his strips. All the cancer ones are online along with explanations[1]. I have some favorites, that help me during this process[2][3] and especially [4] which came out last year. Read #4's panel 19 and 20 explanation to understand why this encouragement is so important.

[1] https://www.explainxkcd.com/wiki/index.php/Category:Cancer

[2] https://www.explainxkcd.com/wiki/index.php/931:_Lanes

[3] https://www.explainxkcd.com/wiki/index.php/881:_Probability

[4] https://www.explainxkcd.com/wiki/index.php/1928:_Seven_Years

Currently riding the train for my mum who is doing fairly well after ~9 years with one recurrence, and my dad who isn't doing so well @ 4 years and two recurrences.

There's another xkcd strip that, while not as optimistic, is just as important: https://xkcd.com/931

Another person who wrote about this stuff, was Richard Feynman. His first wife, Arlene, had non-Hodgkins Lymphoma. He has a fairly long section of "Surely you are joking Mr. Feynman" on this. He talks about what they went through, the antics, and so forth. And he talks about the whole end of life process, as her cancer wasn't one that many people survived in the 1940's. His world view of making every second count, and not sweating the little things was fantastic.

Just make every moment count with them, regardless of the prognosis.

My mom was diagnosed 1.5 years ago, and the uncertainty of it all is the worst part. It soothes my heart to know I'm not going through this alone.

You aren't alone. You need to stay strong and positive for her, but you need to be able to talk with others who've tread this path before. Many have.


Thanks for your comment. I appreciate it :)

I was diagnosed with Stage 4 cancer in September/October. Basically it aggressively grew within a few months. I can tell you without any doubt that huge advances are coming. I expect a hockey stick in the survival rates based on new understandings of all cancers. We're right on that edge of a manageable disease. Hang in there.

I would have thought that smoking cessation and earlier detection are - in a sense - medical advances?

Medicine pushed for these, as the most effective ways of fighting cancer. And it's working.

The same will hopefully happen to air pollution in cities, as a public health problem.

The same will also happen due to diet changes in the next years. More and more people are severely cutting back on meat intake, specially smoked and processed meats, and it will likely have an impact on a multitude of types of cancer down the road.

It's not meat, it's sugar -- specifically fructose. See Dr. Robert Lustig's important work on this topic.

Which types are caused by which meat though? I haven't seen a strong correlation between the two (unlike smoking and cancer). The charts on that article don't really show much change in most types of cancer over the past 50 years, despite changing diets and a higher availability of meat and processed foods.

on the "negative" side: we're living longer so odds of cancer increase, and we're obese. Or close to /above it.

I think "the War on Cancer" is a great teaching moment for technologists: be careful with huge promises that millions latch on to: they won't forget, and when you fail, it will be (very, very) hard to win their trust back (thus, in the case of cancer, the continued fallacious narrative which the author so nicely deconstructs that we aren't making progress on cancer)

Well that's fantastic news, and I'm glad to hear it. I expect we will more or less beat cancer someday, and heart disease too. Probably long after I'm gone, but I'm happy for future generations that will no longer have to fear this scourge. (Assuming we don't wipe ourselves out some other way first.)

The sad thing is that most instances of cancer and heart disease are down to diet and lifestyle. We already have the "cure" to these conditions, but people won't take care of themselves.

No, you're absolutely incorrect that most cancer is down to lifestyle. As someone with cancer, I don't smoke, I never really drank, there's no history in my family for my type of cancer, and I haven't eaten meat in years. I've exercised more than the average person. Most cancer is due to dumb luck. The only cancers we know of that are attributable to lifestyle are basically lung cancers. Breast, Colon, Pancreatic, Stomach, Leukemia, etc. are all just dumb luck or family history. You obviously didn't read the article. We're only better at things we know how to screen for. Survival rates in Asia for Stomach cancer are significantly better than the US because they have regular screenings for it. The typical cases are salty diets, but Americans still get this even though we've changed our diets.

There's no "cure" for cancer as of yet, you can't even guarantee prevention. It's a combination of genetic damage and an immune system that gets prevented from stopping the proliferation of cancer cells because the cells mask themselves as healthy cells. In effect cancer is an immune disease, but that's not fully accurate. Saying it's down to lifestyle is like telling people who got the plague they should have just moved. We have barely more understanding about cancer causes than doctors at the time did about disease.

No, we can barely convince people to take prescribed drug; we don't have effective intervention to change diet and lifestyle. If we had effective intervention to change people's diet, that would qualify as "cure" and would be a great boon; we don't.

Is that really true?

I mean, it might well be true that most people could live an extra 20 years with a suitable change of diet and lifestyle.

But even if that's the case, isn't everyone going to die of cancer or heart failure at the end of those extra 20 years?

We now (since 2015) have approved PCSK9 inhibitors. We didn't know about PCSK9 until 2003. Gene to drug in 12 years! Yes, it's very exciting and I am very hopeful about heart disease.

I'm not sure I buy the logic of ignoring improved detection when it comes to survival chances. As the article and pretty much everyone else tells you, early detection is the key to a better prognosis in most cancers. If all malignancies proceeded to stage IV before detection, survival rates would be very low despite the medical advances we've made. Successfully treating a cancer, usually indicated by being cancer-free 5 years post-treatment, is not discussed in this article at all and seems like it would provide some evidence for or against cancer treatment getting better from the perspective that matters most -- the likelihood that a diagnosed cancer will kill you.

As someone who has quite unsettling health anxiety and somewhat of a cancer-phobia, these articles kinda calm me. But what doesn't calm me are usually the comments because it seams that cancer is really all around me, and it's just around the corner, just waiting to snatch me. I don't have any cases in my closer or further family, but I'm frightened I'm going to be the one... The 1-in-2 statistics doesn't soothe either...

I have 1000s of moles and get checked once a year. I live in crippling anxiety that one they, melanoma will pop up and I won't notice early enough. So I get your fear _very_ well.

All we can do is to reduce the probability by causing less DNA damage through all the various means. Even then, pure chance can still strike because this is what life is at its core: a cumulation of (seemingly?) random chance events.

Yep, I have quite some moles myself and get them checked. Although, thanks to the place I live in, I don't get too much UV exposure and try to protect myself well. That's the thing, I'm not too frightened by the things I can control, it's those that do not have so much prevention, such as pancreatic or leukemia... I think I did some permanent damage to my psyche by just reading too much about those things and too much statistics. Sometimes I see myself as just a number when reading these articles.

Don't get me wrong, I'm going to therapy and I'm not depressed. Just scared :) .

> The only contrary data point I can find is this study of laryngeal cancer, which finds worsening stage-specific survival rates for high-stage laryngeal cancer since 1977.

I seem to recall someone theorizing that the increased prevalence of oral sex is responsible by putting HPV deeper into the throat and thereby increasing the prevalence of throat cancers.

HPV vaccinations should cause a corresponding down tick shortly if that is true.

> Recent data suggest that individuals that come to the disease from this particular cause have a significant survival advantage,[20] as the disease responds better to radiation treatments than tobacco caused disease.

From Wiki + https://www.mdanderson.org/newsroom/2010/06/hpv-positive-tum...

It does seem indeed that there is more prevalence of laryngeal cancer due to HPV however according to that link above, it should not worsen survival rates but rather make them better, wouldn't it?

can't wait to see how much improvement whole genome analysis will bring to patient outcome.

I wonder how widespread was cancer in, say 17-18th century. Obviously there would not be much technology do diagnose it, but still, if someone has any references, that'd be of interest.

It would be by necessity less of a killer as people didn't live as long. The average age of death in those times was late thirties, far lower than the common age of cancer in the mid to late 60s. Keep in mind that sanitation, food preparation, medical science, and clean water were all huge advances in life spans and general health.

The PBS series "How We Got to Now" is a great resource for the huge advances in health made in the 19th and 20th centuries.

I always thought this plot: https://slatestarcodex.com/blog_images/cancer_tobacco.jpg

Looked eerily similar to this one: https://i.ibb.co/cLbTRfM/nuctest.png

Source for nuclear test data: https://www.armscontrol.org/factsheets/nucleartesttally

And before there were nuclear tests, there was a big fad to expose yourself to radiation early in the 1900s: https://www.popsci.com/scitech/article/2004-08/healthy-glow-...

Is the argument that the nuclear tests, etc had zero effect on lung cancer rates? If not, what was it?

They're not very similar. The nuclear tests graph starts at 0 at 1950 (when the cigarettes-smoked graph is three-quarters of the way to its peak). Lung cancer incidence has been rising steadily and pretty much linearly since the 30's, until the 80's.

If nuclear tests are the main reason for lung cancer, then why did lung cancer incidence start rising before nuclear weapons were invented?

If nuclear tests are the main reason for lung cancer, and their effect is due to background radiation, why weren't all cancers increasing steadily rather than simply lung cancer?

>"If nuclear tests are the main reason for lung cancer, then why did lung cancer incidence start rising before nuclear weapons were invented?"

It became easier to diagnose, or perhaps more trendy? It is actually really hard to distinguish between lung cancer and tuberculosis.[1] At ref 2 you can see tuberculosis rates dropping from 200/100k in 1900 to near 0 in 1980. In the meantime lung cancer rose from near 0 to ~100/100k deaths.[3]

Also, as I mentioned, there was a fad to expose yourself to radioactive stuff starting about 1900 that only died out in the 1960s.[4]

So, could be lots of things in combination.

>"If nuclear tests are the main reason for lung cancer, and their effect is due to background radiation, why weren't all cancers increasing steadily rather than simply lung cancer?"

My understanding is they were.[5] The lungs may especially vulnerable to radioactive dust though, if they had already "accounted for", as the lung cancers due to smoking, then this could have been missed.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3876596/

[2] https://en.wikipedia.org/wiki/History_of_tuberculosis#/media...

[3] https://slatestarcodex.com/blog_images/cancer_types.jpg

[4] https://www.popsci.com/scitech/article/2004-08/healthy-glow-...

[5] https://www.cancer.gov/about-cancer/causes-prevention/risk/r...

The usual strategy is to control for smoking. If nuclear test rate correlates with lung cancer rate, but does not correlate with non-smoker lung cancer rate, nuclear test is not causing lung cancer.

Epidemiology is a field of study about how to do this right. One of the great achievement of humanity in the 20th century was advance of epidemiology.

But here you go:

>"the lung cancer mortality rate has risen substantially between 1914 and 1968 among persons who never smoked cigarettes. For white males the relative increase for ages 35--84 years has been about 15-fold; the relative increase for ages 65--84 years has been about 30-fold." https://www.ncbi.nlm.nih.gov/pubmed/285289

Yup, epidemiology of non-smoker lung cancer is an active area of research. (Epidemiology of smoker lung cancer is not; it's very well established.) I haven't seen this old article, but it sounds perfectly right; the quoted result is probably due to diagnostic criteria.

Lung cancer in never smokers (2009) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3170525/ shows the field in confusion. Radon, indoor air pollution, second-hand smoking are named as possible sources, but it's still not satisfactory.

That's interesting -- doesn't seem to have a graph, but states a significant increase between 1914 and 1967 in nonsmokers. There doesn't appear to be any control for secondhand smoke, however, which could also be a significant factor over that time period.

Sure, another thing to consider is fallout incorporated into tobacco and then smoked...

But do you see why I didn't care if a correlation existed? There are a bunch of ways to explain such a correlation. That line of reasoning was based on the affirming the consequent fallacy[1]:


  H: Nuclear tests are a significant factor in lung cancer rate 

  O: We should still see a correlation between lung cancer incidence and nuclear test frequency [in non-smokers]
We can't conclude "O is true, therefore H" (since there are a million potential reasons for O besides H), so we don't learn anything from it. Now if O had been some very specific prediction unlikely to arise from other hypotheses, I would care.

In terms of Bayes' Rule, I only care if p( O|H[0] ) dominates all the other possibilities (H[1:n]) in the denominator. The mere existence of a positive correlation isn't specific enough for me to care:

  p( H[0]|O ) = p( H[0] )*p( O|H[0] )/sum( p( H[0:n] )*p( O|H[0:n] ) )
[1] https://en.wikipedia.org/wiki/Affirming_the_consequent

Yes, epidemiology is not deductive, and in the end it depends on whether observation is "specific enough" for you st unlikely to arise from other hypotheses, which in turn depends on your prior, which is informed by laboratory results in addition to field results. Epidemiology is a truly fascinating subject.

> "epidemiology is not deductive"

-Not sure what you mean by this, it is quite possible to deduce precise predictions from an epidemiological model. See, eg, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2007940/

>"it depends on whether observation is "specific enough" for you [is] unlikely to arise from other hypotheses, which in turn depends on your prior"

-This has to do with the likelihoods, p( O|H[i] ), not the priors: p( H[i] ).

>"[the prior] is informed by laboratory results in addition to field results"

- Yes

Affirming the consequent is a deductive fallacy. It is not a fallacy in probabilistic reasoning, because the consequent is an evidence of the antecedent.

Re: laboratory results, it is noteworthy smoking and non-smoking lung cancer have different molecular signature. Lung cancer probably should be considered two separate diseases, and this probably overstates smoking connection because some lung cancer by smokers is of non-smoking kind and not caused by smoking. This can be better understood by including such signatures in epidemiological study.

>"Affirming the consequent is a deductive fallacy. It is not a fallacy in probabilistic reasoning"

Nevertheless, people commit that fallacy all the time in epidemiology. They say stuff like "if smoking caused cancer, then people who smoked would get cancer more often". Then if there is a (statistically significant) positive correlation between cancer and smoking, they conclude smoking causes cancer.

That is quite obviously affirming the consequent. I explain the (completely different and correct) probabilistic way to think about this situation using Bayes' rule in my post.

It has nothing to do with arbitrary cut-offs for telling us whether a correlation exists or not. Instead it requires comparing the relative performance of the various explanations (at least the top few candidates since extremely unlikely ones can be dropped from the sum in the denominator).

I don't think anyone serious about epidemiology says stuffs like that. On the other hand, it is true many people interpret what epidemiology says as such.

Saying smoking is a risk factor of cancer is very different from smoking causes cancer. As you said, you must consider alternative explanations. On the other hand, people can disagree about what alternative explanations are "extremely unlikely" and "can be dropped", that's what I meant by prior. For example, nuclear test causes cancer, is probably extremely unlikely.

Smoking being a risk factor of cancer is an evidence of smoking causing cancer. Being evidence of causal relation is completely different from being sufficient to prove causal relation.

>"I don't think anyone serious about epidemiology says stuffs like that."

Pick a paper, I bet they do.

> We can't conclude "O is true, therefore H"

Definitely, but if O had been false we might have learned something. More interestingly though:

> another thing to consider is fallout incorporated into tobacco and then smoked

That's an interesting thought, I'd never considered that.

My expectation for beta and gamma emitters is that they would be unlikely to cause cancer specifically in the lungs, because lungs aren't very dense and would have a pretty low interaction cross section for such penetrating radiation. For comparison, I-131 -- a beta emitter common in fallout -- is a significant risk for thyroid cancer only because iodine is biologically concentrated in the thyroid.

However, if there were significant fallout of alpha emitters, which were deposited in tobacco, this could conceivably deliver a significant localized dose to the lungs. Alpha emitters are more or less only dangerous internally, but smoke particulates that become lodged in the lungs could do the job. There appears to be a fair amount of literature on this -- Naturally occurring Po-210 is apparently still considered a danger today [1], and has been known to the cigarette companies since the 60s [2]. Some apparently comes from decay of natural U (via a few intermediaries) in phosphate fertilizer.

A figure in one paper I was able to find [3] suggests up to 100 pCi/kg Pu-239 and Pu-240 (both alpha) in leaves and grass in Finland circa 1963, declining to less than 5 pCi by 1967. The modern Po-210 alpha activity in tobacco appears to be much higher than either of these values (perhaps several hundred pCi/kg in dry tobacco, if not more), but I couldn't find good data on variation over time. For Pb-210 (beta) there is more data, but seems to be a lot of scatter and not so much of a clear trend through time [4].

[1] https://scienceblog.cancerresearchuk.org/2008/08/29/radioact... [2] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2509609/ [3] https://www.osti.gov/biblio/5271041 [4] https://doi.org/10.1016/j.jenvrad.2011.01.005

My impression is that second-hand smoking is considered insufficient to explain epidemiology of non-smoker lung cancer; see my other comments.

>"The usual strategy is to control for smoking. If nuclear test rate correlates with lung cancer rate, but does not correlate with non-smoker lung cancer rate, nuclear test is not causing lung cancer."

I'm not really concerned with whether there is, or is not, a correlation, I already know there is a correlation.

I am concerned about: How much of the increase in lung cancer incidence we see in the data could be accounted for via exposure to radiation from nuclear tests? I doubt it is zero, but it could be relatively negligible I suppose.

The timing of it does overlap quite well though.

This is missing their point: you can distinguish the two potential causes by comparing lung cancer incidence rate through time for smokers versus non-smokers.

If nuclear tests are a significant factor in lung cancer rate, we should still see a correlation between lung cancer incidence and nuclear test frequency when looking at people with no exposure to cigarette smoke. If the correlation in the general population just a coincidence, we should not.

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