The future of medicine, as seen in 1987
May 25, 2012 8:55 AM   Subscribe

 
If only money hadn't been more important.
posted by batmonkey at 9:13 AM on May 25, 2012 [4 favorites]


That's funny, because I was just reading an article in the Health section on Yahoo! News mentioning that accurate predictive disease risk profiles based on genetic data are still quite a ways off.
posted by infinitywaltz at 9:18 AM on May 25, 2012


That is long and complicated but well-worth reading. I have read nearly all of what Darwin has put on his chronosphere blog since he began putting stuff up there over a year ago. He is not the most concise writer and he is definitely a fringe thinker but he is a very provocative thinker. We could use a bunch more like him if you ask me.

The prediction and retrospective post specifically was covered in overkill level detail on LessWrong in December 2011. Link.

My favorite post of his with conclusions I wholeheartedly disclaim was his post fucked. Needless to say that link is NSFW. This is the most eloquent most modern survivalist manifesto I have ever seen. He doesn't just talk the talk either. Right now he is living off the grid halfway between Phoenix and Los Angeles.
posted by bukvich at 9:24 AM on May 25, 2012 [1 favorite]


Someone else can fill in the blanks for non-psychiatric disorders, but let's take a look at the mental illness predictions.
"Fairly predictive tests for Alzheimer's disease, schizophrenia, depression, some malignancies, heart disease, and most of the rest of the major killers and disablers will probably be in place by 2000 to 2010. Many if not most of these ailments will be assessable in terms of a very sophisticated genetic risk profile which it will be possible to generate in infancy or childhood (or in utero)."
A recent genome-wide association replication study failed to replicate the vast majority of single-nucleotide polymorphisms (SNPs) for major depression, and recent a genome-wide meta-analysis found essentially no strong evidence for predictive SNPs in major depression. The genetic picture is slightly more optimistic for schizophrenia, but there are still no known biomarkers for schizophrenia (i.e., predictive tests), and the genetics of schizophrenia overlap substantially with the genetics of bipolar disorder and affective psychosis. To date, the single best predictor of conversion from mild cognitive decline to Alzheimer's disease is not a blood test, genetic profile, or any other biomarker, but rather pencil-and-paper cognitive tests. To summarize, the NIMH dubbed 1990-2000 as the "Decade of the Brain", yet to date there are still no reliable and well-substantiated, let alone widely employed biomarkers, genetic tests, or imaging tests that are predictive for the vast majority of psychiatric illnesses. I say "vast majority" only as a hedge; so far as my personal knowledge goes, there are none, full stop.

Needless to say, "diagnosis by brain scanning (metabolic MRI) and chemical analysis of cerebrospinal fluids" is anything but "commonplace" 20 years later. To have invested in that fantasy in the early 1990s must have involved taking for granted that our understanding of psychiatric disorders was on solid conceptual footing and that the diagnostic categories of DSM III, IV, IV-TR, and soon, DSM-5 manage reasonably well to "carve nature at the joints". If that were right, then those categories would be a good foundation for biological research. If that was wrong and we were really whistling in the dark, then it ought to come as no surprise that noise came out of noise, and that biological psychiatry has yet to put its money where its mouth is.

One prediction that is true regarding psychiatric disorders is this one: "Expect 2nd and 3rd generation drugs and combinations thereof for treatment of depression and psychosis by the late 1990's." What the prediction failed to get right was that second generation antidepressants and antipsychotics are on average no more effective than 1st generation drugs, and that they are barely more well-tolerated.

Setting all that aside, what I'm most disappointed about is the unsatisfied expectation of "real aphrodisiacs and potentiators of sexual pleasure to be available by the late 1990's, if not before". This failure, my friends, stings most mercilessly.
posted by mister-o at 9:37 AM on May 25, 2012 [6 favorites]


Setting all that aside, what I'm most disappointed about is the unsatisfied expectation of "real aphrodisiacs and potentiators of sexual pleasure to be available by the late 1990's, if not before". This failure, my friends, stings most mercilessly.

I'm pretty sure the internet exists.
posted by srboisvert at 9:49 AM on May 25, 2012


First of all, thanks mister-o, that was really interesting.

Secondly, this post makes me feel sort of out of it. From what I gather from the link, this is a transhumanist (and a founder of cryogenics) who turned out to be too optimistic in his predictions. This seems pretty common (see Dean Kamen). Is there a reason this prediction is particularly salient? I mean Mike "Darwin" isn't a doctor, or anything, so why do his predictions matter?
posted by blahblahblah at 10:32 AM on May 25, 2012


They're not aphrodisiacs, but we did get ED drugs that help with sexual function.
posted by Pruitt-Igoe at 11:29 AM on May 25, 2012 [1 favorite]


Most of the easy diseases, one gene one disease type, have been identified. Understanding mechanisms involving multiple genes affecting the way a disease is expressed is much more difficult. I remember that twenty years or so ago we thought that the genes involved in Deep Vein Thrombosis were all discovered: now we are back at "We have a partial understanding on the mechanism." And if we have to consider epigenetics or environmental influences....
posted by francesca too at 12:16 PM on May 25, 2012


Yeah it turns out that shit was way more complicated than we thought.
posted by humanfont at 12:50 PM on May 25, 2012 [1 favorite]


"A large number of illnesses which are poorly understood today will be well-characterized the next decade and will be easy to diagnose very early in their development or even before they develop because they will be found to have direct or indirect genetic cause."

Sadly, a lot of this is a crock. There are tons of places, like 23 and me, that can map your genes, but so what? The number of diseases you can predict getting with any certainty based upon your genes, is a tiny, tiny, tiny subset of all diseases (did I mention "tiny"?). The problems is that for most gene associated diseases, it's the interaction between genes and the environment - an interaction that in many cases cannot be controlled (by f.ex. diet or lifestyle factors like exercise or drug abuse). So you know you have genes X that predispose you to a disease. Now what? You can't control the factors which might trigger the activation of those genes, the degree of activation, the consequences of activation in the presence of variable environmental inputs, and really, at best you can find out that, well, you're "22%" more likely to get disease X compared to the rest of the population. What do you do with this information? Just having the genes means very little, since there is literally no way to predict the expression of those genes given random environmental uncontrollable interactions which may involve thousands of variables - it may be mathematically impossible.

It was a simplistic dream to imagine that if you could sequence your entire genome, you'd have essentially the blueprint for all genetically driven diseases you could then control. It transpired to be NOTHING like that.
posted by VikingSword at 1:33 PM on May 25, 2012


I wonder if Darwin realizes that tests for the three major killers -- HIV, TB and malaria -- have been available for years.
posted by docgonzo at 1:44 PM on May 25, 2012


It was a simplistic dream to imagine that if you could sequence your entire genome, you'd have essentially the blueprint for all genetically driven diseases you could then control. It transpired to be NOTHING like that.

Actually, I think this kind of pessimism is probably a little premature. We're still very, very much in the infancy of personalized genomics and GWAS is just one approach to making sense out of the genome.
posted by en forme de poire at 1:51 PM on May 25, 2012


Genetics is one thing - very complicated and still in its infancy, yes - but there are tests available for more conditions and disorders and diseases than we can imagine, but it doesn't make any difference now and it won't later because no one can afford them. We continue to fund medical research, at tremendous cost, but when the research develops a test or screening method for a given disease, it's almost never put to use for the benefit of the general public - and it's always about the money. Only those with the most expensive insurance coverage can take advantage of the newest technology and medicines - assuming they're available to anyone at any cost. Prenatal genetic screening has been around for many years, but only a select few are given the option to use it. What, really, would be the benefit of having such a full genetic profile - one that specified a large range of disorders and gave one's odds of each - when the cost of that profile would overrule its use?

I think this is another of those wonderful ideas for the future that we all had oh-so-long-ago, before we became aware of how much control the banks and insurance companies would have over each of our lives. I remember how we used to actually believe that the real goal was to help people have healthy, good lives instead of to spend as little as possible to keep the numbers looking good so there could be little complaint.
posted by aryma at 6:49 PM on May 25, 2012


We continue to fund medical research, at tremendous cost, but when the research develops a test or screening method for a given disease, it's almost never put to use for the benefit of the general public - and it's always about the money. Only those with the most expensive insurance coverage can take advantage of the newest technology and medicines - assuming they're available to anyone at any cost.

Right now, sequencing a genome today can be as cheap as $5,000. The cost per basepair is falling at a rate faster than Moore's Law, and while extrapolation is obviously non-trivial it seems reasonable to predict that the goal of a $1,000 genome is not that far into the future. And there are people who want to take it even lower, into the hundreds of dollars. I understand that's still expensive (especially out of pocket), but the researchers developing the technology want sequencing to be as cheap and ubiquitous as possible, because the predictive power of the genome increases with the number of genomes you have. In other words, the benefit to any individual scales with the availability. I share your frustration with banks and insurance companies, but I think the overall picture is not so bleak.
posted by en forme de poire at 9:47 PM on May 25, 2012


They're not aphrodisiacs, but we did get ED drugs that help with sexual function.

What I find interesting about this is how the availability of these treatment has effectively redefined sexual dysfunction in physical terms, rather than the long-popular psychological. I remember a major article published in, I believe, Playboy about impotence, in which the author had to first surreptitiously find a doctor who would talk to him about it, and ended up with a diffident urologist who mostly seemed to want him to go away. At the time, the prevailing opinion seemed to be that therapy was the key (or perhaps, one suspects, sexual orientation) and body doctors weren't interested in finding causes or cures. Then Viagra was discovered, infamously by accident, and the whole thing flipped back around to physical overnight. There's still a role for couples therapy, but it's no longer regarded as a secret shame that implicates one's lack of manhood (or, again, suspected sexual orientation issues). Sociologically, I find this fascinating.

Staying on point, this sort of reframing goes on all the time, and makes life difficult for futurists. Will your prediction even make sense in twenty years?
posted by dhartung at 12:51 AM on May 26, 2012


Knowing Genetic Makeup May Not Significantly Improve Disease Risk Prediction

"Harvard School of Public Health (HSPH) researchers have found that detailed knowledge about your genetic makeup -- the interplay between genetic variants and other genetic variants, or between genetic variants and environmental risk factors -- may only change your estimated disease prediction risk for three common diseases by a few percentage points, which is typically not enough to make a difference in prevention or treatment plans. It is the first study to revisit claims in previous research that including such information in risk models would eventually help doctors either prevent or treat diseases."

[...]

"For breast cancer, the researchers considered 15 common genetic variations associated with disease risk and environmental factors such as age of first menstruation, age at first birth, and number of close relatives who developed breast cancer. For type 2 diabetes, they looked at 31 genetic variations along with factors such as obesity, smoking status, physical activity, and family history of the disease. For rheumatoid arthritis, they also included 31 genetic variations, as well as two environmental factors: smoking and breastfeeding.
But, for each of these disease models, researchers calculated that the increase in risk prediction sensitivity -- when considering the potential interplay between various genetic and environmental factors -- would only be between 1% and 3% at best."

The study appears online May 24, 2012 and will appear in the June 8, 2012 print issue of The American Journal of Human Genetics.

Well, that was fun while it lasted. Now about those flying cars...

It's hard to make predictions, especially about the future. - attributed to Yogi Berra.
posted by VikingSword at 12:22 PM on May 26, 2012


Well, that was fun while it lasted. Now about those flying cars...

VikingSword, that paper (Cell Press, therefore paywalled, sorry) is about one approach to getting more information out of one type of genomic mapping, the Genome-Wide Association Study (GWAS).

The idea behind GWAS is that you genotype a bunch of people (though usually the entire genome isn't sequenced, just "markers" along the genome), then look for an association between a locus and a disease. GWAS studies have come under fire lately because they typically have low power, so-so reproducibility, and usually only explain a small fraction of the heritability for complex diseases. One hypothesis was that interactions between loci would be significantly more powerful. This paper does simulations to show that even in the best case, interactions aren't going to get you very much additional power. From the paper:
Although multiple biologic interactions among GWAS-identified risk loci and clinical risk factors are likely to contribute to the etiology of many common diseases, this study suggests that the identification of statistical interactions among these factors might have a modest impact on risk prediction and discrimination for common complex diseases.
This paper is interesting and potentially very helpful to the field, but it is in no way evidence that personal genomes don't contain useful information for preventing/treating disease (see, for example, the counterexample I linked above). GWAS is one of a variety of strategies to understand the genome, and looking for nonlinear gene-gene and gene-environment interactions is one of a variety of strategies used in GWAS.
posted by en forme de poire at 3:16 PM on May 26, 2012


« Older "You look nice. Who dressed you, the Great...   |   Pun intended. Newer »


This thread has been archived and is closed to new comments