Tuesday, August 31, 2010

The wishing well: money down but wishes unanswered

A sobering study of studies reported in the NYT by Gina Kolata, one of their best science reporters, has shown definitively that no findings about prevention of Alzheimer's Disease (AD) can be shown to have worked in any substantial way.

AD is a devastating disease and NIH and other research funders have poured money into the wishing well, essentially chasing every idea that somebody said s/he thought (wished) would lead to a way to avoid this disease.  Keep exercising.  Keep doing crossword puzzles or mental challenges.  Eat lots of veggies.  Nothing.

In some senses this really has been wishful thinking.  It never seemed highly credible that some simple lifestyle change would have such a major effect (but those of us who do crossword puzzles and eat a balanced diet could hope, of course!). There are many problems.  They begin with problems diagnosing AD, especially before a post-mortem.

Even then, some of the chemical evidence, such as deposits of beta-amyloid molecules in specific parts of the brain were thought to be chemically responsible, and diagnostic.  Various trails of genetic evidence (genes involved in or affecting amyloid or other particles in the brain) were followed.  The diagnostic criteria are apparently not even that definitive.

Possibly AD is, like most things with complex effects, a mix of similar traits.  Almost certainly AD is along a spectrum of mental-function, a point on a continuum that we define as disease.  It may have many causes because there's not actually an 'it'.  'It' is more likely 'them.'

This is the challenge of complexity.  Evolutionarily, even if we can believe evolutionary stories in situations like this, AD occurs past reproductive ages by and large, so natural selection could not remove those who would get it.  One can contrive, or contort, stories to get around this:  those who are doomed in our long-living society to get AD have some reproductive charisma earlier in life (yes! that kind of argument has been raised).

But if the likely fact is true that AD is past the effects of natural selection because it strikes too late in life, then anything that could lead to it could circulate in the population without being removed by selection. If there are many ways to lose memory function, perhaps thousands of ways, then that's what we'll see today.  It seems consistent with the evidence.

However, if some physiological pathways are truly causally involved, then it may be possible for preventive (even gene-product-based) therapies.  Many investigators are trying and the news media (perhaps especially the less responsible reporters) regularly tout the next miracle cure.

Anyone with a relative suffering from AD wishes that some such miracle will in fact be found -- and now.  If we can keep ourselves alive many years beyond our proper sell-by date, then it will be important to keep at least our mental functions in decent order, even if everything else falls apart.  Naturally, rather than accept earlier Exits, we keep throwing our money down the wishing well.

Whether we're wishing for something that's not possible in the face of causal complexity, or a stunning treatment will answer our wishes, nobody knows.  That's what keeps us tossing resources down the well.


Dan! said...

"Almost certainly AD is along a spectrum of mental-function, a point on a continuum that we define as disease. It may have many causes because there's not actually an 'it'. 'It' is more likely 'them.'"

I'm often confronted with this same thought when I consider the challenges facing autistic spectrum disorder researchers. It seems very clear to be that autism is likely many disorders - with many types of causes - yet with symptoms manifesting themselves in very similar ways. Some causes may be genetic, while others are probably influenced by environmental factors - which is probably why association studies keep coming up short.

I wonder if researchers of this type of complex disease need to change the focus of their research. Maybe they need to begin, not by looking at how diseased individuals are similar to each other, but instead at how they differ from healthy relatives. Although for any significant revelations to come out of this type of study, we'd have to understand the healthy state of the body alot better than we currently do. My guess is, there are many ways to not have a disease.

Anne Buchanan said...

Yes, autism and many other diseases and disorders. In fact, not long ago, a paper on asthma in the Lancet concluded by saying that every case of asthma should be treated as unique.

The environmental factors are certainly one complicating factor but, to complicate things even further, probably these kinds of complex traits are due to many genes with weak effect, which also plays havoc with genetic studies. As we've said before, we're prisoners of Mendel -- still looking for single genes with large effect. We need to change our focus.

occamseraser said...


(not so great news on the AD front)