Scientific American has a pretty good article this month by Francis S. Collins and Anna D. Barker that is cautiously supportive of the Cancer Genome Project. If you read it, you will learn that the consensus view for some time now is that cancer is caused primarily by genetic mutation, and that they believe that by mapping every cancer's genome we can learn more about how to prevent or cure cancer.
However, this week's Newsweek has a pretty hard-hitting piece by Sharon Begley that fairly thoroughly blasts the project:
When the government dangles $1.5 billion in front of scientists, they rarely say, oh no, please, keep it, there are better ways to spend the money. But as the biomedical establishment gears up for yet another megaproject, some leading scientists are doing exactly that, making the heretical suggestion that this latest extravaganza is poor science and bad policy.
Prominently quoted is our friend George Miklos. It touches on other issues we've discussed here on Dean's World in the past--and by the way, Miklos has said he'd like to write an article on this for us on this, so stay tuned.
Here's the problem in a nutshell: what if genes have little or nothing to do with cancer? If so, what are we pouring all this money into this project for? Especially when we are spending very little research money on alternative theories?
Miklos will tell you that he believes, as a growing number of biologists do, that the entire search for the "oncogenes" (cancer-causing genes) and mutation has been a decades-long multibillion$$ boondoggle that's wasted countless lives, because that's not what causes cancer.
You might reason that, because we don't know for sure, research on cancer genetics is a good idea. That is a very defensible argument.
Repeat: that is a very defensible argument.
But, virtually all the money the government and private industry now spends on cancer research is going into that one area. Which means that we may well be learning all sorts of interesting things, but we aren't learning much that will actually cure cancer.
This goes to a quote I've picked up from Al Gore, which applies well here:
"It is difficult to get a man to understand something if his salary depends upon his not understanding it."
The cancer research establishment today works under a system they call "peer review" which is quite different from what peer review was 30, 50, or 100 years ago. What "peer review" used to mean is that if a scientist did some research, she'd write it up and then submit that writeup to anonymous review by some well-qualified peers, who would critique it before publication and give her a chance to make changes or corrections, or even occasionally say "you've made a major error, you need to rethink and re-do this whole thing."
However, in the last few decades, that system has changed, so that now the "anonymous" peer reviewers not only critique each others' work, they also control who gets funding for their research and who does not.
Which creates an obvious conflict-of-interest problem that's never been properly addressed: if a scientist comes along with brand new ideas that challenges the establishment in a major way, that scientist can become a threat to the income--as well as the prestige--of many of their peers.
What's most insidious about it is that if you take up the cudgels to defend one of those mavericks, you can get branded a "conspiracy theorist" and a "pseudo scientist." Worse, you might even be accused of "letting politics interfere with science"--as if people who accept taxpayer money should not be subject to scrutiny and hard questions from voters.
Worst of all, if you're a practicing scientist you may find your own grant money suddenly drying up if you even express support for a maverick. Neat trick, eh?
Peer review is a good system for vetting papers. As it's currently constructed, however, especially when it comes to money, it's in desperate need of reform. And taxpayers have every right to demand reform, by the way.
You might be wondering what, other than genes, might cause cancer. Well there are several alternative theories, but the one making the most waves right now is the aneuploidy theory. A growing number of respectable scientific publications have begun to acknowledge that aneuploidy may just be the most important area of research in cancer that's come along in decades. It's still getting relatively little funding--after all, it's not about genes or viruses, and there's very little funding $$ for non-genetic, non-viral causes of cancer right now--but any computer programmer can instantly recognize aneuploidy as a very promising theory. To put it in computer terms, the aneuploidy theory is simple:
You have this long chain of DNA in every cell, that DNA being divided into sets of chromosomes. Whenever a cell divides, it replicates the DNA. Think of the genes as individual bytes of data. Think of the chromosomes as the data demarcation points--this set of data starts at point A and ends at point B.
You've got millions of cell replications happening all the time. Every time a cell divides, you have a chance for replication error. Most errors will be minor, but once in a while you'll have an error that destroys the structural data integrity. And, with all these millions of replications going on all the time, occasionally the badly-replicated data is not just corrupted, but it interacts with the replication process in such a way that error compounds upon error. You wind up with more than just corrupted data, you have an out-of-control Frankenstein monster that corrupts the replication process itself, with bad data spreading and overwriting the good all over the place, blowing over sector boundaries and corrupting not just data but code.
The individual bytes of data (the genes) haven't got much to do with it. It's an out-of-control replication error.
If so, then the fact that you've got weird mutated genes in cancer cells would be completely unsurprising. Furthermore, spending billions of dollars and years of effort to study the corrupted data ("Oh look, here's another error! Let us study the significance of this individual bit of bad data!") would be ridiculous when what you really needed to look for was what corrupted the replication process in the first place. Or how to detect a replication error and correct it sooner.
I suppose I'll have lost most people on the above explanation, but I'll bet most of my hard-core computer buddies get it just fine.
The establishment wants 1.5 billion dollars and several more years to find and study individual errors in replication, because each of those errors might point somehow to the cause of the problem. They haven't bothered to ask whether the errors are the symptom rather than the cause.
Is the aneuploidy theory correct? I can't say. Something else might be correct. But it's fascinating to look at how all competing theories have been locked out by this supposedly pristine "peer review" process of funding, and the obsession with individual genes.
The need for reform of the system has never been more urgent. It's not just taxpayer money that's at stake, it's lives that are at stake.
