DNA Cancer Codes Cracked By International Effort

Enigma23 writes "As reported on news.com.au, scientists from the International Cancer Genome Consortium of 12 institutes around the world will today release the first DNA profiles of some of the most prevalent types of tumors. While the story asserts that 'A new era of cancer treatment has dawned,' I'm a bit more skeptical, given that gene therapy and immunotherapy are still very much in their infancy at the current time."

Re:I'll take whatever advancement we've got.

[Wyatt Earp]

I was diagnosed with Acute Lymphocytic Leukemia (ALL) in the fall of 1980, it doesn't have stages like tumor cancers, but at the time I had a 5% chance to make it 3 years. I was clean after 3 months, it relapsed in the fall of 1982, 5% chance to make it 5 years.

Spring of 1991 I was diagnosed early with Stage 0 (Carcinoma in Situ) Testicular Cancer on both sides, surgery and some chemo.

Fall of 2006 I was diagnosed with a non-cancerous tumor in my neck, that left a nice scar.

I'm the only one in my NIH tracking group of ALL from back then to be alive.

Doctors freak me out, IVs of anytime scare the hell out of me, my fear are lumps from a lymph node.

As A Cancer Researcher...

[00Sovereign]

I've been following this for awhile. Looks like I get to update my "hit list" of gene targets to investigate. And that's what this will ultimately be...a list of interesting genes to look at for further investigation. No cures right away, it will take time to absorb this data into the collective intelligence of the medical research community and years to turn it into new treatments.

Re:As A Cancer Researcher...

[graft]

It also occurs to me that you can use this to identify specific alleles associated with increased risk of cancer. Not that genetic screening is generally very useful yet, but more maps like this will give a better idea of an individual's chance of developing specific types of cancer, maybe even associated with specific activities. E.g., "if you have these particular variants, you're much more likely to develop lung cancer, so you should smoke even less than a normal person." (I hope to GOD we (in the US) have public health insurance before this becomes widespread.)

Re:I'll take whatever advancement we've got.

[Anonymous Coward]

The great thing about this is that ICGC [icgc.org] is releasing all their data under what they call a confederated database DCC [icgc.org]. If you browse to their samples they have T4N2M1's and more. I think it's great that the data is getting out there!
However they do seem to be lacking the actual science/criteria behind the submission.

Re:It is not a code! There is no crack!

[Anonymous Coward]

Well, that's not entirely true. There's also some weird stuff going on with methylation that still needs to be cracked. That's a part of the "code" if there were to be one. Also, you could say there's a histone code, with it's own methylation. I'm sure there's other mysteries in there as well.

But, as the parent says, sequencing is sequencing --- not code cracking.

Also, can I make a plea to the world that we stop saying scientists "map" a genome when they really sequenced it. Mapping is a completely different thing. They may have done some mapping when they sequenced it, but likely you really mean "sequenced" when you say "mapped". You can't throw these jargon terms around like they're synonyms, because they aren't.

And I'm not nit-picking here (nor is the parent poster). To put it in terms here that everybody can understand, it's like when my wife refers to the tower case of my PC as the "hard drive" of the computer.

Also on the issue of sequencing cancer "genomes" I think we'll find it to be a big a waste of money, as the HapMap project was, in a couple years. Maybe they'll find a few drug targets, but I predict the more cancer "genomes" that are sequenced, the more we'll find that anything is possible in cancers. What we really need to find is the driver genes that start the cellular changes leading to cancer, and lead to progression of the cancer. I'm not sure sequencing a genome that's a total mess because it's *already* a cancer cell is going to make it immediately apparent what those driver changes were. Not doubting it can be done, but it is a big effort, and I'm not sure it's the best way to do it. I think more likely you will end up with a lot of sequence data that's really difficult for most people to interpret, and contributes not that much to cancer biology. Good to know what the deal is with the cell lines they've been using all these years as models, but beyond that, who knows. (Heck, maybe I'm wrong and we'll find a bunch of previously unknown viral sequence integrated in the genomes.)

Now, I'm not trashing personalized medicine, where you sequence a patient's tumor line, and then use that to determine a treatment tuned to the cancer cell line they have. I think that's exactly how things need to go. But I am trashing this idea of creating a catalog of *all* cancer "genomes" (for the reason I mentioned above) as the goal of a "genome project". Try sequencing some new and useful genomes instead.

Overall, I don't like "gold-rush" biology, and there seems the be a lot of that going on. Be the first one in, and leave a big mess of data for the next guy to deal with.

Re:It is not a code! There is no crack!

[RoadNotTaken]

You're right - that's probably where it started! That said, codon refers to the only (that I can think of) true code in biology. The other potential codes just mentioned (DNA methylation, histone mods, etc) are really more biochemical features that interact with proteins to regulate gene expression. The triplet code (code) is basically universal. TAA means STOP to every organism on the planet from worms to fish to bacteria to man. Histone mods and DNA methylation matter, but they're much more specific to particular cells/organisms.

Re:As A Cancer Researcher...

[morty_vikka]

I hear ya, it's exciting stuff... Just to add a twist, it should be noted that the vast majority of mutations found in tumours that have been fully sequenced so far occur outside protein coding regions. Not to say protein coding genes are not important, but it's likely that mutations in intergenic regions, promoters, noncoding RNA 'genes' and introns are also likely to play a major role in transformation to a cancerous phenotype in many cases.

The issue is going to be getting enough data for statistical robustness, even before any candidate mutations are investigated in the wet lab. I wonder how many thousands of each type of tumour should be sequenced? In any case, there will be buckets of data getting pumped out of the next generation of next-gen sequencers pretty soon, so plenty of work for bioinformatics nerds in the near future!

Re:I'll take whatever advancement we've got.

[Anonymous Coward]

die a little every day.

Hurry the fuck up before you bankrupt the rest of us.

Lest that stand completely heartless, look here. I feel for you. Really. Cancer must suck, and I've seen people melt from it too.

If there was a $1,000,000 permanent fix, I for one would be among the taxpayers happy to pay for it.

But, if you cost $1,000,000 per year to keep alive for the rest of your life, the rest of us would be far better served to look after your family on your behalf. Where do you think the $1,000,000 per year comes from? Do you seriously think your+employer's $800/month plus $20-$75 copays actually cover the checks going out? Is there some reason to expect your costs to drop?

Health insurance is nothing else but the most colossal Ponzi scheme of all time. The very notion of insurance is to spread risk, where rising health costs aren't a risk, but a certainty. Everyone is going to die, and most of us are not going to enjoy feeling it come.

Previous poster was right. Top end health care cannot be provided to all while the profit motive remains; rephrase, while the system has a vested interest in -not- scaling well.