Slashdot Mirror
Open Source Drug Discovery Prompts a Fundamental Heart Failure Breakthrough
An anonymous reader writes "Case-Western researchers, led by Saptarsi Haldar MD., have made a fundamental discovery that could prevent heart failure after reviewing the "chemical recipe" for a cancer-treating molecule made open source by Jay Bradner MD. (whose TED Talk articulates the open source approach to drug discovery) This cross-discipline discovery, which was published in the August 2013 issue of CELL, is a fundamental breakthrough in heart failure research, and highlights the value of an open source approach outside of software development."
What does it mean for a molecule to have source? How is the "source" of the molecule distinct from whatever passes for a "compiled binary"?
What's this?! Everyone jumping on the Socialist-Commie-Pinko Open movement?!?
WTF!
Before you know it, IP with be severely weakened and all of us will have increased standard of living - except for the poor poor billionaires!
Won't someone think of the billionaires?!
Without the billionaires lording over us, what will inspire us?
We need IP to keep up the carriers to entry! We need to impede progress in order to preserve the billionaires! Our way of life will be destroyed!
Researchers from Case Western Reserve University School of Medicine and the Dana-Farber Cancer Institute have made a fundamental discovery relevant to the understanding and treatment of heart failure
It does not say "prevent" heart-failure, anywhere in the article. It is implied in the article that treatment could be greatly improved by this therapy, however, I'm not sure where the line in the summary about prevention comes from.
Most heart disease is caused by eating animal products (which humans aren't supposed to eat), lack of exercise, and smoking.
Does this magic pill cure all of those too?
Most people want to have their cake and eat it too.
It sounds promising, but there is a lot of work ahead before it hits the market, if ever. Dosing, administration mode, side effects, when it can be used, what other drugs it will interact with and which it won't are all a part of what needs to be determined. The "drug" may have been discovered, but is really just a tiny part of what needs to be known before you can safely prescribe the use in people under all the varying conditions of use where it might be needed.
Thus, just because the molecule is "open source" doesn't guarantee it will be viable or commercially developed. Now the real money has to be spent to justify it being approved by the FDA and released as a safe commercial product.
What does it mean for a molecule to have source?
It can refer to what Eric S. Raymond referred to as the "bazaar" model, or it can refer to a license that grants rights to the public analogous to those listed in the DFSG or FSF definition of free software. I see hints of bazaar in the transcript of the TED talk:
And here I see the spirit of publishing a discovery instead of locking it up behind secrecy and exclusive rights:
This leads up to the benefits of bazaar and publication:
And finally, a direct answer to your question as to what is the source code of a molecule:
it's not linked to an actual scientific article
The transcript of the TED talk mentions articles in Cell.
The only thing drug patents do is make drug companies rich. If we as a nation (USAians here) truly wanted to maximize progress in medical treatment, we'd nationalize all drug research. Don't even bother arguing that profit motivates progress. The overwhelming majority of researchers and engineers are motivated by the joy of success, not crushing the opposition and getting filthy rich.
As we've seen over and over again in nearly every technology area, the greatest progress occurs either in "open source" areas or when patents expire and everyone can innovate.
(Yes, I'm a socialist. No, I don't think that in any way invalidates the fundamental claims I'm making here.)
https://app.box.com/WitthoftResume Code: https://github.com/cellocgw
the drug in the article has been proven to do exactly nothing.
no trials yet.
put it back in your pants.
When it comes to science, there is no need to license a discovery to make it available to all. Simply publish and don't attempt to patent it. Scientific knowledge is public once published.
For a long time, commercial activities have claimed that only through commercial enterprise could quality and value be achieved. This has been claimed of software for a long time as well as with other industries. Among these is the drug/pharmaceutical industry because only they can afford the R&D needed to make important things happen. (Conveniently we forget that many of the most important drugs predate the big pharma industry.)
With 3D printing edging into device manufacturing space, just about everything is going to face free/public competition, not the least of which will be energy production. Big business with its dependency on having the public dependent on them has its days numbered. I look forward to those days... I wonder if I will live that long?
Either your thinking is confused, or you are doing this deliberately...
In the software world, there is a very big difference between source code and executable binaries.
In chemistry, there is no such distinction.
Absolutely. And guess what would happen to any drug company that did not patent their drugs? How would they be able to compete with companies that did not have to pay the hundreds of millions in research to get the drug approved? The only reasonable alternative to patent systems that I can see are (a) trade secrets, which means that the discovery is not made available to all, or (b) go back to a patron system where a generous benefactor foots the bill for research, in which case the research that the scientists do is only what the benefactor wants, which may be the ultimate cure for baldness or a little dick. I think patents are better than the alternatives. That is, unless you can come up with a better idea. Just be sure to think it through...
What a fool believes, he sees, no wise man has the power to reason away.
So there's no difference between the instruction on how to create a substance and the substance itself?
The Tao of math: The numbers you can count are not the real numbers.
trade secrets, which means that the discovery is not made available to all
Which is extraordinarily difficult for drugs, because everyone will simply buy a bunch of their competitors' pills, and figure out exactly what they're made of down to atomic detail. A typical university chemistry lab could do this in a few days. There are some aspects that are more tricky - the exact packaging is sometimes key to getting the drug absorbed by the body at the desired rate, and the chemical synthesis can be messy - but figuring these out is still way cheaper than coming up with your own drug.
Yeah, just as there is no difference between an algorithm/source code and machine code that implements it. Both are equally readable, right?
you can charge for open source software... it's just if you get the program, you get the source.
The "patron system" is already in place - the gov't foots the bill for nearly all the research, and private corporations add the last 1% of the bill before patenting and reaping 100% of the profits.
In the software world, there is a very big difference between source code and executable binaries.
Authors of computer programs stored in files whose names end in .js, .py, or .pl might disagree with you. Glance up at the location bar to see "comments.pl", then look at this page's markup to see "engage.js".
Collaboration and sharing results is standard in science and has been for the last few hundred years. Practically speaking, patents on their own are meaningless for new scientific advancements and do nothing to prevent or encourage scientific advancement. They're a non-issue. What does matter is how much it costs to read the published results.
The equivalent to open source in science is open access. CELL is not open access. As a result, I (as a scientist without access to CELL) don't have access to their results in the final form. Nor do I have access to their raw data or any discussion of the analysis. How is this open?
Can you get away with not telling the FDA what is in a drug?
Seems counter-intuitive.
Sig Battery depleted. Reverting to safe mode.
I was wondering how long a drug patent lasts right now and if it is tied at all to how long it took to research whatever the drug patent is for? A solution could be simply to put a time limit on drug patents, after which it becomes a part of the public domain.
Agrisea Tsunami - Epyc Servers... https://agrisea.net/products
Perhaps government patents would be a solution. Let the government license its patents freely to education and research but take a percentage of profits for corporate use.
This space intentionally left blank
The traditional way is append the word "herbal" onto anything to avoid having to disclose the contents.
Inheritance is the sincerest form of nepotism.
Yeah, just as there is no difference between an algorithm/source code and machine code that implements it. Both are equally readable, right?
Readable for who?
To a person well versed int the arts? Absolutely.
To a decompiler, you bet. The code you get out of a decompile may not look like the original, but it is still the code.
Sig Battery depleted. Reverting to safe mode.
they would compete in the way they used to compete.. by being the best in making the compound.
world was created 5 seconds before this post as it is.
What's claimed is pretty impressive. They say they've gotten 60% improvement in heart function from a month long treatment course in mice and even quicker protective effect against declines in function. The caveat, as always, is that many things work well in mice, but don't translate into human therapies.
This is still paywalled, but it has many of the figures from the article as well as the abstract.
http://www.sciencedirect.com/science/article/pii/S0092867413008842
This is of direct interest to me as I have some right side heart enlargement (precursor to failure) and take medicine for it.
So there's no difference between the instruction on how to create a substance and the substance itself?
Yes, there is a big difference.The instructions are much more valuable. The instructions never go bad. The substance itself will eventually go bad.
If Slashdot were chemistry it would look like this:Cadaverine
The vast majority of government funded biomed research is a waste of time. It may be useful for other social reasons (get more people to know "a scientist") but most of that research money goes down the drain and creates confusion rather than advancing knowledge.
The first patent act was in 1790. The Constitution only permitted Congress to have patents. Congress had to decide to do it.
Let's see. This is a molecule in pre-clinical testing. I would give this specific molecule about a 1 in 1,000 chance of actually being marketed. Those are damn good odds for a molecule at this stage. This is why you always have to take the word "potential" with a boulder of salt.
I am neither a doctor or a chemist (IANADOAC) but it's actions like these that bring me some measure of hope about the human spirit. Sure the odds are stacked against them with the evil Pharmaceutical conglomerates circling round like evil Pharmaceutical compagnies, but still.
I wish for more.
The inventor of the drug, could be the only one that was able to state that they developed the drug. Most people have a sense of fairness and will pay extra to reward the discover of the drug. Not to the point where they go bankrupt or die because they can't afford the drug, but in that case they person making the drug would not get there money anyway.
If you don't believe me that reputation can make money for the drug industry simply go down to the chemist and look and the difference in price between Panadol and generic paracetamol (same thing, 3 times the price).
I find it amazing that people believe that the only way to develop drugs for large drug companies to develop them? No matter how big the company is, to say that the rest of the people in world are unlikely contribute more than them seems the height of arrogance.
Also a public company structure seems like the worst possible way of developing drugs. Public companies by their very nature must optimize for profit, (otherwise they are not doing their job) therefore it is never in their interest to cheaply an effectively cure a disease, it is much better the bottom line to keep the person alive while charging them an ongoing fee for living, until they run out of money and die.
Yes, there will be a mexican standoff between drug companies. They will simply 'agree" not to pilfer, and then ALL drus will be administered via careful monitored 'health centres' where as machine will be used to do the injection so the nurse can't game it for a small start up company.
what people for get to ask is how much the 95% of drugs that never make to market actual cost to develop and go through trials.
Wrong.
Private corporations are the ones who supply the $ to do research. Without that money either student fees would more than double or the government would have to pay more than double.
By careful planning you can show government funding does NOT go into research, on teaching and infrastructure.
As someone who lost his mother to heart failure when she was 59, I applaud these researchers who put patients before their own financial gain.
... go back to a patron system where a generous benefactor foots the bill for research, in which case the research that the scientists do is only what the benefactor wants, which may be the ultimate cure for baldness or a little dick. I think patents are better than the alternatives. That is, unless you can come up with a better idea. Just be sure to think it through...
One of the things governments were invented for WAS to be the benefactor which foots the bill for research (usually, in the civilised parts of the world, they don't insist on micromanaging where the research goes).
It's NOT a conspiracy... it's a plot.
Drug companies don't have any "their" drugs. Scientists who works for drug companies do. The drug companies are the "generous benefactor" in the current scenario. Ditch the patents, change the FDA oversight to something closer to what you have for food... check production is as it should be, confirm what they are selling is what they claim and not something known to be dangerous. Then take away all legal immunity that is granted by going through the current process. Toss out grants and funding to encourage the efforts.
If scientists were working for themselves, pooling resources to get and share equipment and distribution lines. The money they make just being the first one to market with a drug would likely be much more than they are paid by drug companies.
In this case it probably means a similar (but patented) molecule will probably be the first to market. It could also mean that their JQ1 inhibitor, while successful in model trials as a lead or tool compound, has already failed a preclinical test. Either way: 50:1 odds against a new drug candidate succeeding in preclinical trials, 10:1 odds against a new drug compound succeeding in clinical trials. 500:1 odds against success, on average. The big discovery here isn't the inhibitor they waived the rights to; it's the pathway and the target (which aren't patentable anyway).
Nope. API (active pharmaceutical ingredient) and excipients (stuff in the formulation that has some function: stabilizer, solubility, crystallization aid) have to be disclosed.
But then you can't call it a drug and have to dance around saying it treats anything.
Bayer still makes aspirin. Sure, it might not be their biggest profit margin drug, but they still sell a lot of it.
Bullshit. The vast majority of spending necessary to turn this discovery into a drug hasn't even been started yet. Even in cases where the drug is invented and patented in academia (~15-20% of drugs, and we won't know if that has happened here with JQ1 for at least another 5-10 years), universities license the molecule to a biotech or pharma to move it through clinical trials. The NIH is starting to spend more on translational research (the preclinical/clinical stuff Pharmas typically do), but it will be at least another decade before we can tell if the NIH is doing a better job at it than Pharmas.
Government (university) patents and "taking a percentage for corporate use" has been happening for over 30 years. the "freely to education and research" part is a bit more tricky.
It's useful for discovering pathways and targets that are later determined to be useful for developing drugs. Also about 15-20% of drugs are invented in academic labs. Those drugs are more likely to be first in class drugs and more likely to focus on unmet medical needs than drugs that are invented in Pharma companies.
Three times the price, is nice and all, but you'll need more like 100x the price of a generic small molecule pill to recoup the R&D investment cost and turn a profit. The multiples aren't nearly as high for biologics (antibodies, peptides, etc), but would you willingly pay 3x as much for a drug that costs $14K just to manufacture per patient-year?
I was talking about the large percentage of non-reproducible studies in the literature
And you don't think someone came up with the idea before implementing it in the patent act?
I think government funded medical research is a completely viable alternative.
Which in most cases is 100% OK because it's completely ineffective.
Really it costs $14K to manufacture? Without counting NRE? I have strong doubts that any drug costs that much to manufacture given how much automation is in place for drug manufacturing.
NRE = Non-Recurrent Engineering (just so I'm not accused of being obscure)
Only I can judge you.
Imagine what a GPL like approach would cause here. The major secretive pharma conglomerates would be disadvantaged because they wouldn't want publish their contributions, therefore they wouldn't be allowed to start research on the already published stuff. While numerous smaller national firms would receive the benefits simultaneously. Way to go.
> The only reasonable alternative to patent systems that I can see are (a) trade secrets,
Public funding, crowd funding, public reward from success, etc. There are many alternatives that are better if your only motive is to get better health with least resources. Current system is good for making money. Cure for all would be a disaster to money system.
And this is new? Academic scientist gives a terrible compound with no bioavailabilty to his friends to test in irrelevant assays? Call it 'open-source' and it's somehow a TED talk?
I dunno, patronage system sounds like the surefire fastest way to develop biological immortality....
And in all seriousness, what else should the ultra-wealthy be spending their coin on? Cars, shiny rocks, too-big houses and boats?
It costs $1 billion I believe to bring a drug to market, that's not engineering costs but rather FDA costs and costs of failed drugs. Drug trials are not cheap and you don't really know which ones will work beforehand.
Someone needs to pay that or the drug can never be manufactured and sold. Do you have a billion lying around and are you willing to hope people pay you back?
This generous benefactor could be the government, you know.
If Pandora's box is destined to be opened, *I* want to be the one to open it.
Please precisely define "executable by hardware". Modern x86 chips don't run x86 bytecode; it gets recompiled to micro-ops inside the CPU, and it's been that way since the Pentium Pro.
None.
Drug companies as they exist today don't deserve to. Putting them all out of business would be a benefit to the world. Capitalism has no place in health care. It's a violation of the Hippocratic oath.
Restricting access to medication in the name of profit is immoral. Period. I don't give a wet fart if they spent money to develop the cure once it's known it's free game.
The solution is publicly funded research. The free market will reward whoever can manufacture the drugs the cheapest.
We USAians ?! wtf are you smoking? probably the first thing one would learn living here is how we refer to ourselves, you're not even past that step buddy.
Rakishi, you are proceeding down a line which is irrelevant to the question I asked. My question was specifically aimed at pepty's absurd assertion that a drug costs $14K to manufacture per patient per year. I understand the absurd costs to get past the FDA hurdles, those are a different question though. ...Anyone? ...Anyone?
Again:
a drug that costs $14K to manufacture per patient per year?
Only I can judge you.
Look, we already have so many drugs that have been used for so long, and RESPONSIBLE researcher will tell you that MOST "new" drugs are unnecessary and dangerous, and developed primarily for the purpose of making corporate wealth.
The system is broken to the core, and open sourcing it like this is probably the only thing that will mitigate that third leading cause of death in the US. (Doctors treating you with new drugs they get a kickback on).
There are drugs that have fallen into disuse that are effective, well established, and so inexpensive that Big Pharma has no desire to make them anymore, because they do not yield the OBSCENE PROFITS for the shareholders.
The entire system is based on fraud and lies and deception. Taking a known drug that's come off patent and fucking around with it a little so that you can claim it's novel enough to repatent and charge the public up the ass for is great strategy for the stockholders, but it KILLS people at worse, and rips them off when they could take something generic and established at best.
Hell, BY FAR most drugs people take in the US are prescribed because people didn't take care of themselves to begin with.
I don't believe corporatism/capitalism/wahtever you're calling it today is inherently evil; just that our version of it clearly IS.
Homogenized milk. We know an inflammatory molecule is absorbed from homogenized milk. This molecule inflames arterial walls and plaque is laid down as protection..
What is so ironic about this article is that Bradner didn't "invent" anything. JQ1 is a minor change to compounds first discovered by Mitsubishi Pharma, who told the world about them in patent WO/2009/084693. Bradner's group made one compound based on this and gave it to academics - fine, but he's also set up a company to exploit it, www.tenshatherapeutics.com. Doesn't sound so impressive now, does it.
There is automation in chemical synthesis and biological production of actual production batches of drugs, but it's not the kind that leads to unattended operation. Most of that type of automation is way downstream at the packaging stage.
Drug manufacturing costs:
small molecules (example: ibuprofen): $1 per gram
peptides (synthetic insulin): $50-1000 per gram
antibodies (Herceptin, Avastin): $500 - 5000 per gram
The $14K figure was the cost of materials for 1 patient year of a peptide AIDS drug. The starting materials and reagents for peptide drugs are expensive compared to those used to make small molecule drugs, yields are low, purification and QC costs are higher too.
Biological drugs (mostly antibodies) tend to be very expensive to manufacture. The processes are extremely difficult to carry out and have low yields, the products tend to degrade quickly. These costs are dropping fast though: they are much better at scaling up production of antibodies than they were a few years ago. Also, while it might count as NRE, the physical infrastructure needed to produce large quantities of many of these drugs means sinking several hundred million and several years into the manufacturing plant.
See above.
huh? You stated "costs $14K just to manufacture per patient-year" (italics mine). I take issue with the statement as I don't believe any drug costs that much just to manufacture per year. Care to elaborate on your "See above"? Citations supporting such stupendous manufacturing costs are what I am looking for.
Only I can judge you.
Here's an article on the AIDS drug I was talking about. My number was from a personal communication.
http://www.aidshealth.org/archives/news/birth-of-aids-drug-is-10-year-tale
All along, producing Fuzeon on a commercial scale had been a concern. The cost of producing the drug for Phase I clinical trials was thousands of dollars per gram, said Michael Recny, Trimeris vice president of corporate development. With a prescribed dosage of roughly 80 grams a year, that translated into hundreds of thousands of dollars per patient.To streamline the process, the company recruited three manufacturing experts from what today is GlaxoSmithKline, a giant London-based pharmaceutical company with a U.S. headquarters in Research Triangle Park. Those experts came up with a shortcut. T-20 is a chain of 36 amino acids, and the company had been constructing it by adding a single amino acid at a time. The new team found a way to produce large quantities of three shorter chains, which could then be combined to create a fully-formed T-20. Even so, making the drug requires 106 steps, more than 10 times the norm. About 45 pounds of raw material are needed to produce 1 pound of Fuzeon. The production cost is cited as a driving force behind Fuzeon’s high price. Bolognesi, who remained at Duke in the company’s early years but was recruited to be CEO four years ago, said that because the cost of making Fuzeon is at least 10 times that of existing AIDS drugs, the profit margins on Fuzeon will be “significantly less” than with other AIDS drugs.
http://slashdot.org/comments.pl?sid=4050587&cid=44499733
http://slashdot.org/comments.pl?sid=4050587&cid=44499185