8 ms·
Medicine needs to embrace open source
- jussij 7y ago> Because Pfizer believed it would lead others down a wrong path. Call me a sceptic but I don't believe that for a minute. Instead I'm sure they acted on some other ulterior motive.
- gumby 7y agoBelieve me: if you had a drug that had any positive effect on Alzheimers you'd be rich beyond the dreams of Croesus. What possible incentive could Pfizer have to suppress something they thought could be useful? Read the link to Derek's Loew's analysis in Science for deeper analysis: it's posted elsewhere on this discussion.
- bboygravity 7y agoValacyclovir http://www.virology.ws/2018/11/08/herpes-simplex-virus-and-alzheimers-disease/ http://www.virology.ws/2018/11/08/herpes-simplex-virus-and-a...
- SketchySeaBeast 7y agoAt the end of your article: "I understand that AD affects many families who would very much like to have an intervention for the disease, and are anxious to see this trial clinical proceed. Nevertheless, I do not feel that this trial is warranted based on the existing data. Additional studies are needed to understand the mechanistic connection between HSV-1 infection and AD, and specifically to clarify whether infectious virus, or viral proteins are involved." Sounds like it's not yet in the realm of proven effectiveness.
- outlace 7y agoPfizer has no incentive to axe a promising drug that would make tons of money, but journalists always have the incentive to kick up dust and make sensationalist articles.
- jussij 7y ago> What possible incentive could Pfizer have to suppress something they thought could be useful? Without seeing the results of the test it would be impossible to know. But hypothetically the reason could be as simple as them not wanting to provide any kind of help to their competitors, no matter how small. Their explanation that they did not release the data because it could lead to other scientists heading down another dead end path of research is just stupid. The field of science is full of such dead ends and is exactly the scientific process that has got us to where we are now. I would say it is more likely Pzifer did not release the data to help ensure their competitors would head down that same dead end should they try.
- agotterer 7y agoIf you are interested in open source medicine there are two amazing HOPE talks you should watch: 1) How to torrent a pharmaceutical drug - https://youtu.be/Py5TkirrO-U https://youtu.be/Py5TkirrO-U 2) Torrent more pharmaceutical drugs (follow up 2 years later to video 1) https://youtu.be/cfrfs7W-K_k https://youtu.be/cfrfs7W-K_k
- _def 7y agoThis news site is absurd! Reminds me of https://news.ycombinator.com/item?id=20172214 https://news.ycombinator.com/item?id=20172214
- g00s3_caLL_x2 7y agoMedicine in America will never do this, at least not for the foreseeable future. They moved from 'cures' decades ago. Plus add a big heaping spoonful of the FDA and the red-tape that must be jumped through. Anytime someone ingests something, someone, somewhere, will die from it. When you work in pharma, you learn that there are teams of lawyers and budgets just to deal with the inevitable deaths. Just getting to human trials during the experimentation phase takes years. I'd say look more toward real-working natural cures and improving our lifestyles and environments would be more realistic in the short term.
- nradov 7y agoThat's obviously false. See sofosbuvir for a recent counter example.
- thepaperone 7y agoSuch a good example! It only costs ~$84,000 to take a full coarse of sofosbuvir in the United States......................................................................... ... .....
- trimbo 7y agoCounterpoint: Sovaldi was basically a miracle drug for a widespread, expensive disease and even priced at $84K, that represented huge savings for the US healthcare system. At the time it came out, the destiny for a Hep C patient was a $1M liver transplant. And that's not even including all of the other costs and misery Hep C patients incur as their liver fails. Now those people are treated quickly, saving tons of money and time, as well as improving their lives. Also, because of the high price, it inspired competition and innovation. Within a couple years of Sovaldi and Harvoni, several other improved and cheaper drugs appeared on the market. The cost dropped to a quarter of the price of Harvoni and with half of the treatment cycle and better cure rates (Mavyret).
- postpawl 7y ago“According to industry watchdogs, reducing clinical trial costs will not help curb list prices because those prices are not determined by investment costs; they’re determined by what the makers think the market can bear, which helps explain why profit margins are so high.” https://www.nytimes.com/2018/06/08/opinion/drug-approval-cutting-prices.html https://www.nytimes.com/2018/06/08/opinion/drug-approval-cut...
- throwayEngineer 7y agoGovernment creates a monopoly in pharmacy. Government creates a monopoly in Physician services. Now that these healthcare monopolies are well established, they lobby for stronger monopolies. Obamacare was a prime example about how the established can increase their wealth by using the government.
- pg_bot 7y agoWe should start with an open source drug interaction database...
- iak8god 7y agoDrugBank contains interactions: https://www.drugbank.ca/interax/drug_lookup https://www.drugbank.ca/interax/drug_lookup And is downloadable under a CC license: https://www.drugbank.ca/releases/latest https://www.drugbank.ca/releases/latest The US National Library of Medicine provides an API to query drug interactions based on DrugBank: https://rxnav.nlm.nih.gov/InteractionAPIs.html# https://rxnav.nlm.nih.gov/InteractionAPIs.html#
- pg_bot 7y agoThis is not useable for electronic prescription of drugs in the US. That is the problem I want solved.
- iak8god 7y agoCould you explain more what you mean, both about what problem you want solved and why this is not useable?
- pg_bot 7y agoThere is currently a monopoly in electronic prescriptions in the US. Surescripts runs the only network that connects electronic health records to pharmacies. (over 99% of electronic prescriptions are sent using them) They dictate that you must have a contract with an approved commercial prescription drug database vendor. I want the industry to come together and make an open source drug database so that it can become the world standard. This way a lot of legacy businesses will no longer have a reason to exist and charge exorbitant licensing fees.
- iak8god 7y agoI am still having trouble understanding the specific problem you want solved, and what an additional open drug database would add that DrugBank lacks but that is required to accomplish your solution. It seems like your frustration is mostly with the Surescripts monopoly on electronic prescribing. You say that they dictate an approved commercial vendor. This is a very different problem, and would have very different solutions, than the problem We don't have any open source drug interaction database, which is where this thread started.
- jcranmer 7y agoI'm not an expert in drug development, but even I was really skeptical when reading the Washington Post article on Pfizer "burying" the Alzheimer results. Derek Lowe (an actual drug discovery chemist) did put out on article with his thoughts on the article, which were similarly damning: https://blogs.sciencemag.org/pipeline/archives/2019/06/06/a-missed-alzheimers-opportunity-not-so-much https://blogs.sciencemag.org/pipeline/archives/2019/06/06/a-... The impression the original article gave was that the company ignored compelling evidence because they couldn't make any money off of it. But the "compelling evidence" actually comes across more like the sort of thing you'd find here: https://www.tylervigen.com/spurious-correlations https://www.tylervigen.com/spurious-correlations . As Derek Lowe puts it, "One of the biggest objections to the idea inside the company was that this finding was unlikely to be real because Enbrel doesn’t really penetrate into the brain." A final coda: the actual data may not have been public, but the potential correlative results were published in 2010: https://www.ncbi.nlm.nih.gov/pubmed/20478733/ https://www.ncbi.nlm.nih.gov/pubmed/20478733/. So much for being buried?
- wutman 7y agoExactly. People tend to overstate the importance of spurious, surprising findings that don't make scientific sense. Some of these may lead to breakthroughs, sure. But the overwhelming majority are just noise. And it takes experts to decide which are worth investigating. To wit, anti-vaxers love subgroup analysis without supporting scientific logic (ie more autism in this population of 5 boys from the study). I'd hate to see old, frail people getting Etanercept (an immunomodulator) to prevent Alzhemier's based on very questionable evidence when all they really get are the costs (increased risk of serious infections leading to death-- an FDA black box warning).
- xvilka 7y agoThere is an Awesome List[1] of open source healthcare projects. Also a quite nice blog[2] about various FOSS initiatives. Some of the important medical projects were accepted[3] in the Google Summer of Code this year. [1] https://github.com/kakoni/awesome-healthcare https://github.com/kakoni/awesome-healthcare [2] https://medevel.com/ https://medevel.com/ [3] https://summerofcode.withgoogle.com/organizations/ https://summerofcode.withgoogle.com/organizations/
- neural_thing 7y agoFun fact, it costs hundreds of dollars, just to LOOK at some safety standards in medical devices, for example: https://webstore.iec.ch/publication/2606 https://webstore.iec.ch/publication/2606
- derefr 7y agoWeirdly, this can also be true in FOSS software. Despite e.g. DBMSes like Postgres being open-source, SQL-the-standard costs ~USD$180 to read: https://www.iso.org/standard/63555.html https://www.iso.org/standard/63555.html
- butteroverflow 7y agoHere's a related discussion: https://news.ycombinator.com/item?id=20174847 https://news.ycombinator.com/item?id=20174847
- behnamoh 7y agoI know it's off-topic, but with all the attention FOSS and other Open-Source projects get these days, I've been thinking about the final consequences and paradigm changes they impose on businesses and societies. In a world where FOSS has won, what new businesses emerge and how do they operate? It's an interesting question for me, because I believe one reason we're not there yet is not that business owners have philosophical problems with the idea of Open-Source, but rather most don't know how that'll be profitable for them. I know there are successful examples in which Open-Source has helped the business model (RedHat, etc.), but the bigger question is how can it be successful in other businesses and contexts (such as medicine)?
- rebelrexx858 7y agoyou might look at the novel Autonomous by Annalee Newitz for her take on such a world, it's quite interesting
- behnamoh 7y agoThanks for telling me about it! I'll sure give it a look.
- pbhjpbhj 7y agoIt's basically communist/collectivist, it's not going to be a profit producing model in the long run; those profiting are selling other services around their OSS AFAICT. How it can help in medicine is by saving people's lives or improving their quality of life, it can't line people's pockets - indeed that goal is contrary to the aforementioned benefits.
- noobermin 7y agoThis is a case of reading the headline, but really, no one other than developers embrace open source, if were honest with ourselves.
- s0rce 7y agoPhysicists seem to be doing a pretty good job, mathematicians as well.
- noobermin 7y agoFor some areas of physics. In my field(plasma physics modelling), non-open source codes abound.
- phkahler 7y agoWhy? Who owns them?
- noobermin 7y agoThere are some understandable reasons, some of them related to them containing potentially export controlled models. Even then, you could strip out those bits but for some reason, no one does it because they see no need to once they have access to the code. Often a hefty (for academics) license of 4 figures a year gives them access to modify the source code but they can't share their modifications with others.
- forgot-my-pw 7y agoI think it will be cool if there's a pharma company that would use crowdfunding to decrease the medicine costs. So each tier/goal will unlock additional benefits. For example, if they created drug X with the research cost of $10M. When sales/crowdfunding reaches: - $1M: drug margin is reduced by 10% - $5M: drug margin is reduced by 50% - $10M: drug is sold at cost & recipe is now open-sourced These clear goals will be big incentive for people and organizations to donate. A successful fundrasising could drive new medicine cost way down or open-source it. But of course, the system can still be exploited by inflating cost numbers, etc.
- froindt 7y agoMaybe it could work for a non profit, but the medical world has so many methods they start exploring but find are unfruitful (with many thousands to hundreds of millions invested). You essentially have to make boatloads of money on some products over a long time frame to continue the research and development efforts.
- forgot-my-pw 7y agoFor sure. This will only work if someone wants to fund an "ethical" / non-profit pharma research company.
- Jedd 7y agoThis is one of the better examples of where the use of 'open source' rather than 'free' distracts everyone from the end goal. This particular case may well be a beat-up, but there's undeniably a regular habit within pharma to not publish neutral or negative results, and we don't know how useful those data may be simply because we don't have access to them. AFAIK no country has laws requiring this, so it's not a regulatory / enforcement problem - it's a legislative one.