What Is Open Source Pharma (and Why Should You Care)?

Andy Updegrove writes: Humanity today is almost completely dependent on huge pharmaceutical companies to create the drugs we need. But these companies focus exclusively on drugs that can be sold at high prices to large populations — in other words, to patients in developed nations. This means that those who live in the emerging world that suffer from the remaining 'neglected diseases,' like Malaria and drug resistant TB, have no one to depend on for relief except huge charities, like the Gates Foundation. They also have no way to afford many of the patented drugs that do exist. But there is another way, modeled on open source software development, which relies on crowd sourced knowledge, highly distributed, volunteer efforts, and advanced open source tools. That methodology is called Open Source Pharma, and it has the potential to dramatically drive down drug development while saving millions of lives every year.

Comments

Pharma development is hard and expensive

[hawkeyeMI]

Here's approximately how the drug development process works, to the best of my knowledge. I'm not in pharma but I've discussed some research projects with pharma folks and done a few projects on contract with them. I am not a pharmacologist or chemist. I am involved in medical device development and clinical trials for that purpose.

1. Pick a target condition (based upon need, existing treatments, size of market, whatever)
2. Based upon pharmacology, look at possible pathways to address said condition
3. Develop/find compounds that might address those pathways
4. Do whatever possible to narrow down these compounds by screening for safety before doing any trials, animals or otherwise. Select very carefully for screening techniques that won't give you false dangerous results becuase even if the compound is actually safe in the end, if you have a screen that looks bad at this stage the FDA is not going to like it down the road and it exposes you to liability.
5. Do animal trials with the promising compounds and hope some both work and don't cause harmful effects. Depending on the animal model this can get very expensive and time consuming.
6. Do human trials with promising compounds. A well-powered study will be VERY expensive (easily tens of millions of dollars or more), and depending on the condition being targeted, may also be very time consuming. Hope that what worked in animals works in humans, and no harmful effects crop up.
7. Assuming you make it through FDA approval, and they don't make you do more trials or enroll more patients before you get it, now you can produce the drug.
8. Try to sell enough of the drug to recoup your R&D expenses on ALL the compounds you checked out for the condition, all the trial expenses, etc, before the IP protection expires and the drug goes generic.

This "open source" model is neat and it may help a lot, especially in places where you can get away with less regulatory approval, but the way it's done is not because pharma companies are evil. It's because drug development is hard and expensive, and anything less than a blockbuster drug carries a high risk of never recouping the R&D expenses.

I think there's a lot of hubris to the idea that it can be done so much better this other way, but I will be happy to be proven wrong, because it really is a problem that needs solving.