Hacker News new | comments | ask | show | jobs | submit login
Computer program that could bypass patents to produce synthetic drugs (europeanpharmaceuticalreview.com)
125 points by zhamisen 28 days ago | hide | past | web | favorite | 54 comments

> In the global industry, some of the best kept secrets are those necessary to make life-saving medication and other pharmaceutical products.

What? Patents are not secret. Patents are the opposite of secret. The very word "patent" means "open". Are drug patents some kind of trick to do the opposite of what a patent is supposed to do?

> However, when the researchers designated some atoms and bond as untouchable, the program proposed plans that avoided the patented ones.

Can someone explain just exactly how do drug patents work? Can this legally work? I understand that normally patents cover processes or working inventions. I didn't think that the precise chemical bonds were required to be produced in the final output of a patent.

Also, since apparently the researchers have read the patents in question and designed their software to produce drugs that avoid those patents, have they willfully infringed? Since they were aware of the patent and the process by which the drug worked, so using this knowledge they produced similar drugs that do the same thing?

I assume that the researchers know drug patent law well enough and are confident that their method avoids patents, so there's obviously a lot I don't understand.

I'm a patent holder for a pharmaceutically relevant class of compounds. There are two types of patents typically seen in pharma. One is procedural (how the drug is synthesized, isolated, formulated or administered) and another is composition of matter (the atoms in the molecule, their relative coordinates and the lengths and types of bonds that connect them). This program takes a target compound and does two things. 1) produces a library of similar compounds by structural diversification (making substitutions of atoms or small groups of atoms in the target compound with other atoms or groups which are known to behave similarly) with restrictions based on what has been patented under composition of matter. 2) It then takes each molecule and looks for ways it can be synthesized. It does this by breaking apart the molecule piece by piece until it obtains commercially available building blocks. These piecewise separations correspond to synthetic a steps in the other direction, which the program screens for literature precedence. It does this for every possible combination of piecewise separations until it finds a set of viable pathways from available compounds to the target. This is called retro synthesis. It then filters the potential synthetic routes for those that are covered by procedural patents until it has a list of non patented, commercially viable synthesis plans.

As you can imagine there are thousands of ways a moderately complex molecule could be deconstructed and the bank of known reactions is more than any one person can really grasp. That and the myriad patent literature and how cryptic and dense it can be make the problem particularly suited for algorithmic treatment. This is only being done now because cataloging all the research and patent literature (which goes back to the late 1800s) and digitally formatting in a way that allows it to be computationally analyzed and processed requires manual translation of each report by a human being. Not many people have the expertise required for this and those that do usually prefer less menial work. So it has taken this long to amass a digital library of sufficient size to give one confidence that the answers it provides are comprehensive and that further searching would be pointless.

> The research team hope that their software could help pharmaceutical companies in the protection of their intellectual property, and to hasten research into organic chemistry.

It seems what they are attempting to sell is a IP protection service: let us show you other molecules and synesis routes you should also patent defensively, without actually attempting them in practice. Pretty soon you will have algorithmically defined patents.

It's IP anti-science at it's worst.

Depends on how many possible patents there are. Patents do cost money to maintain/file. It's possible that there could be life saving drugs that have so many possible patents that compared to the actual likely financial return it would be uneconomical to patent them.

Which to be fair is probably very unlikely given how healthcare costs and whatnot work in America. Can always bump the price up to compensate for expenses.

BUT - this is still a very useful tool for locating possibly cheaper/simpler methods of producing drugs that have expired protections. There is a project out there that is working on a bioreactor kit that allows the safe home production of several common drugs that are not patent protected. There are also projects that could have been life saving but were already not financially viable that could be rescued by this program.

And if someone is fast enough and has the funds to burn, they could patent some of those processes and then do what Tesla did with their patents.

I think this may open up more opportunities in the long run than it closes.

A not-so-fun fact: My country used to be on the 301 Special Report (IPR "rogue country" list) until 2015 because American pharmaceutical companies didn't consider procedural patents good enough. The last patents in that category were issued in 1995...

Apparently the income of Big Pharma investors is more important than the affordability of generics.

> Apparently the income of Big Pharma investors is more important than the affordability of generics.

You say this as if it is surprising.

One thing I don't get: if you're trying to avoid patents, why go back to the 1800s? Don't you just need to look at recent patents that haven't expired?

The papers going back that far are to determine which synthetic steps are experimentally viable. You can look at a molecule and think how could I make this? Well if I broke this particular bond it would make A and B, which I can order from a supplier. So let's search the literature to see if anybody has successfully carried out a reaction that connects an A-like molecule and a B-like molecule in a way that would yield the kind of bond(s) that connect them in our target. That's a lot of searching because for a typical drug, there are 10s to 100s of bonds and a total synthesis can take upwards of 20-30 steps. Thankfully, making and breaking most of the bonds is just not possible given the fundamental nature of the compounds (more specifically, it would take a ton of energy to put the molecule together this way, and if that much energy was applied, it would react at more reactive sites long before the desired reactivity occured). For others it is less clear, which is why having a comprehensive literature sure is useful. If your database goes all the way back to the 1800s you can say with some confidence that a certain step has never been done before and therefore assume its not possible (perhaps an incorrect assumption, but pharma isn't interested in developing novel types of reaction methodology for a target screen because if it hasn't been done already, it would probably take several years and the chance of success is low. And besides, that's what graduate students are for ;)

So the literature search is to find out what's possible. Among the possibilités, the only ones of interest are those that are not covered in patents, hence the second screen.

Thanks so much for your extensive and informative answers.

Depends what you want: If you want to find new processes and compounds that you can patent, then avoiding older patents is essential, as they're prior art.

On the other hand, if you want to find new processes and compounds that aren't patented, then finding a past expired patent covering a part of the process would mean that part of the process is "safe" (not finding one might mean it is safe, or might mean you've overlooked something)

> Can someone explain just exactly how do drug patents work? Can this legally work? I understand that normally patents cover processes or working inventions. I didn't think that the precise chemical bonds were required to be produced in the final output of a patent.

A big part of it is that drug patents cover the process of production that gives the final molecule with viable yields. Pharmaceutical companies try to write patents broadly enough, or just file enough different patents, that there's no way to get to the final product without infringing one or more of their patents. The idea here is to come up with a sufficiently novel synthesis pathway that the original patents don't cover it, and thereby don't apply.

Derek Lowe has a good discussion: http://blogs.sciencemag.org/pipeline/archives/2019/01/18/aut...

>>Pharmaceutical companies try to write patents broadly enough, or just file enough different patents, that there's no way to get to the final product without infringing one or more of their patents.

The patent can also be thrown out if it's to broad, no? The broader it is , the better the chances that there's prior art. Sure it costs money, but then there's money to be made so someone will sue

In theory yes.

In practice we except the patent to be rejected by a government department which value is measured in number of granted patents, or a special court which is infamous known to favor patents over rejecting them and which previous chief Judge was himself a previous patent lawyer.

The cost of getting a patent invalid is also very minor. There is no requirement to pay back the extracted rent by the time the patent get invalidated. The author also has many chances to rewrite a challenged patent and remove over broad claims. The risk vs reward of making broad claims in patents is evidentially in favor of risk.

> I understand that normally patents cover processes...

> since apparently the researchers have read the patents in question and designed their software to produce drugs that avoid those patents, have they willfully infringed?

the patents cover process, not the resultant molecule. the program finds a different process to make the same thing.

>The very word "patent" means "open".

I was really surprised by that. So when Elizabeth I sold a patent for the monopoly of salt to some duke in order to raise gold for the crown the word "patent" meant that it was a public rather than a secret decree.


Yea it’s important to note these guys are risking breaking the industry. If for some reason pharma companies were unable to at least recoup the cost of their (expensive) R&D efforts, it would disincentive new drug development.

Pharma companies don't do their own R&D anymore, for the most part. The documentary, Drug$ (2018) does a tough takedown of the industry. Pharma companies acquire drugs from indie or academic researchers. What pharma companies do is get the drugs FDA approved then market them, and at the highest price the market will bear (meaning that some people can't afford them). In other words, there are no R&D expenses. https://www.imdb.com/title/tt7394770/

It depends on what you consider a "pharma company". If the major companies acquire drugs, then it could still be argued (and I'm sure people do) that the drug prices, being dependent on patent rights, are indirectly paying for R&D. Just because it's not the same company doesn't mean that the ecosystem doesn't depend on the patent rights to finance the research.

I'm not completely confident of the necessity of patent rights as they are, but it seems logical enough on the surface. I can't assume that entities which engage in transactions together are independent.

It also doesn't mean the ecosystem is dependent, either. Disney made lots of money selling movies based on stories in the public domain. Just because a middleman exists doesn't mean the middleman is paying the ultimate source in proportion to the money the middleman takes.

See also the story of the insulin patent: https://www.thelancet.com/journals/landia/article/PIIS2213-8...

> [Frederick] Banting felt “…that as a physician who had taken the Hippocratic oath he could not be party to any patenting of a discovery. ... John J R Macleod, with whom Banting shared the 1923 Nobel Prize in Physiology or Medicine, similarly declined to be named on the insulin patent. However, neither [Charles] Best, a medical student who had yet to swear the oath to Hippocrates, nor [James] Collip, a PhD biochemist, faced any ethical dilemma and were named as inventors, although for $1, Collip and Best transferred all rights, title, and interest to the Governors of the University of Toronto. Banting, Best, Collip, and MacLeod all believed that insulin should be made as widely available as possible, without any barriers such as cost....

(ht https://www.vox.com/science-and-health/2017/5/12/15621952/in... ht 'steveklabnik yesterday on Twitter)

I think the proper conclusion is "no information" - the fact that redistributors are being paid what they're paid does not tell us anything positively or negatively about the necessity of them being paid what they're paid. You'd need to look at their actual costs, and more importantly the market of those costs: if they stopped paying researchers, what would the market effect be? (Perhaps they could fund research grants instead of discoveries, incentivizing everyone to work on the problems most interesting to them regardless of profitability, and thereby increasing research into cures for rare diseases.)

It depends, on the chemistry research side the pharmaceutical companies still do a lot, if not most of the heavy lifting. Cheminformatics benefits from the big scale in these organisations and also do the of the type of thing discussed in this article. The same with other parts of this research, like having access to the fancy spectrometers etc. Finding and understanding new compounds with potentially useful properties is only part of R&D process, then developing the biology all the way through to making it a drug fit for human consumption is also quite risky because most of them fail and that's where the small companies come in who takes up that risk.

>at the highest price the market will bear (meaning that some people can't afford them)

If anyone can't afford your drugs, this is an issue of insufficiently optimized price discrimination (as long as what they can afford is above the marginal cost of manufacture). There's often manufacturer's coupons for exactly this reason - make a compelling case about what you can afford, and the manufacturer will drop prices until you can give them some money instead of none.

I'm no economist, but I strongly feel perfect price discrimination would be a rather terrible thing. You seem to be implying that it would be "sufficient".

Imagine if the water you drink was priced at the maximum amount you would be willing to pay not to die of thirst. This would palpably be a terrible thing. All of us go around every day buying things that we think are worth substantially more to us than the money we pay. It seems to me like this is the whole value proposition of capitalism for human beings.

I have always thought that free markets are useful, but I think that is only the case if they tend to drive producer surplus to zero and consumer surplus to the maximum. If there were no consumer surplus, as with perfect discrimination, then I don't see the point of economic activity.

I sort of thought (because I read it somewhere) that orthodox economic theory implied that perfect free markets should drive economic profits of businesses to zero. And profits should be roughly equivalent to producer surplus. So we shouldn't be balancing consumer and producer surplus, let alone allocating the surplus all to producers.

I agree with you there, lots of things should get competed down to low or no producer surplus. The problem is that you have to pick one of price discrimination, unaffordable medication, or unavailable medication.

If you don't allow for price discrimination, then many new medications cost so much per patient with the relevant disease that your choices are to have an unaffordable average price charged to everyone, or just don't make the medication in the first place and let nobody have it. With price discrimination, you can keep the average price high by gouging those who can pay, and using that to subsidize those of more modest means.

There's no choice here that's an unalloyed good. There's always tradeoffs. Price discrimination is just sometimes the lesser of several evils.

Well that's one way to describe "any lifesaving treatment costs the sum total of resources that you can command".

It's kinder than extending treatment on credit and then doing price discrimination via bankruptcy proceedings.

> What pharma companies do is get the drugs FDA approved

That's the D part of R&D, isn't it?

> If for some reason pharma companies were unable to at least recoup the cost of their (expensive) R&D efforts, it would disincentive new drug development.

I'd like to see any evidence at all for this.

The closest thing we have to a controlled experiment is the software industry, where patents are a) difficult to obtain, at least in a way analogous to patenting a molecule and b) not culturally common among large portions of the R&D community. There is a huge amount of R&D into better databases, better operating systems, better distributed systems, better UIs, better programming languages, better compilers, etc. etc. that happens with no patent protection, and often with no license fees or other direct monetary compensation at all. Microsoft being willing to keep their source code secret and patent their techniques didn't help NT defeat Linux one bit, and certainly has caused no loss of people willing to work on Linux.

Book Bad Pharma thoroughly debunks this idea. These companies spend more time and money in lobbying, marketing, and rehashing the old research.

Well, and now they're going to spend a lot of computer time finding parallel constructions for the same molecule they want to patent, so they can patent those as well.

This is how is it in every industry. Almost every startup spends more money on boosting traffic and sales than on coding the product itself, too.

There are some differences between a company charging $270,000/year for a medication you need to survive [1] and a company trying to sell you on 'Uber for dry cleaning,' or whatever.

[1] https://99percentinvisible.org/episode/orphan-drugs/

Fascinating read. I am amazed how they have allowed to create monopoly on the back of goverment subsidies.

But wait, these drugs won't even exist if they didn't charge as much. They are needed by very few people (so if the price was $1000 they won't boost the number of patients so much, as the diseases are very rare), and drug research and regulatory process of approval is very expensive.

I'm not convinced drug research is expensive. I've met grad students. I've seen how much they do and how much they get paid.

That leaves the regulatory process, and if the problem is government intervention in the free market with overtones of regulatory capture (drug regulation being too excessive), the solution is certainly not more government intervention in the free market with overtones of regulatory capture (patents).

That leaves the regulatory process

Well, that and rampant price gouging.

the solution is certainly not more government intervention in the free market

You should really listen to the podcast I linked to above.

Not sure I understand - are you agreeing or disagreeing with me? The podcast describes how a well-intended government intervention in the free market and how it led to rampant price gouging.

I would be more than thrilled if we got rid of the free market entirely. But that seems unrealistic, and if we're going to have it, the government needs to be very much on guard against saying "We can fix these market inefficiencies by letting these people make more profit." I maintain that if the government had instead caused pharmaceutical companies to make less profit from their regular drugs, orphan drugs wouldn't have looked so unappealing in comparison.

> I would be more than thrilled if we got rid of the free market entirely.

It is a market controlled by various lobbies focused on increased the shareholder value. How is it a free Market? In a free market, there shouldn't be monopolies.

Goverments first sitting idle and allowing the companies to grow like cancer and then rushing in to save them when they crash in the name of saving the economy is anything but free market.

Fine, I'll rephrase that as "I would be more than thrilled if we got rid of the allegedly-free market entirely." Let's admit that truly free markets are an unstable equilibrium and that government regulation can actually do some good, and let's stop being afraid of government-run markets.

> But wait, these drugs won't even exist if they didn't charge as much.

Why they should not charge fairly? How much is as much?

On the issue of innovation, profit seekers does not have enough courage to do this. Board of big companies live quarter to quarter because their compensation is directy linked to stock value. It doesn't matter how much money they already have. Here is quote from the ted talk[1] linked below.

> In the drug industry, for example, 75% of the “radically innovative drugs” in the United States are researched in nationally funded laboratories. Whereas drug companies focus their efforts on “me-too” drugs — varieties of proven, profitable drugs — governments will take on potentially game-changing “new molecular priorities,” despite the risks involved.

[1] https://blog.ted.com/government-investor-risk-taker-innovato...

Edit: Point about charging.

Sure, but startups also tend not to patent their work. When one startup dies, another can make another attempt at the same thing.

(I'm not saying patents in the tech industry don't happen ever - at the last startup I worked for, I got three patents - just that they don't meaningfully limit the ability of other people to replicate the work or compete directly, the way pharma patents seem to.)

Lots of misinformation in these comments. Both the process and end product are patentable, but some companies do not disclose their processes and opt to keep them as trade secrets.

As for the article, changing the molecule can have unpredictable effects on how the compound acts in the body. Even changing one bond/atom can result in dramatically different properties. The article doesn't address that.

The Cell article and the In the Pipeline post explain more clearly: this system plans synthesis of the same final molecule by different pathways. The only patents it can detour around are process patents. Since the final product is the same, its properties also remain the same.

More relevantly, this scheme doesn't address the prescription side at all.

Doctor: I judge you to have [medical condition]. Here is a prescription for [expensive on-patent drug].

Patient: I would prefer to take this less expensive drug, which a computer program thinks might be equivalent.

Doctor: Hmm. Doesn't look like there are any clinical studies for this analogue drug. Since I don't want to risk a malpractice lawsuit, I'm not going to prescribe it. Without a prescription, your insurance won't pay for the analogue drug either, so it'll have to be cheaper than your copay in order for it to make sense for you.

The conversation goes more like this:

DOCTOR: You have [medical condition]. We can manage it with [medication which is on the appropriate tiered formulary for the patient’s pharmacy benefits manager].

PATIENT: I would prefer to take this medication they advertised to me during the last football game.

DOCTOR: This medication is only a slight incremental improvement over the generic one approved by your insurance. If you want the newer one, it may have higher copays and will have more paperwork hassle as we try to get the insurance to cover it. In the end, it may not be covered. Do you still want it?

PATIENT: Yes. [Leaves angry Yelp review over how the physician wouldn’t prescribe them X medication].


As for the final point, if there are no clinical studies (at all) for a given drug, then it is not FDA approved and would be illegal to prescribe unless under the auspices of a registered clinical trial with a filed IND with the FDA.

If it is FDA approved but not for that specific indication, then it depends on the clinical practice patterns at large whether this is a reasonable course of action vs malpractice. Case in point: Avastin is way cheaper than Lucentis for macular degeneration and works via similar mechanism, but was never tested or approved for that indication. Many Retinologists will inject Avastin anyway because some trials have shown benefit for cost conscious patients or patients from whom insurance will not cover Lucentis. This is not malpractice.

Using another drug from another class entirely without any clinical evidence to back up its use outside of a trial could be considered malpractice, but only provable if there is a bad outcome.

Without a prescription, they aren’t getting the drug legally. With a prescription without insurance/PBM approval, then yes, it could be very expensive for unconventional treatments.

But isn't a modified molecule a reformulation, which cannot be assumed to have the same effect, or dose-response, or even safety, as the original molecule? Say if you can synthesize something _like_ garblezolib or whatever, aren't you then faced with all the safety and effectiveness trials that were so expensive for the OEMs?

The end molecule is not patented, the process and pathway to creating it is.

If this software can come up with sufficiently novel pathways other processes can be used to come up with the same drug.

This is the idea anyway.

It could bypass patents, but instead it is owned by Merck, Giant Pharma so they can now bypass all the patents for themselves

Exactly! I’m sure the first customers of this software will be the big pharma patent departments who will use it to further expand their patent space.

I love it when sci fi and sci fact seem to merge https://www.npr.org/2017/09/20/548664924/in-a-future-ruled-b...

This article is not complete. In order for a drug to be approved by the FDA for a particular condition, it must demonstrate (through expensive clinical trials) superior efficacy to existing treatments unless the drug is considered a generic. It's hard to say if a drug with slight molecular modifications could be said to be a "generic" form of another. There's a good chance the answer would be no, because slight molecular modifications can cause significant differences in efficacy.

Some commentary from my favorite chem-blogger: https://blogs.sciencemag.org/pipeline/archives/2019/01/18/au...

Sounds like this is for finding different synthetic routes to the same molecule. So it might help avoid method claims, but not claims for the product itself.

Information wants to be free.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact