Hacker News Reader

Comment by fivesigma | Hacker News Reader

Not US specific:

* Ridiculously long patent validity periods. Let's put aside the fact that patenting a fucking molecule is even a thing, 20+ years is way too long a term to guarantee profit to a private corporation. Enforced by governments.

* Heavily subsidized (read: price artificially propped up) by insurance laws. This is student loans all over again. Again, created by governments.

* Astronomically high R&D costs, most of which is the cost of obtaining regulatory approval. Enforced by government agencies.

Free market? Yeah, right.

View on HN

Is 20 years actually that long for a drug? Drugs tend to have 12-15 years of actual patented time in market given the time eaten up the approval process. It's not like software where that time lag might make the patented drug obsolete by the time it becomes generic.

What "insurance laws" "subsidize" drug costs?

The cost of regulatory approval is high, but that's justified because the economic cost of a bad or even just ineffective drug piles up very quickly. Reduce the life expectancy of 1,000 people by 10 years (or give them an ineffective alternative to a drug that could extend their life by 10 years) and that's a billion dollars of damages right there. Is there a cheaper solution than extensive clinical trials that accomplished the same result?

It's not clear there is. Remember, what really makes approval so expensive is that most candidate drugs don't get approval. So the investment into all the drugs that don't work has to be recouped from the ones that do. But the fact that so many drugs fail at the clinic trial stage suggests we don't have better ways of predicting whether drugs will be safe and effective. Drug companies don't invest into drugs that don't pass approval for funsies. If they could determine a drug wasn't going to pass approval before the clinical trials, they'd do it. It'd be pure profit for them. Also, the fact that so many drugs don't pass approval suggests that the FDA is keeping a bunch of ineffective and harmful drugs off the market. The cost of getting rid of the FDA has to account for the cost of those bad and ineffective drugs.

Also, getting rid of these things would not make the drug industry a "free market." In any advanced society, the government will pay for drugs for elderly and poor people. That creates an entity that can exercise monopsony power. Get rid of patents on drugs, then you get the free rider effect undermining R&D. And get rid of the FDA, and you end up with a market with huge information asymmetries that allow drug manufacturers to sell products that don't work without customers having a way to know that. These would destroy any hope of a functioning competitive market. You can't talk about markets while pretending only the parts of economics that comport with your world view exist.

> The cost of regulatory approval is high, but that's justified because the economic cost of a bad or even just ineffective drug piles up very quickly

I wonder what the result would be of vastly reducing regulatory approval requirements, but in exchange giving pharmaceutical companies unlimited liability. So, sure, put that poorly-tested drug out there, but every person who has horrifying side-effects is going to sue you for millions

> I wonder what the result would be of vastly reducing regulatory approval requirements, but in exchange giving pharmaceutical companies unlimited liability

I don't think lawsuits are the best "first line of defense" for protecting consumers against harmful drugs. Because the usual measure of damages is the cost of harm, the actual recovery (once discounted by litigation risk and diminished by legal fees), is a fraction of the cost of the harm. That discount is particularly significant in medical cases, because proving causation is very difficult (how do you know that this drug caused your dad's heart attack instead of the three others he was taking?).

Moreover, drug-related litigation these days relies heavily on the factual record developed by clinical trials and FDA investigations. Class action lawsuits are often precipitated by an FDA action. E.g. major litigation like Fen-Phen and Vioxx, that netted tens of billions of dollars in recovery for consumers, were kicked off by FDA recalls. Individual litigants are poorly positioned to develop the scientific record necessary to support a lawsuit.

For one, drug companies would probably start making people sign liability waivers before agreeing to sell them a lot of newer drugs.

Define your expectation of an advanced society please. They don't come out of a mold.

Perhaps an advanced society should involve one generation taking care of the next instead of the government?

That's what we do, this generation buys drugs for the previous generation. We take advantage of economies of scale though by pooling a bunch of money together and paying some people to coordinate it for us because that's more efficient and effective than expecting 150million do it individually.

> That's what we do, this generation buys drugs for the previous generation.

Not really. Indeed, given voting patterns it's more like the older generation empowers enforcers to rob the younger generation and then empowers others to use those stolen funds to buy them drugs.

Note that companies have also been fighting for such standards in order to prevent new entrants and restrict competition. So it's not just governments at fault, there is complicity between the two.

It's the same in every regulated industry: auto, insurance, finance, education. All it does is drive up the cost like crazy which gets passed on and people still get scammed, but pay extra for the privilege.

It defaults to higher costs, but that's not enough to say it's worse. I don't care if milk could cost half as much if it has a significantly higher chance of making me sick.

On the other hand if I had less money I might be more willing to take risks. Thus, regulation is not an independent thing and depends on the economy it's operating in.

PS: Historically milk was surprisingly deadly. Without regulation it would likely be much safer than that today, but not as safe as it currently is.

You do realize that milk is heavily subsidized to be priced where it is, right? I would much rather pay a local farmer the market amount and have a relationship with him than to pay into the bureaucracy and support those subsidies with all kinds of taxes.

I don't actually drink / buy milk directly so the sticker prices has little meaning to me.

As to having a personal relationship with a farmer, I grew up next to a farm. It's very much an industrial process.

You know that's the first time I've ever heard that argument. People usually just freeze and don't know how to respond. I'll have to think about this

> Let's put aside the fact that patenting a fucking molecule is even a thing

Assuming you don't find fault with the mere concept of patents, I am legitimately curious why patenting a molecule shouldn't also be possible.

Because it's a discovery rather than an invention. Though the mere concept of patents is already a fraud, as is clear when you start to look closely the arguments. See http://hintjens.com/blog:31.

I would argue that any invention is a discovery of sorts. The inventor of the electric coffee maker didn't fundamentally alter the universe to cause it to be possible for a device to brew coffee.

And even if you don't buy that argument (it's admittedly tenuous) the fact that it's a discovery wouldn't convince me that the discoverer shouldn't still get exclusivity rights, given the amount of research involved in discovering new molecules and shepherding them through the approval process.

Could you possibly find something a little less conspiracy theory looking? Also something a little less manifesto.

It's my own blog. I spent two years fighting software parents in Europe as president of the largest NGO involved (FFII), and it was pretty clear that we were fighting a well organised and funded lobbying machine. After some years of analysis my conclusion was that software was not in fact a special case, and that all patents were a problem. Sorry to not have made it clear that I was linking to my own work.

> Could you possibly find something a little less conspiracy theory looking? Also something a little less manifesto.

Clearly his own blog.

Oh, so it is. Good catch.

The question is whether that reward benefits society or not. Do we get innovation without it? Are there studies that show patents preceding innovation rather than following it?

you're absolutely wrong.

if you think finding out which molecule fights cancer cells vs kills you outright and how to produce it in controlled fashion is trivial, you're a moron.

if you think creating genetically enginnered bacteria to produce insulin is trivial, you're an utter moron. lest you like injecting harvested insulin from pigs or dead humans.

HN likes to pride itself on techie knowledge, but the cluelessness around actual hard tech is staggering. modern medicine is up there with rocket science, worthy of nobel prizes.

they just CURED hep C for FUCKS SAKE. but yeah, no reward for that, but sharing catpics - here, have billions.

Nobody's questioning whether these things are non-trivial or high-value. There are very legitimate questions to be asked, though, especially regarding patentability and profitability.

There are many other fields with very high barriers to entry and which require very large R&D budgets, but which - unlike the prescription drug market - nonetheless manage to attract competitiors: The semiconductor industry, for example, comes to mind.

hardware is full of patents. your point?

pharma is full of competitors. exactly because it rewards risk. the whole point of the system. and the patent cliff hits and generics and biosimilars take over.

Most of the patent drugs never existed until somebody first synthesized them. How is that not an invention?

This is an absolutely incorrect assertion: drug discovery is but one part of the drug development process. You're ignoring (among other things) the work in medicinal chemistry, retrosynthesis, and scale up/chemical engineering that go into creating a drug. The compound at the end of the development process rarely looks like the one identified in initial screening, and the work done to manufacture it and bring it to market unquestionably count as inventive.

Great write-up. More as a reference work to draw on. There's simply too many angles and information in it. Do you mind if I present a condensed, high-impact version of it should I run into those that can influence the patent system? I'll credit you as the original source with mine being a re-hash or revision.

Go for it.

What if it's an engineered molecule?

You don't patent the molecule.

You patent the method of synthesis or extraction, and you patent the method for treating a human disease or ailment.

If you discovered a novel use for acetylsalicylic acid (aspirin), you could patent that. You are not re-patenting aspirin. It can still be used as an analgesic.

If you discovered a novel process to produce ethanol, you could patent that. You are not re-patenting alcohol. People could still use the traditional fermentation and distillation process. You could even patent a process step to remove the remaining water from azeotropic distilled alcohol.

Obviously, such patents are more monetizable if the molecule in question is expensive to produce by the public domain processes, has no other commonly known uses, or is a controlled substance.

There are composition of matter patents that allow you to patent a novel molecule: https://en.wikipedia.org/wiki/Composition_of_matter

That's perfectly in line with patenting inventions, but not discoveries, and implementations, but not ideas.

So in order for me to accept a patent on a molecule, it would have to be found nowhere else in nature or in prior art, and the applicant must have synthesized a measurable quantity of it.

In practice, this means patentable molecules have to be rather complex, because there are a ridiculous number of compounds that have been created once, for someone's dissertation, or to create catalog entries for boiling points, freezing points, dissociation constants, and such, and then never looked at again.

Quite a lot of "new" drugs were actually first described decades ago and then shelved. There are just too many possibilities in organic chemistry to examine each in detail without a specific application already in mind.

Would you actually want drug research to _not_ be regulated?

I want drug research (and approval) to _not_ cost billions of dollars per drug just to justify the salary of a few bureaucrats.

Not only it's expensive, but tons of existing drugs have been approved with loose standards (a single trial showing better efficacy than placebo, wow, big deal!). Every new drug should be benchmarked against the best drug in class, and period. Instead of that you end up with doctors who have no idea which drugs work the best and who rely on companies to "educate" them, with obvious conflict of interest.

"Best" is rarely a clear cut answers. There is pretty wide variation in the several billion people on the planet. Some have serious side effects, others have none. Some drugs work well as part of a cocktail, but not as stand alones. This idea of selecting a single best option and making everything relative to it would have a strong adverse impact on the market. It would be like saying we shouldn't clear a Ferrari when a Corolla can get groceries just fine at a lower cost.

That is the current standard (Ph 3 clinical trials against standard of care). There are rare exceptions to this, but the ones I can think of are drugs for which there is no current standard and drugs for which the current standard doesn't match the intended indication. In modern medicine, we don't just take diagnosed patients who meet criteria for a trial and let them progress without disease.

Thats why I mentioned "have been". But such drugs are still being sold.patented and used.

Yes; our system makes it very difficult to put horses back in the barn. I heard one physician say that Tylenol would not make it through today, or at the very least it would be prescription, because of the slim toxicity margin.

Tylenol causes ~56,000 emergency room visits, 2,600 hospitalizations, and ~450 deaths per year.

IMO, at a minimum it should have much stronger warning labels on it.

Out of curiosity, is there any reference that breaks down the costs of approval process? Like how much goes to fees in trial, staff, etc.

You think the drug approval process is only to justify the salaries of a few bureaucrats? there is no other reason? safety doesnt factor in at all? I must be misunderstanding you here

Don't forget to add stuff like banning Canadian drug imports. Medicine absolutely should be validated before being sold in a country, but other than that, I see no reason why Canadian companies, or companies from other countries, couldn't sell a competing drug in the U.S. for whatever price they want.

Obama played a big role in banning this type of imports. Just saying. And TPP/TTIP would make this problem much worse, and virtually irreversible, because then they can just say "well, we wish we could change the monopoly laws now, but it's in the treaty, you see...so tough luck."

There are ridiculous levels of regulatory capture. E.g. epi pens are off patent and there are many competing companies in Europe, but the FDA refuses to approve any of them. Or cancer drugs that have been used for decades in Europe very effectively with tons of data there to support them, but the FDA refuses to allow them. One time they even required a drug to have a trial with more patients than exist in the US with that disease. They kill way more people than they save by preventing terminally ill patients from getting life saving drugs. And raising the cost developing them.

The FDA needs to be shut down. Perhaps replaced with something like a prediction market, where people can bet on how likely a drug is to pass a trial or even work in specific patients.

> epi pens are off patent and there are many competing companies in Europe, but the FDA refuses to approve any of them

Not entirely true: there's adrenaclick. But if a doctor writes "epipen", the pharmacist isn't allowed to give you an adrenaclick, even though that kind of substitution is usually allowed. http://slatestarcodex.com/2016/08/29/reverse-voxsplaining-dr...

It's not allowed in this case because the FDA, rightly or wrongly, wants the patient trained in the use of the specific injector they are being prescribed. Having never needed one I don't know what the practical differences between the operation of an Epipen injector and Adrenaclick injector are, but the FDA seems to think they exist.

One reason Pharma companies in other countries can sell drugs for far cheaper is that actual manufacturing costs of most drugs are pennies. Once an American company discovers a safe and effective drug, foreign companies can simply look at the publications on how to synthesize the drug and don't have to spend any cash on R&D.

> 20+ years is way too long a term

I've heard that the approvals process eats up almost half of those years. (Or if not the approvals process, then something that's going on after they get the patent but before they're allowed to sell it.)

This is usually true. The patent is filed before you start clinical trials, because it must be described in publications. Those trials together often take 5 years or more. Only after that can you even file the New Drug Application or foreign equivalent.

Edited to add: Also, people often lump the market exclusivity extensions in with the patent. The FDA offers companies and incentive to try their drugs in children. Without this, almost nothing would be approved for kids because the trials are ridiculously expensive (even compared to adult trials).

> Astronomically high R&D costs, most of which is the cost of obtaining regulatory approval.

And yet, IIRC, pharmaceutical companies spend about 20% of their revenues on research, but about 40% or more on marketing. These costs are galaxonomical!

The average number of emails you must send before a subscriber starts to consider you safe to buy from.. Eight.

Assuming one's health has an even higher threshold, and that a pharma company can't exactly cold email everyone in the U.S, well... That's a lot of commercials to get the word out. That's 800,000+ doctors to educate on a new option..

Marketing always goes hand in hand with product development.

I'd rather have a pharma company that's still in business charging $30,000/mo for a solution vs a company that charged the minimum, and thus has thrown away any chances of making successful products in the future.

Reply on news.ycombinator.com