Hacker News new | past | comments | ask | show | jobs | submit login
Carcinogens Have Infiltrated the Generic Drug Supply in the U.S. (bloomberg.com)
254 points by refurb 32 days ago | hide | past | web | favorite | 135 comments

> By 2014, the FDA had closed its offices in Shanghai and Guangzhou, leaving only the Beijing office with inspectors who could visit Chinese factories on short notice.

> The FDA checks less than 1% of drugs for impurities or potency before letting them into the country.

For a regulator body who's job it is to make sure medical products are safe and available. These comments, if true, lend credability that the FDA is more regulatory capture than actual useful regulation.

> The FDA’s relationship with manufacturers like Huahai, on the other hand, isn’t simple at all. If Huahai wants to make its own version of a generic drug and export it to the U.S., it needs FDA approval. But if Huahai supplies the main ingredient to a company that finishes the drug and sells it in the U.S., it’s required only to keep the FDA informed of any changes to the manufacturing process.

Good, now you have a US company you can actually prosecute severely. The $/mass of phamacuticals means testing a small sample of each batch is super easy and if you know the manufacture process it's not hard to know what secondary products/contaiminents you need to test for. This is pure negligence on the FDA's and the US importers part.

> “The only element who cares in this whole global supply chain is patients”

Not if you actually fine the American importers into the ground. And ban the at fault over seas companies.

>> The FDA checks less than 1% of drugs for impurities or potency before letting them into the country.

> For a regulator body who's job it is to make sure medical products are safe and available. These comments, if true, lend credability that the FDA is more regulatory capture than actual useful regulation.

I wonder though if this statement misrepresents the facts. If they check only 1% of the entire import volume at random, that is totally normal and what I'd expect them to do.

I agree, I think the statement is ambiguous.

1% of all products (e.g. one pill from every bottle of 100 pills) being tested seems like plenty.

1% of shipments (1 out of 100 shipments is sampled) seems insufficient.

Very ambiguous.

Generally how it works is you submit a proposed purity level to the FDA, with rationale and backup by your own validated analyses.

The FDA dives into it and says “this is reasonable” or “this won’t work” or “based on the chemistry, you need to test for X also”.

Then they require you to “release” any product you make, showing that it conforms to the agreed upon purity. If it fails, they expect you to follow up and figure out why.

Note, most of this is done in good faith - the FDA isn’t repeating the science in a lab, but they pay enough attention that if you’re making things up, it’ll be apparent in the data. The CMC (chemistry manufacturing controls) section of the NDA can be hundreds of pages long.

When it comes to inspections, they are typically random (but if you have a bad track record they happen more often). They can happen any time and they’ll look at all your processes, not just for a specific drug. Don’t have a record that you cleaned your equipment between runs? That’s a paddlin’.

Repeatedly run tests until you get a “pass” result? Yeah, your product will get banned from importation.

But yes, much of the regulation is based on good faith simply because the manpower required to verify everything made would be massive.

“much of the regulation is based on good faith simply because the manpower required to verify everything made would be massive”

This is where I wish there were more tolerance for enough government spending to employ as many people as it would take to test properly. We’re talking about medication being used multiple times a day for long term-amongst many uses.

The FDA has been downsized many times over the past three decades. So what we're seeing here is not at all unexpected.

The FDA’s budget is more than 6x what it was in 1992.[1]. An increased every year since then.

What do you mean downsized?


That's no inflation adjusted. And it's also not adjusted for processing load. But it does show the greatly increased role of user fees, which I gather has facilitated regulatory capture.

Inflation hasn’t been 600% since 1992.

Not sure what processing load is, but if they’re asked to do more that could be an issue, but I wouldn’t call it a cut.

I do remember reading about cuts in FDA resources. As an excuse for why they'd missed so many violations.

Checking, I see that it was Reagan era cuts.[0] So the timeline that you cited begins in 1992, which was probably the minimum funding level.[1,2]

Also, Hatch-Waxman was approved in 1984, which increased the workload considerably. And then there was the first generic drug scandal in the late 80s. That's what I was thinking of.

0) https://academic.oup.com/shm/article-abstract/29/1/202/24722...

1) https://www.everycrsreport.com/reports/R44576.html#_Toc82198...

2) https://sci-hub.tw/https://doi.org/10.1093/shm/hkv106

Even 1% shipments would be okay, IMO. I interpreted it as "99% of drug products never get inspected", which is way worse.

It’s more like the later than I’m comfortable with tolerating.

The think the accurate statement is that 1% or product lines have had a shipment sampled. The FDA is not in the business of performing QA testing, that is the job of the manufacturer. It is better to think of them as a legislative body.

Maybe a special duty office for this should be created. (Like for meat and meat products in EU.)

I agreed with that statement, but from the perspective that the FDA should drop its regulatory capture activities, and increase that 1% to a bigger sampling N. That is, they're focusing their efforts on the wrong problems.

The problems in the article amounts to fraud. Police that and let people decide what they want about what they're seeking.

1% of the entire import volume would far more than necessary to test so it made more sense that it's only 1% of batches or shipments.

> These comments, if true, lend credability that the FDA is more regulatory capture than actual useful regulation.

It lends more credibility to the fact that the FDA is notoriously underfunded when it comes to ensuring the safety of drugs[1].

[1] https://www.ncbi.nlm.nih.gov/books/NBK52926/

Underfunding certainly can(has?) interfere with their ability to do their job. But proper funding might not help if regulatory capture is also in play.

I'm of the persuasion that governmental organisations need to be shaken up from time to time to prevent ossification, stagnation, inefficiencies, regulatory capture, ideological capture, etc... but the new agency needs to utilize the pieces from the old agency (not to start from scratch).

An industry that sells drugs would want the regulating authority to be underfunded when it comes to insuring the safety of drugs.

Really? Back when the FDA was underfunded in the 80s, there was a massive backlog in drug approvals that cost drug makers billions. They happily coughed up fees to better find the FDA.

Seems like "every company in drug supply chain must independently test each batch of received drug via GC/MS. That company is liable for any impurities found downstream in un-rejected batches" would be a nice regulation.

Companies don't want to do their job, they say this is "excessive regulation that is killing the economy".

This already kind of happens. You declare a starting material and then test it to make sure it’s pure. You do the same for the next step all the way to the finished product.

In this particular instance, the process caused the impurity, so it wouldn’t be reasonable to punish the company that sent them starting materials.

Companies that import drugs or precursors from China should independently test the product. Use a block chain to enforce trust.


It's a jobs program for GC/MS technicians! /s

> Not if you actually fine the American importers into the ground.

Sounds like an invitation for the importer to be a nearly judgement proof shell that just gets abandoned at the first sign of trouble.

> And ban the at fault over seas companies.

That part sounds more effective.

Jail time for high level execs will make sure that they don't cut cost by neglecting quality.

Sure, but it's not clear-cut. Executives can be (in good faith) fooled and that's a problem. You can't just jail people just because. I'm happy seeing the Purdue Pharma people get strung up, but I think we need due process and reasonable doubt as well.

They aren't being strung up, by any stretch.

Considering the fine American tradition (dramatically amplified in the last decade) of underfunding every important regulatory agency and then hollowing it out through other methods, it doesn't seem very surprising that the FDA is bad at checking products. Hard to imagine closing offices because of "regulatory capture", if anything the big US companies would want more enforcement on foreign generic drug manufacturers to secure their hold on the market.

Dunno how much the US FDA is to blame here: these API plants pretty much supply the world with generics.

If anything, it's the FDA that's doing foreign plant inspections and every other country's regulator just does what the FDA does.

My real question is: Why is each country doing its own inspections? Why not one global inspection?

Everyone else is getting a free ride from the FDA's work.

> For a regulator body who's job it is to make sure medical products are safe and available. These comments, if true, lend credability that the FDA is more regulatory capture than actual useful regulation.

This feels like a typical right-wing playbook item: defund and de-scope government organizations until they can't be successful, then point at how bad they are, and privatize them or shut them down. Government agencies don't work unless you pay for them.

Regulatory capture is a separate issue from privatization.

Regulatory capture is when there is a powerful government regulator, but it has been compromised by the companies and is used to prevent competition to those companies.

For example - Boeing successfully convinced federal government regulators to prevent Bombardier from selling a 737 competitor in the US.

Indeed, although I suggest that that comes about in part because the government fails to fund these organizations to the level required for self-sufficiency forcing them to outsource more and more of the work onto the only "competent body" willing to take on the job free of charge: the companies themselves. This creates a relationship where the companies have a disproportionate share of the power, and can dictate things like anticompetitive practices. It's basically a case of "give us what we want or your costs skyrocket."

That's all beside the point.

As long as outside entities can dangle future low-effort, high-compensation posts to government officials in exchange for present favors, regulatory capture will remain a fact of life.

It's an incredibly tough problem. Even if we prohibited regulators from ever working for the regulated, it's so easy to get around that with nonprofits and shell corporations.

Really? It's 2019, there is no argument to be had over this.

Impurity problems have occasionally come up over the years, but they have largely been addressed quickly by existing regulations. And at no point have they ever approached the level of risk posed by the illnesses they are prescribed to treat.

While the risk of impurities in the generic drug supply certainly warrant research, the percentage of issues again doesn't approach significance with comparison to these pervasive illnesses.

Nothing is risk-free. Eating food has risk, taking paracetamol has risk, getting blood drawn has risk.

Saying something has risk without talking about the amount of risk and the context of the risks it mitigates is pointless.

Drug impurities in generics introduce negligible amounts of risk. Eating peanuts in public or going for a drive is much riskier, and we almost all do those things without a second thought. By contrast, access to affordable drugs mitigate huge risks: the illnesses we medicate can, did and do cause horrific harm up to and including death, and do so very well.

The pharmaceutical industry is not immune to criticism, it is just they have weathered all the reasonable criticism leveled at them. If you are willing to drive in a car, taking generics that could should be a complete non-issue for you.

In an era of fake news and increasing doubt in the medical establishment, publishing reports like this is not only irresponsible, but unethical. This only adds fuel to the fire of unwarranted skepticism towards our medical system.

I thought the article did a decent job of explaining exactly what the problem is in this case and communicating the scale of the risk. Your comment, on the other hand, sounds like a kneejerk dismissal entirely based on the headline not agreeing with what you thought was true.

Yes generics have proven to be broadly safe. That doesn't mean there aren't still issues that need to be addressed by tweaking or adding regulation. Yes even with carcinogens the contaminated generics are probably better than taking nothing. That doesn't mean that the contamination is acceptable.

Exactly, I sort of failed to highlight this directly in my comment but the fact that the article focuses on a single case of tainted generics vs who knows how many people die each year waiting for drugs to be approved and/or imported is pretty insane.

Apparently NDMA has also been found in Zantac (ranitidine), a common OTC heartburn medicine.

The FDA has issued an alert - https://www.fda.gov/safety/medwatch-safety-alerts-human-medi...

And it is not just the generic, but the original too:


And I would have never learned about this unless I was surfing hacker news. Bleh.

From the NDMA wiki: [NDMA] is also used to create cancer in rats for cancer research.

That is super alarming.

Yep. Just happened to see this [1] article out of the corner of my eye while browsing. Really concerning. And from the sound of it, this Valisure place, not the FDA, is the one who discovered the issue. One wonders what use regulators are these days.

My infant daughter was prescribed ranitidine for reflux from age 3 months to about age 6 months. Anyone with more chemistry knowledge than me want to venture any guesses about how concerned I should be?


Do you have any of the pills left? You could get the leftovers tested. But I would ask your pediatrician. I would guess you probably don't need to be specifically concerned.

Yeah, I replied to another commenter, but I have some of the syrup left over and I'm sending it to be tested. Turns out the folks who discovered this have publicly-available testing. I'm thinking that the limited duration is a good thing. Fingers crossed.

It's a different issue than the valsarten issue. Valisure is suggesting that ranitidine breaks down itself into NDMA, which, as an organic chemist, I guess could happen to a small degree based on the structure[1], but it would be hard to mimic the conditions inside the body in a lab.

The Valisure petition speculated that the source of the NDMA was the result of the “inherent instability” of the ranitidine molecule, which can degrade under certain conditions, such as when it is digested, to create NDMA.


> I guess could happen to a small degree based on the structure

There's no N-N bond in ranitidine. And nitration/amination isn't a metabolic pathway for good reason.

Must be a funky re-arrangement reaction if it is from breakdown.

My money is still on side-reaction from synthesis.

Ranitidine has a nitro group which can be cleaved and reduced to nitrous acid and combine with the dimethylamino group to give NMDA.

Whether that’s a reasonable metabolic pathway I have no idea.

They offer that but little additional information. Is it just Zantac the commercial brand? How do I know if I’m affected?

Not an ad, just my experience: I googled the company, Valisure, that was mentioned in this article and another in the NYT. In addition to selling drugs that they've tested, they'll also take a sample and test your own prescriptions. Price is $40 for their "impurities/carcinogen" analysis. Worry may well be overblown at this point, but my kid was prescribed ranitidine and some peace of mind is worth 40 bucks to me.

Not sure. It is indeed pretty alarming! I take it every day (and have for years) for GERD! :/

Guess I'll try switching to something else and see how it goes.

I would pay good money for a supply chain based “consumer reports” so I actually know what’s in my products eg

- medicines! - supplements - food - clothing

Clearly in the USA their is gigantic demand for organic / harmful chemical free products.

The article talks about a pharmacy that does just that - tests all the products before dispensing them and rejecting those that don't pass.

For food at least I think it may be somewhat amenable to some of the newer fingerprinting techniques (e.g. using DNA microarrays). I've been wondering why origin/content/quality validation for food based on actual biological science doesn't exist for years. I did biochemistry for undergrad, and some of those labs could be done in the average RV kitchen, and others could be packaged into test kits.

For supplements you should look at labdoor.com. They do independent lab tests for label accuracy and ingredient safety.

> I would pay good money for a supply chain based “consumer reports” so I actually know what’s in my products eg

That's already existed for decades, so why aren't you paying for it then?

Seems like a good differentiator for a retailer... Maybe online only retailer. Maybe someone like Walmart trying to compete with Amazon.

Amazon (ironically enough, given their counterfeiting problems) has been doing this for their Elements supplement brand.

Example: https://www.amazon.com/Amazon-Elements-Multivitamin-Cultured...

You can scan a QR code and get the test results for the specific batch included in the bottle.

The FDA is intended to be this with teeth - clearly it has rather corrupted. What makes you imagine a private organization would be better?

It's not corrupted, but too small and underfunded, while the scope is huge. Really.

This could be an interesting case for blockchains. I normally am not a proponent of seeing everything thru that lens, but having a proper chain that is publicly verifiable is in the public’s interest. That said, there’s not a lot of incentive for people to reveal their supply chains — it’s often trade secret or hidden for a reason.

If there was a market behind it, then there might be a way. But good luck getting Samsung to tell you where their cobalt came from, or if some generic drug was made in a dangerous lab. Even if you knew it was Generic Factory 1736, it’s hard to know if that’s good or bad.

I carry a lot of guilt for having "oh, blockchains could help" thoughts quite a lot. A federally supported blockchain used as a public notary and other official functions, as you mentioned, seems like it may have a place in the future.

Someone really needs to create a version of the cloud to butt extension that replaces "blockchain" with "database".

What you're describing is a useful application for a _database_. Mixing "blockchain" into it would only hurt the viability of the database, not bring any obvious benefit.

Your comment makes so much more sense with the word database:

"A federally supported database used as a public notary and other official functions, as you mentioned, seems like it may have a place in the future."


Normally I’d agree with you, but given this is all about verification it might make sense to distribute both trust as well as usage.

You know, of course, blockchains do not magically solve the garbage-in garbage-out problem.

Of course not. But that’s true of any system. Instead, you’d have to rely on each piece of the chain feeling confident about what’s before them... if they are professional there should be non-technological ways to attack that.

I wonder if the point of this, is to have us weigh getting "brand name" more seriously. There is no sensible reason to pay a premium for brand name if the generic is chemically equivalent.

This another in a long string of failures in generic regulations that underscore broader problems pertaining to regulation in the US healthcare system. There's this theoretical land of regulation, what the consumers and citizens are encouraged to presume, and what actually happens.

It's been common, for instance, for people to repeatedly be berated for paying more for name brand. Studies in fact, of the public health economics of it, why people pay more for name brands, trying to understand it as if there are deficiencies in consumer reasoning.

This is despite repeated problems with generic suppliers overseas, mainly India, diluting medications. This was uncovered, problems were corrected, and then it happened again.

And now this.

So who exactly is the rational purchaser? The person who buys generics under the assumption that they are chemically equivalent, or the skeptic who prefers to have transparency about where their medications are coming from, and purchases brand names?

My point isn't to berate the FDA as incompetent -- I think they're overburdened -- but I do think there are broader problems involving a disconnect between how we discuss regulatory systems in the US and what actually happens. To me, this is just the tip of the iceberg. My personal opinion is the FDA needs to focus solely on this kind of problem -- ensuring drugs are what they are labeled to be -- with maybe some substantial research into effectiveness -- and to drop its regulatory role as drug police.

I only buy generics from known reputable manufacturers, these are mostly locally made EU ones, bit more expensive than big pharma, say Teva or even maybe Sanofi (they're very likely local though) but at least predictable and it helps local business.

(Polish, Czech and Slovakian.)

Yes there is.

It's primarily based around different characteristics of how the medication dissolves in the patient's stomach, or the differences between fillers and binders.

Doesn't that mean they're not chemically equivalent?

My understanding is that only the "active ingredient" is the same. Everything else is subject to cost-cutting, as there are many fillers and binders to choose from, and they're priced differently.

The active ingredient is the same - the binders and fillers and capsule materials are mostly inert. This isn't true for all drugs - for the blood thinner Coumadin, you're meant to stick to generic or brand name and not switch between the two since there are known differences in efficacy.

I do not think the results for Coumadin/Warfarin are at all clear: https://www.ncbi.nlm.nih.gov/pubmed/21449627

Thanks for the update - interesting study.

I understand that. OP said:

> There is no sensible reason to pay a premium for brand name if the generic is chemically equivalent.

If they're not chemically equivalent and have different medicinal effects as a result, then it's much easier to justify a price difference. I understand chemistry is more than just a list of ingredients. It's how those ingredients are combined that gives them their unique interactions.

For sake of the discussion here, bio-equivalence is the more helpful way for meaningfully comparing generic to brand names. The filler might be different, but if the biological effect is identical at half the cost, one would be prudent to chose the generic option.

Yet we see drugs that are, say, 100 times the cost of the generic. So how can that happen; how is that sustainable? Presumably nearly everyone starts on the generic, and then switches to the brand name if they have problems with the generic. So there is actually not competition so much as price discrimination! You can't tell someone who the generic has failed that they are equivalent, because their only hope is that they are not! Not only is the generic no longer an alternative, but psychologically, it's easier to put your hope in a super-expensive option because of the cliche "you get what you pay for".

So I think telling people continually that generics are identical is not helpful, because it's not information. It's an assumption, which might be true or it might not be, but is essentially never helpful to any individual in making a decision.

Chemically equivalent but formulated differently. Think hot dog vs. corn dog: the meat is the same but the packaging is slightly different.

The packaging can have a surprisingly large effect on drug absorption, and therefore efficacy. Notably with difficult to dissolve drugs a lot of work typically goes into formulating them to be more efficiently processed by the intestinal tract.

The FDA approves generics based on bioequivalence, not just chemical composition. In order to be an "AB" rated generic (and used as a substitute for the branded product), the generic has to fall between 80-125% bioequivalence of the brand, which measure dissolution and absorption.

The fillers in the generic don't have to be the same as the branded product, but the drug does have to behave the same as the branded product.

There was an issue I read about with a drug where they did not test all the different dosages of a generic, and I think the behavior of one of the doses was not as expected because it didn't dissolve well or something.

This was a psychiatric drug, and so of course, reports of problems went unheeded for years because patients were assumed to be imagining things.

Saying "the drug does have to behave the same as the branded product" is like saying "the product as delivered does have to match the spec".

Yup. There have been a handful of approved generics that didn’t have bioequivalence to the branded.

It’s not a perfect process, but it works in nearly all the cases.

We don't know that "it works in nearly all the cases".

We know that in nearly all the cases, any issues don't rise to the level where the FDA or the public identifies them. I doubt that the system is constructed in such a way that nearly all the problems are identified.

It seems to me that because drugs are developed to meet the absolute minimum standards for proving efficacy and safety, there's inherently going to be little if any margin for error in the manufacturing process and hardly any way to analyze problems. When reporting potential side effects, most drugs are going to be just above the noise level in their principle effect, so how are you going to determine what is a random unrelated issue and what is a negative side effect, and what is the consequence of a QA or design failure?

"...pay a premium for brand name..."

My physician had prescribed "Diovan, not Valsartan" last year because of this. My pharmacist called me up as I was driving home from the physician's office. He assured me that he wasn't buying anything from that Chinese company, and observed that I'd be going from $65 per three months to $365 per three months. OK, generic it is...

If you haven't read it, I highly recommend the book 'Bottle of Lies' which was released a few months ago. It details systemic corruption in a number of generic pharmaceutical manufacturers.

It also details how the FDA is struggling with monitoring overseas manufacturing.

I just came to write this :-). Here’s the link for anyone interested


Wow, to formulate a generic drug and export it to the U.S. requires the FDA stamp, but if you generate components used in manufacture of US-made drugs get literally no FDA oversight. How can this be?.

I made this account to respond to exactly this question.

The FDA's legal basis to regulate medicines is actually incredibly weak and really hasn't changed much since the 1930s. One of the examples is exactly this case. The FDA only regulates finished medicines in two cases a) brand name or b) generic equivalents. Leaving aside brand name drugs for the moment getting approval to sell a generic equivalent doesn't look anything like what you would expect. What people generally think is that generic drugs are 'copies' of existing brand name drugs and the FDA cares about making sure the generic guys are producing the same drug as the brand name one.

This isn't what happens at all. A manufacturer has to demonstrate similar blood level uptake as the FDA approved equivalent drug using a very similar approval process for the brand name drug. This requires an incredible effort that is substantially the same as getting an original drug approved.

The bottom line is that the FDA doesn't examine the on-going manufacturing of the drug much or supplier reliability much at all. If a manufacturer can pass blood level equivalence nothing else really matters. This equivalence is only valid for the direct manufacturting producer of the drug, not the company marketing it. The reality is that if you want to get FDA approval for a generic you might as well partner with a firm outside the US that knows how to handle this process because its a waste of money jumping through FDA hoops to certify a new facility in the US.

Once you have proved equivalance you can start selling the drug. At this point, the FDA is basically done. The FDA basically dones't care anymore and has very weak sanctions to enforce safety compliance.

The FDA can't do anything about this situation on its own; the way drugs are approved is directly established by law that dates back to the 1930s. I notice a lot of people in this thread suggesting that FDA leadership is bought off and does the work of the drugs companies. I assure you this is not the case.

FDA oversight of pharmacutecial ingredients is accomplished under an entirely different legal regime that has nothing to do with drug approval whatsoever.

In principle the FDA should inspect the US-made drug after manufacture so inspecting all of their suppliers would be redundant.

For what it's worth many things are like this. It's the same thing with genetically engineered products, also under FDA purview. There is 0 mandatory oversight. If you have the clearance to sell a potato, then you can produce any engineered potato you like and sell it with 0 oversight, testing, or additional requirements. All the FDA offers is a purely voluntary opt-in consultation. [1]

Perhaps unsurprisingly this [2] was the head of the FDA under our last president. He was appointed a newly created executive role, not entirely jokingly referenced as the 'czar of foods.' A Monsanto vice president and lobbyist whose career highlights included arguing that companies should be allowed to knowingly allow at least a very small amount of carcinogenic chemicals into foodstuffs. Seriously, that was the head of the US Food and Drug Administration for nearly 8 years.

[1] - https://www.fda.gov/food/food-new-plant-varieties/how-fda-re...

[2] - https://en.wikipedia.org/wiki/Michael_R._Taylor

Are these problems only limited to the US? Do Canada or the EU have similar issues? Do they not import chemicals from similar suppliers or do they screen the better?

What about India? Are Indian generics affected as well?

Very likely too, including other issues previously found I'd expect anything to happen there.

It's an issue everywhere. Almost all generics are manufactured in India and China. And most with Chinese precursors.

Yeah but the article and some others are making it seem like something unique to the FDA. It seems to be a global issue and they all need to improve their quality control and testing processes.

To be honest, QC and testing has never really been the FDA's job. They perform a one time assessment for safety and effectiveness to prevent the sale of snake oil equivalents. They also have the power to inspect manufacturers periodically (maybe 1 in 100 manufacturers per year). Otherwise testing, QC, and release is self-regulated by the manufacturers.

Pretty much worldwide, since there's usually only a few manufacturers of the Active Pharma Ingredients supplying the whole market.

Scale matters.

In this case, the manufacturer came up with a new technique for manufacturing the API. They failed to account for these very small molecule impurity side-products in their model/testing.

They probably filtered out a lot of it in their purification process and still met USP standards, but 99% or 99.9% pure still leaves a risk for potent side products.

Sometimes these side-products are tested for, but you have to know what to look for. A tiny blip may not be elucidated.

This whole episode of "self regulation" of pharma companies by the FDA reminds my of the FAA's acceptance of Boeing's "self regulation" on the 737 MAX.


Probably the best source of information is from this long-time blogging Medicinal Chemist here:


This story has some speculation about why different molecules of the same therapeutic category are implicated:


From the Wikipedia page https://en.wikipedia.org/wiki/N-Nitrosodimethylamine

"NDMA is a yellow, oily liquid with a faint, characteristic odor and a sweet taste."

"characteristic" doesn't say much, but I often wonder how we know about the taste of a highly toxic chemical like this one --- and if someone died for it.

You wouldn't taste it in a drug. Low concentrations and bound in binders or hidden by taste additives in syrups.

Typically researchers tasted small samples, not too risky. Usually characteristic has a specific chemical kind of taste.

Yes, people have tasted solvents, glue and more nasty stuff.

Adding to this some people may not know unlike here in Canada supplements in the US are not regulated. Herein Canada they are treated as a drug and regulated by Health Canada.

There have been recalls of supplements from the US due to contamination or mislabeling (for example, adding stimulants or steroids but not disclosing that). Although supplements here in Canada can still be questionable in quality.

Technically FDA handles supplements and foods too, but they're really too small to handle the volume of this market.

Most of the gardening implements, and most of the cheap shower heads in my local home improvement store say they cause cancer. I assume people buy them anyway, or the store would stock something else.

So are these drugs with carcinogens of more concern than the other stuff? Should we be more concerned about the other stuff?

All of my fishing weights and some lures have these warnings. It’s because of lead. I certainly don’t put them in my mouth and neither should you.

Shower heads are very concerning. Never seen those warnings.

Lead isn't particularly known as a carcinogen, but a neurotoxin. I guess it might be, but is it "known to the state of California"?

The things I noticed that almost all had California warnings were shovels, rakes, etc. I don't really know why these would need to have lead in them, if that's what it was.

I did buy a shower head, and I did buy one without the warning. But, I also thought - maybe this one has as much or more carcinogens as the cheap ones, but it doesn't have a label simply because it's not sold in California?

Quick question. Title says "generic", but obviously there is a massive economic interest in switching people from generics to non-generics. Is this contamination also possible in non-generics???

Sure, it is possible, but because the brand name passed the carcinogenicity tests in animals, and if the manufacturer kept their same suppliers and formulation, the risk should be much lower.

Whether that's true or not in practice is a different matter. But yes, I would have more confidence in a brand name after this report, which makes me cynical about the intentions of the reporting in the first place.

Maybe I'm totally being manipulated by big pharma, but I'll switch my medication to brand name as soon as possible, and that's just because my insurance is good. Very few people have that option.

> the brand name passed the carcinogenicity tests in animals, and if the manufacturer kept their same suppliers and formulation

I really doubt this is the common case. The studies with animals and the clinical studies with human for safety are very small, so I guess they use a special batch. For production at scale they probably use some industrial method that is cheaper. They may also change the provider, or build it in another factory they own in other country.

Also, every drug has precursors, and has precursors have precursors, you can't produce all of them so you must buy some. And with many of them there are more possibilities of changes that introduce contamination.

What you get with a brand name is more quality control in the incoming products, more quality control in the process and more quality control in the final product. (And also more marketing, sometimes it is only marketing.)

The problem is that big pharma can (and do!) change their chemical synthesis process for its APIs. Without going through all the same testing again.

According to TFA, Novartis, the inventor of (non-generic) Diovan, found the NDMA when they bought some (generic) Valsartan from Huahai. They were the first customer to test thoroughly enough to find it. So, contamination in non-generics is possible, but probably less likely. Actually the NDMA wasn't produced in the original manufacturing process designated for Diovan and Valsartan, but Huahai used a different process to save money.

Now imagine what level of contamination is going on in supplements which are completely unregulated in the US (unlike the EU).

Don't get generic drugs unless you absolutely have to. You are legally unable to sue if you get life altering side effects from the drugs. Which seems even worse after this news story.

Welcome to the future of warfare.

This deserves many more upvotes. External agents find cheap ways to destroy the country, and some internal agents sacrifice common man for quarterly bonus.

A virally popular device (non-prescription and cheap, for god’s sake) to detect toxicity levels, and connected app to visualize the health impact will serve as a wake-up call.

Completely agree. You'd think that when something like this happens, it would be followed up by a string of arrests. There is a total lack of accountability in this country. This country is totally corrupted.

> Some of the contaminated valsartan contains as much as 17 micrograms of NDMA in a single pill. That’s equivalent to eating 48 pounds of bacon. The FDA estimates that for every 8,000 people who took the highest dose of contaminated valsartan daily for four years, there would be one additional occurrence of cancer.

Given the relatively small significance and the fact that this article was published in an issue with the blurb "Can You Trust Generics?" on the cover, I'm gonna say this is just more FUD from Bloomberg.

> Given the relatively small significance and the fact that this article was published in an issue with the blurb "Can You Trust Generics?" on the cover, I'm gonna say this is just more FUD from Bloomberg.

It's not a small significance, and it's not FUD. It's admittedly possible to take some measured numbers for the prevalence of one of the nitrosamines, put that through a dose-response model (of controversial accuracy) and come up with low-sounding numbers of expected cancers, but that's not enough to dismiss the issue.

Nitrosamines are notorious carcinogens, the side-reactions that produced them is well-known (if this comment (https://blogs.sciencemag.org/pipeline/archives/2019/01/04/th...) is to be believed, known since the 1930s), and detecting these genotoxic impurities is feasible without exotic analytical equipment.

The mechanics of the synthesis routes should never have allowed production of those impurities, the active pharmaceutical ingredient should have been monitored for these impurities as well, and this aberration should have been discovered far earlier than it actually was.

It's a lot like foreign object debris in aircraft, or retained surgical instruments inside a patient. Neither are a guaranteed death sentence, but they're harmful, utterly avoidable, and their prevalence is a good indicator of careless/inappropriate working practices and inadequate monitoring procedures. If their procedures couldn't detect these nitrosamines for years -- what else is being let through? Heavy metals? Pyrogens? Some other class of toxic chemicals? Inadequate/overly-high active ingredient dose/concentration?

I've installed avionics inside other people's aircraft and I was careful to not leave behind any tools / bits of wire insulation / snipped-off wires / screws / etc floating around. Yes, the probability that a single bit of wire will find its way onto uninsulated connectors is small, but that argument is not an excuse for sloppy working methods. With life-critical systems, it's not OK to be careless just because someone else might catch the problem or because there's other subsystems that'll attenuate the severity of the fault.

> retained surgical instruments inside a patient.

> If their procedures couldn't detect these nitrosamines for years -- what else is being let through? Heavy metals? Pyrogens? Some other class of toxic chemicals? Inadequate/overly-high active ingredient dose/concentration?

Leaving behind instruments in a patient is a "never event". The accepted amount is literally zero.

When it comes to heavy metals, there are standards of acceptability, because you'll never get all of the lead out of a sample of calcium.

Same for pyrogens.

No active pharma ingredient is 100.000000% pure. And the cost of making them so would defeat the purpose.

At some point you say: "Well, the air you're breathing isn't free of cancerous pollution either, when should we spend dollars on that instead of safer pills?" because there is a line between the two.

One is discrete while the other is continuous, so that's different. Still, the proposition "a single pill is equivalent to 48 pounds of bacon" seems to put us back in the discrete domain...

Even the continuous curve has a minimum threshold. One molecule of anything isn't going to kill you.

When it comes to bacon, it's not only the NDMA that can kill you :)

For a blood pressure medication, 1 in 8,000 is not a small risk. Serzone, an antidepressant, has a similar risk rate for inducing liver failure, and was taken off the market.

For nefazodone, we had other antidepressants (so why take the risk?), and it was only taken off the market when it went generic (lol).

For BP medication, hopefully at least 1 in 100 recipients are benefitting from it. Probably more like 1 in 10. Overall, don't stop taking it. Reducing the cancer risk is possible and feasible in this case without changing therapy.

We'll probably see further testing of APIs and validation of synthetic route changes in the future because of it.

It demonstrates a not insignificant risk but it would be from taking the most contaminated batch of the product for 4 years. If a patient took the most contaminated sample found for 4 years there would be a 1/8000 chance of cancer from it. It illustrates this level of contamination should be a concern but it seems unlikely anyone could be exposed continuously to this highest detected level.

Good to know that the US health care system will impose these individuals with the cost. Shame on them for not properly researching the product. They should have considered their consumer choices against the possibility of cancer years later. No doubt the medical bankruptcy some will have is deserved.

Obviously sarcasm. This is, however, how the current US medical system behaves. It's impossible that these patients would have the capacity to evaluate medication sources like they would any other product. Many likely had no opportunity or choice. Yet the US supports a system that punishes an individual for aspects well beyond the individual.

You should able to sue the pharmacy and your insurer for sourcing and requiring tainted products that are not quality controlled. I would expect some class action lawsuits to appear shortly.

I hope so. After reading the Bloomberg article, I looked up my Losartan Rx. It's from a manufacturer's lot that was pulled by the FDA on Feb 28 of this year.

The kicker is, my (major chain) pharmacy sold it to me this July...

I would get a lawyer. Sounds like you are due a nice settlement.

Ideally yes. However, that's assuming a lot of privilege that the average consumer wouldn't have at all in this case.

The solution is to have three classes of quality for every drug: lower class, middle class, and upper class.

The lower class drugs must have pills with Chinese characters on them, and packaging with Chinese writing, and an English insert. They get minimal testing.

The upper class pills have American flags impressed on them, cost a metric tonne more, but are tested rigorously by private agencies (that are paid from the extra cost).

The aim here is propoganda: have people associate poor quality with Chinese goods and good quality with American goods.

(There is no middle class of drugs: if you are a middle earner you either decide to be either cheap, or to keep up appearances).

> have people associate poor quality with Chinese goods

Nobody is going to associate poor quality with Chinese goods while iPhones and most luxury continue to be made in China.

> keep up appearances

Nobody should know what drugs I'm on, much less the brand I take.

Too many Bloomberg articles make it to the front page of Hacker News and that is unfortunate because Bloomberg is corporate media at its most diabolical. Bloomberg has a long history of publishing articles to earn moral credit and then cashing out on that credit from time to time with articles that clearly favor a small group of special corporate interests. Today, the story is that generics are harmful and cause cancer -- we better pay 20x for those name brands. Before, there were unsubstantiated claims about microprocessors used by Apple that were spiked with surveillance hardware: https://www.bloomberg.com/news/features/2018-10-04/the-big-h...

Don't be a Bloomberg shill. Ask yourself when you see an article published by Bloomberg who they are really serving with these stories.

This comment was downvoted in about 5 seconds from when it was published. This was a comment at the bottom of a comment thread. There is no way for that to have happened other than by a bot. Bloomberg uses third party-controlled bots on HN to control the discussion. Sad.

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