Comment by kortex | Hacker News Reader

Comment by kortex | Hacker News Reader

Why don't we just have a schedule/tier system?

Category A - approved by the FDA, basically what we have right now

Category B - approved by some other established regulatory body (EU). Comes with all the warnings, doctors can write it off-label, insurance companies may not cover it but doctors can ask for a variance

Category C - approved by some non-OECD body. Insurance companies under no obligation to cover

Category D - experimental, this is stuff maybe still only in animal models, but pharmacies can still order and dispense it

Category E - experimental and basically limited run from pharmaceutical companies. These essentially need to be tailor-made or produced by a GMP kilo lab. There are plenty of drugs in this category - I worked on them - and the intended recipients are entirely animals, QA, and regulatory agencies. But maybe if some crazy S.R. Hadden type (billionaire in Contact) wants to guinea pig themselves, let em.

The latter category also opens the door for custom therapies (gene/mRNA) that you basically can't test the active pharmaceutical ingredient for efficacy on.

> Category B - approved by some other established regulatory body (EU). Comes with all the warnings, doctors can write it off-label, insurance companies may not cover it but doctors can ask for a variance

We can already prescribe off-label if the FDA has approved it for at least 1 indication, and it mostly gets reimbursed.

> Category C - approved by some non-OECD body. Insurance companies under no obligation to cover

There's very little that's approved by a non-US body and approved by the EU or non-OECD body that warrants clinical use, and if it is it gets reviewed quickly. The US is the largest market for manufacturers so they almost always start here.

> Category D - experimental, this is stuff maybe still only in animal models, but pharmacies can still order and dispense it

Pharmacists and physicians are ethically bound to prescribe to the best of their ability and avoid harm, by prescribing something only validated in animal models it means we are not prescribing/dispensing something validated in humans and therefore not meeting or exceeding the standard of care.

This sounds like a recipe for killing people.

> The latter category also opens the door for custom therapies (gene/mRNA) that you basically can't test the active pharmaceutical ingredient for efficacy on.

Huh? There are many ongoing gene-directed and mRNA studies being tested.

> This sounds like a recipe for killing people.

Do you know what else sounds like a recipe for killing people? Not allowing people to access therapeutics that might save their life because it hasn't yet gone through regulatory approval yet for whatever reason (delays, too expensive to submit), etc.

> There's very little that's approved by a non-US body and approved by the EU or non-OECD body that warrants clinical use, and if it is it gets reviewed quickly

LOL. What are you talking about. There are so many examples.

One of the most tragic is amisulpride. Amisulpride is an antipsychotic medication used to treat schizophrenia and other psychiatric conditions. Some key notes about its regulatory status:

- Amisulpride was first approved in France in the 1980s and is widely used in Europe.

- It was never approved by the FDA for use in the United States, and at this point there's not organization that can afford to go through the approval process because there's no patent.

- The reason often cited is that the manufacturer did not apply for approval with the FDA. It was likely not considered commercially viable for the US market at the time.

- Amisulpride is believed to have comparable efficacy to other second-generation antipsychotics like olanzapine and risperidone, but with a lower side effect burden according to some studies.

- In Europe, amisulpride is considered a first-line treatment option for schizophrenia, but American psychiatrists do not have access to it. According to some sources, it is literally recommended as the best antipsychotic in other countries.

> Do you know what else sounds like a recipe for killing people? Not allowing people to access therapeutics that might save their life because it hasn't yet gone through regulatory approval yet for whatever reason (delays, too expensive to submit), etc.

Should we at least demand more specific criteria than "X _might_ save their life", like threshold of suggestive evidence? There will always be lots of stuff that hasn't been closely studied, the effects of which we can only partially describe. You could isolate any new molecule from some previously unknown bacterium and say it "might" be a treatment for any disease, but that's just a statement of our own ignorance right?

If we say, "so long as it hasn't been conclusively shown to _not_ beneficial for the patient's disease, then it _might_ help them, so it should be fair game", then that seems to open the door to quacks selling snake oil to desperate dying people and their families. And of the unenumerable list of potential "it might work because we haven't yet shown that it doesn't" chemicals, why shouldn't unethical practices pick the most expensive options available?

"Of course you must understand there can be no guarantees with any treatment, and this may be a long shot, and precisely because of the lack of prior studies we cannot even give you any efficacy numbers. But we're at the cutting edge of medical science! Please make out a check for $500k and sign this waver and we can begin treatment as soon as possible."

> Should we at least demand more specific criteria than "X _might_ save their life", like threshold of suggestive evidence? There will always be lots of stuff that hasn't been closely studied, the effects of which we can only partially describe.

We do, it's part of the FDA process and is determined on a case-by-case basis considering alternative treatments, disease course and intervention safety amongst other variables.

It's how the vast majority of stroke and novel cancer therapies are currently being approved.

> Do you know what else sounds like a recipe for killing people? Not allowing people to access therapeutics that might save their life because it hasn't yet gone through regulatory approval yet for whatever reason (delays, too expensive to submit), etc.

You're assuming a therapeutic not tested in humans has a better chance of saving someone's life than something tested and available. Do you have any evidence to support this?

> One of the most tragic is amisulpride. Amisulpride is an antipsychotic medication used to treat schizophrenia and other psychiatric conditions. Some key notes about its regulatory status:

There is little randomised evidence comparing amisulpride with other second generation antipsychotic drugs. We could only find trials comparing amisulpride with olanzapine, risperidone and ziprasidone. We found amisulpride may be somewhat more effective than ziprasidone, and more tolerable in terms of weight gain and other associated problems than olanzapine and risperidone. These data, however, are based on only ten short to medium term studies and therefore too limited to allow for firm conclusions.

This review compared the effects of amisulpride with those of other so called second generation (atypical) antipsychotic drugs. For half of the possible comparisons not a single relevant study could be identified. Based on very limited data there was no difference in efficacy comparing amisulpride with olanzapine and risperidone, but a certain advantage compared with ziprasidone. Amisulpride was associated with less weight gain than risperidone and olanzapine.

What's so tragic about this? Equally efficacious antipsychotics are available. Once again the alternative to non-approved drug isn't nothing or inferior substance.

I will concede that legacy off-patent drugs are part of the "very limited" gap with the FDA, this doesn't hold for new drug discoveries as discussed in the article.

Even then, there is an ongoing US trial for amisulpride and the substance is approved for nausea/vomiting (granted not in oral form).

[0] https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD...

The issue with antipsychotics tends to be the side effects, not the efficacy

Amisulpride, for a lot of people, would be one of the best antipsychotics.

Antipsychotics are incredibly important medications for a lot of people. It really matters that amisulpride is not available.

https://psychnews.psychiatryonline.org/doi/full/10.1176/appi...

"A comprehensive meta-analysis published in 2019 in JAMA that compared 32 oral antipsychotics helped solidify the sentiments shared by Kahn and other investigators who have conducted clinical studies with amisulpride. That meta-analysis identified amisulpride as the second most effective antipsychotic at reducing overall symptoms in schizophrenia patients (behind clozapine) and the most effective in terms of reducing positive symptoms. The analysis also ranked amisulpride better than clozapine in terms of tolerability and side effects."

> Antipsychotics are incredibly important medications for a lot of people.

Agree.

> It really matters that amisulpride is not available.

Not sure about this, I'm not a psychiatric expert but my cursory lit review shows conflicting meta-analysis as to whether amisulpride is better than 2nd gen. Although your source is newer Cochrane is generally the gold-standard on SRs and the included studies in the 2019 JAMA article predate the Cochrane review, they detail the limitations of comparison.

Looking at Canada, amisulpride is limited to special access and is also not first line.

The UK pharmacotherapy guidelines are also waffly and cite limited evidence to guide firs-line decision making.

A systematic-review from China showed different side-effect profile for both, hard to say which is better. Amisulpride was cheaper.

In any case old drugs that were never approved are part of that "very little" I was referring to that fall through the cracks.

Not sure I'd call this one tragic though given that other countries also don't use it or limit access and there are good alternatives.

The problem with "there are similar efficacy rate alternatives" is that, for most psychiatric drugs, it's not necessarily the _same_ population that is helped by two different drugs at similar efficacy for the same issue, even if the total success rates are comparable.

A number of people have tried many different individual or combinations of treatments before finding something workable, or close enough to workable that you can paper over the cracks. I would strongly suggest against concluding the lack of access to something isn't an issue for some patients because they have alternatives, without very compelling evidence.

A story from my recent past - I've had, historically, no issues swapping between different generic manufacturers, or name brand and generic, if someone stopped making something I was taking, or I moved and the pharmacies stocked different versions, or whatever. So a little over a month ago, when I noticed my meds looked visibly different after getting them filled, I checked to be sure they hadn't given me the wrong thing, but no, same dose, different generic manufacturer, and I stopped worrying.

...until I started getting extremely nauseous to the point of being debilitating, for 6-8 hours every time, when I take this medication twice a day. So I went back and tried the third generic manufacturer the pharmacy had, after a few days of making sure it wasn't a fluke, and the third one had the same issue. We finally ended up switching to the extended release version of the med, which had still more different manufacturers, and did not have this issue, since the first manufacturer had a shortage for over a month and counting.

Medications are a lot less interchangeable than we might hope.

This are already dead people walking.

Reply on news.ycombinator.com