The problem developing new antibiotics is not the science it is economic and regulatory. Let's say you are crazy enough to pay the enormous cost of bring a new antibiotic through the regulatory process, your new antibiotic doesn't get used by doctors. They put the new antibiotic on the shelf to be used in emergencies when all the other antibiotics haven't worked. Your sales are then near zero. This is why all the major pharmaceutical companies have shut down their antibiotic divisions.
What we need to work towards is a large cash payment on approval of each drug or a guaranteed minimium annual payment. Once we have this we will start to get the new antibiotics we need.
Why don't we use for medicines the same procurement procedures we use for aircraft carriers? The USG should just hire someone to develop antibiotics for them. It is a matter of national security too - in case of biological attack having a couple more potent will help a lot.
Yes this would be one way of doing it. I think we should try a range of different funding models to see which one works best. What we have learned over the last 25 years is the traditional laissez faire model has failed us when it comes to antibiotics.
One thing I should mention is we don’t have to keep developing totally new antibiotics we can develop antibiotics that are tweaks of existing antibiotics. Antibiotic resistance is an arms race that we can never win, but we can stay ahead with continued investment.
>What we have learned over the last 25 years is the traditional laissez faire model has failed us when it comes to antibiotics.
What's the best way to quantify the failure? Like a chart showing the bacterial infection death rate for the past 100 years? Costs for the average antibiotic dose over time? Something else? Those data sets seem like they should be available, but are apparently harder for me to find than I would have thought, given the prevalence of all the information available on anti-microbial resistance that pops up for my collection of search terms.
Do you see this new work as a departure from, or a continuation of the antibiotics development of the past?
How fast people are discovering new antibiotics is the important thing, I would think. Assuming resistance is going to happen no matter what, we need to keep discovering new antibiotics at a steady pace to keep up.
I would think it is important to know the magnitude of the problem. Has the number of death from bacterial infections increased from 1 per million people per year in 1990 to 100 per million people per year today? Has it stayed the same? Has it actually decreased over time, and were just trying to be proactive? I'd don't think it is too odd to want to know.
How about a measure of depth of the defence pipeline:
total number of efficacious treatments available per million infections, plotted over time?
The million infections might need to be weighted by severity of clinical outcome.
The problem with this is it is often not known how many other antibiotics might have worked. It might be better to measure drugs known not to work as a percentage of the drugs that work for that class of organism.
Getting this data is likely to be really hard to get and be very labor intensive.
The problem seems to be that an assertion was originally made that we are supposed to accept as true, without any way of quantifying whether it is accurate.
Just got me thinking, shouldn't we try to have a spectrum of antibiotics switched regularly as some form of 'wear levelling' for bacterial resistance ?
>Do you see this new work as a departure from, or a continuation of the antibiotics development of the past?
It is a continuation of some great science that doesn’t ever make it to market. There have been a lot of interesting new antibiotics discovered over the last 20 years or so that just have not been picked up.
What we have learned over the last 25 years is the traditional laissez faire model has failed us when it comes to antibiotics.
I disagree.
Think of it this way. If it were truly laissez faire, then drug makers would charge a price high enough to still make it worthwhile. Unfortunately, the way antibiotics are paid for in US hospitals, the hospitals would lose money each time they used the drug.
Hell, we pay ~$80K to cure hepatitis C (millions have it in the US). If you have an antibiotic that cures a highly resistant infection that happens maybe a few hundred times in the US, why not charge something similar? What is the benefit to society to save someone's life?
> If you have an antibiotic that cures a highly resistant infection that happens maybe a few hundred times in the US, why not charge something similar?
Because a few hundred times 80k is still not much. And even so, many people still can't afford it.
If you can't provide enough value to society that you can spare a few tens of thousands of dollars to cure yourself, are you really worth the cost of developing these drugs, testing them, training doctors, etc.?
All of this over-regulation is why medicine is declining in the first place. That shouldn't justify more regulation.
That easily makes the top 3 list of most sociopathic things I've read all month, and between mass shootings and debates about refugees it's been a banner month for sociopaths on Facebook. Did you intend to measure the value of a human life in terms of enhancing shareholder value or is this just poorly worded?
Agreed and I find this utterly appalling. For me this is a damning indictment of capitalism and a clear indicator that there has to be a better way to handle task prioritization and resource allocation in a modern society.
Is there any system where you can avoid it? Even in a command economy system trade offs have to be made. If you have resources X to spend, you need to figure out what to spend it on. That means you're putting a dollar figure on a human life. I don't see how it could be avoided.
> If you can't extract enough value from society so that you can spare a few tens of thousands of dollars to cure yourself
Fixed. I also generally feel casual disregard towards human life and suffering as long as the human in question is not me, but lets call things with their real names.
With a price of drug per year being twice the average income in the richest country in the world - that creates impossible situation by definition for half of its populace.
Also value provided is not always correlated with value extracted - a janitor for a modest fee keeps my house clean. A hedge fund manager ruins the global economy just when I must enter my prime growing years in earnings and make sure I am at a disadvantage at both the cohort before and after me.
"With a price of drug per year being twice the average income in the richest country in the world - that creates impossible situation by definition for half of its populace."
Even worse. There are a few gigantic outliers at the upper end of the scale. That means that the number of people below average income will be >50%. What you want to check against in such cases is the median (50:50 split by definition).
"Also value provided is not always correlated with value extracted"
That is exactly one of the main problems society is struggling with. How do we even measure value provided? It's a subjective measure to begin with. In the case of the hedge funds manager, it's probably net-negative for most people (but I'm sure her friends would disagree).
Then again. True laissez fair would mean no patents. So I think its fair to assume that it would be radically different in which actors are involved and in which constellations they collaborate. In such a system I wonder if "drug makers" would be able to charge much more than the production cost. Is first mover enough to cover RnD? I suspect not in which case "drug makers" would probably not be "drug inventors"
We certainly do have to keep developing totally new antibiotics. Problem is many of the new antibiotics that we have are tweaks on existing antibiotics. We're in such a rush to have something useable that side effect profiles get nastier and bacteria develop resistances to entire families of antibiotics.
Many antibiotics are already derivatives of other antibiotics. The problem is that bacteria are developing resistances to whole families of medication. You can tweak erythromycin and get clarithromycin or azithromycin, but that means nothing when the mechanism of action they exploit has been patched in resistant bacteria.
Based on how most government procurement of bespoke products turns out, they would pay far more than it should cost and get a product that doesn't work.
The episode took a turn, however, when they looked at Medieval versions of antibiotics -- a treatment for eye sties, I think they discussed -- they found the described treatment was wildly effective.
The researchers they interviewed postulate resistance wanes when a treatment leaves the population for a while.
The conclusion at the end was that it would be beneficial to have a big bevy of antibiotics and such, that we could rotate through as immunity came and went.
In this ^ way, it would seem these pharmas, and the pharma market, needs to adopt a second perspective on antibiotics. They should look at a new drug not for revenues over the next 6 quarters, but over the next 6 decades. In that time frame the drug would hit, be shelved, hit, and repeat.
Perhaps this would be one place that more permissive or advantageous patents would help, by encouraging growth in the antibiotic industry. That probably sounds unpopular given the general "no patents!" sentiment, but I there could be reason in it for at least some period of time.
It's more complicated than that. Just because resistance happens, doesn't mean it happens on a broad scale. A lot of those older antibiotics you mentioned are still broadly used as a first line therapy. Since only a fraction of people are infected with resistant bacteria, they work for most infection.
It's when the first line antibiotic stops working that you need to go to the newer antibiotics.
Exactly. If there was money to made selling antibiotics they would have keep investing in this area. They have started again recently, but there is an awful long lead time to develop new drugs - even longer when you have laid off all the people skilled in this area.
> This is why all the major pharmaceutical companies have shut down their antibiotic divisions.
No, but the other side of it is that for the money that they had invested, before shutting it down, they got very little at all. That is the other explanation here — that they invested money into antibiotics research and simply couldn't make it work. E.g., http://blogs.sciencemag.org/pipeline/?s=antibiotics
Yes they had invested money and they could not make any money out of antibiotics for the reasons I listed. This is why they shut them down.
There is some evidence that some of these divisions weren’t very good at finding new drugs, but even when they were found they weren’t brought through to market because there was no profit.
More and more it seems that pharmaceuticals are a poor fit for free market capitalism. The basic research is already mostly government funded, they should fund all of it. From trials to sales. Maybe they could use private actors for manufacturing.
One possible solution would be to split the world into regions, with different front-line and reserve anti-biotics in each region. In other words, antibiotic A would be front-line in North America and B,C & D would be held in reserve for resistant cases, while B would be front-line in South America, C in Europe, D in South Asia, etc. New anti-biotics would be allocated a new front-line territory by splitting existing territories. So when antibiotic X is developed, then e.g. North America is split in to East (uses A) and West (uses X), providing a market for X. This solution also has the advantage that it shrinks the population which is using each antibiotic regularly, increasing its useful lifetime.
Of course, this solution ignores issues of varying toxicity and side effects (not every antibiotic is equally appropriate for front-line use) and anti-trust issues. If an ordered schedule for reserve antibiotics was used, the same principle would apply, just one step farther up the reserve schedule. So, if no one wants to use your antibiotic because it is very harsh, but if your antibiotic is first choice among the harsh reserve antibiotics in some region, it will at least get used some, perhaps enough to pay for development. Anti-trust issues could be dealt with by splitting the regions along jurisdictional lines, or by international agreements. Because this proposal is pro-competition (because new entrants are granted new territories) it also may suffer less regulatory ire than most collusion agreements.
Please give a better cite. Yours does not seem to support your claim. The only part of that link that mentions "millions" is
> Treatment failure to the last resort of treatment for gonorrhoea–third generation cephalosporins–has been confirmed in Austria, Australia, Canada, France, Japan, Norway, Slovenia, South Africa, Sweden and the United Kingdom. An estimated 106 million people are infected with gonorrhoea every year (2008 estimates).
Only 1,000 people die a year from gonorrhoea worldwide, the vast majority of whom live outside the US and EU.
To the best of my knowledge, the anti-biotic resistant bacteria that currently poses the greatest public health risk is Methicillin-resistant Staphylococcus aureus (MRSA).
> Many of these infections are less serious, but the Centers for Disease Control and Prevention (CDC) estimates that there are 80,461 invasive MRSA infections and 11,285 deaths due to MRSA annually
Parent didn't say that he or people with gonorrhea in general "would be happy to go without any treatment" -- so why do you ask him that?
He merely said that people with such needs are a few tens of thousands, and not many millions, so they don't constitute a viable market.
If your new point is that, "ok, they might not be millions, but they would still pay any price for a treatment", that would also be wrong: They might be WILLING to pay any price, but for most of them it wouldn't be possible to pay that much. There are lots of people that need stuff from surgeries to dental work, that are NOT happy not having them, but that nevertheless don't have them, because they can't afford them -- and expensive gonorrhea drugs wouldn't be much different.
Who said anything about being happy about it? You said:
> These are serious, life-threatening, life-limiting illnesses where the seller can set high prices.
This after referring to millions of cases. The response was that the only disease in your reference anywhere in that ballpark is gonorrhea. While I definitely don't want to get the clap, it is none of those things you mentioned.
Yes this is because we have not been developing new antibiotics because of the non-use problem. Now of course because all the antibiotic divisions have been shut down we don't have the infrastructure to bring new antibiotics to market at the rate we need. You have to make the decision to develop a new antibiotic 15 years before you need it.
We need to develop new antibiotics before we have a problem and where they will sit on the shelf, rather than wait until we have none and then panic.
These are serious, life-threatening, life-limiting illnesses where the seller can set high prices.
Unfortunately not. The way that public payers (Medicare, Medicaid) pay for antibiotics levels very little room for high prices. Even private insurance is going in that direction.
If drug makers charged a high price, then hospitals would be left holding the bag.
The ability to culture organisms is the single most important roadblock to understanding. Even if a new culturing technique only enables you to grow .00001 % of the bacterial species out there that is still a 100 fold increase in the number that we can now study at scale. DNA sequencing has made it possible to identify new species by looking at their 16s ribosomal RNA, but if you are looking for small molecules and proteins you have to be able to amplify the whole organism, not just its nucleic acids.
Huge deal, expect much more to come from microbiologists using this culturing technique.
Pretty much. We are completely blind to the other members of the micro ecosystem that are probably living happily on and around us under non-pathological conditions. There are theories out there that suggest that many infections are caused not by too many bad bacteria but too few or unstable populations of good bacteria (vast over simplification).
In addition, we actually have quite a bit of trouble culturing bacteria that might be pathological (our inability to culture them is quite related to our inability to determine their pathological status).
If we could simulate what's happening in just a single bacterium, we could pre-screen drugs in a virtual petri dish. Promising compounds could then be put to real-life tests.
The article gives the impression that a lot of biology involves ad-hoc, artisanal techniques that have been somewhat standardized and automated over the years. And that the standardization and automation involved can actually stand in the way of changing an approach.
There was an article about the standardization of laboratory mice - it allows large scale, reproducible results but guarantees you won't things that can't be learned from the single, genetically uniform mouse type that is used.
>guarantees you won't things that can't be learned from the single, genetically uniform mouse type that is used.
This is a misrepresentation of the state of mouse biology research.
We use many different strains of inbred mice (check out the JAX Labs list of available inbreds)[1].
Inbreeding reduces the genetic variability of our mice, allowing us to control more of the variation in an experiment. This allows us to perform pilot studies with smaller samples sizes while still reaching sufficient statistical power.
This is important -- mouse biology is expensive, and we try to discover as much as possible with limited funding.
If a result merits generalization to multiple backgrounds, it is often tested in multiple inbred strains and an "outbred" strain that is genetically diverse.
I think you're confusing my comment, which addresses the inherent limitation of automating particular experimental processes with a criticism about how biological research ought to be conducted. It may well be that uniformity is needed for reproducibility - even adding more controls for the uniformity, biology seems broadly limited by a relatively few particular starting points. As the OP points out, because the vast diversity of biological phenomena, it seems there are variety of things that simply aren't seen with only these starting points (again, not to deny that as practical matter one might still need to limit one's starting points).
Currently our understanding of biochemistry is so poor that we still have a lot to learn about the very basics. I don't think the diversity of our lab mice is currently a limiting factor.
Our understanding of biochemistry is actually quite good, it is our understanding of biochemistry diversity that is poor (i.e. why are you and I different). Using inbred mice does not help solve this problem.
I'm no expert, but it seems to me that as long as there is still basic research being done on the behavior of very simple proteins, we're a long way from understanding the complex interactions that happen in cells.
Yes we have a lot to learn about all the interactions between proteins in a cell, but this typically does not fall under the heading of biochemistry.
Slightly off topic the traditional divisions between different areas of biological sciences that built up in the 20th C have all broken down with the rise of molecular biology. Once biochemistry was a different field to say microbiology, now both are part of the super-field of molecular sciences.
The important practical results (e.g. many particular drugs) that we get and need are obtained not because of understanding why they work but despite that, empirical testing is a key part of current process.
We have considered and discarded a lot of candidate drugs with great potential before starting human testing. As a rule, if some molecule would cure an important disease but for whatever reason doesn't work on the popular strains of mice, then our current process would cause us to discard it.
I would like to see inbred mice banned from all research not about inbreding. No study should use inbred mice - sure is lets you use less mice, but at the expense of generating results that are limited to one individual. If we can't afford to use outbred mice then we should not have done the experiment.
The use of inbred animals isn't a cost saving technique. Inbred strains are developed over many generations so that individuals are genetically homogeneous. You know that any different results you get are due to different treatments for instance rather than the genetic background of the animals you're using.
Yes it is. Using inbred mice allows you to use fewer mice as the variability due to the genetic background is reduced. The more variability in your system the more individuals you need to see a statistically valid difference due to treatment. The more you have to test the more the cost.
None of this would matter if we were developing cures for diseases of C57BL/6 mice [1], but we are actually interested in developing cures for humans who are not inbred (well in most places). We need our animals models to match as closely as possible to humans and using inbred mice is not a good match.
>"You know that any different results you get are due to different treatments for instance rather than the genetic background of the animals you're using."
I don't think this is really true. Perhaps that makes it unlikely genetic differences are the cause of a relatively large apparent treatment effect, but not just any difference.
What we need to work towards is a large cash payment on approval of each drug or a guaranteed minimium annual payment. Once we have this we will start to get the new antibiotics we need.