Skip to main content


Myths of Vaccine Manufacturing

In the last few days, the question of why more drug companies haven’t been enlisted for vaccine production has come up. It’s mostly due to this tweet:

The problem is, as far as I can see, this is simply wrong. There are not “dozens of other pharma companies” who “stand ready” to produce these mRNA vaccines. To me, this betrays a lack of knowledge about what these vaccines are and how they’re produced. Even though I’m not a pharma manufacturing person, I am indeed a pharma researcher in general. So I would be glad to fill in this gap, and here’s why it’s not possible to suddenly unleash dozens of companies to crank out the Pfizer/BioNTech and Moderna vaccines.

The first thing to understand is that these are not, of course, traditional vaccines. That’s why they came on so quickly. mRNA as a vaccine technology has been worked on for some twenty to twenty-five years now, from what I can see, and (as I never tire of mentioning) we’re very fortunate that it had worked out (and quite recently) several of its outstanding problems just before this pandemic hit. Five years ago we simply could not have gone from sequence to vaccine inside of a year. And I mean that “we” to mean both “we the biopharma industry” and “we the human race”.

At this point, let me briefly dispose of an even less well-founded take that’s been going around as well. I’ve seen a number of people say something like “We had the vaccine back in February! It only took until the end of the year to roll it out because of the FDA!” The main thing I’ll say about that idea is that no one who actually works on vaccines, in any capacity, has any time for that statement. Not all vaccine ideas work – we’re already seeing that with the current coronavirus, and if you’d like to talk to some folks about that, then I suggest you call up GlaxoSmithKline and Sanofi and ask them what happened to their initial candidate, and while you’re at it, call up Merck and ask them what happened to their two. Note that I have just named three of the largest, most experienced drug companies on the planet, all of whom have come up short. So no, we did not “have the vaccine” in February.

One of the other reasons we didn’t have it back then is the whole problem of figuring out how to make the stuff, and that brings us back to today’s discussion. How do you make the Moderna and Pfizer/BioNTech vaccines? And what’s stopping “dozens of other pharma companies” from doing the same? Let’s get into those details, stopping briefly again to imagine asking James Hamblin above to actually start naming “dozens” of pharma companies. Anyone have a good over/under on how many names would get rattled off?

OK, let’s look at the actual supply chains. The single most informative piece I have seen on this is from Jonas Neubert – I’ve recommended it before, and this is absolutely the time to recommend it again. I also have to mention this detailed article at the Washington Post, which focuses on the Pfizer/BioNTech vaccine, and this one at KHN about manufacturing bottlenecks in general. You should also read this Twitter thread from Rajeev Venkayya, who knows what he’s talking about when it comes to vaccine manufacturing, too. All of these will cover details that I’m not even going to get to today!

It’s not in my nature, since I’m an early-stage drug research person myself, but I’m going to totally sidestep all the R&D questions behind the various components and just treat this as a manufacturing process that fell from the sky in its final form. To distill a huge amount of background and detail down into the simplified steps, we have:

Step One: Produce the appropriate stretch of DNA, containing the sequence that you need to have transcribed into mRNA. This is generally done in bacterial culture.
Step Two: Produce that mRNA from your DNA template using enzymes in a bioreactor.
Step Three: Produce the lipids that you need for the formulation. Some of these are pretty common (such as cholesterol), but the key ones are very much not (more on this below).
Step Four: take your mRNA and your lipids and combine these into lipid nanoparticles (LNPs). I have just breezed past the single biggest technological hurdle in the whole process, and below you will learn why it's such a beast.
Step Five: combine the LNPs with the other components of the formulation (phosphate buffers, saline, sucrose and such) and fill those into vials.
Step Six: get those vials into trays, into packages, into boxes, into crates, and out the door into trucks and airplanes 

OK, you have now produced the mRNA coronavirus vaccines and shipped them out into the world, so sit back and open a cold one. You will not reach that stage, though, without some significant challenges. Let’s take those step by step. The DNA production in Step One is not too bad. As the Neubert article details, Pfizer does this themselves in Saint Louis, and Moderna outsources this to the large and capable Swiss firm Lonza (update: a good part of the Lonza work is being done in Portsmouth, NH). DNA plasmid production on an industrial scale is pretty well worked out (and keep in mind that “industrial scale” for DNA means “a few grams”. It’s not something you can do in your garage – as with every step in this process there’s a lot of purification and quality control to make sure that you’re making exactly what you think you’re making and that it looks exactly within the same specs as the last time you made it. But that’s what biopharma manufacturing folks are good at, and there are a lot of people who can do it. That said, a goodly number of them are occupied doing that for just the vaccines, but if we needed more of this DNA, sure, we could produce more.

But we don’t. That’s not the rate-limiting step. Nor is Step Two, the transcription into mRNA. Pfizer and BioNTech do this in Andover, MA and at BioNTech facilities in Germany. They have manufacturing in Idar-Oberstein (a town I recall visiting in the cold rain one weekend in 1988 during my post-doc!) and last fall they bought another facility in Marburg which is just getting revved up for such production now. The Moderna mRNA step is also handled in Switzerland by Lonza. Now this is not so common as an industrial process, for sure, because it’s only relatively recently that people have been treating RNA species as actual drug substances themselves, worthy of scale-up manufacturing. If I had to ask someone else to make me some more bags of bespoke mRNA, I might turn to Alnylam (who have a manufacturing facility in Norton, MA although to be sure, they’re using it for their own drugs!), but doing so would not increase the number of vaccine vials coming out the other end of the process. RNA production is certainly closer to being rate-limiting than Step One, but it’s nothing compared to the real bottlenecks that are coming.

Now to the lipids in Step Three. This doesn’t have to be done in sequence like the DNA/RNA step, of course – the lipids needed for the formulation are an entirely different production process. As the Neubert article will show you, Pfizer and BioNTech are getting all of theirs from a UK company called Croda, with production likely going on in the town of Alabaster, Alabama, which (unlike Idar-Oberstein) I am certain that I have not visited. Now, each of these vaccines needs some odd lipids with positively charged groups on them; that’s a crucial part of the formulation. These are surely not trivial to make on scale, but they’re still small molecules with relatively straightforward structures. I’m sure that barrels of these things aren’t stacking up at the factory for lack of demand, but I don’t believe that they’re the limiting reagent in manufacturing, either. If you had to, you could surely get some other manufacturers up to speed on the process.

I’m going to skip ahead to Step Five and Step Six. These are surely running at a good clip, but they are more traditional functions of a drug company (or of any manufacturing company). It’s true that pharmaceutical vial fill-and-finish on this scale narrows you down to fewer players than would be involved in, say, canning tuna. But these folks are already involved. Pfizer is doing this in Kalamazoo and in Puurs, Belgium, and BioNTech is doing this in several locations in Germany and Switzerland, both at its own facilities and via at least two contract firms. Moderna, meanwhile, outsources this to some of the big players in the US and Europe: Catalent, Rovi, and Recipharm. Everyone in this part of the manufacturing business has known for months that a Big Vaccine Push has been coming, and has been cranking up vial manufacturing, bringing all available production lines up to speed, and signing deals all over the place with everyone who has any kind of advanced vaccine effort.

Ah, but now we get back to Step Four. As Neubert says, “Welcome to the bottleneck!” Turning a mixture of mRNA and a set of lipids into a well-defined mix of solid nanoparticles with consistent mRNA encapsulation, well, that’s the hard part. Moderna appears to be doing this step in-house, although details are scarce, and Pfizer/BioNTech seems to be doing this in Kalamazoo, MI and probably in Europe as well. Everyone is almost certainly having to use some sort of specially-built microfluidics device to get this to happen – I would be extremely surprised to find that it would be feasible without such technology. Microfluidics (a hot area of research for some years now) involves liquid flow through very small channels, allowing for precise mixing and timing on a very small scale. Liquids behave quite differently on that scale than they do when you pour them out of drums or pump them into reactors (which is what we’re used to in more traditional drug manufacturing). That’s the whole idea. My own guess as to what such a Vaccine Machine involves is a large number of very small reaction chambers, running in parallel, that have equally small and very precisely controlled flows of the mRNA and the various lipid components heading into them. You will have to control the flow rates, the concentrations, the temperature, and who knows what else, and you can be sure that the channel sizes and the size and shape of the mixing chambers are critical as well.

These will be special-purpose bespoke machines, and if you ask other drug companies if they have one sitting around, the answer will be “Of course not”. This is not anything close to a traditional drug manufacturing process. And this is the single biggest reason why you cannot simply call up those “dozens” of other companies and ask them to shift their existing production over to making the mRNA vaccines. There are not dozens of companies who make DNA templates on the needed scale. There are definitely not dozens of companies who can make enough RNA. But most importantly, I believe that you can count on one hand the number of facilities who can make the critical lipid nanoparticles. That doesn’t mean that you can’t build more of the machines, but I would assume that Pfizer, BioNTech, Moderna (and CureVac as well) have largely taken up the production capacity for that sort of expansion as well.

And let’s not forget: the rest of the drug industry is already mobilizing. Sanofi, one of the big vaccine players already (and one with their own interest in mRNA) has already announced that they’re going to help out Pfizer and BioNTech. But look at the timelines: here’s one of the largest, most well-prepared companies that could join in on a vaccine production effort, and they won’t have an impact until August. It’s not clear what stages Sanofi will be involved in, but bottling and packaging are definitely involved (and there are no details about whether LNP production is). And Novartis has announced a contract to use one of its Swiss location for fill-and-finish as well, with production by mid-year. Bayer is pitching in with CureVac’s candidate.

This is all good news, but it’s a long way from that tweet that started this whole post off. There are not “dozens of companies who stand ready” to produce vaccines and “end this pandemic”. It’s the same few big players you’ve already heard of, and they’re not sitting around and watching, either. To claim otherwise is a fantasy, and we’re better off with the facts.

129 comments on “Myths of Vaccine Manufacturing”

  1. James Turner says:

    Just a comment regarding Lonza’s facilities. I know a significant piece of the production is happening in their Portsmouth NH facility, not in Switzerland. I have family that works on it there, and they’re pulling heroic shifts to get it out the door.

    1. Amity Morin says:

      I just hope we get the vaccine and this will all come to an end. We still have the whole year of 2021 for companies to be able to do it. In the meantime I’m always taking natural supplements to stay healthy in this time of global health crisis.

      1. DataWatcher says:

        We may have the whole year of 2021 left, but with mutations occurring at their current rate, I’m not sure we can wait that long. In just the past two months, we have discovered at least three mutations that are more contagious and possibly more virulent than the “original” virus that was our standard for the vaccine trials, and at least two of them appear able to evade immunity — meaning that most of the data from those trials need to have an asterisk appended to them, at best. The nightmare scenario is the development, through mutagenesis, of a virus as contagious as measles that then continues to mutate at a rate outpacing treatment and vaccine development.

        1. CommentGuy says:

          Agreed, though I don’t think vaccines alone will necessarily resolve this issue. More widespread rapid testing will be needed to contain spread of the disease. More radically, a larger epidemiological system that monitors disease spread on a national/international level (analogous to weather forecasting) will likely be needed to dampen/prevent pandemics like this in the future. Although privacy is the foremost issue with this concept.

          1. Stupidosaurus says:

            To CommentGuy’s point on surveillance, there is already an international epidemiological surveillance network for influenza called GISRS, the Global Influenza Surveillance and Response System. I believe that this has also been tracking and recording Covid-19 cases.

      2. Charles H. says:

        According to most recent new (popular press, so believe it if you will) vaccine resistant variant strains are already starting to show up. Even if true it’s not clear what that would mean. Does it mean the death rate doesn’t go down? That long COVID continues? Or that everyone will get bad cases of the sniffles?

        Still, the vaccine creates a new selection pressure, so it’s not appropriate to assume that evolution won’t deal with it. Hopefully the results of “dealing with it” will be something less threatening, but I don’t think there’s any guarantee of that.

        P.S.: I’m a programmer, not a biochemist or medic, so take this with the appropriate grain of salt…but *I* think it’s a reasonable projection.

    2. Johann Miescher says:

      Lonza is has three production lines in Visp, Switzerland, for “world ex. US” production (3/4), and one production line in Portsmouth for US only production (1/4).

  2. Joseph M Severs says:

    All excellent points and all true. Validation of such processes is also a big, big deal.

  3. Joseph M Severs says:

    My spouse also adds that you need to incorporate obtaining necessary regulatory approvals.

  4. Mic says:

    Thanks Derek! I was just getting the same questions by friends and family and, as usual, you have done a great job.
    Another question that is often asked is: “why do we not suspend patents, so other companies would be free to produce the vaccines?”.

    1. Not-an-epidemiologist says:

      Although I’m not sure how general this is, in Australia CSL is manufacturing the AZ vaccine, and I believe other countries may be doing so as well. Under license, but clearly at very little cost — so almost as good.

      I agree with most of this post — certainly we didn’t have the resources to make the mRNA vaccines at large enough scale a year ago (and we still don’t even now). That said — we probably could have made more of the AZ vaccine earlier (Australia’s CSL, again, is just one example — production not licensed until September, production not commenced until November).

      To extend this further, you could argue that the current situation warranted a war-time-style industrial re-purposing at the start of the pandemic, to allow much larger scale adenovirus-based vaccine production now. Given how long this is stretching on I suspect it would have been justifiable on an economic level alone, let alone the millions of lives it would have saved. (If you’re concerned about potential vaccine resistance and the need to have global vaccination coverage ASAP to prevent this, then all of the above applies even more so.)

      Pfizer, Moderna, AZ and the rest were all very clear by mid-2020 about their production estimates for this year. We knew back then that there would not be nearly enough doses this year. We could, with hindsight, have done more.

      Also, re. the throw-away line on GSK and Merck vaccines — those weren’t complete failures based on immune response (all induced a reasonable level of response), and may well have been much better than nothing if we hadn’t already had so many strong candidates by the time they were through Phase I trials. Let’s not forget that we were in an extremely fortunate situation with SARS-COV-2, in that so much vaccine research had been conducted with the original SARS. We knew the right antigen to target right from the start, and it provided great confidence that almost all vaccine production efforts would work (and that’s been right — Pfizer, Moderna, AZ, J&J, Novavax, Gameleya, Sinopharm, Sinovac … the successes massively outweigh the nominal failures here). We could have justifiably bet the pharm on one of these.

      1. Adrian says:

        You got the scope of this blog post wrong.
        And Derek could have made it clearer since this limitation is less obvious for the 95% of humans living outside the US.

        This is a US discussion, with both James Hamblin and Derek ONLY discussing the two vaccines currently authorized in the US.
        These are the most high-tech and most difficult to produce vaccines.

        The Oxford/AZ vaccine is easier (and cheaper) to produce but not (yet?) authorized in the US.
        The Gamaleya/Sputnik V vaccine is like a proper version of the Oxford/AZ vaccine, without the stupidity of using the same vector twice.

        The Sinopharm and Sinovac vaccines are relatively low-tech inactivated virus vaccines.

        Most of the vaccines approved anywhere in the world right now are low-tech enough that companies in India could start producing them within 3 months, but the 2 vaccines currently approved in the US are exceptions.

        1. Some idiot says:

          I agree with your general points that doing tech transfer of an mRNA vaccine to a completely third party would be very, very far from trivial, and would take a long, long time to complete (and get approval from the relevant authorities etc), whereas for other more mature technologies it would be easier and faster (but by no means “fast” along the lines of “we could begin to produce and sell in two months” fast…).

          However, taking the Oxford/AZ vaccine as an example: there will be a huge shortfall in the delivery of vaccine to the EU in the next few months, at the very least, apparently due to “production difficulties” in their EU site. So even for a “mature” technology, being produced within a company which has developed the vaccine, things can still go pear-shaped (apparently), so, again, once a “winner” (i.e. approved vaccine on the market) has been announced, trying to get a third party up and running to produce it very quickly is going to be problematic, to say the very least.

          Incidentally, in this case, CSL and the Australian government had an agreement with Oxford/AZ quite a while ago (6-9 months? I don’t remember precisely) so I would imagine that their tech transfer would have started around that time. So that will not be an extremely fast transfer (i.e. < 3 months or similar…).

          BTW, on a related subject, I would be interested to hear from those in the UK on an issue ( eg Chris Phillips?). This is regarding Oxford/AZ's "production problems". The word we hear on this side of the channel is that they have production problems in their EU site, which is going to significantly reduce the number of vaccines they can produce to the EU in Q1. Reading between the lines, it is pretty clear that many within the EU top administration believe that the "production problems" is a lie, and that instead they are exporting much of the EU production to the UK. Apparently, EU people talking to AZ say that AZ refuse to actually tell them what the problem is, which deepens the suspicion that AZ is not coming clean, and have hidden motives. Hence the saber-rattling from EU as regards "export controls" on vaccine production (I don't really believe they will do anything there, but the threat is pretty bloody obvious).
          Now, I am old enough and ugly enough to not believe in the literal truth of stuff which has politics involved, and especially when one is clearly only fed one side of the story. Therefore, I am honestly interested to hear what "the word on the street" is in the UK… Any thoughts/reflections would be greatly appreciated!

          1. Andy Dawson says:

            The process for making the AZ vaccine, as an adenovirus-vectored treatment are somewhat different.

            That involves culturing human cells in bulk in a bioreactor, then infecting them with the modified adenovirus. This results in production if large volumes of the virus (as it invades the cells). The resultant virus is then extracted and taken through various stages of filtration and purification to make the final product.

            AZ has told the EU (indeed, has put it in the public domain) that the issue is with process yields particularly in he Belgian subcontract plant run by Novasep. It’s gone further and noted that the problem is with the filtration stages. The end result is yields about 1/3rd those attained in the UK plant and by SII in India.

            So no, it’s not being secretive. It’s notable that the high profile raid on AZ’s plant last week didn’t attempt to investigate production processes – apparently the EU may still investigate it (who they’d use would be an interesting question).

          2. Some idiot says:

            Thanks! I appreciate the feedback! 🙂

          3. Adrian says:

            Already before COVID-19 the majority of vaccine doses produced in the world were made by companies in India.

            The Serum Institute of India is the largest manufacturer of vaccines in the world (1.5 billion doses per year not counting COVID-19 vaccines), and last summer it took them around 3 months to start producing the AZ vaccine.

            All Indian vaccine producers combined plan to produce around 3 billion doses of various COVID-19 vaccines this year.

            CSL Australia started to setup production of the AZ vaccine in November and plans start deliveries next month. It is no surprise if it takes such a small local company longer to start production than the big and far more experienced vaccine manufacturers in India.

          4. Anon says:

            A lot of times, “production difficulties” and delays often relate to timely release of input raw materials with proper QC Release documentation, intermediate step QC testing delays, and equipment turnaround and cleaning times. Take that from someone who’s deeply involved in the production commercialization of greenlit approved products. Never attribute to sinister motives delays caused by adherence to cGMP.

          5. Some idiot says:

            Thanks! Probably not far wrong, either… says he who has been pulling his hair out due to delays due to QC release of some (quite common) solvents…! 🙂

    2. fajensen says:

      Often, the really tricky stuff in manufacturing a product is not patented. Because patents are public and readable by The Competition. Sometimes one will publish an article about it in an obscure trade journal that nobody reads to block The Competition from patenting the tricky parts and then suing the inventor of the process for patent infringement.

      I would imagine that the exact setup of those fluidics devices are exactly one of those tricky parts that are only documented internally, the documentation and the operations protocols are known by very few people, who are treated very well and probably signed an NDA and Non-Compete contract. .

      If I was The Competition, I would try to worm myself into operating that step, even if I didn’t make any money from it. Just to learn from it.

    3. Gordon Croucher says:

      That is not a question that should be raised by anyone with even a small understanding of economics. After reading the extremely difficult requirements for producing vaccines, and the risk of failure (Merck et al.) and the investment just Pfizer alone made, without help from governments ($2 billion) to remove the ability to recoup the investment and risk by invalidating patents is not a solution, rather a dis-incentive to hasten R&D, improvements to production and distribution etc.

  5. Lorenzo Lucchini says:

    Is the process potentially any simpler for Novavax, which is a protein-based vaccine, not an mRNA one, but appears to have very good efficacy?

    I guess in theory, producing the protein would be harder than producing the mRNA that then causes the proteins themselves to be produced… but given the mRNA production process doesn’t look simple at all, maybe my “guess in theory” is wrong.

    1. Sam Weller says:

      Canada actually announced today that it reached an agreement with Novavax to produce its (yet to be approved) vaccine in Montreal. Even in this case, the first vaccines are only expected towards the end of the year. Granted that this is not an established pharma company that is pitching in, so maybe the path could be a bit shorter in principle.

    2. Mantis Toboggan says:

      Not a process chemist or anything, but I’m pretty sure that protein drugs are easier to make than mRNA. Instead of making the mRNA through cell free In vitro transcription as you would for the mRNA vaccine candidates, you transfect cells with the DNA, which they will convert to mRNA and then protein. Making more protein just becomes a matter of growing more cells and purifying more protein at the end. There are lots of protein drugs on the market. Antibodies for example, are just large multisubunit proteins. Or as a more direct example, Pfizer by 2015 apparently prodcued over 1 billion doses of Prevnar ( which is a recombinant protein that is conjugated to 13 different sugars. I’m not quite sure how to estimate the production of that because I don’t have very good numbers, but seems like it scales quite well.

      I don’t know how involved the formulation of Novavax’s multivalent vaccine is though, and that could definitely complicate things. They had production issues even for making enough for US clinical trials and had to push back the trial start date.

      1. Eric says:

        Of course, each different protein will need a different, custom purification process developed for it. Whereas there is already an established purification procedure for mRNA–even if the sequence changes, it’s still RNA that you’re separating from proteins and other cell debris.

        1. Tharik says:

          That would be true if the cells were synthesising only the mRNA sequence that you want. In reality, I believe, you would end up with a mixture of many different mRNA sequences.

  6. BA says:

    Most of these analyses come from outside people looking in (e.g., Neubert). I find the lack of transparency from the manufacturers is eroding my trust. I’ve not seen anyone that will definitely say: no, not even an extra $10b will speed things up.

    1. Anonymous says:

      Have you seen someone try to claim that extra $10billion. I’m sure someone could cough it up were it that easy to step up.

      1. Roland says:

        WHO is currently asking for $23.2 billion for ACT Accelerator. Not all of that is for vaccination, and only part of what is for vaccination is for actual manufacturing. But some is. (Though probably not for the mRNA ones as Derek explained very well).

    2. FrankN says:

      You should have listened to Sierk Poetting, BioNTech CFO, yesterday night on German Prime time news saying excactly that (ca. 5:40 into the video, after discussing the NLP shortage in quite some detail)

      1. JOSE VARGHESE says:

        I wish I knew German

        1. Jan says:

          In the short version he mentioned that
          a) they took over the plant in Marburg and startet to modify it in september. It was a plant which was already ready to produce biocemicals and had a Filly trained staff.
          And even that took 6 months till they could start producing. (Normaly it would take a year)
          b) It’s a realy new technology so there is no standing supply chain which they have to establish alongside.
          c) there are no doses in hold or stored, everything produced and filled ist instantly send out
          d) Q: for the economy in Europe every additional dose is worth around 1000$ would it help if europe paid that much per dose to get vaccines faster?
          A: No, everything is already producing at maximum. (Including every supplier)
          e) India can produce a lot of vaccines but they only have the equipment for vector vaccines not for mRNA. To modify one of their plants it would take about a year (See point a)

  7. Malthus_2024 says:

    Ideally, someone should be planning how to vaccinate the entire global population of more than 7 billion people. Is WHO, or anyone else doing planning and implementation for what will be needed to fully vaccinate everyone world wide? Does it make sense to engage in the multi-year process of bringing new production facilities on-line? if so, how soon will that begin if it hasn’t already?

    1. mous says:

      Yes, WHO is looking into that (the 7 billion question) via Covax / GAVI and the ACT-Accelerator (link in name). And now that the recent changes in DC means the US is on board with those efforts (alongside pretty much every other country in the world who have been there since the beginning of the pandemic) it may actually work…

  8. Willy Chertman says:

    Is this true across the board, for all vaccine candidates, or is this most true for the mRNA vaccines, and less true for the adenovrius-based candidates and Novavax’s candidate?

    1. E Ray says:

      It’s specifically true for the mRNA vaccines.

  9. Earlier today there’s been some reporting that for Moderna at least it’s actually step 5 that’s their current binding constraint. That is, they’re asking for permission to putt 15 rather than 10 doses in each vial so that they can speed up production.

    1. Karl says:

      I can think of a couple ways to read that. The article doesn’t actually say that vial fill is their overall rate-limiting step today. It’s possible that Moderna is looking to bring more nanoparticle production online, and they want to make sure that later steps aren’t a bottleneck when that happens. It wouldn’t make sense to open up step 4 and then bump your nose on fill, when it’s something that can be alleviated by a regs change pursued in parallel.

      1. CMCguy says:

        I agree this does not imply fill/finish as rate limiting as appears more as option to gain higher output using same vials and operations thus increases efficiency. The key question for such an alteration would involve what stability data supports greater volume per vial. Does help manufacturing plus possibly distribution as get more doses in the supply chain. I think Derek is correct that Step 5 is a less prevalent technology that requires specialized equipment/facilities therefore can not just drop in to most pharmaceutical manufacturing plants (unlike most the other Steps)

  10. CB says:

    April 4, 1947, New York: a smallpox outbreak and announced plans to vaccinate everybody in the city. The mayor called an emergency meeting with the heads of the seven American pharmaceutical companies involved in vaccine production and asked them for a commitment to provide 6 million doses of vaccine asap. The pharmaceutical companies accomplished the task by putting the vaccine into round-the-clock production. One year later most New Yorkers were vaccinated.

    1. Mammalian scale-up person says:

      Um, no.

      Pharmaceutical companies knew how to make smallpox vaccine in 1947. I used to work at Wyeth back when we still had a farm in PA full of critters. Herds of cattle were not hard to come by, we didn’t have to import them from China, we had loads of the things just laying around (literally). You know what the bill of materials for smallpox vaccine was in 1947 (it’s not now)? “grass” and “vials” and “steel scratcher prongs”. That’s it. Occasionally you need to buy a new cow, but they also tend to increase on their own if you add enough grass.

      I wish we could make covid vaccine out of grass, since I too would love to go back to normal life and eat in a sit-down restaurant indoors, travel and go to the spa, but it ain’t happening. Pharma employees are as miserable as anyone else, I assure you. The more I see restrictions lifted due to political pressure instead of due to dropping infection rates, the more I think we may be cursed with this for a long, long time.

  11. Chris Phillips says:

    Apparently James Hamblin is billed as “Journalist, improv comedian, and physician”.

    Are we sure which hat he was wearing when he wrote that tweet?

    1. Ogamol says:

      Improv: “If you don’t have a time machine, all the money in the world won’t make this faster. If you do have a time machine, lead with that.”

  12. Smokerr says:

    And why would the EU not accept the UK and US test results?

    While not huge, another delay so their regulator people can put their finger in the pie and justify their existence and yell Hallelujah !

    A Western population si a Western population, sheese.

    Of coure Japan takes the cake for drag feet and we are dong the Olympic come hell or high Covd rates.

    1. Sean McDonough says:

      Possibly because they got burned badly in the recent past when they took a major US company at their word (different industry, but even so…)?

  13. JB says:

    “We had a vaccine in March” reminds me of the complaint about HTS and medchem, “Why did you screen 1M compounds and make 5k analogs when you could have just made the molecule that works?”

    1. Watson Ladd says:

      Challenge trials would have gotten effacity data and correlates of protection. Now we see the FDA refusing to even look at foreign trials, as thousands of Americans die.

  14. I’m firmly of the opinion that this could have been avoided – not by speeding things up now, but by having paid to scale up manufacturing 9 months ago. None of the steps – including the constraint on customized microfluidics devices – would have been hard to accelerate given an early expenditure of a couple billion dollars. It’s just that drug companies couldn’t or wouldn’t spend themselves into potential bankruptcy if the vaccines didn’t work out – and no-one paid them to do it.

    As I said in early April – “In about 12 months, the world will need to start producing massive quantities of a COVID-19 vaccine, but we don’t have enough vaccine production facilities to quickly produce the billions of doses we need globally.” The simple suggestion was to have governments actually guarantee the funds for scaling up quickly across the board, well before we knew which vaccine(s) worked. (Yes, I was still learning about the complexities of the manufacturing process, but the suggestion still stands up. And the final paper based on the suggestion is now published: )

  15. 이웅견 says:

    Now this post comes just in time for posting about one component that I had heard worries about being a potential bottleneck(sic) over the last few weeks:
    “The tough little bottles crucial to fighting Covid”

    Astonishing how much care has to go into seemingly trivial parts.

  16. RK says:

    I found this (somewhat older) BioNTech talk regarding LNP manufacturing:
    “Nanoparticle engineering by microfluidics” – Ferdia Bates, BioNTech, 2016:

  17. Ezra Abrams says:

    Is it true that mRNA reagents are not in short supply ?

    RNA polymerase
    the pseudobase

    and stuff for purification (? industrial scale HPLC or sephadex) RNAse qualified

    I mean, 100 liter sephadex columns (yes, they use them for insulin mfr) do exist, but they don’t grow on trees)
    see the figure on page 9 of this PDF

  18. Scott Underwood says:

    Derek, this is great information! You have explained it well enough to understand where the bottleneck is, and by what you are saying it’s obviously solvable. Our real problem is the whole industry, of course not because they are evil or lazy by any means, but because of the limitations they are operating under. It is the system, it is the system of Just In Time manufacturing. It works well for a corporation if they make things when they are needed. The cost of storage, spoiled product, the expense of machines that sit around and do nothing, it all makes perfect sense in that system. The world works with this method and has for a long time. That is the real problem because it is built on profit and it is that way to avoid losing money and decreasing profit. Derek your explanation is an excuse really for what our system is ABLE to produce. So the person you rail against is a person that is really saying the same thing I just did and hasn’t explained that part for you. Not in any way do I believe that you need this explained to you and yet you excuse what we have instead of screaming at the top of your lungs that it is not right for human beings’ health, especially in the face of a global pandemic. It’s not the best of what humanity can produce. The best we can do is far away from what this system can produce. It is the system that is the problem and I’m tired of people trying to rationalize an inhuman system of heath. Nothing personal.

    1. Lappan says:

      The possibilities for prophylactic spending are limitless: the richest country could devote all of its spare capital (both financial and brain-power) to stockpiling things that seem likely to be useful and still be caught out. We could build a fire station on every city block and probably succeed in reducing the damage from fires, but now that’s a larger proportion of the population being firemen rather than teachers or doctors, and when the hospital needs a new radiography unit perhaps a choice must be made: reduce the fire dept budget? increase taxes? and are more fire stations the best answer, or instead should we spend the same additional funding on fitting smoke detectors everywhere and drilling the population in safe evacuations, so that fires damage buildings not people?

      That’s not to say that a more determined pursuit of the technologies wouldn’t have been wise: after SARS faded out most countries shelved their research on novel viruses, vaccines, treatments, and civil preparedness.

    2. Pete says:

      What you’re basically saying is, if we had a few extra Step 4 machines sitting around ready, there would be no (or at least less) problem.

      The point though is that the relevant machines aren’t off-the-shelf kit. First you need to figure out exactly what your particular vaccine needs, then you need to build the machine that can produce that, which on its own probably needs quite a bit of time and experimentation to get right. If you tried a few, or even a few thousand, guesses in advance of machines that might be useful, the chances are very high that you will just have a pile of machines you can’t use, worth several billions or trillions of dollars and taking up all your space (or, if you went for the thousands option, a small town). That is a pure waste of money, which is why we do not just throw more resources at the problem.

      1. Charles H. says:

        It’s actually probably worse than that. If you leave the machines sitting around they probably won’t work when you need them. Dust will get in a crucial place. Valves won’t flip because their lube either evaporated or got too thick. Etc. Etc.

        And that’s not considering the surfaces of those micro channels which must not oxidize/dry out/breakdown.

        IIUC, microfluidics devices are quite picky about all manner of things that won’t stay the same while they’re being stored..or even used. They probably need to be replaced fairly regularly…and each one is(was?) a custom manufacturing job. The electronics are probably fairly stable. (It’s been a long time since the fancy vacuum tubes needed to be replaced frequently.)

      2. SteveH says:

        Coming from electronic manufacturing that’s what seemed odd about the slow start in the AZ production in Europe. Once you have the process right, which AZ had in UK in Autumn, then you should be able to replicate those high production rates in any new factory with little delay. As long as you have the gear, and this article seems to confirm the gear is not the rate limiting factor

        1. E Ray says:

          I wish biopharmaceuticals were as straightforward as electronics manufacturing. Things often don’t scale up well, different sites may produce different results, contaminations can be common… Certainly gives me more of an appreciation of the complexities of biology.

    3. Johnathan Hughes says:

      Scott: sure, just in time manufacture is part of the problem, but it’s far from the ONLY issue.

      Your thought would be great if the only product needed from the pharmaceutical industry was the COVID-19 vaccine – if we could gear all the resources to produce this one item. Unfortunately, we can’t. There are all the OTHER vaccines, drugs new and old, vitamin and mineral supplements. Homeopathic substances which are utterly useless, but demanded by the public anyway. Some resources which could be repurposed for the vaccine cannot be, because they’re used for other projects [some of which are literally a matter for life and death]

      All of this is to say that even if you solved that problem – if we snapped our fingers and now manufacturing of pharmaceuticals was no longer a for-profit industry, and we’d somehow made that work – there are other issues that get in the way of doing this. Some of the resources are inherently scarce. If suddenly we had unlimited money for manufacturing vaccines, we would STILL not have a huge number more people working in that industry (there is a limit to the number of people who are both interested and talented in that area; some chemicals, equipment and feedstocks can’t be manufactured any faster no matter how much cash you throw at them). In my opinion this is a greater problem.

    4. sort_of_knowledgeable says:

      Storing a year’s supply of stable PPE for emergencies would have been prudent.

      Storing excess chemicals has an environmental cost. Over years they deteriorate even in refrigerated conditions and disposed of as hazardous waste if not used in timely fashion. You can read stories on this blog where a bomb squad literally had to be called when old chemicals were found and known to decompose to expulsive substances.

  19. mymagoogle says:

    Also add all the QC testing along the way. I presume for the in process (between step) testing, given the urgency, that they are always proceeding to the next steps at risk, before the tests are completed with a thumbs up to proceed. Or maybe process time is so scarce that they don’t want to waste that slot just in case that batch is bad.
    Regardless, the release testing is always at the end – or in Derek’s scheme somewhere in the middle of step 5, and let us not forget that sterility testing takes a month.

    1. A Nonny Mouse says:

      Under the EUA here in the UK, the regulators have to test each batch before release. I believe it is now down to about 5 days with 24h working.

    2. Mammalian scale-up person says:

      Correct, we proceed at risk unless there is some obvious indication of contamination in in-process samples; release testing at the end. However, sterility testing does not strictly speaking have to take a full month as there are now rapid micro tests. If qualified properly, we can do batch release in more like a week. Unfortunately I’ve not seen them widely adopted – only at a few sites that are focused on cell therapies where you have to return that packet of frozen CAR-T back to the hospital ASAP.

  20. Julien says:

    Maybe a dumb question, but don’t you think that drug companies that are able to make liposomal drug products (i.e. AmBisome = liposomal amphotericinB by Gilead; Doxil = pegylated liposomal doxorubicin by Baxter) could quickly be taught how to encapsulate these mRNA strands in the right lipid nanoparticles?

    1. Mammalian scale-up person says:

      No, the problem is that the process == the product in biologics, to a very large extent. There is limited acceptance even of biosimilars identical in structure, sequence and posttranslational modification – there is certainly no “just as good” for RNA. Once the process methodology is filed, it’s really set in stone, for all biologics, even when it’s clearly nonsense that has no impact on the finished product.

      I’ve seen startups without previous experience in regulatory filings misunderstand the requirements and put in the number of doors and windows in the building in the actual BLA, under the section that asks them to describe the facility. Worse, they counted wrong, and when the inspector came through they put paper and a large piece of equipment in front of the extra door…and then they were unable to transfer manufacturing to another site, even when their original site was fit for nothing but a bulldozer, so the company that bought them was stuck trying to re-file. You have to really know what you’re doing when you write that manufacturing process out on paper….

      1. fajensen says:

        The TBTF-project I work on did exactly that by SIL-classifying all electrical systems and making all of them part of the radiation safety approval records – instead of the conventional, sane and even normal alternative, which is to keep all systems with “regulatory requirements” separated from everything else as far as practical. Because of The Paperwork.

        In time they will find out what I failed to get them to understand; when they have to apply for re-approval of their operational license by the authorities after changing the size of some fuses!

    2. Skybolt says:

      Actually, Baxter has signed an agreement with Pfizer/BioNtech to produce some part of theurvvaccine in its Halle facility in Germany. Theynplan to start in March and dekiver the equivalentbof “hundreds million doses”.

      1. Marko says:

        Wow. Spellcheck is having a bad day.

  21. Mammalian scale-up person says:

    LNP manufacturing isn’t all THAT different from liposome manufacturing – and there are a handful of ways to do that, but at commercial scale we do not use microfluidics chambers, ever. That would be ridiculously inefficient, we’ve known how to make emulsions at scale for a long time.

    I am 95% sure Moderna and Pfizer have not even tried the more traditional liposome compounding methods (membrane shear, sonication, high shear blenders) all of which are off-the-shelf and most definitely make uniform particles of whatever size you like, provided you play around with flow rates and the spinning shear head designs a bit. Everything I’ve seen from them is…highly academic and has zero consideration for scale or normal engineering stuff like “Class 1 Div 2 limits” or “semi-continuous vs batch process.” They have not had time to develop a platform – THAT is the problem and creates a bottleneck, everything they do is a one-off and highly specialized. Playing around with different methods takes time, usually upwards of 5 years. That shouldn’t be surprising – most startups don’t even bother to make a platform until they’ve produced several successful commercialized drugs, because it’s financially more rewarding to put all your resources into meeting investor milestones and getting some revenue streams going. Years later, is when you worry about platform development because all this capex and overhead for little one-offs is starting to cramp your cash flows.

    At commercial scale, we like to (need to, after a while) have things on a platform: standardized equipment modified only slightly to meet the new process requirements, with off-the-shelf reagents and no weird catalysts or single source components. If there’s only one vendor in the whole world (e.g. who makes GMP quality polymerase, or GMP quality capping enzyme) then you are going to find yourself in a world of hurt very quickly, as soon as they have the slightest delay. Can we make these things? Sure, just like you can use phenol and hexane as starting material, but it’s adding an awful lot of work to your life and you probably would much rather buy something more ready to use. You can definitely buy dNTPs and enzyme from Sigma and NEB, but these are extremely large quantities that are NOT lab quality crap – they have to be produced under much more rigorous conditions, and that takes time. Basically whatever would take you an hour or two at most in the lab, takes a solid day or two at commercial scale in GMP conditions.

    Unfortunately, anything you create on a new platform has to go through clinical trials. AGAIN. So for this vaccine it is just not going to happen. We will not get a platform for RNA anything for many years and it will continue to be miserable to make. Hence why I think we should focus on what can be done to accelerate adenovirus-based and Novavax’s stuff, because that we CAN make lots of.

    Adenovirus, in comparison, does have a platform: viral vector development has been ongoing for some years now. It’s made on pilot-scale standard equipment we do happen to have laying around, doing pilot scale development stuff, but that doesn’t mean it can’t be repurposed or that we couldn’t get more with 3-4 months of lead time. Re: large IEX columns, I have…let me think… a few 2m x 20-25cm columns, three 1.4m x 25 cm in use at the moment just at one site (multiply by 4 sites), probably a few thousand additional liters out in cold storage. IEX columns are cheap cheap cheap for me, and there are also now charged membranes we can use for the same purpose, made by Pall and Sartorius. All standard reagents, media, buffers I already have laying around the warehouse. Adenovirus vaccines are great! We can make those as quick as we make anything, which is to say we can have batches out the door in 6-9 months.

    Novavax uses a baculo expression method that is…slightly weird? We mostly only used baculo in the mid-late 2000s, and only when mammalian expression failed to produce enough material for HTS / X-ray crystallography. The posttranslational modifications are highly immunogenic and it’s a legitimate question to ask “if they are not normal modifications, how relevant a model protein is this?” That said – it’s made in similar equipment as mammalian cell culture, of which we have lots, but it does need to be made at larger scale than adenovirus cultivation, so there is a real question about who in the world has capacity to run X000L systems for a couple of years, making nothing else. I mean, I don’t, because of the expense of running the things we tend to keep them fully occupied.

    But, basically: source material for the RNA vaccines is just as much of a bottleneck as the emulsion manufacturing. They both suck. Let’s not focus on RNA vaccines please as there is no serious chance we will be able to scale them quickly or meaningfully in the quantities needed, while we CAN! DEFINITELY! make loads of adenovirus and with some moderate effort of capacity sharing probably come up with some locations to make baculo stuff too.

    1. Marko says:

      “Unfortunately, anything you create on a new platform has to go through clinical trials. AGAIN. So for this vaccine it is just not going to happen. We will not get a platform for RNA anything for many years and it will continue to be miserable to make.”

      So you think that an mRNA vaccine for a new variant will have to go thru a full set of clinical trials? I’d agree that that’s what probably SHOULD happen, but I don’t believe it will. Fauci has already said as much.

      I can see that you may mean that if anything about the platform except the payload changes significantly that it would require a whole new approval process, in which case I’m sure you’re right, even in this rushed environment.

      1. Mammalian scale-up person says:

        The second one – if anything about the platform (equipment, lipid, home-brewed enzyme as opposed to purchased) changes, clinical trials must be re-done.

        1. Marko says:

          OK, thanks.

        2. sgcox says:

          But surely not the full scale Phase 3 trials as for the original approval ?
          That would be insane. And yes, I do not know much about it but is very curious.

          1. Marko says:

            Yes, you have to do the full Monty if you change the platform. The same would occur with the flu vaccine if they changed from chicken eggs to ostrich eggs.

          2. Mammalian scale-up person says:

            Yes, unfortunately. There have been a lot of nasty surprises about biologics over the years, so it’s extremely hard to prove equivalency.

          3. sgcox says:

            What exactly is classified as “biologics” ? Is it simply something produced by living cells, like like mAbs ? But in that case many antibiotics are biologics but pobably not held to the same approval biosimilarity standards as mAbs. Bio washing powder to which we all exposed regularly should also be called biologics. But mRNA are produced in cell-free system and for me looks closer to normal chemical drugs from its production process.

          4. Andy says:

            I would hope that, at least under present circumstances, a smaller study showing safety and immunogenicity would be sufficient.

          5. Mammalian scale-up person says:

            Re: what constitutes Biologics
            -A large molecule pharmaceutical or medical device containing a pharmaceutical (e.g. the BMPs which come in a kit with other parts for repairing bones like InductOs or Infuse, insulin pumps) which is going to go into a person or animal.
            -Usually a protein, peptide, virus, modified cells (CAR-T or allogeneic), whole bacteria or bacterial “ghosts”, large polysaccharides (e.g. Prevnar), DNA and now RNA

            You are correct that many of the larger small molecules (cyclic peptides leap to mind) are sort of borderline, so we tend to classify them also by manufacturing method. Tetracyclines, beta lactams, ivermectin are considered small molecules by virtue of size alone, even though these are produced using similar methods – they could be made completely synthetically, but it’s a lot easier from a process standpoint to make them with semi-synthetic methods (ferment something that’s mostly the right molecule, then put some finishing touches on the purified result).

          6. Some idiot says:

            The regulatory side of them is also quite defining as well, no? I am a small molecule scale-up guy… I understand that the regulatory side is somewhat different from biologics, and that (eg) purity and assay for things like tetracyclines (being classified as small molecules) are quite different to similar quality specs for (eg) mAbs…

          7. Mammalian scale-up person says:

            Re: regulatory / QC definition, a lot of this comes from contamination being different in nature than small molecules and in a lot of ways can be much worse. Contamination from small molecules is usually a slightly-different small molecule from a side reaction product which may or may not be toxic vs simply inactive in small amounts, plus a great many small molecules are oral or transdermal dosing which means sterility is not a huge concern. Often you can rebatch small molecules to remove more side product if it doesn’t all come out in a single recrystallization or whatever.

            When a mAb is contaminated, it’s usually a bad bad bad scene and you can’t just re-purify the thing – the bacteria / fungi will chew up the product completely and use it as a carbon source. Contamination is usually by environmental or waterborne bacteria, and since the drug is certainly going to be injected, frequently into a person with a suboptimal immune system, the risks of endotoxemia and sepsis are high. As my old ChemEng professor taught me, if you let impurities build up in a chemical reaction, the danger is you’ll be lulled into complacence because this reaction has proceeded normally 599 times and on the 600th time it finally builds up enough crud to go boom. In biologics, a single cleaning / sterilization failure means you take down the whole system, double-clean it, possibly take down the water system and clean that too, start a new WFI sanitization protocol that means additional down time, change your PPE methods and retrain staff…it’s an ordeal.

      2. E Ray says:

        There is a new flu vaccine every year too, for the variants that are deemed to be most prevalent for the upcoming season. I wonder what kind of clinical trials (if any) are needed for those…

    2. Robert says:

      Great comments on manufacturing advantages with the virus vector vaccines. Do you think AstraZeneca is having manufacturing issues with their Belgium plant producing their Covid-19 vaccine for the EU or is it more just getting the materials sourced and delivered to the plant?

      1. RC says:

        It is more about cells dying and producing less yield. And problems in filtrations.

        1. Mammalian scale-up person says:

          Very common problem in adenovirus and I’ve also seen it in the whole killed virus type vaccines – have also seen it successfully addressed by starting the initial production scale culture as perfusion to drive up cell density before viral infection, then at harvest running sloooowly through a continuous centrifuge to get out more of the debris before running through depth filtration so there isn’t quite so much lost on the filter media and holdup volumes.

    3. FrankN says:

      “Unfortunately, anything you create on a new platform has to go through clinical trials. AGAIN. So for this vaccine it is just not going to happen. We will not get a platform for RNA anything for many years and it will continue to be miserable to make.”

      The next clinical trial of a mRNA vaccine is going to start maybe this year already, lastest next year: The flu shot Pfizer/BioNTech have already been working on since 2018. And lots of things they are doing now have been done with that shot in mind, and being based on their joint flu vaccine research. So, don’t over-dramatize, please.

      And what makes you think that some of the firms mentionned by you aren’t already involved in upgrading mRNA production? One of them is certainly in, from what a leading sales manager has told me.

      1. Mammalian scale-up person says:

        They probably are trying to develop a platform – but as I said this is not by any means a short term undertaking. At all. By the time their platform is developed, we will have more adenovirus based vaccines to choose from, and the shipping and handling of those will give them a much more widespread adoption.

        I will also point out that if you dig through clinicaltrials(dot)gov you will see that Moderna has run a great many clinical trials including about a dozen for vaccine candidates in conjunction with Merck, and has had neither Phase 2 success until now, nor a platform developed despite being in business 9+ years. There was a WSJ article about their management practices July of last year that may help to explain this…

    4. MrRogers says:

      I note that each of the methods you suggest as alternatives to microfluidic encapsulation (“membrane shear, sonication, high shear blenders”) are very efficient at shearing nucleic acids like mRNA, thereby making them useless. There may be better methods to make these nanoparticles (not just an emulsion, mind you), but you haven’t suggested one.

      1. Mammalian scale-up person says:

        Not at all, polycarbonate extrusion has been used (DOI: 10.1007/s11051-013-1960-3) and scaled up processes routinely have relatively high shear compared to lab scale (chromatography where there’s a sudden jolt from turbulent mixing in the bulk to laminar through the resin, where the material “sees” high velocity against an immobile resin wall) – but if you don’t like those, there’s also detergent dialysis (I have TFFs laying around, no problem), spontaneous LNP formation by calcium phosphate condensation of the long RNA strand (DOI: 10.1016/j.jconrel.2009.11.008 )prior to the lipid coating which turns the whole thing into a simple mixing problem, centrifugation (I can get one from GEA with minimal headaches) and two-phase deposition methods that are quite scalable.

  22. Mandark says:

    “Five years ago we simply could not have gone from sequence to vaccine inside of a year.”

    But it’s not just RNA vaccines that proved capable of achieving this feat. The viral vector-based vaccine by the University of Oxford was developed at the same time as vaccines from Moderna and BioNTech. Of course it’s also a fairly new platform and the Oxford team had an advantage in that they had been working on a vaccine for MERS-CoV when the pandemic hit, so 5 years ago its development might not have been as fast. But vaccines based on viral vectors were also developed at comparable speed by the Gamaleya Research Institute in Russa and by CanSino in China (the latter is still in phase 3 trials, but phase 2 results were published already in July 2020).

    More traditional, inactivated whole-virus vaccines, also saw rapid development in China.

    1. Marko says:

      Agreed, but you have to admit, the name “Operation Warp Speed” WAS catchy.

      1. WarpDriven says:

        Think of a rich Scottish brogue, and then

        “I’ve giv’n her all she’s got captain, an’ I canna give her no more!”

        It turns out many things on Earth are like that. A fact which is lost on our new class of clueless leaders.

    2. Michele P. says:

      But the Gamaleya Institute also had an advantage because they also worked on a vaccine against MERS (BVRS-GamVac), with the same vectors rAd26 + 5.

  23. DataWatcher says:

    Let us remember, as well, that the research and development of the Pfizer/BioNTech vaccine (which we Yanks chauvinistically insist on calling the “Pfizer” vaccine) had nothing to do with Operation Warp Speed at all; virtually all of the funding (as well as the the messenger RNA technology itself) came from BioNTech, mostly under the largesse of the German government, who gave the company what amounted to a $445 million R&D grant to help things along. Granted, Warp Speed purchased the initial 100 million doses for $2 billion in July, which is no small item. But the science itself was funded by a democratic socialist government. Basically, we owe it to socialized medicine that we now have this vaccine at our disposal.

    1. Steve Jones says:

      The right term is not “democratic socialist”, it’s “social democratic” and there is a world of difference. Germany is not a socialist country, it has a market economy with the vast majority of the commercial sector in private hands. Yes, it has a strong welfare sector, but even there most of the health providers are in private hands.

      Why do Americans (I assume you are such) insist on calling countries with strong welfare systems “socialist”.

      1. DataWatcher says:

        I was not using that term disparagingly; quite the opposite. In the U.S., the terms “social democrat” and “democratic socialist” tend to be used more or less interchangeably by people who approve of the concept — our “Democratic Socialist” party, while not a “major” party by any stretch of the imagination, consists of folks who would most likely be called “social democrats” in most European countries:

    2. Mikk Salu says:

      I think that your description is not correct. Biontech received this grant in mid-September last year. It was meant for production.
      Around same time Biontech announced that they intend to buy production facility in Marburg, I guess, most of this grant went into Marburg deal.

  24. anon says:

    What about the future? If another virus were now to arise how quickly could a new mRNA vaccine be developed and marketed, given the lessons of COVID? Should mRNA vaccine technology be understood as bringing us to a post-pandemic era?

  25. David says:

    Great post. I’m never surprised when somebody not in the industry underestimates the effort involved (or, for that matter when somebody in industry underestimates the effort in somebody else’s line of work).

    You can see the Lonza facility in Portsmouth, if you ride the 50 or 100 mile annual “Cycle the Seacoast” bike ride hosted by the Am Lung Assoc. The route goes right past it. Fortunately, the route ends at, or near, a nice brewery.

    1. warr says:

      Near the Pfizer factory in Belgium (Puurs), there is also a nice Brewery called Duvel (which you could translate to devil). I’ve visited that Brewery a while ago. And not that far off you have also a factory of Capsugel (Lonza).

  26. Chad says:

    Could the process be easier to scale up for viral vector or inactivated virus vaccines? The Chinese are making inactivated virus vaccines which seem to be rolling out not much slower than the high tech mRNA candidates? These are what developing countries will rely on in the absence of the mRNA vaccines. It seems strange to me that not a single American company went with the traditional inactivated virus approach.

    1. FrankN says:

      Inactivated virus vaccines are cultivated in chicken embryos, i.e. eggs, so supply of fresh eggs can be quite a limiting factor. Asides, adaptation to new variants is quite time consuming.

      Both factors in combination have lead to the well-known, quite unsatisfactory situation with traditional inactivated flu vaccines, which are typically short in supply, and of rather low efficacy. mRNA entered the agenda as a way to improve flu vaccine supply – the fact that COVID came first doesn’t change the fact that inactivated vaccines are generally regarded as unfit for the 21st century.

      Check out Chinese production capacities (the actual, not the claimed ones), and reports from Argentina or Egypt on delayed supply, and you understand what I mean.

      1. sgcox says:

        no, that is only for flue.
        SARS-CoV-2 does not infect cheicken (no ACE2 or very different ?) so it is done in cell cultures like adeno vaccines as far as I know. Surely at much more stringent safety conditions.

        1. sgcox says:

          spelling 🙁
          Do not do it before morning coffee.

  27. mallam says:

    I’ve had the same discussion with some very non-scientific persons. So my much simpler explanation is that these are the first two mRNAt vaccines ever approved for human use, and as such, other companies do not have and have not had to build manufacturing capabilities for a vacciine of this type. For more than 6 months I’ve been also saying that the best investment for Warp Speed would be (could have been) in facilities toward making the vaccine(s). These could be a national resourse, available for rent by companies in normal times and used in such national / international crises in the future.

  28. Jens says:

    According to the German newspaper FAZ* (which is a reliable source) reporting the results of a summit-talk of politicians yesterday (including Angela Merkel) Biontech reported that they could produce more of the RNA vaccine if supply of lipids wasn’t limited. These lipids – of course – have to be ultrapure and can only be obtained from a small number of comapnies. Furthermore, the purity has to be certified (of course) and cannot be upscaled within a short time.
    Only in German /you may use for translation)

  29. Jonathan says:

    Interestingly, after reading this blog I saw a piece suggesting just such an outsourcing of mRNA vaccine production is going to happen.

    In this case it is establishing production capacity for an mRNA vaccine still in Phase 3 trials – so regulatory application when it occurs will include the new site – plus collaboration on a “second generation” vaccine designed from the outset to be tweakable for new variants.

    It does establish that it should be feasible to get other manufacturers on board, albeit with long lead times.

    In terms of the AZ Europe problem, what I had heard was that poor production is a “teething problem” of setting up a new adenovirus plant. Despite having optimised production in the UK, replicating it in a new plant has resulted in lower efficiency and there will need to be some troubleshooting before the hoped for yields start to emerge.

    I am not sure whether there is evidence of European AZ production being diverted to the UK, the EU sources were asking why there couldn’t be diversion the other way. Though I am fairly sure I recall that at the point of UK authorisation there was production coming from one of AZ’s European bases (possibly Sweden, or Netherlands) so now UK manufacture is on stream that may be contributing to meeting the EU contract.

    1. A Nonny Mouse says:

      The UK production is at 2 sites in the UK, but some of the material is sent to an AZ facility in Germany for reconstitution into vials. The rest is put into vials at a Wolkart factory in Wrexham.

      As for “teething problems”, we have been trying to get an enzyme produced for 2 years and it’s still not sorted for production purposes. Just like the AZ contract, this is on a “best efforts” basis so we still have to pay after they have made changes that we have not approved. 300,000 Euros for a small company is not trivial.

  30. drug czsar says:

    I would say that, as often, the truth is in the middle. There are certainly a number of companies that have both the expertise and the infrastructure to make one of the vaccine variants. It is more a matter of political will to force them to work as contract manufacturers. Something that those companies are not really interested to do spontaneously, as we live in a reality where making a buck, instead of 100 or a 1000, is considered just not worth the effort

  31. SteveM says:

    I have not read all 75 previous comments. But what is the production process for the adenovirus vaccines? Sputnik V also reports 90+% efficacy and it can be transported and stored using conventional refrigeration.

    Whether Sputnik V or some other adenovirus product, if an adenovirus solution is also superior to mRNA products related to manufacturing and logistics, it seems to me that it should be declared the “winner” and allocated whatever-it-takes resources to ramp up production.

    1. A Nonny Mouse says:

      It is 2 doses with 2 different adenovirus carriers, with the second being quite difficult to produce which is limiting supply. This is probably the reason for a new study looking at the AZ vaccine as the second dose (trial about to start soon).

      Interestingly, the AZ vaccine is significantly cheaper as well.

  32. electrochemist says:

    One aspect that is often forgotten: this isn’t a project to create a few billion doses of vaccine(s) and get large populations immunized. It is a project to manufacture billions of doses of vaccines and get large populations immunized EVERY YEAR from now on. That includes syringe manufacturing, glass tube manufacturing, bright stock procurement, etc., all above and beyond what is needed for current medicines. It won’t work to divert vials used for insulin fill/finish to use for Covid vaccines because not supplying insulin to patients leads to even bigger problems.

  33. Alisande says:

    I don’t understand the DNA process in step 1. The Covid-19 virus is an RNA virus that contains no DNA. What am I missing? What DNA is used?

    1. sgcox says:

      DNA (prepared in E.coli) is a template which is used to synthesize mRNA by in-vitro transcription reaction. Technically, mRNA can be synthesised chemically without DNA step but that is hellishly difficult for such long sequences. No DNA in the vaccine product.

    2. MrRogers says:

      DNA is the template required by the RNA polymerase that produces the mRNA.

  34. James says:

    I assume some people who say “we had the vaccine in February” are making the mistake of thinking we knew the vaccine was effective in February and the FDA was literally twiddling their thumbs for most of last year, but there’s a similar-sounding argument but much smarter argument that I hear that I hope you don’t dismiss out of hand quite as easily: is *10 months* from candidate to fully approved for distribution actually as fast as we can go? Is it really, truly impossible to learn whether a vaccine is safe and effective any faster than that? Or is that something we’ve been hemmed in to just accept by a bureaucracy with perverse incentives that lead to hundreds of thousands of unnecessary corpses?

    1. E Ray says:

      I mean, 10 months is astonishingly fast. There’s not much more that they could shave off. I don’t know how people are starting the clock in February–I guess because they had the covid genetic sequence then, which could be targeted for the eventual mRNA sequence? They had first in human by July, and it took just a few months to get results, which they compiled into a EUA submission, and the FDA took, what, 3 weeks to look it over? Obviously they greatly shortened many steps, but they still have to develop an at-scale process that makes the same thing every time, they still have to run a large scale clinical trial, they still have to submit the authorization paperwork, which was greatly reduced from a typical BLA filing…

  35. Vader says:

    Who is James Hamblin?

  36. E. M. Unfred says:

    Is the Alnylam site in Norton set up for mRNA or siRNA manfacture? Two completely different processes.

  37. PS says:

    In US (both Trump and Biden admin), has tremendously underestimated and continue to underestimate the number of vaccinations required for vaccinating a population of 350 million. It is NOT 350 million vaccinations. The actual figure one needs to plan for is 3 to 5-fold of 350 million.
    Biological manufacturing is not like making wooden widgets. It involves many complex sub-processes that are not well understood. The likely hood of failure of batches, delays during manufacturing, contamination during storage, failure due to storage temperature incursions during shipping and storage, never before seen contamination in starting raw materials and glass particles in the vials which manifests itself upon storage, are all common occurrences even in a fully validated manufacturing process. These failures are unpredictable, random and sometimes the root cause can never be identified One avoids this risk by planning to manufacture large excess number required of doses over a reasonable period of time.
    If the Vaccination process, (shots in the arm) runs into months, 350 million doses will not be enough. Again, one needs several-fold of 350 million doses. We are actually experiencing this now with the emergence of new vaccine variants (UK, Brazil, S Africa and CA and who knows what variants), the need for unplanned booster shots to counter the variants. The longer the vaccination process lasts more doses would be required. This is apart from the 3000 deaths per day in Jan 2021 with total at 450,000 to date!


    1. fajensen says:

      What I don’t understand about the USA is that: The country has been at war since Bush I and yet there seems to be no way to actually pull a “state of war/emergency -lever” that authorises f.ex. mass vaccinations country-wide, top-down, without the process getting snarled up with every regions bureaucrats eager to prove their unique worth by questioning each, and, every, single, step of the program, A.K.A. “The Swedish Model”.

      I actually though this kind of thing was what FEMA was for. Maybe I have watched too much X-Files :).

  38. idiotraptor says:

    Having previously worked at Alnylam, I feel reasonably confident saying that both Pfizer and Moderna prepare their vaccine RNA by chemical synthesis. No one does preparative or industrial scale RNA manufacture with cell free in vitro translation; doing so is not technically or process feasible.

    1. Mammalian scale-up person says:

      Nope…having previously worked at Moderna, I assure you it’s in vitro. The sequences are more like 2000+ residues, not 20-some, is the problem.

  39. A says:

    Step 4 occurs does not require any sort of sexy microfluidics tech, in fact methods for rna encapsulation by lipid nanoparticles has existed since at least 2002 (US20040142025A1). It seems like a good chunk of the process used by Moderna is referenced in US8058069B2.

  40. idiotraptor says:

    Wow, never imagined that was the case. I stand corrected.

  41. Nucleotide Wrangler says:

    The big players, including Pfizer and Moderna, decided back in March that they were going to assume that their vaccine would be approved. They did not wait for the trials, they went right into scale-up. I’d think that most people here would assume that, but I work in an adjacent field, so I know for sure. There are no mRNA vaccine factories sitting idle. There is very little spare capacity of any kind. My company hasn’t been fully staffed since last January; equipment and people were rented out as surge capacity. (The scientists are mostly back, now it’s the logistics team that’s been called up.) Going from lab-scale to “we need a literal billion of these” does not happen overnight, even with an infinite budget.

  42. An Old Chemist says:

    GSK inks deal to produce CureVac’s COVID-19 vaccine—and develop a next-gen version, too (FiercePharma, Feb 3):

  43. An Old Chemist says:

    Vaccine manufacturing greenhorn Bayer to make 160M doses of CureVac’s COVID-19 shot (FiercePharma, Feb 1):

  44. Bert says:

    The tail end of a talk Neubert linked to — — has a good discussion of LNP manufacturing and the challenges in scaling it up, including some nice slides at around minute 44 comparing various traditional and microfluidic approaches. The rest of the talk is also a nice review of mRNA vaccine design and production.

  45. Tom B says:

    I’d be critical of many aspects of the UK government’s response to the pandemic, but one thing (in hindsight) that they did get right was to invest fairly heavily in pre-ordering vaccines and sponsoring plant expansion in the summer of last year
    It’s fair to say that was a political risk as well as a financial one. It wasn’t too predictable 7 or 8 months ago that the vaccines were going to be as effective as most of them turned out to be.

  46. IA says:

    Microfluidics for making nanoliposomes with the special lipids and mRNA sounds like a handicap to vaccine preparation. I wondered whether sonication or extrusion have not been tested, I never thought that lipofection could be so difficult; I suppose it is their candidate and FDA approved. The positive charge lipid and the other lipids can be manufactured by many brands. The upside was mRNA but the downside are the lipids and this weird microfluidics tech, yet it was the first vaccine approved.

  47. Jamil Gregory says:

    This guy decided to start producing the Oxford vaccine before it was approved anywhere and the bet paid off.

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.