Skip to main content


A Close Look at the Frontrunning Coronavirus Vaccines As of May 1 (updated)

There’s plenty of news on the coronavirus vaccine front, so let’s have a look. If you need some details on the different sorts of vaccines in general, here’s the background post, which should help this one make sense. This is a rapidly advancing field, with a huge number of programs. Some of the players are doing a lot more than they’re talking about, while others (as is always the case) are talking much more loudly than their actions really justify. The signal/noise isn’t great, but this will be an attempt to make sense of the landscape as of today.

Update: I should put in the links to the larger vaccine lists, as I did in the earlier post. Here’s a good overview of the coronavirus vaccine world in a recent Nature Reviews Drug Discovery. The official WHO list is here, and at BioCentury they have constantly updated open-access summaries of the vaccines and other therapies that are in the clinic and the ones that are still preclinical. They’ve also recently published this excellent overview of vaccine issues in this area.

CanSino’s Ad5-nCov

Probably the most advanced candidate at the moment is CanSino Bio’s Ad5-nCoV. This one has completed Phase I studies and the company has apparently started enrolling patients for a Phase II trial, making them the first to do that, to my knowledge. That one is of the kind mentioned in the background post as using a different virus entirely (adenovirus, which infects human cells readily) to deliver the DNA for a coronavirus protein (or proteins). You can also look at this as a hybrid of “live virus” and “recombinant protein” approaches, because you have a real infectious virus (just not the one causing the disease) being used to generate protein antigens that will call up antibodies to the real disease virus. Here’s a review of the approach (open access), which has had a lot of work put into it over the years. Adenovirus vector vaccines of various sorts have gone into human trials for HIV, influenza, Ebola, tuberculosis, and malaria, but none have made it all the way through yet. That’s partly because those are some damned hard immunization targets – people have been trying to come up with a decent tuberculosis vaccine since before any of us were alive – but that also tells you how seriously people take this technique. There are vaccines that use this different-DNA-in-another-virus technique (such as the Merck Ebola vaccine) but I’m not aware of any adenovirus vector vaccines that have been approved anywhere for human use yet. CanSino has an adenovirus-vector Ebola vaccine of its own (Ad5-EBOV) that’s already in Phase II trials; work on that one surely provided the boost needed for the company to advance this candidate so quickly.

No one outside CanSino has seen the Phase I results, and the main thing that we know about the Phase II trial is that the company is planning to enroll 500 patients in Wuhan, and that the highest dose from the Phase I protocol has been dropped. 250 people will get the Phase I middle dose, 125 will get the low dose, and 125 will get a placebo injection. We’re likely not going to hear much about this until the conclusion of the trial; any sudden news before then has a better chance of being bad.

Update: here’s more on the Phase II trial – I particularly like the line about how “this time” the participants will be able to stay at home rather than staying in the research facility the entire two weeks. . .

The Oxford vaccine, ChAdOx1-nCov19

Meanwhile, Oxford University has its own candidate, which is in a very aggressive clinical development program. They’re telescoping Phase I safety and dosing and Phase II efficacy measurements into one 510-patient trial, with numerous endpoints. This one is another adenovirus vector which will set off production of the coronavirus Spike protein and (one hopes) raise a vigorous antibody response to it. The vector is a chimpanzee adenovirus from Vaccitech, so although the concept is similar to the CanSino vaccine, this will be a different beast. There is literally no way to know which of these competing efforts will yield a better vaccine, or if either will work at all: that’s why we dose human beings, and in this case those humans are (I believe) beginning to be dosed right now (Thursday!) Oxford is definitely taking a chance with their trial design, but then, everyone else is taking chances of one kind or another here.

The good news is that the Oxford group had also put work into developing a MERS vaccine (yet another coronavirus) using this same platform. Their  ChAdOx-MERS vaccine also expressed that virus’s spike protein, and in Phase I human trials there were no safety problems, and they did indeed elicit the desired immune response. The group has a new preprint out that shows that a dose of vaccine in animals (up to monkeys) also provided immunity against a whole suite of known MERS mutational strains, which is good to see as well. (For more on this vaccine, see the Sinovac section below).

Update: this team has reported work in Rhesus monkeys (at an NIH lab in Colorado) that shows apparent protection from viral spread, so that’s good. Right now it’s just press reports; I’ll put up a link to a preprint when it appears. And they’re pushing ahead with that aggressive clinical plan mentioned above, with a 5,000-patient Phase II/Phase III trial set to start next month. They’re talking about possibly having emergency-use vaccine ready in September (!) That is about as fast as is physically possible, I would say, and it’ll be something to see. It will only happen by then if everything works perfectly the first time, though. India’s Serum Institute will be involved in the manufacturing, among others.

Update 2: AstraZeneca has jumped into the vaccine picture by signing a worldwide production and distribution agreement with the Oxford team. The press release notes that the vaccine has been given to 320 people so far in the trials.

Moderna’s mRNA1273

This is another one that’s progressing rapidly in the clinic, and if you’re keeping score, is the most advanced vaccine candidate from a US company.  Moderna’s expertise is in messenger RNA-based therapies, and this one is indeed an mRNA vaccine, developed in collaboration with the NIH. The hope is that this engineered RNA will enter cells and make them produce coronavirus spike proteins, which will then set off an immune response. As mentioned in the background post, this is a relatively new vaccine technology, and no vaccines have been approved yet using it. It has the advantage of being fast, though, which is why this candidate is in the position it is.

Volunteers have already been given a low dose of the vaccine in a 45-patient Phase I trial in Seattle, and a larger one is enrolling at Emory, dosing 25, 100, and 250 micrograms of the mRNA in various age groups. That will set up the dosing protocols for the first Phase II trials, which Moderna’s management has been saying could begin in the spring. Of course, “spring” is a flexible concept! This is a big bet on a new technology – the company has set up to receive as much as $483 million from HHS’s BARDA to ramp up clinical work and manufacturing in an effort to not miss a beat should the vaccine show promising data.

Update, 4/28: five days after starting the second Phase I doses, Moderna filed for an IND to go on into the Phase II trial mentioned above. It will have 600 patients, divided in to 50 microgram, 250 microgram, and placebo groups. As soon as the safety signals read out from the current dosing, they’ll take off, starting in May. And they’re already planning for a rapid Phase III later in the year.

Update, 5/1: Moderna has just signed a worldwide development agreement with Lonza, which could allow for scaleup of a 50-microgram vaccine dose to 1 billion doses per year. Moderna’s own production facility in Massachusetts is already heading to a 24-hour production schedule, using that BARDA funding mentioned above. 

BioNTech and Pfizer

More mRNA candidates are moving along briskly as well. BioNTech has a deal with Fudan to work on such coronavirus vaccines in China, and they signed up last month with Pfizer for the rest of the world. (The companies had already been working on an mRNA influenza vaccine). Word has just come that the companies have received clearance from German regulatory authorities to start a Phase I/II trial. They’re spreading out the risk by adding to the work, taking four different candidates into the clinic more or less simultaneously.

They’re varying both the payload and the method of delivering it. Two of the candidates use mRNAs with naturally occurring (but less common) modified nucleoside bases in them (presumably things like pseudouridine), a trick that’s been tried over the years to increase stability and to cut down on the problem of developing antibodies to the mRNA vaccine itself (rather than to the protein it eventually produces!) The third has another modification, uridine-containing mRNA (presumably an extra tail of U residues?), which has been shown in some cases to increase the immune response to the protein product. And the fourth is a so-called “self-amplifying” mRNA, which has a sequence for a replicase enzyme in it as well. When this gets translated into protein, the replicase goes to work making more copies of the mRNA, including some double-stranded species that prime the immune system even more. As for the payload, two of these have the Spike protein (a popular choice, and for good reason), while the other two have just the receptor-binding domain from the spike (which came up in a recent post on coronavirus mutations here as well).

The trial will be dose-escalation design (1 to 100 micrograms), doing the usual range-finding for the later trials in up to 200 volunteers. They’re also going to look at the effect of repeat vaccinations and will try some cohorts of higher-risk patients as well. This is an ambitious program indeed.

Update, 4/28: Pfizer has said that they are going into human dosing next week, and say that they could be ready for emergency use “in the fall”, which is getting pretty lively even by the current standards (!)

Update, 5/1: the first cohort of 12 patients has been dosed by BioNTech in Germany, and Pfizer says that they are expected approval any day now to start similar trials in the US and move on to dose-escalation, etc. The first three candidates mentioned above will be dosed twice in patients, while the self-amplifying one will be a single dose.

Sinovac’s PiCoVacc

Meanwhile, back in China, Sinovac has received approval for human testing of an inactivated-virus vaccine (see that background post for more on these). You may recall that these sorts of vaccines often need an adjuvant to boost the immune response, since they can be less like a real infection as far as the body is concerned, and Sinovac has just recently partnered with US-based Dynavax. They have an adjuvant that they’ve used in their own hepatitis B vaccine, and they’re bringing that in for Sinovac.

Sinovac themselves made news this week with a preprint that shows evidence that their vaccine produced neutralizing antibodies in mice, rats, and rhesus monkeys. The latter animals were significantly protected against challenge with the coronavirus itself, which successfully infected the control animals in what is a first report of a possible animal model in primates. Moreover, these antibodies appear to be effective against ten different mutational forms of the virus, which is good news in light of recent news about variant strains.

Perhaps the biggest news is that the company saw no evidence of antibody-dependent enhancement (ADE). This is an extremely annoying effect in which some of antibodies raised by a potential vaccine can actually be beneficial to the virus upon infection, helping it to enter cells such as monocytes and lymphocytes. ADE can be hard to get a handle on; it can depend on the antigen, the antibody titer induced in the patient, what the next viral pathogen is, etc. Indeed, it’s possible that antibodies to other coronaviruses might be helping the current one along in some people due to this effect. This was a major problem with attempted vaccination against dengue – that virus comes in several closely related varieties, and it turned out that immunization against one could make subsequent infection with another one even worse. It’s exactly what you don’t want from a vaccine, and the only way to know if it’s happening is to try it and see.

ADE was seen in some SARS vaccine attempts, unfortunately, where it’s worth noting that those blood cell lines just mentioned don’t even have the now-famous ACE2 protein on their surfaces at all (the virus enters through another pathway, perhaps complement receptors). There is a report of an inactivated-virus SARS vaccine that did not show ADE, though, and in the animal studies mentioned up in the section on the Oxford MERS vaccine, they didn’t see this effect, either. And now it appears that this new nCov-19 vaccine doesn’t have obvious ADE, which is good news.

Update, 4/28: the company says that it is building a new production facility (like so many others in this space, on a risk basis).

The Wuhan Institute For Biological Products

This vaccine candidate was given the go-ahead for human trials at the same time as the SinoVac one, but it’s not easy to find any information about it. All I know at the moment is that it’s another inactivated-virus one, so it will be interesting to see what differences might show up between it and the SinoVac effort. I have been unable to find out more about the size of the trials, etc., but if anyone has information, I’ll be glad to update.

Inovio’s INO-4800

The Inovio candidate is a DNA vaccine, the only one I’m mentioning today. That’s a broadly similar idea to the mRNA vaccine, in that you’re coming into the patient’s cells with genetic material and trying to get to them to make the antigen proteins for you. The company has been working on this platform for several years, and like several others their earlier efforts on MERS and/or SARS have jump-started their efforts on this new coronavirus. They started dosing 40 volunteers here in the US earlier in April, moving from younger, healthier participants now to older ones, and they’re moving into similar trials in South Korea. The regulatory authorities there have set up a number of fast-track procedures for accelerated safety and toxicology approvals in cases like this, where the general vaccine platform has been into human patients before.

One of their challenges is that they’re also developing a new delivery device to administer the vaccine through the skin. They have some Gates Foundation money for that, but if they’re going to get this working on a large scale there will, you’d figure, be some significant manufacturing challenges in producing both a new vaccine and a new handheld device to dose it.

Update, 4/28: enrollment has completed in the Phase I trial, with two doses to be spaced four weeks apart and readout expected in June. The company says that they are planning for a Phase II/III trial to start sometime this summer, but there are no more details on that yet.

Update, 5/1: Inovio has signed an agreement with Richter-Helm BioLogics for scaleup and production of their candidate. This is not particularly surprising; they’re already the ones producing the company’s investigational HPV vaccine (which is also a DNA-based one). It should be noted that the short-selling investors at Citron Research have come out with a very negative view of the company and its history with new infectious diseases.

Johnson & Johnson (Janssen)

Now we get to the candidates that (as yet) do not have human trials set. J&J has another adenovirus platform of their own, Ad26 (see the CanSino and Oxford entries above for more on this technique), and they’ve been working on a number of vaccines with that same vector. They have signed a huge deal with HHS/BARDA to help develop this one, and what we know so far is that they have run a whole series of gene constructs through the adenovirus and selected the most immunogenic candidate for dosing in primates. The company has some pretty serious manufacturing capability, and is ready to partner with others to get up to the number of doses needed – they’re targeting a billion, on a not-for-profit basis.

Their Phase I trial is not going to start until September – the company explains that with the adenovirus vector that you need to get the correct “seeds” to grow more viral vector. The stuff makes itself, but you have to be very sure that you’ve picked the one that you really want, and that it’s stable enough to keep giving you the same material over a big manufacturing run. (This is interesting to contrast with the full-speed-ahead approach taken by CanSino and Oxford – one would like to know the differences between these adenovirus platforms and who’s taking on the most risk with their programs)

One big difference could be that the Ad5 vector being used by CanSino has run into some problems with immunity showing up to the vector itself. (Update: I should clarify that this is because that variety of adenovirus is one that many people are already exposed to, without any particular disease being associated with it). You obviously don’t want that (see the BioNTech/Pfizer entry above for similar worries). J&J has reported that their Ad26 Ebola vaccine candidate avoids this problem, so it’s a potential advantage. (Update: as per that last note, this adenovirus platform and the Oxford one have the advantage of humans not being as exposed to them already).

Update, 5/1: J&J has now partnered with another large biologics manufacturer, Catalent, to assist in  scaleup of their candidate (sterile formulation and vial production, especially). Catalent has announced that they are hiring hundreds more employees and planning to move to a 24/7 schedule as things progress.


And here’s another Big Pharma entry. Sanofi has a vaccine platform that uses insect cells to turn out recombinant antigen proteins, and this has already been used for an approved flu vaccine. They’re bringing this technology and (as mentioned in the earlier vaccine roundup post) combining it with GSK’s adjuvant (as used in their own shingles vaccine). Such adjuvants (immune-response boosting agents) are important when you’re vaccinating with specific proteins like this, because otherwise the antibodies might not reach useful levels. They’re also working with BARDA, and the companies have stated that they plan to move into human testing in the latter part of this year (similar to J&J).

Sanofi’s antigen is a Spike protein as well (the logical choice, as you can see from so many people using it), and their protein production platform is said to be able to deliver hundreds of millions of doses. Manufacturing capability is already being expanded on an at-risk basis, and “at-risk basis” is pretty much the slogan for the vaccine effort across the whole biopharma industry.

Update, 4/28: here’s Sanofi’s CEO talking about the manufacturing challenge. Note that he basically assumes that we’re going to end up with some sort of vaccine (he’s likely correct!), and that we need to be getting ready to produce enough of it. He’s saying that the GSK/Sanofi effort is “the best shot Europe has”.

Update, 5/1: GSK’s CEO has told investors that she doesn’t see large-scale availability (hundreds of millions of doses) of any vaccine candidate until the middle of next year, due to manufacturing constraints. As you can see from the latest round of updates here, manufacturing is (as anyone in the field would have predicted) a crucial step for everyone in this story.


So by my count, the biggest and most advanced programs include two inactivated virus vaccines, three different adenovirus vector vaccines, two mRNA possibilities, a DNA vaccine, and a recombinant protein. That’s a pretty good spread of mechanisms, and there are of course plenty more coming up right behind these. You cannot do the tiniest search for such information without being inundated with press releases about companies working on coronavirus vaccines – not complaining here – and moving on to smaller companies would make this post multiple times longer. I’ll update as more news comes out and add in more companies and candidates.

80 comments on “A Close Look at the Frontrunning Coronavirus Vaccines As of May 1 (updated)”

  1. NJBiologist says:

    “The third has another modification, uridine-containing mRNA (presumably an extra tail of U residues?)…”

    Ah, so that’s the 3′-uridine-translated region, then?

    … I’ll show myself out.

  2. Vladimir says:

    HI Derek,
    Very interesting overview.
    Question: As far as I know, likelihood of approval for each vaccine is very low (~10%). What if all leading candidates fail?

    1. kismet says:

      Then we’ll probably have to win the hard way like South Korea did. Sustaining an R0/Reff <1 for months on end and developing a contact tracing and screening regime to catch reintroduction via international travel. Effective drugs could give a very small but relevant push here. Once an effective system is in place it might be possible to control small outbreaks before they grow, at least for developed nations willing to do so (sadly not clear if the USA is willing).

    2. Shazza says:

      Actually in an analysis published in Biostatistics in April 2019, Dr. Andrew Lo, et al. studied the probability of clinical trial success rates and related parameters of more than 185,000 clinical trials involving 21,000 compounds. This robust analysis compared infectious disease vaccines to other therapeutic areas and found that infectious disease vaccines have an overall 33% probability of success that increases to a 42% probability of success once a Phase 2 study is launched, the highest of all therapeutic areas.

      1. That’s in line with the earlier development risk evaluation of Pronker et al (2013), making for a probability of market approval of around 33%. However, that’s obviously not factoring in risk associated with novel technologies- one analyst has assigned only a 5% chance of success for Moderna’s mRNA candidate, nor any benefit from less rigorous regulatory processes.

      2. Hi,
        I am not sure that statistics will work here. Probability of success is 100%-risk. Statistics may not work due to lack of relevant data. Here is our paper where we simulated portfolio of COVID vaccines
        Only rigorous risk analysis could help to derive “right” POS

        1. Correction. Probability of success = 100% (minus) risk

  3. James Donovan says:

    we were able to define epitopes from the first SARS that elicited ADE effects –
    Do we have equivalent info for SARS II? If not aren’t we just flying blind in vaccine production?

    1. Toni says:

      is it actually known how ADE really works?
      One always reads that it is Fc-mediated uptake via the corresponding receptors, possibly also via complement. I wonder whether this view is not too simplistic, because neutralising Abs should also have to be uptaken. Or is it in the nature of the binding between multivalent antigen epitopes and the Abs. For example, in the way that a kind of opsonization is more likely to occur here?

      1. DTX says:

        Toni asked about how ADE worked and I previously asked about which studies showed that adverse outcomes with Covid-19 were due to an overactive immune system. An excellent review in today’s Science (Wadman et al.) suggests its still unclear how SARS Cov-2 damages the body.

        It notes that whether a cytokine storm actually happens is unknown. It states: “some clinicians SUSPECT the driving force in many gravely ill patients’ downhill trajectories is a disastrous overreaction of the immune system” (emphasis added).

        It also notes that other aren’t convinced : “there seems to have been a quick move to associate COVID-19 with these hyperinflammatory states. I haven’t really seen convincing data that that is the case, says Joseph Levitt, a pulmonary critical care physician at the Stanford University School of Medicine.” He further notes that dampening a cytokine response may suppress the body’s ability to fight off the virus.

        Hence, we still don’t know if concerns about vaccines inducing a subsequent over response to the virus are valid. (or whether using drugs to suppress the immune system is a good idea).

        Thanks for the excellent blog & insightful comments.

  4. TPO says:

    If GSK is relying on the same adjuvant as in their shingles vaccine, then there’s trouble ahead. The adjuvant in that vaccine contains multiple components and both are limiting. the vaccine, which is given mostly to people over 50, is back ordered. I’ve been waiting months.

  5. dwh says:

    Derek – Just a minor point of clarification regarding the adenoviral vector directed immune responses. The rationale behind the use of the Oxford chimp Adenovirus and J&J Ad26 vector is not that they don’t elicit immune responses against the vector (they do), but that there is low pre-existing immune responses directed against these serotypes in the human population. In contrast, the Ad5 candidates are based on a serotype which commonly effects humans. Seroprevalence of Ad5 neutralizing antibodies depends on geographic location, but is quite common (~60% in U.S. and Europe, > 90% in certain regions of Asia and Africa). The concern is that these pre-existing immune responses will inhibit the ability of the vaccine to infect the recipients cells and express the target SARS-CoV-2 antigen(s). While there might be some ability to overcome these pre-existing responses with higher doses, you are left fighting a bit of an uphill battle. The concerns with the other Adenoviral serotypes is that the vaccine will induce anti-vector immunity which can limit the ability to re-dose. This becomes important if a single immunization is not sufficient to elicit high frequency of the target immune response. In that eventuality, an obvious strategy would be to integrate two of the vaccine candidates under development into a heterologous prime boost immunization regimen. If the results from initial immunogenicity studies of any of these viral vector candidates looks like a single dose will not be adequate to achieve target levels of immunity, I would expect them to start looking for partners to team up with. For example, if the Oxford and J&J candidates could be given in succession. This would have the benefit of avoiding the vector directed immune responses and probably provide broader anti-S response (assuming there is some variation in the spike sequences incorporated in the vaccine).

    1. Derek Lowe says:

      Good point! I’ll clarify that in the post.

  6. JeffC says:

    I remain skeptical or all of this (and I get little joy out of this). I don’t really care too much about the platform, what I really care about is what these vaccines are targeting. And that’s where I get really really twitchy. From what I can see they all (?) target the spike protein. That may very well be a terrible target. Such fusion proteins are also found in RSV. The structure and the conformational changes you get are really really cool and we’re still figuring out what happens. But if RSV is anything to go by, the fusion protein is a crap target. While paluvizimab made it to market it’s efficacy has always been considered average and it only works prophylactically. So surely we could develop a newer, better mab or even better, a vaccine? That has turned into a story of epic Phase 3 failure. Medimmune and Regeneron have tanked in P3, with Medi giving it one final go. Novavax have failed in P3 multiple times with a vaccine that appears, at least, to generate an antibody response that should have worked. No dice.

    Why paluvizimab works has become something of a mystery based on the “better” antibodies tanking.

    I think the main issue is the the immune does not really see the lung surface as well as other tissues. The lung surface is really outside the body. So all those nice viruses get an easier time of evading the immune system. A systemic small molecule bathing the cell in compound has got a much better shot of getting int he way of the virus.

    We’ve never been able to get much of a handle on what level of fusion antibodies gives protection in RSV, if at all, and I can’t see SARS being different.

    A vaccine that goes after a different viral target would be much better. It’s not been clear to me if those are in development in the current batch.

    1. Barry says:

      As far as I can tell, a coronavirus whose Spike protein doesn’t bind productively to ACE2 is not virulent in humans. So either:
      – it presents an epitope to which we can develop an effective vaccine
      -it’s non-virulent
      -by cataclysmic bad luck the epitope of the RBD/ACE2 docking surface is identical to some epitope that the host (we) express ourselves, and it is therefore not immunogenic. But that last scenario has not been shown to apply

      1. Some idiot says:

        This is probably a stupid/naive question (I am a process chemist, not an immunologist! 🙂 ), but…

        Putting together JeffC’s and Barry’s comments, would I be correct in saying that:
        (a) the interior surface of the lung does not “display” (or whatever it is called) an immune response, therefore a vaccine would not stop the virus entering through lung tissue, but also
        (b) the antibodies from a vaccine to (eg) spike protein would reduce the spread of virus from the lungs to the rest of the body?

        If these lines of logic are sort of reasonable, then, given that (apart from cardiovascular problems) given that the main problem with COVID19 is that it seriously messes with the lungs, would that not mean that a vaccine would not actually really make much of a difference for many patients?

        Now, I am quite sure I have stuffed up something here logically, probably due to misunderstanding or lack of background understanding, but I would love to have it cleared up!


        Thanks in advance!

        1. Toni says:

          To point A : „the interior surface of the lung does not “display” (or whatever it is called) an immune response“

          The lung with its massive surface area is protected from outside pathogens by a robust and highly controlled immune system and we find there all the immune cells neccessary for an effective immune response.

          1. JeffC says:

            Toni, perhaps I should have been a little more elegant in my comment

            There IS an immune system (we see that after all in the inflammatory response). It’s just that as is said below, it’s different to the more systemic immune response. And viruses have adapted to exploit those differences.

            My main point though, is I think the fusion protein is a poor target for a respiratory vaccine. I think there’s enough evidence from other viruses to support that view.

        2. Toni says:

          Antiviral B cell and T cell immunity in the lungs:

          An excellent article to read, also with regard to the pitfalls in vaccine development for lung infections

          1. Some idiot says:

            Many thanks! (-:

    2. johnnyboy says:

      I think JeffC does raise a good point in terms of the lung immune landscape. All these vaccine approaches work by systemic injection, and therefore will raise a circulating IgM/IgG response (antibodies free in the blood or bound to memory Tcells in circulation and lymphoid organs). Indeed circulating IgG is a major efficacy endpoint in these trials. Circulating IgG may help prevent disease if the coronavirus has a systemic incubation step that is significant in the early pathophysiology of the disease. However to be truly protective, you would need to raise an IgA response in the respiratory tree, where secreted IgA present on the surface of the upper and lower respiratory tract would get to the virus before it replicates significantly in the respiratory epithelium. Perhaps circulating IgG would be sufficient to prevent the more serious systemic phase of the disease (the cytokine release), so people would still get the respiratory infection but it wouldn’t have the serious ARDS complication it has now in a minority of patients. Or perhaps the presence of circulating antibodies could actually enhance the cytokine release effects, via ADCC or CDCC. It’s all a bit of a crapshoot frankly. I wouldn’t want to be a trial patient that gets challenged with the infection…

      1. Some idiot says:

        Thanks! 🙂

      2. lauri bretthauer says:

        I’m not a scientist but a writer of sorts and have been following a lot of the releases of information on different levels. I am curious to read these posts about vaccine development which look at the logic of the ‘immune’ system approach in creating a ‘boosted’ atmosphere of ‘warrior intensity’ (via the immune response as a means of combating the assault of the virus). I wonder if anyone has considered that whilst trials are underway to combat the virus based on past SARS approaches/efficacies, has anyone studied the evidence and current “why” of some individuals having extreme ‘immunity’ to the virus once infected, and others succumbing to mortality? I’m curious why a study would not be underway to look at the differences of these individuals’ bloodwork or immune responses – one example I can think of is a man who was, like, a hundred years old and survived it! I mean, wouldn’t they want to analyze his tissue and his blood??? The virus is always discussed as if it exists on a continuous trajectory of assault; yet, the evidence before us reveals that individuals with ‘pre-existing’ conditions are susceptible. Isn’t there some usefulness in studying the differences within this landscape first? Yet, i never see anywhere a discussion of, or trials involving, learning from this. Is this a dumb thing to point out? It is something I am always wondering in a cart-before-the-horse way. To leap to an overgeneralized assumption that “Oh, well these folks all have compromised immune systems” seems way too simple. Also, I really don’t find any investigations of information in studies that have presented evidence of natural substances resisting viruses. Why wouldn’t scientists want to ask: what is it about this substance that repels/discourages growth of this virus? The immune/vaccine approaches seem to arrive with a lot of unpredictability.

        1. Frank Schaper says:

          Lauri, to give a very short reply: we have absolutely zero ideas why some people, including smokers, live to 100, while others die early. If we do not even have a handle on this, searching the biochemical reason for some outlier survivors would be like searching the proverbial needle in a haystack, blind, with gloves and I am not telling you which land the haystack is in.

          1. MATTHEW CREEDICAN says:

            Lauri, Frank,

            It is likely not worth the time looking at individuals, but this may be impetus to open/release anonymized data on Covid related cases in order to use big data and machine learning to look for correlations. We have plenty of sensing devices that can be correlated in time with one another and given enough patients/data points we can probably start to find useful correlations such as early symptoms, related to changes in resting heart rate or blood pressure, who knows. All of this being valuable to a integrated response, especially if we don’t get a magic bullet in a vaccine or drug. It would be really nice to know who is actually at a high and low risk of death.

        2. x says:

          You can know everything there is to know about chemistry, physics, even human psychology – but none of that will teach you the rules of poker.

          We have a pretty good idea (for some definition of “pretty good”, and this is certainly debatable) how the human body works in a general sense. But much like trying to intuit poker, that doesn’t always mean we know what’s actually going on in there. There are numerous genes, proteins, chunks of RNA, etc. that we don’t exactly know what they do or how. And the only way to find out is with good guesswork and tedious experimentation: knocking out a gene here, adding RNA there, blocking up a receptor, and then trying to figure out what, if anything, has actually changed. If we guessed right, we learn one more thing. If we guessed wrong, then maybe we learn something else – or we learn something WRONG – or we’re just left six figures poorer and scratching our heads.

          There are LOTS of ways to see what’s in blood, but that doesn’t make it an open book.

  7. Rich Stern says:

    Thanks for this Derek. Nice summary. Do you now if there is any work being done on corona virus-like particles? These are the type of particles used Gardisil (HPV vaccine). I think they are relatively easy to make, do not replicate (since they lack nucleic acids), and give the benefit of having “normal” looking structural proteins ( and maybe more than one). There was some work done with these for MERS and the data seemed promising.

    1. dwh says:

      Medicago is a Canadian biotech with a tobacco plant bioproduction platform. They got a substantial investment from DARPA to build a production facility in North Carolina and have worked previously in influenza vaccines. I believe they have the most advanced SARS-CoV-2 VLP candidate (Phase I human testing projected for July 2020).

    2. Derek Lowe says:

      I’m sure that someone is looking at VLPs, but I haven’t seen any examples yet!

      1. Colintd says:

        There was some work from various teams, but they have all swapped to looking at IV disinfectants…

        1. johnnyboy says:

          But I bet they won’t test the IV disinfectants correctly and so will come to the wrong conclusion, just like with hydroxychloroquine. Everyone knows you’ve got to add the zinc to the disinfectant before injecting.

          1. loupgarous says:

            The British anticipated all this back during their colonial adventure – the Gin and Tonic very elegantly combines a disinfectant (ethanol in an astringent infusion of juniper and other botanicals) with quinine, the model of a series of versatile molecules that are ionophores, antivirals, hypoglycemic agents – whatever they need to be this week.

      2. There’s a few in the works: GeoVax/BravoVax (VLP is encoded by a non-replicating viral vector); Osaka U/BIKEN (protein subunit VLP); Saiba GmbH (protein subunit VLP?).

        Probably another couple more banged out by little biotechs/academic groups by the time I post this….

  8. Wilhelm Cody says:

    How do the Chinese trials find enough patients since there is such a low level of new cases reported now? new cases chart shows 20 to 30 new cases per day.

    1. eub says:

      I’m picking up what you’re trying to put down, but back up and read the post. These are vaccine trials, they don’t need or want cases of the disease in order to do the human safety testing. Also: the animal testing is done in animals. Human efficacy trials run in China might have have trouble, but that’s a long way off.

      1. Wilhelm Cody says:

        Yes, the Phase I and Phase II trials are safety and for Phase II, dosing trials. Both trials will attempt to show efficacy in producing neutralizing antibodies. It is usual for Phase II trials also to look at some indication of preventing infection in the vaccinated however. As Lowe noted, “CanSino Bio’s Ad5-nCoV… has completed Phase I studies and the company has apparently started enrolling patients for a Phase II trial, making them the first to do that, to my knowledge….” “…Phase II … the company is planning to enroll 500 patients in Wuhan…”.
        If the infection rate is very low, then they may have two choices to proceed into Phase III.
        1. Show that more than enough neutralizing antibodies are produced that the subjects would have been protected, which might be CanSino’s approach, or
        2. Use trials where volunteers are purposely infected to show vaccinated subjects have some resistance, an approach unlikely with this disease to meet current guidelines for such studies..
        Phase III trials must show efficacy as well as safety. Efficacy must include the ability to protect against the agent in those receiving the virus when compared to those not receiving it. There must be unvaccinated arms in such trials. Only in extraordinary circumstances might approval be given solely on the basis of safety and demonstrated production of antibodies shown in laboratories to be neutralizing. If cases are this low in China now, how likely will there be enough cases several months down the road to initiate Phase III trials?
        To repeat, given the current rate of infection in China, it is unlikely that they will have enough people likely to be infected to do an effective Phase III and maybe even Phase II trials unless there is a recrudescence in China.
        Having enough infected patients in the control arms will eventually present an issue for the other Chinese groups just entering Phase I trials: SinoVac and Wuhan Institute For Biological Products

        1. James Millar says:

          Check out Retraction Watch’s recent coverage of some concern over organ sources. I suspect they will manage to find participants for ph.iii.

          1. loupgarous says:

            If not in Wuhan, China has a million Uyghurs in a concentration re-education camp in Xinjiang who I’m sure could be convinced to let SinoVac or any other Chinese firm inject them with anything.

    2. eyesoars says:

      For vaccine trials, you don’t. You want naive individuals, and perhaps some who’ve been exposed to ensure that there aren’t any safety issues w/ primed immune systems.

      1. Wilhelm Cody says:

        Yes, you want to vaccinate virus-naive individuals. My statement did not parse properly.
        However, you want enough of un-vaccinated individuals to get infected during the trial to show a difference between vaccinated and un-vaccinated (or placebo vaccinated). If the infection rate is 1 per 2 million per month, (20 individuals per day in a country of 1.2 billion) accumulating enough infected individuals in the control arm during the trial, after vaccination, may take a long time to demonstrate, even if you vaccinate hundreds of thousands in Phase III. This was an issue in developing vaccines during the first rounds Ebola and SARS outbreaks..

        1. eub says:

          Well, hey, if North Dakota doesn’t want to run trials of a Chinese vaccine, maybe Brazil will.

  9. thebp says:

    As a curious outsider, how fast are these vaccine candidates going through Phase I trails compared to what you’d normally expect? Should we be concerned about the speed of the Phase I trials?

    1. Barry says:

      Phase I clinical trials are for safety. You’re mostly looking for acute responses. Is the immune response to the adjuvant excessive and dangerous? Does the vaccine provoke autoimmune response to some self-protein?
      These are important trials, but need not take a long time. It is Phase II that first asks whether a vaccine evokes a protective immune response. That can’t be rushed.

      1. dwh says:

        @thebp – The primary readouts for a typical Phase I study of a novel vaccine take 6-12 months. The exact duration depends on a number of factors, but is based primarily on your immunization schedule, the number of groups / doses you plan to test, and how fast you are comfortable accruing subjects. It is common with brand new vaccines to include sentinel subjects with each dose level being tested. This is where you restrict enrollment at the new dose level to a small number of subjects to see if they have any reactions before proceeding to fill out the larger cohort. It is also common to have a committee review the safety data from one group before opening enrollment in the next group. My guess is that investigator’s and IRBs are likely to be less cautious with these types of safety checks in the interest of quicker study completion. Bottom line is that there is probably a mild increase in risk to the volunteers in these studies compared to conventional vaccine phase I studies not in a pandemic situation.

        The bigger question is what size of clinical safety database are we comfortable deploying the vaccine for various populations. There will be a lot of pressure to approve the vaccine once there is indication of efficacy, but that could occur even with a relatively low of of subjects in the safety database. If the vaccine had a relatively rare, but serious side effect even in a small % of the population that had not yet been seen in clinical testing, a wide rollout could be disastrous. That is probably the biggest risk we face with the aggressive push for vaccine deployment. To help mitigate this risk, most likely you will see vaccine deployment in select populations to help build the safety database before trying to move to mass vaccination.

  10. Not a Doctor says:

    On seeing the realistic timelines, I am curious – which steps are safe to skip in the easy-mode case of the yearly influenza virus? The background post mentions that these are of the inactivated virus type, so presumably every step of the yearly process is identical except for the particular pile of deactivated proteins tossed into the mix to match the current year’s strain. I further assume that, since those proteins are only different enough to evade pre-existing acquired immunity, we already have a reliable cocktail of denaturing disinfectants to mash the proteins up enough to not be dangerous but not so much that they are unrecognizable targets. Still, I presume there’s some need to check that the different inputs still produce an effective and safe vaccine.

    How would that vaccine pipeline break down if applied to a different set of target proteins? Is it that we don’t know what proteins to target, that we’re not sure proteins would be inactivated, we’re not sure proteins would be recognizable to the immune system after inactivation, the delivery mechanism might not work with a different protein, or all of the above plus more? How similar would a virus need to be to seasonal influenza for this established vaccine to be a useful building block?

    1. dwh says:

      @Not a Doctor
      The short answer is that the seasonal influenza A and influenza B vaccines are regulated primarily as a process rather than as a product. Each company’s process must undergo clinical testing to demonstrate that the antigens produced by that process can be safely administered and are capable of providing protection against infection. This database of clinical information is leveraged thereafter to support production of vaccines for new variants of viruses of known subtypes (e.g., H1N1, H3N2) using the exact same process as was proven with the previous strains. To obtain licensure for the new vaccine, the manufacturer submits a supplement that includes all of the manufacturing and analytical release data on the product, but is not required to include human clinical data. The exception to this is the live attenuated influenza vaccine which is required to undergo a clinical safety test to verify that it has been adequately attenuated.

      For completely new viral subtypes that lie outside the previous seasonal experience, human clinical testing would be required for licensure. An example of this was the H5N1 vaccines developed to address potential outbreak of avian influenza. The exact scope of this testing would depend on the strain and the circumstances (i.e., if it were an emergency).

      Use of the influenza vaccine strategy for another virus would require full human clinical testing to demonstrate safety and efficacy. You could then propose to regulatory authorities to use that same process to generate vaccines against closely related viral strains, but would probably need to generate efficacy data against multiple variants in order to get your process established.

  11. Buzz says:

    While being a few weeks ahead of one company seems to be a big advantage, I don’t see how any of these small biotechs can compete with say, J&J to get a vaccine that the whole world may need.

    Paul Stoffels has committed to manufacturing at risk, making the vaccine on a nonprofit basis, and manufacturing a billion doses!

    Assuming targeting the spike actually works, How does a small Biotech company that doesn’t make any products yet and is hoping to make a profit have any chance to compete with that even if their trials read out first?

  12. Covidiot19 says:

    But Derek, what about injecting with bleach? Or putting light IN the body?

    BBC NEWS: Outcry after Trump suggests injecting disinfectant as treatment

    We shouldn’t take anything off the table, right?


    1. A Nonny Mouse says:

      Strangely, this was suggested to me by a mini-cab driver about 6 weeks ago though I was a bit worse for wear after a pre-Covid funeral and couldn’t be bothered telling him what I thought (much as the entourage do with Trump).

    2. ThirteenthLetter says:

      Hey, “Covidiot19,” why don’t you buzz off and take your political sneering to literally the rest of the internet, where it would be welcome? Adults are discussing actual science here.

  13. Top PI adviser expert at harvard says:

    Trump suggested that UV may help people…he needs to resign. Vitamin D biology is a pseudoscience, heck UV is a pseudoscience!! — top expert (harvard)

    1. Covidiot19 says:

      But he did say:

      ‘Its just a suggestion from a brilliant lab by a very, very smart, perhaps brilliant man…. I’m just here to present talent’

      -Donald J Trump.

  14. The Man with the Glycans says:

    In Sanofi’s platform, they use insect cells, which have already been shown to have a completely different glycosylation pattern than spike proteins expressed in human cells. As far as I know, the only potent fairly well-characterized SARS Ab that cross reacts with SARS-COV2, called S309, binds specifically to a fucosylated complex type glycan, which is not made by insect cells.

  15. Walter L Wilson says:

    Instead of building up immunity why not seek and destroy defective interfering particles or at least build a DIP stick to get a delta.

  16. Derek- the CanSino Ad5-nCov Ebola vaccine was approved in China in 2007. Concept to approval claimed to have taken only three years.

    1. dwh says:

      FYI, the Ad5-EBOV vaccine was approved in China without any human efficacy data. Approval was granted based on a 500 subject Phase II immunogenicity study and a small non-clinical challenge study. Read into that what you will, but I would not advocate for that as a model to be emulated.

      1. Simple statement of fact, not suggesting that past approval validates the platform technology beyond basic safety data and manufacture to clinical trial batch scale.

  17. Barry says:

    I am reminded irresistibly of Typhoid. Yeah, that’s bacteria rather than viral. But asymptomatic carriers were known who never mounted an immune response, remained infected and infectious for decades. Nonetheless, an effective protective vaccine has been developed. I.e. the vaccine evokes a more consistently effective immune response than does the pathogen itself. (although in fairness, I don’t know that the vaccine evokes protective immunity in 100%)

    1. loupgarous says:

      Are any vaccines 100% reliable in conferring immunity to their targeted pathogens? Some vaccinated people will have unreliable immune response to either the adjuvant or the actual vaccine component. I’ve always assumed that a lower degree of protection has been accepted as proof of efficacy for vaccines.

  18. Nikolaus A. Schaefer MSc ETH, MA HSG, FRM says:

    There are at least three vaccine projects under way in Switzerland. All got a lot of coverage on national SRF TV.
    All of them expect to be ready in four to six months.

    1. Derek Lowe says:

      This are all *very* small efforts using unusual technologies, and I see no way that any of them can be ready for any kind of real rollout in that time frame, unfortunately.

  19. Have you heard of any group looking at using a non-replicating Corona-19 as a vaccine? In principle, it would be made the same way a VGT is done–transfection in cell culture, but with a missing gene, RT would make sense. It should be both intrinsically safe and since it both contains and would express the spike protein in the one infection round, should induce an immune response.

    The only roadblock I see is whether or not it will express well into established cell-lines like HEK-293T suspension cells.

  20. Carl Pham says:

    Derek Lowe, I’d be interested in any observations you may have about the subject of this NYT column:

    I realize it’s may not be entirely in your bailiwick, having zip to do with small molecule therapies, but whatever you can say about the mechanisms involved would be of interest.

  21. Bruce A says:

    Implications for Public Health: It seems that all (non-political) public health responses in USA are founded on the premise that we need aggressive measures to contain Covid-19 spread until a vaccine becomes available and we can immunize our way to herd immunity.

    These posts suggest the uncomfortable conclusion that there is a relatively high likelihood that this basic premise is unfounded, that it could be years before a reliable vaccine is available, if ever. If we strip away the vaccine assumption, then it seems that the only way to develop herd immunity is through a “controlled burn” approach; i.e., relying on a regionally based application of community mitigation rules and contact tracing whenever local hospital capacity (beds, staff, equipment and PPE) becomes strained.

    Thoughts from the scientists? (My apologies for straying from science-focused commentary, but this is a critical unknown in the public health community, and if our fundamental assumption is flawed, then we are causing massive, unproductive economic and social damage by inflicting “lock downs” across large swaths of the country.)

    1. loupgarous says:

      You’re totally right. Physicians I’ve spoken to who remember their reading in epidemiology all say we ought to remember John Snow, and focus hard on detecting and isolating specific cases of Covid-19 and their contacts, not the entire population.

      We’re better prepared to aggressively track cases and their contacts than ever before. South Korea’s had better luck than we have, so far, because they have read marked, and inwardly digested the contents of their epidemiology books, and are doing what we ought to have been from day one.

      This isn’t a partisan swipe at the GOP – I doubt anyone on either side of the aisle could spell epidemiology reliably without using spell check. If physicians in Congress such as Rand Paul are pushing epidemiological measures hard, I honestly have seen that reported. That could be the press’s fault – Google only shows lip-smacking “oh, now he’s caught it!” articles about Rand Paul.

      1. loupgarous says:

        Correction: “If physicians in Congress such as Rand Paul are pushing epidemiological measures hard, I honestly haven’t seen that reported.”

        On to making coffee, blessed coffee.

      2. Barry says:

        India is trying to contain the contagion with very little testing by contact tracing and broad quarantine. I think Iceland succeeded at that before tests were widely available; I’m less confident that India will make it work

        1. loupgarous says:

          When I said

          :Physicians I’ve spoken to who remember their reading in epidemiology all say we ought to remember John Snow, and focus hard on detecting and isolating specific cases of Covid-19 and their contacts, not the entire population

          “detecting” implicitly included “testing”.

          Do you know if our national testing kit problem (too few kits to locate the asymptomatic or subclinical cases, too many false positive/false negative test results, uncertain specificity and sensitivity) is closer to being addressed?

        2. loupgarous says:

          Iceland is an incredibly good place to study the molecular-biological basis for any medical issue. Its government consented to study of 100,000 of its inhabitants’ genomes (out of 325,000 Icelanders) and their medical records (which are more complete and available for study than most other places on Earth). I hope people are investigating Icelandic cases of Covid-19 thoroughly as a way of adding to our understanding of how SARS_Cov2 functions in the human body.

        3. DataWatcher says:

          New Zealand pretty much did it that way, too, if I’m not mistaken. But both New Zealand and Iceland are relatively small countries, both geographically and population-wise, and neither has the dreadful social inequality and crushing poverty of India. I think it’s doubtful that any kind of analogy among these countries can be made.

  22. John Wong says:

    After seeing news that a Canadian company will be help do trials (I’m Canadian) and immo suppressed. I wanted to join the upcoming trial in Canada for Ad5-nCov, but then I see it’s a LIVE VACCINE which immo suppressed people can’t take, I find this very strange that CanSino is going this route (most vaccine’s are NOT live because vulnerable people can’t take live vaccines). I have read that live vaccine’s are MORE effective since you are actually getting the REAL weakened virus, which is maybe why they are going this way.

  23. Jens says:

    Today, CureVac (Tuebingen, Germany/Boston MA) announced that the preclinical trials with their RNA vaccine have proven successfull. Clinical studies are to be started soon.

    Worldwide, there are 123 projects regarding the development of vaccines. An updated overview can be found here:

  24. Chuck Choi says:

    Hi Derek,

    The preprint of the Oxford vaccine trial in monkeys is out. Dr William Haseltine wrote a very sceptical review of what this trial reveals, while the press remains positive. Your opinion would be appreciated and informative. Thanks.

    1. Hopeful Layman says:

      May I ask an obvious layman’s question about the efficacy of the TB vaccine? As the writer states here, it does not provide sterilizing immunity, and in fact over a million people died of TB last year. Nonetheless, it’s rare enough in the U.S. (and most of the developed world) where we rarely think about it, and we certainly do not have to “distance,” wear masks, and otherwise modify our daily lives to protect ourselves from it. How has most of the world managed to “control” TB without these other, more draconian measures?

  25. lotoresult23237 says:

    These lotteries are devided into three parts and from their one part which is published on 4pm called as nagaland state lotterysambad result.

  26. Dark Day says:

    I hate to bring something like this up, but given people’s biases, and given how they’ve been exacerbated by Trump’s incessant use of terms like “The China Virus” and “Kung Flu”, isn’t it likely that Sinovac’s vaccine, even if it succeeds remarkably well in Phase III trials, will be met with resistance by a significant number of Americans, thus dooming its chances to successfully move us in the direction of meaningful large-scale immunity?

  27. Hopeful Layman says:

    I’ll add something here that I posted elsewhere — what do people think about this intranasal vaccine being developed? Any estimates on the timeline of getting these into human trials and moving toward final approval? The researchers say they’re planning on initiating primate studies shortly.

  28. Our reusable gown canada designed to create a physical barrier between the mouth and nose, isolation masks help protect against larger respiratory droplets, splashes, or germ particles. However, the face mask does not filter or block small particles in the air that may be transmitted by coughs and sneezes. Avoid touching the outer layer of the disposable mask when removing it, as it could be contaminated.

Comments are closed.