Good morning and welcome to COVID Transmissions. It has been 441 days since the first documented human case of COVID-19. Welcome back from the weekend. New vaccine news in today’s newsletter, and some disturbing news out of LA. Also, some news about pregnant women.
I keep hearing that various vaccine manufacturers (Moderna, Novavax) expect to be able to easily tweak their vaccines to focus on new variants. Do you have any information on how that might work in practice? More specifically:
What does the testing and approval process look like for an adjustment like this? Will they have to go through all three testing phases for every new version?
Do you expect the updated versions to be used exclusively as boosters? Would they replace the existing versions entirely? Or would it be feasible to mix them - half the shot designed for original COVID, the other half tailored to a variant, or something like that?
Well, this is a complicated question--and a good one!
I think some vaccines take more easily to being modified than others. The examples you cite are in the "easier" category, because they are relatively simple in construction. The Moderna vaccine is an mRNA vaccine, generated by synthesizing RNA molecules and then mixing them into droplets of fat to produce nanoparticles that can be readily taken up by cells. The Novavax vaccine relies on a protein subunit design, which is a protein construct made in a laboratory setting by growth in cultured cells. To do this, they actually employ a whole other virus. They use insect cells, which they infect with a specially-designed insect virus containing DNA instructions to create the protein subunit they want to manufacture. Once they do that, they have a culture dish full of protein factories--in this case, producing a construct made from the SARS-CoV-2 spike protein.
Both of these are essentially programmable. For the Moderna vaccine, RNA synthesis machines can be programmed to make a different RNA molecule relatively easily. For the Novavax option, it is a little more difficult because bioengineering is at the center of it, but a new protein--or a subtle variation on the old one--can be made by changing the virus DNA introduced to the cell platform. Both processes have relatively quick turnaround for changes, though I do think that the RNA vaccines have an advantage by being manufactured in a chemical, rather than biological, reaction system.
Anyway, that's the "how." You wanted more than that. The regulatory process here is indeed a bit different than if the vaccines were being newly invented. For a product like this, the closest analogue is the seasonal influenza vaccine, which is made new every year. For those vaccines, a global organization of influenza surveillance groups determines what strains should be included in a new vaccine, and then various companies produce candidate vaccines. Limited experiments are conducted to demonstrate that these new formulations are what they say they are and work like established influenza vaccine designs. This simple process only works, though, because we have done robust validation on, and have a lot of experience with, seasonal influenza virus.
In an influenza pandemic, things can be a little different. Simple clinical studies are conducted to demonstrate that (1) the immune response to the new vaccine is generally similar to what we would expect from a protective influenza virus vaccine, and (2) no unexpected safety events are observed. These trials will be designed based on negotiation with the FDA based on the degree of similarity between the new vaccine vs existing, validated vaccines. In 2009, the novel H1N1 influenza virus strain was so similar to known viruses that no clinical trials were required--it was treated just like a seasonal strain change.
For the first alterations to COVID-19 vaccines, I imagine that the FDA will employ a balanced approach. The technologies are simple and have been deeply studied, but they are still new and this is still a new virus. I expect that simple, small-scale clinical trials will be conducted to establish that any new vaccines produce similar immune responses as the old designs and that no new safety signals are observed. But, we will see. Either way, I don't expect that the full three-phase approval process will be required.
Regarding the deployment of any new vaccine designs, my expectation is that they will be used as third booster doses. This is going to depend on what variants circulate most commonly in each region, so it's really hard to answer. It may be that we still do two vaccinations, with the prime being with the "earlier" vaccine version and the boost being with the "new." I don't think that we would need to create mixed-dose versions of the new and old. That type of vaccine is called "multivalent" and typically does require fully, new clinical trials because we have never had a multivalent SARS-CoV-2 vaccine before. The fastest approach to get new variations to market would be to release them as separate doses.
Thanks for the great newsletter! Just a semantic point, but one that I think matters... It's probably generally better to say "pregnant people" as opposed to pregnant women. While all of the people in the study may have identified as women (or maybe not), there are certainly people who get pregnant who don't identify as women. I see no reason this research wouldn't apply to them as well. Gender rules are shifting, and it's exciting!
Thank you for your words. I agree with you that this matters.
Generally speaking, I prefer to say "pregnant patients" or "pregnant subjects" when discussing research findings about pregnancy. Unfortunately, the study I was sharing was conducted exclusively in women and reported all subjects as women, so I chose to report the study itself the way that the authors reported it--I have no further information beyond that regarding the gender identity of the subjects. I wish I did, or that the study authors had at least considered this issue in reporting their findings.
Sadly the medical literature on this issue tends to lag best practices in society.
You'll note, though, that in discussing the general principle of passive transfer, I did not use the word "women"--that was on purpose, because I wholeheartedly agree that "pregnant women" is not a phrase that captures the full spectrum of the human experience. In this case I favored the word "parent" because it required the least acrobatics linguistically, though generally I might have used "patient" or "subject."
I keep hearing that various vaccine manufacturers (Moderna, Novavax) expect to be able to easily tweak their vaccines to focus on new variants. Do you have any information on how that might work in practice? More specifically:
What does the testing and approval process look like for an adjustment like this? Will they have to go through all three testing phases for every new version?
Do you expect the updated versions to be used exclusively as boosters? Would they replace the existing versions entirely? Or would it be feasible to mix them - half the shot designed for original COVID, the other half tailored to a variant, or something like that?
Well, this is a complicated question--and a good one!
I think some vaccines take more easily to being modified than others. The examples you cite are in the "easier" category, because they are relatively simple in construction. The Moderna vaccine is an mRNA vaccine, generated by synthesizing RNA molecules and then mixing them into droplets of fat to produce nanoparticles that can be readily taken up by cells. The Novavax vaccine relies on a protein subunit design, which is a protein construct made in a laboratory setting by growth in cultured cells. To do this, they actually employ a whole other virus. They use insect cells, which they infect with a specially-designed insect virus containing DNA instructions to create the protein subunit they want to manufacture. Once they do that, they have a culture dish full of protein factories--in this case, producing a construct made from the SARS-CoV-2 spike protein.
Both of these are essentially programmable. For the Moderna vaccine, RNA synthesis machines can be programmed to make a different RNA molecule relatively easily. For the Novavax option, it is a little more difficult because bioengineering is at the center of it, but a new protein--or a subtle variation on the old one--can be made by changing the virus DNA introduced to the cell platform. Both processes have relatively quick turnaround for changes, though I do think that the RNA vaccines have an advantage by being manufactured in a chemical, rather than biological, reaction system.
Anyway, that's the "how." You wanted more than that. The regulatory process here is indeed a bit different than if the vaccines were being newly invented. For a product like this, the closest analogue is the seasonal influenza vaccine, which is made new every year. For those vaccines, a global organization of influenza surveillance groups determines what strains should be included in a new vaccine, and then various companies produce candidate vaccines. Limited experiments are conducted to demonstrate that these new formulations are what they say they are and work like established influenza vaccine designs. This simple process only works, though, because we have done robust validation on, and have a lot of experience with, seasonal influenza virus.
In an influenza pandemic, things can be a little different. Simple clinical studies are conducted to demonstrate that (1) the immune response to the new vaccine is generally similar to what we would expect from a protective influenza virus vaccine, and (2) no unexpected safety events are observed. These trials will be designed based on negotiation with the FDA based on the degree of similarity between the new vaccine vs existing, validated vaccines. In 2009, the novel H1N1 influenza virus strain was so similar to known viruses that no clinical trials were required--it was treated just like a seasonal strain change.
For the first alterations to COVID-19 vaccines, I imagine that the FDA will employ a balanced approach. The technologies are simple and have been deeply studied, but they are still new and this is still a new virus. I expect that simple, small-scale clinical trials will be conducted to establish that any new vaccines produce similar immune responses as the old designs and that no new safety signals are observed. But, we will see. Either way, I don't expect that the full three-phase approval process will be required.
Regarding the deployment of any new vaccine designs, my expectation is that they will be used as third booster doses. This is going to depend on what variants circulate most commonly in each region, so it's really hard to answer. It may be that we still do two vaccinations, with the prime being with the "earlier" vaccine version and the boost being with the "new." I don't think that we would need to create mixed-dose versions of the new and old. That type of vaccine is called "multivalent" and typically does require fully, new clinical trials because we have never had a multivalent SARS-CoV-2 vaccine before. The fastest approach to get new variations to market would be to release them as separate doses.
Thanks for the great newsletter! Just a semantic point, but one that I think matters... It's probably generally better to say "pregnant people" as opposed to pregnant women. While all of the people in the study may have identified as women (or maybe not), there are certainly people who get pregnant who don't identify as women. I see no reason this research wouldn't apply to them as well. Gender rules are shifting, and it's exciting!
Thank you for your words. I agree with you that this matters.
Generally speaking, I prefer to say "pregnant patients" or "pregnant subjects" when discussing research findings about pregnancy. Unfortunately, the study I was sharing was conducted exclusively in women and reported all subjects as women, so I chose to report the study itself the way that the authors reported it--I have no further information beyond that regarding the gender identity of the subjects. I wish I did, or that the study authors had at least considered this issue in reporting their findings.
Sadly the medical literature on this issue tends to lag best practices in society.
You'll note, though, that in discussing the general principle of passive transfer, I did not use the word "women"--that was on purpose, because I wholeheartedly agree that "pregnant women" is not a phrase that captures the full spectrum of the human experience. In this case I favored the word "parent" because it required the least acrobatics linguistically, though generally I might have used "patient" or "subject."