COVID Transmissions for 4-15-2021

New developments in the J&J clotting situation

Apr 15, 2021

Greetings from an undisclosed location in my apartment. Welcome to COVID Transmissions.

It has been 514 days since the first documented human case of COVID-19. I’m afraid I don’t feel like being cute or pithy about this today; I am focused on the situation with the J&J vaccine, and I want to get right to discussing it.

As usual, bolded terms are linked to the running newsletter glossary.

Keep COVID Transmissions growing by sharing it! Share the newsletter, not the virus. I love talking about science and explaining important concepts in human health, but I rely on all of you to grow the audience for this, which you can do by using this button here:

Share COVID Transmissions

Now, let’s talk COVID.

J&J vaccination safety, continued

I spent a good part of yesterday digesting everything that I could about the J&J vaccine clotting situation, and I’m going to expand on it in even greater length here today. The more I look into this situation, the more seriously I take it and the more I think the decision to recommend a pause was the right one. I’d like to explain why.

First, I want to share this link that describes the type of clotting syndrome that is being observed: it is called cerebral venous sinus thrombosis, or CVST. This is a clot in the veins that drain blood from the brain, which eventually leads blood vessels to break and blood to leak into the brain. This is colloquially known as a stroke. Here is a resource that can tell you more about CVST specifically: https://www.urmc.rochester.edu/encyclopedia/content.aspx?contenttypeid=134&contentid=69

Normally, CVST would be treated like most other thrombotic events, with any resulting hemorrhages treated like strokes are usually treated. Clotting syndromes are often treated with the anticoagulant heparin. However, the CVST pattern being seen with the J&J vaccine and the AstraZeneca vaccine is also being seen in combination with a condition called thrombocytopenia. “Thrombocytes,” also known as platelets, are the blood cells that are primarily responsible for clotting. The suffix “-penia,” which means “poverty” or “need” in Greek, means that they are lacking. So thrombocytopenia is when a patient has a meaningfully lower-than-normal level of the cells that are responsible for clotting.

This is pretty weird. We have a rare clotting syndrome that can lead to stroke happening after vaccination in patients who also present with a reduced level of cells that cause blood clots. Normally, you expect patients with thrombocytopenia to have bleeding problems, not clotting problems. You would not give such a patient heparin, because one of the adverse reactions to heparin treatment can be thrombocytopenia, and if the patient is already thrombocytopenic, heparin can make it worse. At some point, platelet levels can fall low enough that the most minor of injuries will bleed forever. That can be deadly.

So, this extremely rare, potentially-vaccine-related clotting condition is also happening alongside a drop in the ability of patients’ blood to clot, making certain anticoagulant therapies potential dangerous. This is strange and dangerous; not what we look for in medicine.

Now, as of yesterday, we had information that 6 in 7 million patients given the J&J vaccine had experienced this adverse event, and that one had died. As of today we have more information, presented by Janssen (the subsidiary of J&J that actually made the vaccine) at a meeting of the American Committee on Immunization Practices (ACIP) today. I watched it so that you don’t have to. But, if you can find a recording, I think it’s worth watching.

Anyway, the Janssen presentation today described 8 cases of this event pattern. Specifically, 2 more cases were identified from the ENSEMBLE clinical trial program for the vaccine. Given the relatively smaller population from that clinical trial, I am immediately more concerned about the risk of this event than I am by the 6 cases observed so far in post-EUA use. If there were 2 cases in the clinical trial, is this rare event more common than 1-in-1 million? Or was the clinical trial just so unlucky that two 1-in-1 million events happened in a trial with only about 22,000 participants that received the vaccine? That’s possible, I suppose.

However, this adverse event pattern is also quite similar to a pattern seen with the AstraZeneca vaccine, which is based on a chimpanzee adenovirus vector technology while the J&J vaccine is based on a human adenovirus vector technology. For the AstraZeneca vaccine, the odds of the CSVT-with-thrombocytopenia event were about 1 in 200,000, as described in this article: https://theconversation.com/is-it-the-adenovirus-vaccine-technology-used-by-astrazeneca-and-johnson-and-johnson-causing-blood-clots-theres-no-evidence-yet-158944

I want to take a moment here to remind everyone that there is not an established causal relationship between this clotting event and vaccination with either vaccine at this point. However, it is starting to look like there could be one. The fact that two vaccines in the same class of design have had this sort of event, but it has not been detected with the mRNA vaccines, suggests that there really could be something to this as a rare but serious vaccine safety event.

I brought up the AstraZeneca vaccine for a different reason than making this point, though. What I wanted to do was establish a potential high estimate for how common this event may end up being with the J&J vaccine. You see, I think that the 6-in-7 million figure that went around yesterday could be an underestimate of risk. ACIP also felt that this might be the case in their meeting today, based on some of the presentations. One particular point that was made was that 52% of the J&J vaccinations administered in the US (about half of that 7 million) were given in the last two weeks, and it is thought that it takes about two weeks for this event cluster to manifest. So, there may be more cases out there have yet to develop and present for care. Furthermore, these cases were detected because they were serious outcomes that were captured in the Vaccine Adverse Event Reporting System (VAERS) database. It is possible that there were intermediate cases, or cases where the vaccination element of the patient history was not connected with the observed events. In other words, there is a good chance of under-reporting because nobody was looking for this and all the cases we know about come from a voluntary reporting database.

At this point, I think we have a wide set of possible values for the risk level of this event. All of these values are less than the population risk of CVST, but combined with thrombocytopenia? I don’t know, that’s still weird to me.

Still, I think we can estimate the risk of this type of event at between 1 in 100,000 and 1 in 1 million. It may be more common than that, or it may be less common. The risk of death from the condition will be lower than that. All of these values are extremely rare, still. It is hard with the small number of cases and limited follow-up time to approach a real estimation of risk. In the US, COVID-19 mortality is about 172 in 100,000 overall considering all deaths over the entire pandemic. We’ll come back to that figure in a moment.

Now, OK, we are not sure this event pattern is caused by any vaccine, but it’s a pretty strange pattern and we now see it with two vaccines that have similar designs. We also don’t know what the real risk level is, but we can estimate it across a wide range. The simple risk-benefit calculation would say that we should compare the risk of death from this event to the risk of death from COVID-19 and act accordingly from there. You could also compare with the risk of dangerous blood clots from COVID-19, which is quite high (presuming that you have actually become sick). If the J&J vaccine were the only possible vaccine option available in the US, at a whole-population level it would appear to me that the benefits outweigh the risks.

However, there is more to it than that. First, we have two other vaccines approved in the US. These other vaccines have manufacturers that have looked at the incidence of similar clotting events in their patient populations and found no such cases. We have an alternative set of options that can mitigate the risk of COVID-19 without having to increase the risk of this concerning event pattern. Previously during the pandemic I would have been highly concerned about the supply of these other vaccines to replace the J&J doses, but today’s ACIP meeting cleared that up for me. According to information at that meeting, 5% of administered vaccine doses in the US were J&J vaccine doses. Currently, the US is administering around 2.5 to 3 million vaccine doses per day, so we are looking at about 150,000 vaccine doses daily being J&J vaccinations. The remainder are Pfizer and Moderna vaccinations, but a big portion of these are second doses. The 7-day moving average for first dose vaccinations is 1.6 million. I am not 100% certain whether the J&J patients are counted as first-dose or what, since only one dose of that vaccine is administered.

Right now in the US, there are at most 53 million people awaiting second doses of an mRNA vaccine, as I understand the data. There are also 53 million distributed doses available that have not yet been administered to patients. These are, presumably, intended to complete the vaccination courses of the patients who are waiting for second doses. Meanwhile, at the ACIP meeting today we were told that Pfizer and Moderna currently deliver 14 million new doses each week. This translates to 2 million potential first doses per day, and presumably manufacturing capacity will continue to expand—though I am not totally certain how much more it will be expanding in the US. As mentioned before, 1.6 million people are receiving first doses each day. Even if the 150,000 daily J&J vaccinations are not counted among those 1.6 million, there is an available excess of mRNA vaccines that can fill the gap.

On the other hand, we do not know how protective the mRNA vaccines are in just a single dose. Estimates vary from 50% to 80% vaccine efficacy, with the latter number being close to the efficacy of the J&J vaccine. I’ll caution that there has been no head to head trial of these products so it’s totally inappropriate to compare their efficacies. Yes, I am going to compare them anyway, because I don’t have anything better to go on. So, replacing all planned J&J doses for the next couple of weeks with mRNA vaccine doses might provide people with the same level of protection they would have had anyway from just one dose, or it might leave that population with protection that is about 30 percentage points lower. Still, that’s a lot better than nothing, and we’re talking only about efficacy against disease here. Serious disease and death seem to be prevented quite well by vaccinations that do not prevent mild disease in all patients. And of course, eventually, any patient who gets an mRNA vaccine will be offered a second dose of that vaccine to make their protection even more complete. If there is any added risk relative to one dose of the J&J vaccination, that relative risk exposure for any disease at all occurs only over a few weeks to a month.

Based on this information, I feel that we can sustain a pause of J&J vaccine administration for a few weeks. I do not believe that all of the 150,000 people per day who would have received the vaccine will go unvaccinated. They will receive some amount of protection from available mRNA vaccines. I do not expect entirely perfect coverage because the J&J vaccine has been used in some populations that are hard to get 2-dose follow up for, but I do expect pretty good coverage because most states that have reported halting J&J vaccinations have reported that they’ve been able to reschedule these appointments to mRNA vaccines pretty easily. Certainly I am convinced the US has the doses to cover the gap.

Now, I mentioned earlier that COVID-19 mortality is about 172 in 100,000 in the US over the last year. This looks a lot higher than our safety event risk estimated earlier, but it doesn’t tell the whole story. This number includes a lot of deaths that happened before we had monoclonal antibody cocktails, dexamethasone treatment protocols, and added hospital capacity. The actual risk of death in COVID-19 cases has fallen substantially and is holding around 1% of cases, for all cases across all ages and demographics.

Here is where I will begin to bring in some demographic information about the patients experiencing this rare event. So far, 7 of the 8 have been female and 1 of the 8 was male. All have been between the ages of 18 and 48. Even early in 2020, before many treatment modalities were available, patients in this age range were seen to have a case-fatality rate from COVID-19 that is substantially less than 1%, ranging from about 0.2% in the youngest bracket in this range to 0.4% in the oldest bracket. These risks will be lower now.

Meanwhile, in the US, there are about 65,000 new cases per day, for a per-day incidence of 20 per 100,000. These are very estimated numbers, folks, but bear with me here. If the risk of death in each case in the age groups we are talking about is less than 1%, then we are talking about a population risk of death that is at most 1 in 500,000 each day, and probably meaningfully lower. Earlier we estimated that the risk of the potentially vaccine-associated clotting pattern is between 1 in 100,000 to 1 in 1 million, but this was not estimated by age of the patient receiving the vaccine. If we did, the range might have tightened a bit, but either way, I am made uncomfortable by the fact that the risk of death from COVID-19 in the US for this age range is starting to look similar to the risk of this dangerous safety event.

All this having been said, this is all estimation and I am being very fuzzy with numbers. I do not want to estimate the risk of death from the potentially vaccine associated clotting event, because I don’t have enough data. I’m not sure what the risk of death for this specific age group is right this moment, so I’m approximating it with a ceiling of 1%. I don’t know the age distribution of administered J&J vaccines. What I’ve done is my best attempt to assess the relative order of magnitude of risks here, and frankly, they are just too close for me to feel good about the situation.

There are other considerations, though. Vaccines against COVID-19 appear to prevent transmission of disease as well, so each vaccination may prevent more than just a potential case in the person getting the vaccine. However, like I mentioned before, I think we have the vaccine supply to make up for the loss of this rarely-given, low-supply option. I don’t think we are comparing the choice of J&J vs no vaccine here. I think we are comparing the choice of J&J vs an mRNA vaccine, which makes the risks of COVID-19 death I described above even lower.

In light of that, I’ve come to two conclusions:

I think the recommended pause was a good call considering the chances that the risk of this event may be as high as the risk with the AstraZeneca vaccine. That pause has been extended by ACIP and will now continue until more data can be collected. Since it takes about 2 weeks for this clotting pattern to manifest, I think we will have more data within a week or 2, and a new recommendation will be forthcoming. We have the doses of other vaccines to cover a gap of a couple of weeks, and may be able to cover a gap that is even longer.
I think that ultimately the J&J vaccine will be resumed with changes to its EUA to include warnings for use in special populations. At the ACIP meeting today, a representative of the FDA said that the agency is considering adding warnings to the EUA based on the limited available data. ACIP seems to want more specific data before they feel comfortable issuing specific restrictions, so over the next couple of weeks I expect the CDC, FDA, and J&J to work together to stabilize risk factors that should limit the administration of this vaccine and also to develop monitoring protocols to identify this clotting pattern if it should appear. They will also be looking into whether this is really being caused by the vaccine. Once the unknowns are dealt with, the label will be modified appropriately and the J&J vaccine will be back on the market. This safety event will become even more rare than it is now, and the vaccination campaign will continue with minimal interruptions.

Overall, I have to say I am impressed with how well the vaccine safety infrastructure worked here. An extremely rare event—seriously, if this happens to you, buy a lottery ticket—was identified by a voluntary database, and a combination of public health experts, government regulators, and the manufacturer of the vaccine put together a coherent and well-considered meeting about that risk within a day of its identification. That allowed an informed recommendation to be made for a short-term approach, with a strategy being developed to settle on a long-term solution. Overall, I am convinced that the short-term approach will have very minimal negative impacts compared to the potential negative impacts of having done nothing.

We don’t have a solution for this issue yet, but I think we are doing what needs to be done to safely get one. On the other hand, all of this applies to the US, where abundant alternative options are currently available. When we know more, I am going to talk about the potential implications of this issue on global vaccination campaigns, which I think is a whole lot trickier to deal with.

What am I doing to cope with the pandemic? This:

Setting personal records

Some weeks ago, I reached the point where I was able to bench press my own body weight. Today, I reached that same point for deadlifting, which is less impressive but still a milestone. A year ago, I couldn’t do either of those activities at any weight, let alone at more than my body weight. I have also reached the point where I can run 4 miles in the time that it would have taken me a year ago to run 3.

I got to these milestones by incrementally failing a little better each day. There were a lot of crappy workouts that felt bad, in there. I had setbacks. But I did something and learned something almost every day, and in that way, on average, moved forward. I didn’t put pressure on myself to always improve; it just happened. And it happened because I was willing to fail at each thing I tried, and was willing to be OK with that as a process towards being able to try failing at something even more ambitious. Embracing failure can lead to surprising success, as it happens.

The changes to my overall health as a result of this are huge. Also, I never would have gotten my kitchen moved on my own without this. There is something to be said for just doing a little work, each day, over months and years. I started a year ago, and I now feel better than I felt, every day, than I did last year. I’m pretty happy with how the work I’ve put in has paid off. Plus, it has given me goals to focus on during this long, seemingly unending quarantine period.

Yesterday, I said I was going to provide comment responses today. I even wrote out some great comment responses, the longest of which was sadly eaten by a glitch on my computer. However, I have given you a very long newsletter today already, and I think it is best if we keep our focus on this developing safety story. When a lull in this developing story arrives, I may do an issue that is primarily concerned with comments and responses. There have been some great comment threads over the past few days, and I think they’ll be worth walking through. The Talk Back section may have been on hold for the past couple of days, but it is not going away.

You might have some questions or comments! Send them in. As several folks have figured out, you can also email me if you have a comment that you don’t want to share with the whole group.

Join the conversation, and what you say will impact what I talk about in the next issue.

Also, let me know any other thoughts you might have about the newsletter. I’d like to make sure you’re getting what you want out of this.

Viral Transmissions

Discussion about this post