COVID Transmissions for 9-9-2020

AstraZeneca/Oxford trial suspended

Sep 09, 2020

Good morning! It has been 296 days since the first documented human case of COVID-19.

Today I am using the “Talk Back” section a little differently, to share some comments and questions that were left by other readers and reply to them for the whole group, since I think everyone should have the chance to see these comments and the responses I’ve written to them.

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

Keep the newsletter growing by sharing it! 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:

Share COVID Transmissions

Now, let’s talk COVID.

AstraZeneca “suspends” trial

OK. Many of you will have heard that due to an “unexplained illness,” AstraZeneca has suspended vaccinations in the trial of the Oxford vaccine. This is CNN’s story on it: https://www.cnn.com/2020/09/08/health/coronavirus-vaccine-astrazeneca-pause/index.html

Let’s take a beat to understand what this does and doesn’t mean. First, it doesn’t mean the trial is over. Second, it doesn’t mean that vaccine is unsafe.

It does mean that there was an unexpected “safety signal” in the vaccinated population. It does mean that the trial is being paused while the safety data are reviewed. It also means that new patients are not, for the time being, receiving the vaccine—in the interests of caution, the trial administrators wish to review the data before they proceed.

The exact details have not yet been released. People get sick all the time, and there are a huge amount of patients being enrolled in these trials, so we should not jump to conclusions.

Here is what happens in a situation like this in a clinical trial. The relevant documentation for the patient is collected from the study site where the event occurred; this involves all kinds of patient records. Generally, patients in clinical trials have many more records than typical patients because they are being monitored closely. We know more about their overall health than about the typical patient.

Once this documentation is collected, it is typically sent to an independent safety board for review. This group is usually selected before the trial begins and is on hand to review just this sort of situation. They will assess several things—the severity of the event, the nature of the event, and whether the event is treatment-associated. They will second-guess decisions of both the trial investigators as as well as the trial sponsors, and they will come to a conclusion. In many cases, the investigation is straightforward. Keep in mind that safety monitoring in clinical trials captures every event, because every event may be relevant. A car accident can be considered an adverse event in a clinical trial. Why? Because it’s possible the treatment causes fainting, or erratic behavior, or a loss of judgment that could lead to such an accident.

The safety committee for a trial usually looks at the event and assesses whether a treatment effect could have contributed. We don’t know any details of this patient’s situation, so I can’t even begin to speculate as to whether it is a treatment-associated event or whether it is a coincidence.

What I can say is this: if this is not related to the vaccine, great! We can continue to pursue this candidate vaccine. If it is related to the vaccine? That’s proof of why it’s important to conduct large-scale, Phase 3 clinical trials. To make sure that the treatments that make it to market have been studied in depth and at scale so that we can evaluate whether they are safe and effective.

Kidney injury during COVID-19

Something that has struck me—and many of my colleagues who volunteered on the front lines during the peak of the NYC outbreak—is the amount of kidney impairment that was seen in patients with COVID-19.

Now, a study published in the Journal of the American Society of Nephrology has looked at whether Acute Kidney Impairment (AKI), is common in patients hospitalized with COVID-19. The researchers looked at disease in 3993 patients who were in the Mount Sinai Health System in New York, and found that 46% of them had AKI during their COVID-19 infection.

Disease in patients with AKI was more likely to be fatal. Patients without AKI had 8% mortality while patients with AKI had 50% mortality.

Additionally, of those patients with AKI who survived the disease, only 35% of patients had a full restoration of kidney function by time of discharge from the hospital. An additional 36% later recovered full kidney function.

This is clearly a somewhat under-appreciated aspect of the disease. There’s no evidence that SARS-CoV-2 infects the kidneys of healthy adults, and it is not known to circulate in the blood so I’m not even sure how it would get there. But in a lot of diseases, byproducts of the disease can put strain on the kidneys. In Multiple Myeloma, for example, kidney failure is a common complication and cause of death because the disease causes overproduction of proteins that the kidneys must filter out.

Something similar may be going on with COVID-19, and I am sure we will learn more as the disease is studied further.

Study here: https://jasn.asnjournals.org/content/early/2020/09/02/ASN.2020050615

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

Job stuff

I was told today that I will not be impacted by the layoff! This is good news, aside from all the work I still have to do.

Too nice to Pfizer?

In response to the issue last week where I lauded Pfizer’s statement about not releasing a vaccine that they are not confident about, reader EH commented the following:

Is it possible that you're being a bit generous with Pfizer? Perhaps they don't want to open themselves to massive liability if something goes wrong. Some of the smaller, newer companies that are getting these huge grants from the government don't have as much to lose as a powerhouse like Pfizer. $2 billion is a very different number for Pfizer ($11.8 *billion* in revenues for quarter ending June, 2020) than it is for Novavax ($35.4 *million* in quarter ending June, 2020).

This was my response:

There's a substantial liability shield in place for vaccine makers, especially in these circumstances. Approved vaccines that are marketed have injury claims paid out by the government and not the manufacturers, due to a problem of frivolous lawsuits killing the vaccine market in the 1980s. To ensure that companies would stay in the business, the government stepped in to take over liability for approved vaccines. So, they're not really open to any liability if the FDA approves the vaccine and there's some problem.

This liability shield is rather important for vaccine manufacture. Contrary to popular belief among the anti-vaccination movement, vaccines are not a cash cow for the companies that make them. There are several vaccines that do make notable margins, but these are an exception and very much not the rule. Most vaccines are made with thin margins and low budgets, because they are infrequent preventive treatments that do not provide a steady revenue stream. If manufacturers were not shielded from some of the lawsuits that might arise from vaccine adverse events (regardless of the merit of such lawsuits), the resulting legal fees would likely get several vaccines pulled from the market. It just wouldn’t make sense for the companies to make vaccines anymore.

There may be a political conversation to be had about whether this means vaccines should be made by for-profit companies or by a national or nonprofit health infrastructure. That conversation won’t be had in this newsletter.

Testing for antibodies

Reader Robert Berger submitted the following question in response to yesterday’s issue:

New York State's Covid-19 dashboard focuses on tests for Covid-19. Widespread testing for Covid-19 in combination with contact tracing is an effective tool for minimizing the spread of the disease. Given what we know about coronavirus in general (not SARS-CoV-2 in specifics), would widespread testing for antibodies be an effective tool in assessing the risk of opening various segments of the economy?

This is a great question. What we know so far about antibodies in this case of this disease is that they are generated in most people in response to infection and that they are detectable for a period of at least several months after that. There appears to be substantial variation in the decay period of antibody levels, from individual to individual.

We don’t know if antibodies are protective, and we don’t know for certain (though I believe it to be the case) that recovery for infection implies a likelihood that you are protected from subsequent infection.

For this reason, I do not think that seroprevalence studies (the technical term for large-scale antibody surveys) will really give us any meaningful information.

For example, in New York City, we know that about 25% of the total population has evidence of infection with SARS-CoV-2 per antibody testing. We could use this as an assessment of how close we are to herd immunity, but it would be meaningless because we do not actually know if antibodies confer immunity in humans.

Right now the main value of antibody tests is in epidemiology. They give us a sense of how the virus has actually spread through the population, which is important to understand dynamics of the pandemic and calibrate the models that we’ve come to rely on for forecasting and control measures.

In a sense, then, antibody tests do tell us when reopening is safe, but at an abstract remove. They allow us to understand how the virus spreads. We can then calibrate that to prevalence detected by PCR to understand the way in which positive PCR tests correspond to overall population prevalence of disease.

Put in other words, if we know that X PCR tests returned Y positive results at a time when 20-25% of the NYC population was getting infected, then we could potentially estimate what the overall population prevalence might be if we start to see positive PCR test results climb again. This could be informative as to what new disease control measures might or might not be useful if such a situation arises.

So, directly, I don’t see much applicability of antibody tests to decisions like this. However, indirectly, I see uses for them at a population level.

Join in

As you can see, I’m happy to increase the degree to which this newsletter is a dialogue between its readers and myself.

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