Source: Figure 5 from: Mercado, N.B., Zahn, R., Wegmann, F. et al. Single-shot Ad26 vaccine protects against SARS-CoV-2 in rhesus macaques. Nature (2020). https://doi.org/10.1038/s41586-020-2607-z
I admit to being an unabashed fan of Johnson and Johnson (J and J). That’s an informed opinion, having worked as a technical consultant to numerous Fortune 500 health care companies. From what I could tell, in my limited perspective as a consultant, they were, with one possible exception, the best and the brightest of US health care manufacturers.
And consummate business people. I’m a fan, as I said.
And so, just to prep you for the conclusion here, if J and J thought that Moderna (the officially anointed US vaccine manufacturer) had some sort of amazingly effective and patented approach to producing vaccines, they’d have bought them already. Or at least bid for them, up to what they believed the value the firm was worth.
To me, in this circumstance, the fact that J and J didn’t try to buy them speaks volumes. It tells me that J and J thought they’d do better, relying only on expertise within the J and J family. Not that this is some point-of-pride thing, but merely adhering to a least-cost make-or-buy decision. Just the way that is taught in business textbooks.
Understanding that I’m an economist, and not an expert on the science, let me try to read the tea leaves, on what’s now being reported on the J and J vaccine.
The basic approach for the J and J vaccine sounds the same as every other current candidate. You take a harmless virus, you insert (RNA, DNA — take your pick) that encodes a protein that is specific to the coronavirus. You inject that in somebody and cause an easily-conquered viral infection. In the process of beating that back, the body produces antibodies to the same key protein that sits on the surface of the coronavirus. And so, when you are exposed to the coronavirus, your body is already prepped to kill it.
You know the term GMO, as in GMO food? This is a GMO virus. It’s a genetically modified organism, programmed to put a key protein on the surface of the virus. It’s a mock-coronavirus.
You need a basic idea of what antibodies do, to understand this. Antibodies are shaped so that they “lock onto” a specific protein. Lock, as in lock-and-key. They travel around your body, doing nothing until they happen to bump into that specific protein that they were created to recognize. And then? Then, they lock onto it, and up springs a molecular flag that says to the body, “kill whatever I’m attached to”. And the remainder of your immune system does the rest.
So, the anointed US vaccine (Moderna) works by more-or-less the same principles as the J and J vaccine. That said, a Yugo works by the same principles as an F150. Having the same underlying basic science does not guarantee equality of outcomes.
Even as I have been hawking whatever dribs and drabs of information the US public health bureaucracy will let slip, I have been looking for any indication of how the J and J vaccine is coming along.
And now we’re getting the first little bits of information about it (via the Washington Post). And some of that information clearly must be mis-reported, because it’s just so weird. Or this vaccine is exceptionally effective. Pick one. Nothing else would appear to explain what was said. So let me attempt to read the tea leaves.
To recap: For some unnamed US vaccine (presumably Moderna), based on what has been released so far, it takes two injections four weeks apart, and even with that, it’s a complete dud in 30% of cases. And it’s going to be tested on 30,000 individuals. See Post #815.
The J and J vaccine, by contrast, takes a single injection. And, to a degree, it makes some people sick. The main side-effect, from the Phase I (safety) trials is that it caused fevers in some people, for a couple of days.
Translation: Looks like a much deeper immune system reaction to the J and J vaccine. Instead of causing some soreness at the injection site (localized infection), and having to repeat it in case the body didn’t get the message (two injection protocol), the J and J vaccine is a one-and-done, and in a non-negligible number of cases, invokes a whole-body infection response. That is, fever.
Which is just what you want. I’d gladly trade a couple of days of low-grade fever for immunity to COVID-19.
We don’t know how well it works in humans. But trials on monkeys showed that a single dose of the vaccine worked quite well. Pretty much perfectly, in fact. That’s the graph at the top of the page. It shows how much coronavirus was left, in the nose, two days after exposing monkeys to a large dose of coronavirus. The column of dots on the left is the controls — no vaccine. Lots of virus remains. All the other sets of dots — all with more-or-less zero surviving virus — are monkeys who got variations on the J and J vaccine.
That said, this was monkeys. We don’t know, yet, if we’ll see the same results in humans. And this vaccine potentially faces the same hurdle that the Russian and Chinese vaccines face. It boils down to the use of a common human virus as the “carrier” for the vaccine. Many people may have already encountered that virus and have antibodies for it. Those individuals have immune systems that will attack and (may) kill that virus before it has done its job. Many people (as opposed to monkeys) may be literally immune to the vaccine. It’s too soon to tell.
Now for the mildly odd part.
J and J Phase II/III trials involves 60,000 persons. Or double that of Moderna. If this were some other company, I might think this was simply corporate one-upsmanship. Our trials exceed yours. But meaningless gestures were not the J and J way, in so far as I ever observed it.
And that choice of a larger number makes little immediate sense, given that J and J could afford to make the trial more or less however large they cared to. If they cared to waste shareholder dollars on meaningless display of corporate prowess. Which, as far as I could ever tell, they did not.
So, what’s the point of such a large sample size?
If I had to guess, this reflects a desire to understand effectiveness and potential side effects in greater detail than other manufacturers. This greater sample size provides an opportunity for finer sub-group analysis.
To be clear, I don’t think they upped the sample size as a hedge against low overall effectiveness, due to the square-root rule that governs statistical power. Nobody reading this will care, but the fact is, 60,000 person trial is not all that much better than a 30,000 person trial at determining overall effectiveness. That difference would only matter if the effectiveness of the J and J vaccine were marginal. And I don’t think that’s the case. See below.
If so, that is an excellent risk-limiting business strategy. It maximizes the likelihood that, even if the vaccine is merely average overall, J and J can identify sub-groups for which it provides superior performance. And, because the target population is more-or-less the population of the world, identifying profitable niche markets where this vaccine offers superior performance would be an obvious way to limit downside financial risk from the production of the vaccine.
In other words, I think they were already looking ahead of the game, toward marketing this, under the assumption that it would work well. Which is, IHMO, classic J and J.
What can I say? To me, in my very limited role as a low-level technical advisor on Medicare payment policy issues, this is how I perceive J and J decision-making. Simply the best business people in the business.
And now for the complete weirdness. Emphasis mine:
Anthony S. Fauci, director of the National Institute of Allergy and Infectious Diseases, said that once there were 154 cases of covid-19 in the trial, it would be possible to tell whether the vaccine was effective.
I mean, I get the sentiment of that. Read Post #798. But for a guy who stated the facts completely accurately before, that’s a really odd statement, unless it has been mis-quoted by Washington Post reporting. Because that’s just not true, unless that vaccine works, well, pretty much perfectly.
Note that this was not a round number. Somebody did some sort of detailed calculation to back that up.
Referring back to Post #798, for 15,000 participants, with a background rate of infection of 20/ 100,000 population/ day, you’d reach 157 infections in the placebo group in about eight days. And if that were adequate to determine effectiveness (i.e., you could exclude 50% effective from the 95% confidence interval), that would imply that the vaccine was more-or-less 100% effective. In my example, that’s the only way that a mere 157 case of COVID-19, in the trial, would allow you to conclude that the vaccine effectiveness exceeded the 50% effectiveness threshold set by the FDA for vaccine approval.
So I don’t quite know what to think about this. I believe that we’ve seen public health officials quietly and obtusely signaling true information over the past few days. Presumably, they are doing this to do their best to inform the American public while staying under the notice of the Trump Political Officers (in the Russian sense) assigned to these various bureaucracies.
When I integrate all of that, with what I think I know about J and J, my conclusion is inescapable. Not only am I holding out for the J and J vaccine, I’m going to see if I can get enrolled in the J and J clinical trial. Per Post #814.
That’s my reading of the tea leaves. That’s the vaccine I want.
