> Which means to me that while this vaccine isn't a magic shield, if you do get it, it will keep you out of the hospital and probably make it a very mild experience.

Is that conclusion warranted given the small number of people who got it? Perhaps in another population one of those 5 will indeed have severe, hospital-needed symptoms?

If the immune reaction in the vaccinated was good enough to avoid symptoms but a few still tested positive then those five are surely not the only ones. What are the chances that a random person without symptoms would be getting tested exactly those days that a very mild infection lasts? Those studies don't observe the subjects with daily PCR or something like that (impossible for ten thousands) and most mild infections remain undiscovered. And mild, hard to discover infections aren't exceptionally rare even without a vaccine, as antibody testing studies have shown again and again.

The pressing follow-up question is this: how many of the control group were discovered with (and despite of) equally weak symptoms? The answer could be anything between many more and many less. If it's more (more discovered very mild cases amongst the placebo group) then the vaccine apparently prevents most infections from happening at all, but if it's less then the vaccine doesn't really reduce the number of infected (and infective), it only prevents bad outcomes (which usually remain undiscovered, even in a phase III trial group).

A vaccine that only prevents bad outcomes would still be very valuable, but only to the vaccinated themselves because it would not create a herd immunity effect. This virus is very good at spreading from a mild case, so if the vaccinated still get unnoticeable mild cases they would still serve the virus as stopovers.

Not a doctor, but I thought that asymptomatic is not as big of deal as we thought it was based on contact tracing.

http://www.emro.who.int/health-topics/corona-virus/transmiss....

Contract tracing is more likely to identify sources of infection when the people remember being in the same room with someone who had symptoms. This is also why public transport is virtually absent as a infection source in contact tracing data, but private parties are very prominent.

From what I've heard, virus concentration in the upper respiratory tract (i.e. where the aerosols come from) peaks two or three days before symptoms start.

Here in Germany there's the statistic of "75% of infection sources are not found by tracing" circulating hard through public discussions. I doubt that many of those untraceable infections could be coming from symptomatic carriers, given the level of awareness. Coughing people basically do not exist in the 2020 public. And tracing should still be good enough to recognize most cases when the source became symptomatic after the contact (which would very likely be before the receiver became aware of the infection, triggering tracing). I consider this a pretty strong indication of asymptomatic spread, or very weakly symptomatic (thresholds are very subjective).

I would be surprised if they didn't specifically test all participants in the trial for covid antibodies, symptoms or not. Does anybody know what validation was done?

Depends a lot on the question of whether the antibody test can distinguish the immune system's reaction to the virus from the immune system's reaction to the vaccine (or in this case: to the proteins built from mRNA blueprints in the vaccine). If they can't, and this is quite likely given that they all target the same spike protein, the outcome would hopefully be positive for all of the vaccinated.

In the Pfizer trial, all the subjects in the interim evaluation had symptomatic infection confirmed by PCR test. I'm not sure about Moderna, but it's likely to use the same or similar methods.

> Which means to me that while this vaccine isn't a magic shield, if you do get it, it will keep you out of the hospital and probably make it a very mild experience.

With an n of 5, isn't this a premature conclusion?

If you compare the endpoint "severe illness"... 11 out of ~15000 ended up with severe illness on placebo, and 0 out of ~15000 ended up with severe illness in the treated group.

This corresponds to binomial 95% confidence intervals of (0.0004, 0.0013) and (0.0000, 0.0002). So it seems to prevent severe illness.

Whether it reduces the odds of severe illness IF YOU ARE INFECTED or purely reduces the odds of severe illness by preventing infection--- you're right, there's not nearly enough n to know.

> so the Pfizer vaccine would be an A+ grade, and this Moderna one is a solid "A".

Right now, ease of rolling out a vaccine is more important than its effectiveness, within reason. As long as the vaccine is effective enough to slow the spread of the virus it will save lives -- the more people we vaccinate, the fewer potential spreaders there will be, and that should help us protect people who have not been vaccinated yet.

Today, we would probably be better off with an A grade vaccine that is easier to store and ship across the country than we would be with an A+ vaccine that is more picky about storage temperature.

Of course, we can probably have both vaccines at the same time once the manufacturing capacity spins up.

Can you explain a little bit what the difference between the two T-Cell types are?

The immune system is extremely complicated, but very broadly: CD8+ T-cells (also known as killer T-cells) kill infected cells directly, whereas CD4+ T-cells (also known as helper T-cells) release signals that guide many aspects of immune response, including activating CD8+ cells

They used different assays and at different time points looking for CD8, so we can't say for certain yet that they are different.

It can already be a very mild experience for some people. The question is WHY?

The immune system at its core is basically rolling dice like crazy for new random RNA source code that is compiled into molecule hardware that will hopefully work against the invasor somehow. It even has a mechanism to deposit samples of the invasor were the freshly rolled candidates are let loose. Apparently it's buying into the TDD paradigm. The dice are probably not rolled to try completely random RNA sequences with a D4 for each base, more like mashups of the existing library from previous infections. Different immune systems can come up with different solutions to the same virus and even the same solution can be stumbled upon very early or reached very late, only hours before EOL. The content of the existing library likely plays a decisive role and even more does the general bandwidth of mashup attempts. This apparently shrinks drastically from children to elderly. Maybe an evolutionary tradeoff to compensate for the weak library children have, maybe a side effect of the existing library.

(source: mostly stuff I read linked from the hn page, hopefully without excessive misreading)

This was a pretty interesting explanation, thanks!

Most of the time its because the person was young and fit. Unfortunately the majority of people are not healthy.

Totally anecdotally, the 4 people in my family who got it are all 65+ with multiple underlying issues. It was a bad head cold for a week and half, all recovered. I know, sadly, that isn't the case for so many. I just don't know what to think.

Even in the 80+ population most people are okay. A sample of four is not big enough to see the catastrophic effects even fairly small hospitalization numbers have when millions get infected.

This is exactly my point, and yet my comment gets downvoted.

People seem content with vaccines, meanwhile I want to understand WHY some people fair better than others. There's a far better solution than vaccines to be found, if people are willing to ask the question, and look for it.