It’s cool tech but it’s not magic.
Easy to measure correlates of protection like antibodies wane over time, as expected from any vaccine. T-cells say a lot more about long term protection but are harder to measure. Generally, alarmist antibody research makes the headlines instead. I recommend the TWiV podcast by virologists and immunologists, which notably existed before covid.
Two examples:
1) Israel data shows high level of breakthrough infection. This is the basis for their booster policy:
https://www.reuters.com/world/middle-east/israeli-doctors-fi...
2) Moderna just released their own study showing reinfection rates at 2x at 13+ months relative to <8 months post second dose.
https://investors.modernatx.com/news-releases/news-release-d...
Now, if there is any prior exposure to the disease among participants (this isn't always known, esp. for something like COVID), and if there is any natural immunity that accrues, this will pull that ratio towards 1 (VE=0)
Why that happens is less clear though. Is it a fault of mRNA tech itself? Or the dosing schedule? 21 days is very close together for a 2 dose vaccine, most are 6 months or years apart.
Using 21 days was a good way to get the trials done fast, if they'd used 6 months we'd be where we were 5 months ago, but this was one of the risks. Canada, which spaced doses out further, isn't seeing nearly as high of a rate of breakthrough cases.
I'm not saying we can conclude definitely that dosing is the problem, but it's a question worth investigating before coming to any firm conclusions that mRNA vaccines aren't good enough.
Moderna while still seeming to suffer from this problem seems to be suffering less than Biontech, they are very similar in general, biggest difference is Moderna having already a longer minimum waiting period between the shots and a way bigger dose of mRNA per shot. So this may already be a hint.
Also currently it looks like a third shot is giving a really great boost compared to the second one, at least in the short term. Same with using mRNA as a booster shot after a vector based vaccine.
Probably the worst part about Covid-19 is that Sars-CoV-2 was a totally new Virus with a high attack rate, on top of the danger to the individual patient. If most of the population has seen "something like" it, even very distant variants will not be as bad.
AZ is conceptually similar to the mRNA vaccines actually. They all target the Spike protein.
The Chinese vaccine is an inactivated virus kind of vaccine. That's what you would call traditional. Low protection from infection, but supposedly high protection from dangerous disease onset.