You may remember that the UK measured vaccine efficacy for Pfizer against alpha and delta during the time that they were transitioning from the former to the latter. Two shots went from about 93% to 88%, and one shot went from about 50% to 33%. We can map that onto the AZ data for efficacy of Pfizer to find approximate efficacy against delta of different antibody levels.
The puzzle here is that Israel is seeing a much bigger drop than what the UK data would indicate is expected. Israel previously announced that vaccine efficacy against symptomatic infection dropped from above 90% to 64%. That implies that average antibody levels are about 200, and that seems implausible given the data we have so far on antibody decline. As I've written before, it's possible that the method Israel used is missing cases in unvaccinated people, making it look worse than it is.
However, there is a new report out today that for those vaccinated first (mostly over 65) in Israel, efficacy against severe disease is also dropping dramatically. It's harder to miss severe disease, which is a point against the argument that they are missing data.
I don't have any hypotheses to explain the discrepancy, but am definitely paying attention.
Recent news report out of Israel:
https://www.ynetnews.com/health_science/article/bytq34n0u
Report out of the UK :
https://www.medrxiv.org/content/10.1101/2021.05.22.21257658v1.full.pdf