Thread by @ToddZywicki

Another early and important paper on NI, vax, and boosters from the canary in the coal mine (Israel). Some novel and original data here on so-called "hybrid immunity" (the first I've seen at least). medrxiv.org/content/10.110…

Baseline finding reaffirms yet again uncontestable conclusion: NI provides superior and far more durable protection against infection than vaccination. Protection of NI at 8-10 months is equivalent to vax at 0-2 months and protection of NI at 12+months to vax at ~3 months.

Figure 3C is booster data. Authors definition of "Booster" is interesting: "Booster: Individuals who received a third (booster) dose 12 or more days previously and had not been infected before the start of the study period." Why 12 days?

Recall that as I noted last week, VE for Boosters is highly NEGATIVE for the first month after boosting (probably focused on the dreaded 2 weeks of doom). By dropping out 12 days, the authors ignore most of the period of negative VE, so overstates value of boosting.

Specifically, "An individual in the
Vaccinated cohort who received the booster dose during the study period, exited that cohort on
the day of the booster dose and entered the Booster cohort 12 days later." If infected during first 12 days, then count as NEITHER vax nor boosted!

Of particular interest, is the bottom 4 cells which report "hybrid" immunity in two separate modalities, NI+vax then vax+NI (breakthrough infection). What we are looking for is whether the two forms of hybrid immunity are similar.

The concern is the possible presence of OAS and that infection post-vax will not provide the full benefit of NI in light of some evidence of lower rate of seroconversion of Nucleocapsid abs and lower N abs levels after breakthrough infections. medrxiv.org/content/10.110…

Differences are slight and probably not statistically significant, findings here suggest that the two directions of hybrid immunity may not be identical. NI+vax appears to be slightly more protective than vax+NI at 4-6 months and difference increases slightly more at 6-8 months.

Oddly, the authors provide different definitionsfor "rec+vax" as "previously infected individuals who later were 7 or more days after receiving a second vaccine dose." Unless I missed it, I can't see why they define "booster" with a 12 day gap and rec+vax with a 7 day gap.

Somewhat more worrisome, the authors find some difference in severe cases for NI+vax (0.5/100k days) compared to 1.1 for vax+NI (more than twice as frequent). NI alone was 0.6 and vax+booster was 0.4. NI, NI+vax, and vax+booster are all therefore pretty much equivalent for severe

It is hard to see the apples-to-apples comparison so here’s a yellow line added to highlight the comparison (as @MedleyRob suggested).

Given hint here that vax+recovery might be less protective than recovery+vax, it will be interesting to see how vax+booster+recovery fares. If OAS is applicable, you'd predict booster would further suppress long-term protection from eventual (and inevitable) infection.