#160 - Paul Offit, MD: Latest on COVID-19 vaccines and their safety, herd immunity, & viral variants
- Guest
- Paul Offit
- Read time
- ~4 min
- Uploaded
- 3 May 2021
- Duration
- 1:26:49
- Views
- 29.0K
KEYPOINTS
- 01
Two vaccine platforms were in wide use by spring 2021
The mRNA vaccines (Pfizer and Moderna) deliver instructions for the spike protein. The J&J vaccine uses a replication-defective adenovirus 26 vector, which had already been given to about 200,000 people in West Africa against Ebola. More than six million US doses of J&J had been given, and more than 20 million AstraZeneca doses in the UK and Europe. Russia's vaccine used Ad26 followed by Ad5. Offit traces the lineage back to cowpox in the late 1700s and Pasteur's killed rabies vaccine about 100 years later.
- 02
Rare vaccine side effects show up within about six weeks
Offit says serious vaccine side effects historically appear within weeks of a dose. Examples include narcolepsy from one adjuvanted 2009 pandemic flu vaccine (about 1 in 55,000) and Guillain-Barré after flu vaccine (about 1 per million). Others are vaccine-derived polio from Sabin's oral vaccine (about 1 per 2.4 million) and yellow-fever-like illness from the yellow fever vaccine (about 1 per million). That's why he discounts fears about '10 years of data' for mRNA, which cells translate into protein and can't reverse into DNA.
- 03
Why HIV beat vaccine makers, and SARS-CoV-2 hasn't
HIV mutates its surface glycoprotein again and again within a single infection, and it grows in T helper cells, the orchestrators of the immune system. Several billion dollars, including an adenovirus-vector vaccine that may have made things worse, failed to produce an HIV vaccine. SARS-CoV-2 mutates far more slowly, more like measles, whose 1960s vaccines still work 50 years later.
- 04
Variants so far are close enough for current vaccines
The original D614G strain killed about 570,000 Americans and is the target of every vaccine. B.1.1.7 (UK) is close enough to be covered. Offit watches whether a variant arises that natural or vaccine immunity can't touch, which would put us 'back to square one'. He thinks SARS-CoV-2 won't be a 'one and done' vaccine but needn't be yearly like flu.
- 05
Sterilizing immunity is unlikely, but disease protection should last
Viruses with short incubation periods, like this one, let infection get started before memory B and T cells can respond. Diseases like measles, with 10–14-day incubation, can be eliminated, as measles was in the US by 2000 and rubella by 2005. Offit expects antibody levels to fall over 6–18 months, while memory cells keep protecting against severe disease for a few years, not decades.
- 06
Get the second mRNA dose
One dose gives about 80% protection and two give 90–95%. Offit says the second dose also gives more complete immunity and more breadth against variants that have drifted from the original strain.
- 07
Relative and absolute risk reduction were poorly communicated
When J&J was authorized, headlines compared it with Pfizer's '96%' and Moderna's '95%' efficacy. In Pfizer's roughly 40,000-person trial there were about 165 cases among 20,000 placebo recipients versus 8 among vaccinated people, an absolute risk reduction of about 0.9%. J&J's absolute reduction was about 1.7% and Moderna's about 1.2%. By absolute measures, the 'weaker' J&J vaccine looked at least as good, and all three are great vaccines.
- 08
The J&J pause was an overreaction that chilled confidence
With about one case of cerebral venous sinus thrombosis per million doses, Offit estimates roughly 1,850 deaths prevented per million vaccinated. He understood the goal of warning doctors not to treat these clots with heparin, but believes the pause sent 'a chill' that would spill over to every COVID vaccine.
- 09
Herd immunity needs about 80%, and hesitancy is the obstacle
Offit estimates 22–30% of Americans had been infected and that about another 120 million people needed full vaccination. Surveys then showed about 30% of Christian evangelicals refusing, and about 14% of Black Americans saying they definitely wouldn't get vaccinated. He expected vaccines approved down to age 12 soon and to age 6 by early the next year.
EXAMPLES
- 01
Michigan's surge without a death surge
Cases rose but deaths didn't, because about 55–57% of over-65s were fully vaccinated and about 75% had at least one dose. People over 55 account for about 92% of deaths. The conversation keeps the claim tied to named mechanisms, numbers, and caveats so a listener can act or ask better questions without flattening the nuance.
- 02
Why a healthy 50-year-old should get vaccinated
Offit points to long-term neuropsychiatric complications, reported in up to a quarter of patients 6–12 months out. He also notes about 500,000 Americans can't be vaccinated because of chemotherapy or biologic drugs, and depend on everyone around them.
- 03
Polio as an analogy
Only about 1 in 200 polio infections caused paralysis, so silent spread was rampant, as with COVID. The US hasn't seen natural polio since the 1970s because children are still vaccinated.
- 04
Testing failure
The viral sequence was known by January 2020. By the time US testing problems were sorted out, South Korea had run about 500,000 tests versus fewer than 5,000 in the US. The US has about 4% of the world's population and about 20% of its COVID deaths.
OTHER RELEVANT
- 01
Judge vaccine safety by timing and scale
Most rare harms appear within six weeks, and well over 100 million US doses by then had revealed no rare serious mRNA signal at that point. Use that framework rather than 'it's too new'.
- 02
Interpret efficacy numbers carefully
Compare vaccines on outcomes that matter (hospitalization, death) in similar settings. Use absolute risk reduction and number needed to vaccinate (100 ÷ 1.7 is about 60 for J&J against any illness in its trial) alongside relative efficacy.
- 03
Prepare for the next pandemic
SARS (2002), MERS (2012) and SARS-CoV-2 came within about 20 years. Offit urges stockpiles of protective equipment and ventilators, and global vaccine capacity, since 'we're only as strong as our weakest nation.' Peter and Offit also weigh the natural-spillover and lab-leak theories of the virus's origin.