#115–David Watkins, PhD: Immunology, monoclonal antibodies, & vaccine strategies for COVID-19

KEYPOINTS

• 01

  The adaptive immune system is ancient and underpins vaccination

  Watkins, an AIDS-vaccine researcher who spent decades on SIV in monkeys, separates the fast, nonspecific innate system from the adaptive system of T and B cells. The adaptive system creates memory and goes back at least to amphibians. He calls vaccination, which relies on that memory, the public-health measure that has saved more lives than any other.

• 02

  A positive antibody test doesn't mean you're protected

  Only antibodies that cover the part of the spike that binds the human receptor neutralize the virus; antibodies elsewhere on the spike may not stop infection. IgM appears first and fades over about two to three months, IgG follows and carries the eventual neutralizing response, and IgA protects mucosal surfaces. Standard antibody (serology) tests detect IgG or IgM, not whether those antibodies neutralize.

• 03

  Responses vary hugely between people

  Watkins cites a Rockefeller preprint of about 70 people with prior COVID-19 in which nearly 20% made no detectable neutralizing antibodies. A few people could neutralize the virus even when their serum was diluted 1 in 5,000, while most managed about 1 in 100. Whether low responders can be reinfected, and how soon, was unknown; common-cold coronaviruses do reinfect people repeatedly.

• 04

  Neutralization is hard to measure routinely

  The test requires mixing a patient's serum with live virus, adding it to cells and watching for infection over two to six days. Labs do this for HIV, Zika and dengue, but there was no clinical version at the time, which is why ordinary antibody tests dominated the news.

• 05

  HIV shows what a vaccine-proof virus looks like

  In the first weeks of infection, HIV reaches 10–100 million copies per milliliter, and almost every copy differs. Killer T cells and antibodies select for escape mutants over and over because the infection is chronic. Its envelope protein is sparse and covered in a sugar shield, and no vaccine has ever induced antibodies that neutralize it. Watkins can make monkeys produce huge amounts of binding antibodies, but none neutralize. HIV has infected about 75 million people and killed 32 million, and Watkins calls its vaccine challenges probably insurmountable.

• 06

  SARS-CoV-2 should be an easier target

  It isn't chronic and shows far less escape, which Watkins puts at orders of magnitude less than HIV, so a protective antibody response is achievable. He cautions that he'd been wrong about every prediction about this virus since January. Asked by Attia, he estimated it at perhaps ten times worse than seasonal flu, depending on age, diabetes, obesity and initial viral dose. Attia's own reading of age-stratified infection fatality rates was two to eight times.

• 07

  Live attenuated vaccines work best but carry rare risks

  The 17D yellow fever vaccine came from a virus taken from a patient in Africa and weakened by repeated passage in monkeys and lab culture. In Watkins's Brazilian collaboration, recipients' serum still neutralized it at a 1-in-5,000 dilution, though protection against wild virus taken straight from a monkey was weaker. About 1 in 300,000 recipients get sick from the vaccine, and some die.

• 08

  Monoclonal antibodies are Watkins's favourite strategy

  Find the people who make the best neutralizing antibodies, clone the genes from their memory B cells, test the antibodies in animals, grow them in large vats and inject them. Engineered mutations can keep protective levels for three to six months. Attia frames it as giving people the product of B cells directly, ideal for nursing homes, high-risk patients and healthcare workers, perhaps two to four times a year. Watkins adds that a cheap vaccine is still the best route for most people.