fifteen: How I (accidentally) invented a COVID-19 vaccine
I swear it was a “vaccident”.
Today, my nonprofit’s innovation offshoot, Cure Motive Research Accelerator (CMRA), released a preprint on Research Square describing a potentially cheap, scalable, and stable COVID-19 subunit vaccine approach that could be a great solution for the 3rd world and for anyone who cannot or will not take an mRNA or DNA vaccine.
This discovery was possible because I accidentally contaminated myself with a SARS-CoV-2 protein in the laboratory. Whoops!
SERENDIPITY
How this entire episode unfolded is rooted in one of the most fundamental skills (IMHO) of a scientist: embracing serendipity. More specifically, poking at outlier data rather than dismissing it out of hand. I will be discussing this further at length elsewhere (i.e., my book) but the importance of this skill cannot be understated. Every major discovery I have EVER made, including the work leading to the cancer drug I invented (now in clinical trial), has come out of odd, but provocative, data that I felt compelled to pursue out of pure scientific curiosity. These are data that didn’t quite fit the hypothesis and couldn’t otherwise be readily explained. For many scientists and trainees, these data are destined for the garbage bin (or at least trapped in the lab notebook in perpetuity). For me, I’ve come to learn this is truly where the good stuff is.
I RELISH unexpected findings.
[Dear Undergrads, Grad Students, Postdocs, and Trainees: NEVER EVER throw away unexpected results. Repeat the experiment. If you get the same thing then it is probably real. It could still be a meaningless technical or methodological aberration but it is real and should be explored. >99% of the time it will go nowhere, but the experience is still valuable and worth it for the <1% of victories. If your PI or supervisor ever tells you to ignore surprising/outlier data or tells you to keep repeating the experiment until you get the results they are expecting I suggest you find a new PI immediately. They are not interested in real science. They are interested only in protecting their nut. RUN AWAY!]
As a fun exercise, let’s look at another accidental invention of self-inoculation in the past:
INVENTOR: Constantin Fahlberg (1877)
WHAT HE WAS TRYING TO DO: study compound radicals and substitution products of coal tar at Johns Hopkins.
WHAT HAPPENED INSTEAD: He made benzoic sulfinide (the artificial sweetener, saccharin).
HOW DID HE FIND OUT: He spilled some chemicals on his bare hand in the lab, ignored it, didn’t wash up before rushing off to dinner, noticed a piece of bread he ate with his contaminated hand was sweet, and then tasted his thumb which was even sweeter.
HOW DID HE CONFIRM IT: He immediately ran to the lab and tasted the contents of everything on his bench(!) until he found the beaker of impure saccharin (EH&S entering the chat…).
OUTCOME: He patented several methods of manufacture and commercialized it.
Constantin Fahlberg, the inventor of saccharin (circa 1910).
Now, me:
INVENTOR: W. Shawn Carbonell (2020)
WHAT HE WAS TRYING TO DO: show proof of concept for a new antibody discovery platform by targeting the S1 subunit of the SARS-CoV-2 spike protein.
WHAT HAPPENED INSTEAD: He made a SARS-CoV-2 subunit vaccine.
HOW DID HE FIND OUT: Two months after starting the work, he began collecting his own serum to be used as a negative control matrix for the analytic assays to study the anti-S1 antibodies but realized his serum was instead highly POSITIVE.
HOW DID HE CONFIRM IT: He developed several assays and found he had high titers of anti-S1 antibodies, neutralizing antibodies, and a putative cellular immune response.
OUTCOME: He patented the method on behalf of his nonprofit but isn’t sure what to do with it.
W. Shawn Carbonell, the inventor of ocular vaccination among other things (circa 2021)
DEBRIEF
To address the obvious question directly: no, I did not do this on purpose.
The route of vaccination was most likely mucosal… probably by accidentally getting S1 protein on my gloves, then possibly my iPhone where it dried, then to my bare hand, and—finally—directly on my goddam eyeball (I have allergies and rub them a lot). I can assure you that—until possibly today—no one in the world would voluntarily vaccinate themselves IN THE FUCKING EYE with a viral protein. Thankfully, other than a brief and mild stint of conjunctivitis upon the second exposure, there were no other adverse events. Indeed, the first exposure was completely undetected and only discovered retrospectively months later as recounted above.
[The second exposure may be blamed on (or credited to) of all things, TikTok. My nightly ritual. Full disclosure, I am a paid science influencer for TikTok through the Creative Learning Fund as well as the unrelated Creative Fund.]
IMPLICATIONS
What makes this discovery truly remarkable—again, other than the fact I got it in the eyeball, TWICE—is the amount of protein exposure would have to have been extremely miniscule… possibly two orders of magnitude less than a typical subunit vaccine (e.g., typical influenza vaccine is ~30 µg of subunit protein). And yet, the titers of S1 antibodies in my blood, including neutralizing antibodies, was roughly equivalent to that seen with the mRNA vaccines. More perks: we have human safety data (i.e., me again), the methods of manufacture of protein subunits are well established and are globally scalable, costs are pretty low at scale, and proteins are much more stable than mRNA and DNA. In addition, it begs an intranasal administration (rather than an arm jab or an eye poke) which is advantageous because it may promote mucosal in addition to systemic immunity and, frankly, people don’t like needles.
[A note on occupational exposure. Thousands of scientists have worked with the SARS-CoV-2 spike protein in various forms, including pseudotyped viruses. I’m not the only one guilty of touching their phone with a gloved hand… let alone door knobs, chair seat backs, incubator doors, microscope knobs, sink handles, pipettes, pens, tape, etc. I’ll just say this… I can’t be the only one.]
And, I don’t mean to be a dick during what is surely an amazing year for translational science (esp the rapid emergence of the mRNA technologies), but perhaps the amazing efficacy of the EUA-cleared vaccine candidates merely reflects the very high native immunogenicity of the spike protein and virtually any candidate based on it will potentially be effective. In other words, maybe we all got lucky that coronavirus spike protein is so immunogenic to us. My data seems to suggest so. Why this is important is that we may not be so lucky the next time a virus spills over so we honestly need to wake the fuck up. It’s the classic hammer-nail situation… yes, we have a shiny new hammer in the mRNA vaccines, but not all viruses will be broad-headed nails set perpendicularly in plywood. Some may be bolts in concrete, embedded at 45 degree angles.
So, it begs the question: how can we leverage these data? I believe there is big promise but my nonprofit is focused on brain cancer. Which is, of course, why we’ve decided to go public with it and release the preprint ahead of peer review. So, please reach out if you have ideas.
Finally, we (the nonprofit) did file a provisional patent application for the composition and method and are not sharing details at this time. The raw ingredients are commercially available and we do not want to hear about people trying to intentionally vaccinate themselves with crude protein. Critically, not all forms of recombinant S1 will have the same effect. Everyone should definitely wait instead for vaccine candidates with EUA that have been fully tested for safety and efficacy. In fact, I have already received the first dose of one of the mRNA vaccines because although I showed a good immune response to S1 protein exposure it certainly was not an optimized regimen. In fact, it was completely stochastic. Speaking of…
2020 was absolutely difficult and—as with any such situation—I’ve found the best thing to do is to double down on gratitude… and make some goddam lemonade.
CHEERS!