A Covid-19 vaccine paradox and a few related speculations

To the extent we are all waiting for a vaccine to let us “get back to normal” — whatever “normal” is for a traumatized and transformed world — we will wait much longer if we do “too good” a job controlling the spread of Covid-19 now. This fact leads to a number of interesting and troubling speculations, of which more below after some background.

Earlier this week we received a report from one of the Wall Street analysts (Michael Yee at Jefferies) who covers Moderna, Inc. (NASDAQ: MRNA), the developer of one of the front-running vaccines for the prevention of Covid-19. It came by email and has no public link, so we shall provide a fair use reproduction of its key point, which is that the pace of the clinical trials to ascertain the safety and effectiveness of the various vaccines for Covid-19 depend highly on the rate of infection.

Big Picture: With MRNA’s recent initiation of the Phase III, we’ve fielded many investor questions as to (1) the different enrollment and infection scenarios and (2) various timing at which we could see data. Based on a statistical model and range of scenarios, we think fast enrollment and above-average infection rate in hot spots will drive a good chance (70%) we could see MRNA data by Oct/Nov. We think PFE/BNTX [Pfizer’s program – Ed.] will be similar if not earlier into September due to global enrollment and other minor details (shorter 3-wk dosing regimen) and AZN [AstraZeneca’s program – Ed.] may be Fall w/ more EU based enrollment and infection rate sensitivities.

Perhaps a short bit on vaccine trials would be useful. Basically, the trial sponsor rounds up thousands of volunteers who are at theoretical risk to catch the disease being studied, in this case Covid-19. These volunteers receive either the vaccine candidate or a placebo, and nobody — neither the volunteers, nor the people administering the injections or taking care of the patients — knows who got which. The study has a “statistical plan” — essentially, the scientific hypothesis being tested in the experiment — that depends on a certain number of people getting the infection over a certain period of time. Once you have hit that number of new infections, plus a few more to be on the safe side, you crack the code and figure out whether the volunteers who got the placebo had a greater propensity to be infected than the volunteers who got the vaccine candidate. If so by a statistically significant difference, voila! You’ve got yourself a vaccine.

Therefore, the faster the disease is spreading, the faster you will get to the “end point” of the trial. You would not conduct a Covid-19 trial in New Zealand or most European countries right now, but Mississippi might make sense.

Yee’s model, which includes such things as assumptions about infection spread under a “bear case” (low rates of infection, which would be good for the people not getting sick but delay the data readout for the Moderna trial and therefore bad for Moderna’s stockholders and “getting back to normal”) to a “bull case” (the reverse), suggests that we could see an “interim” readout — some statistical plans allow for an early peek at a subset of the patients — under a high infection scenario as early as late October 2020 and final analysis by Thanksgiving.

If, however, we break the back of the virus and infections decline dramatically, Yee’s model delays a readout — an essential predicate for FDA approval, even on an emergency basis — until March 24, 2021.

As we all have learned, five extra months of this shitshow comes at a huge cost, not just economically but emotionally, socially, and culturally.

Nobody planned it this way, but the spread of the infection in the United States right now (as opposed to earlier in the year when there were no vaccine candidates in Phase 3 trials) has the potential to be of enormous value to those parts of the country which have already been through their big spike. Because science. Let that sink in, Andrew Cuomo.

All of this arouses interesting political thoughts which will please nobody. President Trump’s fortunes may depend on infections declining substantially. There is at least some evidence that Trump’s poll results are correlated with daily infections in the United States. At the same time, though, he and his team have made all sorts of rosy predictions about the speed with which a vaccine may be available.

Going out on a limb here, but it doesn’t seem as though either Trump or the reporters bathing him in snark every day realize that a fast track to a vaccine depends on a high rate of infection somewhere in the United States. We have not seen anybody on the left or in the press accuse Trump of promoting infections to speed the vaccine trials. We find it difficult to believe that in the current climate that accusation would not surface, at least as a “some critics say” smear, had anybody thought of it. And, for that matter, neither has Trump raised the faster vaccine timeline as a silver lining of the high infections, which one could easily see him doing. When that tweet comes, you read it here first!

One can also imagine that the arc of infections will let through a vaccine — say, Moderna’s — that is not the best ultimate vaccine, and then delay or prevent other vaccines that might be better. No vaccine will be effective on everyone (which is why a Covid-19 vaccine will provoke complex political and social questions). Yee imagines that Moderna’s vaccine might range from 50% effective to 90% effective, so we don’t really know anything other than that no vaccine works on everyone. Well, a 50% effective vaccine first out of the gate would be a lot better than nothing, and it could easily happen. Why? Because the lower the effectiveness of the vaccine the faster the clinical trial will experience the required number of infections to hit the end point and final readout!

But that will drive another paradox: What if the first approved vaccine drives infections to the point where it takes forever to prove the effectiveness of subsequent superior vaccines, but not so low as to move us to “herd immunity”? That would suck, because it would make it politically much harder to require universal vaccination, which will be essential if we are going to protect people whose immune systems do not generate resistance to Covid-19 even after vaccination.

But that’s a whole ‘nother thing.