Skip to Main Content

Pfizer revealed Tuesday that researchers have not yet conducted an analysis of the efficacy of the vaccine it is developing against Covid-19.

The announcement is both good news and bad news. 


Umer Raffat, a senior managing director at the investment bank Evercore ISI, wrote in an analyst note that the fact that Pfizer hasn’t conducted an interim analysis was “a good thing” because it means that, based on the details of the trial protocol, the vaccine had not failed to prevent more than 77% of Covid cases, the benchmark for success at this early juncture.

But it also suggests that cases of Covid are being reported less frequently among participants in the Pfizer study than in the U.S. as a whole. That means that the study is progressing more slowly than Pfizer and its partner, BioNTech, originally expected. In August, Pfizer had thought that the first interim analysis could occur as early as September. 

Pfizer and BioNTech are under a white-hot spotlight because their Covid-19 vaccine, by design, is likely to be the first to have any efficacy data. But that first analysis will come when there have been a total of only 32 cases of Covid-19 across the company’s entire 42,000-volunteer study. It would be considered to be positive, Pfizer has said, if six or fewer of those 32 cases occurred in the group that received the vaccine, with the rest occurring in the group that received the placebo. The trial is expected to continue until 150 of the volunteers in the study have had Covid-19.


Pfizer CEO Albert Bourla set the goal of having a Covid-19 vaccine ready by October this spring, in what appeared to be a bid both to combat the pandemic and to make a statement about the company’s research prowess. Pfizer is in the midst of spinning off older, slower-growing parts of the company in a merger with Mylan, which will leave Bourla running a firm whose future will depend heavily on its research laboratories and the prospects of its experimental drugs.

The vaccine itself is expected to be a significant but not huge financial opportunity. Analysts at SVB Leerink estimate sales could peak at $3.5 billion in 2021, and then stabilize at $1.4 billion a year. In its third-quarter earnings, also announced Tuesday, Pfizer reported profits of $2.2 billion on sales of $12.1 billion.

But the October time frame has put Pfizer squarely in the political maelstrom of the U.S. election, as President Trump repeatedly predicted that a vaccine would be available before Election Day. It has been clear for months that other companies developing Covid-19 vaccines, including Moderna, AstraZeneca, and Johnson & Johnson, had no chance of hitting that mark.

On Oct. 6, the Food and Drug Administration released new guidance on safety data that would be required before it would consider clearing a vaccine through an emergency use authorization, or EUA, its fastest path to making a vaccine available. The agency told drug makers not to file for an EUA until at least half the patients in their studies have been followed for two months. In an Oct. 16 public letter, Bourla confirmed that this would prevent Pfizer from filing for an EUA until the third week of November.

At an advisory panel of outside experts the FDA held last week, the agency signaled that it was not certain it would grant an EUA for a vaccine until a sufficient amount of data had been accumulated to more fully understand the safety and efficacy of a vaccine prospect.

Against that backdrop, there is a huge amount of attention on Pfizer’s interim analyses, which were originally designed as a way to make a vaccine available as soon as possible. Such analyses are a standard part of clinical trials, and are conducted by an independent committee called a data and safety monitoring board, or DSMB.

However, it’s unusual that Pfizer will be getting reports from the DSMB on the efficacy of the study as it is still ongoing, and it is also unusual that those results will be announced as the study continues with patients and doctors remaining unaware of who has received the vaccine and who has received placebo. It might also have been possible, for instance, for Pfizer to have told the DSMB only to give it information if there is enough safety data to file for an EUA. That appears to be the approach, for instance, that is being taken by Johnson & Johnson with its Covid-19 vaccine. In many but not all circumstances, interim analyses are made public because a treatment or vaccine is so effective (or ineffective or unsafe) that it is no longer possible to continue a study.

Receiving these reports puts Pfizer in the position of having to disclose them for both financial and ethical reasons. That’s certainly true if the DSMB says the vaccine is effective or definitively not effective, but Pfizer may also feel the need to announce if the DSMB conducts an interim analysis and allows the study to continue.

Bourla said on a conference call Tuesday that he is “cautiously optimistic” that the study will have positive results, and that those could come in an interim analysis. “You never know until you have a study result,” he said. “Let’s all have the patience that is required for something so important for public health.” He later said that he knows that the U.S election is raising the tension around the data readout. “This is not going to be a Republican vaccine or a Democratic vaccine. This is going to be a vaccine for the citizens of the world. I hope that it is going to be effective.”

Making educated guesses about what is happening in a clinical trial is normal practice for investors who follow pharmaceutical companies, but not for a general public that is hoping a vaccine will be able to help slow a pandemic. A warning, then, from past attempts from investors to read the tea leaves on clinical trials: They are frequently wrong.

For instance, on Monday, analysts at Morgan Stanley wrote that they suspected the first interim, which Pfizer said requires 77% vaccine efficacy, had happened in mid-October, and that Pfizer could be getting its second planned report on the data, which occurs after 62 cases of Covid-19 occur in the trial. That was smart speculation — and, it turns out, incorrect.

In an unrelated update, Novavax, another closely watched vaccine maker, said the U.S. study of its Covid-19 vaccine would start in late November, after it was expected.

  • The fact that they don’t have 32 cases yet may well be good news. The more effective the vaccine is, the longer it will take to get 32 cases because more of the 32 will have to come from those getting the placebo.

  • Also what on earth is taking the FDA so long to give EUA’s for Regeneron and Eli Lilly’s antibody treatments for use in high risk patients recently infected by COVID-19? They have EASILY the best data yet for an anti-viral therapeutic to prevent progression of disease and resolve symptoms.

    What does the FDA not get about the massively upwardly skewed risk in high risk patients? Based on the statistics I have seen, people over 80 appear to have an infection fatality rate approaching 20%! While it might be lower now given better understanding of steroid usage, anti-coagulation and ventilator use it is still staggeringly high. You should be giving physicians the option of prescribing this treatment to ultra high risk populations before you convince yourself there isn’t some very, very rare risk for people with HUNDREDS of times lower risk of mortality.

  • The whole approach of forcing companies to wait until two full months after the last vaccine dose is administered is bizarre. Newsflash: Over 6,000 people are DYING PER DAY from COVID-19. Many, many more are spending weeks in ICU. Many, many more are suffering debilitating long term health effects. Worse still, the numbers are rapidly accelerating because this disease has an exponential growth factor. Every week you delay an effective vaccine dooms a minimum of 30,000 people to death and many more to severe illness with long term impacts to health.

    If the vaccine is shown to be effective in the interim analysis of a well run clinical trial it should IMMEDIATELY be granted an EUA for use in high risk patients who have orders of magnitude higher risk of dying and being admitted to an ICU unit. The simple fact is that the risk of taking an effective vaccine that has been through phase 1 and 2 trials with no safety issues and reached an interim analysis in a phase 3 trial with no safety issues has to be VASTLY lower than that of leaving our vulnerable completely unprotected against the virus. Let’s think on a population wide basis rather than obsessing about very, very rare potential side effects for low risk populations. These people wouldn’t be getting the vaccine for months in any case!

  • Time to get the politics out of this . FDA Stop coming up wIt’s excuses to slow or limit approval . You don’t need medieval trials to know its effectiveness and safety . Science actually progressed since trials were invented

  • Tremendous amount of biased “yawn” drivel about just 1 vaccine – of the 30 or so in trials. Pfizer’s takes 2 shots and needs to be stored at minus 70C = too cumbersome for most of the world. STAT writers should dig into OTHER trials: a one-shot (JNJ), or a PILL (Vaxart), or another vaccine platform (Inovio), etc etc.

    • DHL put out a piece specifically on this and found that 2.5 BILLION people live in countries that are highly likely to be able to distribute the vaccine effectively. It isn’t rocket science. Low temeperature freezers at distribution centres and dry ice coolers are all that is required. Set up inoculation stations at the tens of thousands of small regional airports and you have yourself an effective distribution network.

    • OC – your notion would only work in the western / developed world – not in the many really remote areas in underdeveloped countries without airports and with even dry ice evaporating. Thus Pfizer’s vaccine causes discrimination issues. I agree with Dean: for ease, cost and fairness other vaccines specially the 1-dose kinds are not to be overlooked.

  • Why is this taking so long? It is obvious both Moderna and Pfizer should have had 10 times the number of trial participants that they currently have. They should have lined them up over the summer and ask the govt. to pay the up front costs (in case they did not pass phase 2 trials). They also should have scaled up production after phase 1 trials (again the govt should have covered up front costs with pfizer remimbursing them if things worked out) over the summer so we would have all the vaccine we need by the end of the year. Maybe scale up was not possible but I can’t help believing if we had a president that believed the virus was a serious problem from the get go that some of these things would have been possible.

    • Pfizer has scaled up production and announced that they would have 100 million doses by the end of the year (roughly half will be for the US). They did, in fact get federal funding to do this, with the government placing a massive order for vaccines several weeks ago, and Pfizer indicated that they didn’t want funding for the research part of their costs.

      There are 4 massive phase 3 trials for COVID vaccines enrolling in the US right now. I don’t think it is too surprising that all are finding it difficult to get all the volunteers and study sites they want as quickly as they want.

      Ultimately, this appears to be a trivial setback. The safety signal won’t be available for 3 weeks, so if the efficacy signal isn’t available for another 2 weeks, it doesn’t mean anything.

      By all indications, whichever COVID vaccine makes it across the finish line first will absolutely smash all prior speed records, taking less than a third of the time as the prior record holder.

      Fair criticism can be made of the federal response. However, I have a hard time seeing the vaccine development efforts as anything but a bright spot.

  • Understandably politics are a big factor, and I wish this never happened at all (let alone during an election year). Pfizer must feel pretty confident in its product to start testing on kids. I would imagine they do have some data indeed, but like people are speculating, they’re waiting until after the election.

    I have a hard time believing that among tens of thousands of people, 30 or 40 haven’t been infected with COVID.

    • Pfizer is blinded as to who received the placebo. By all accounts, everyone in the trial has remained healthy, so you can have increasing confidence in the safety. The efficacy is still unknown.

      Keep in mind that it is not just 32 people who need to have been infected, they need symptoms in order to count, and I don’t think they count until the second week after their second injection, so it hasn’t been that long that tens of thousands would ‘count’.

  • Say it with me: the election is next week.
    Once you had doctors firing off op-eds saying wait until after the election and you had so many people saying any results prior to the election can’t be believed, you made it so there could be no announcement before next Tuesday. Otherwise, it would simply turn into a spitting match.

    This decision regarding no data has zero to do with science. It is all about the election.
    Tell me this wouldn’t be exactly what would happen:
    Pfizer this week: Hey guys! We have a vaccine that’s 80% effective. We really think this works.
    Swing Voters: Hmmmm maybe that Trump guy is on to something…
    Crazed conservative right: Trump is amazing! MAGA!!!!!!!!
    Crazed liberal left: Don’t trust it!!!! They’re lying!!! Its a political vaccine!!!

    I could be entirely wrong. My bet is 1-2 weeks after the election a truckload of big, positive news comes out.

    • You are essentially accusing Pfizer of lying about the conduct of their trial. That seems extremely unlikely. Particularly as this information was discussed at a shareholder facing event. Pfizer has withstood the barrage of accusations of politics for months. Now that the results are meaningless (the EUA cannot be applied for until a couple weeks after the election, regardless), Pfizer now decides that they are going to lie to the world in order to create the perception they aren’t playing politics? That seems highly unlikely.

  • Topol is a hardcore liberal but that aside, could you provide more clarity on why this is good and bad news? Like, it’s good because in theory it’s working because there are fewer reported cases, but it’s also not because they actually need more cases to prove its efficacy?

    • He’s churning out copy. The writer is required to write x-number of pieces per week, month, whatever.
      Its about the election. Any big announcement, especially if its positive, would be skewered as being political because that’s what we’ve come to.
      This has absolutely nothing to do with science.

    • You want the placebo group to get it. When nobody gets it, all that you proved is that people who sign up for COVID vaccine trials are careful to avoid getting COVID.

      So, the bad news is that we still have no hint as to whether the vaccine does any good at all. The good news is that the prior assumption had been that the vaccine had already had is interim analysis, but hadn’t passed it, which would have implied that its efficacy couldn’t have been better than 80%. We learned today that such interim analysis has not been performed, so a highly effective vaccine remains a possibility.

  • I suspect that the meeting Pfizer had with prominent physician scientist, such as Eric Topol, may suggest that they are intentionally holding back the information until after the election so that the federal government cannot politicize the vaccine, which may cast doubt as to its efficacy. Their nature paper published earlier this month, suggest that there were potent neutralizing antibodies and T cells raised against SARS-CoV-2 S protein at all doses. I do not see why this would change with a larger trial.

    From a scientific perspective, the assays needed to prove neutralization takes several days if done the most rigorously. Processing upwards of 40k+ neutralization test and T cell tests is a tour de force and will likely be delayed as research materials are also experiencing huge delays compared to pre-COVID times. I would know because I am an MD/PhD student at Yale in the midst of my last PhD year.

    • This is an efficacy trial against the disease, not a sero-conversion based surrogate trial for efficacy. As there is no correlate of protection against COVID-19 disease yet, COVID-19 vaccine cannot be licensed based on a surrogate marker of sero-conversion (development of neutralizing antibodies). Having neutralizing antibodies during phase 1 and 2 clinical trials is encouraging, but does not prove that the vaccine is efficacious. You may like to read an article, I published recently, on COVID-19 vaccines (follow the link below), which discusses briefly about various types of clinical trials for demonstrating efficacy of vaccines – efficacy against the disease, sero-conversion and human challenge trials. Best wishes for your MD/PhD program.

Comments are closed.