Skip to content

This past week saw a bit of infighting at the FDA spilling over into the public. Two respected and seasoned staffers at CBER announced their resignation, apparently fueled by disagreements with White House statements about booster plans, and then they signed on to a Lancet article opposing booster approval. This was in advance of the FDA/VRBPAC meeting on September 17. I urge public officials on all sides to tone down their emotions and just get back to evaluating the evidence.

Mask Slackers

I'm always interested in the history of medicine and thus delighted to see this quick graphic article in the NEJM posted on September 18. The article is free without subscription, take a few minutes to look at it!

The Boosting Evidence

I was able to attend a good deal of the FDA/VRBPAC meeting, including the important discussions surrounding a vote about whether to approve boosters for all individuals age 16 years and above. The event showed some evidence of the earlier drama in the week but mostly was under control. I was very interested to hear discussion of data from Israel and I was in stats nerd heaven with the very cogent discussion of pitfalls in interpretation of "real-world" data. As you probably know by now, the advisory group voted 14-2 against approving the vaccine for everyone > 16 years of age, and in the ensuing discussions what emerged was a unanimous recommendation for authorization of boosters for individuals > 65 plus front line providers and others at high occupational risk like teachers. I think the committee made the right move for now.

Here are 3 takeaways from the meeting, some of which I don't think made it through clearly in the lay press. First, the data presented didn't have long enough follow up information to know whether boosters truly served to dampen the surge curve and significantly lessen morbidity and mortality in Israel. Similar data are lacking for what the US curve is doing now and whether the surge will decrease without immediate boosters. It will be only a few weeks until that information is available. Second, although the Pfizer data clearly show a robust "boost" to antibody in booster recipients, it's based on a small number (around 300) of individuals with limited follow-up. Historically, that number of study subjects is at the low end for what FDA has approved in the past for vaccine boosters, Japanese encephalitis vaccine being an example for that low number. Other booster approvals tend to have involved closer to 1000 subjects.

The third takeaway involves the important concern of safety. In particular, why expose young people to the possible myocarditis risk with a booster dose if the benefits to them and society are a bit fuzzy at present? I did learn that the best estimates for highest mRNA vaccine-associated myo/pericarditis is in 16-17 year olds at about 1 in 5000-6000 recipients. The rate goes down sharply with advancing age. This complication seems very mild at present, but we're still awaiting long-term follow-up of this potential side effect. I am less concerned about this factor than I am about the lack of high-quality evidence that boosters are highly beneficial this soon after primary series.

For those interested, I'll be discussing the booster considerations along with a general COVID-19 update and information about studies in younger children at a Children's National Hospital Pediatric Health Network Town Hall meeting at 12:15 PM EST on Tuesday, September 21. Organizers have told me you don't need to be a PHN member to attend the meeting. As I write this the morning of September 19, we await word from FDA as to official authorization of boosters and then the ensuing deliberations from the ACIP on more specific recommendations. Those meetings currently are scheduled for September 21 and 22.

1

The COVID-19 pandemic is confusing at baseline, but now others who should know better are aggravating the situation. My post this week is a little verbose, I'll try to return to concision next week.

To Boost or Not to Boost, That is (Not) the Question

Most evidence points to a need for a booster vaccine dose for the mRNA vaccines, but the timing is far from clear. So, I was dismayed to see that the main message this week was that everyone (some more than others) needs a booster 8 months following completion of the 2-dose series. Buried in the pronouncements was the fact that FDA and ACIP will look closely at the data to make a judgment. It is absolutely appropriate to plan for the possibility that booster vaccine doses will be administered in September, but don't use the need for advanced distribution planning as a surrogate for an 8-month booster timing. Let FDA and ACIP do their jobs.

I say this after having gone over the evidence with my best fine-tooth comb. Three reports appeared as early releases August 18 in MMWR. All are well-designed and well-reported studies utilizing different databases and study designs. One looked at hospitalizations in New York state from early May to late July, another at duration of effectiveness of mRNA vaccines March through July throughout the country, and the third (and only CDC study) looked at vaccine effectiveness in US nursing home residents March through July. The most important take-home point from all 3 studies is that the mRNA vaccines are all still tremendously effective at preventing severe illness and death. (The Janssen/J&J vaccine hasn't been in use long enough to make any concrete conclusions about changes in efficacy.) There was a slight decline in effectiveness in some groups recently, but whether this is due to waning of vaccine immunity, the appearance of the delta variant, or both isn't possible to sort out yet.

What really has had the steam coming out of my ears lately is the discussion of Israeli data, as yet unpublished in a peer-reviewed journal. It seemed to me to be a clear error in statistical analysis and reporting, something I would use an example in my old EBM classes. Just this week I found support in the form of an online site that explains the problem far better than I could. It is an example of Simpson's Paradox (Edward Hugh Simpson, not Homer). Briefly stated, Simpson's Paradox is "a trend or result that is present when data is put into groups that reverses or disappears when the data is combined." More simply, if one is not careful to take into account confounding variables, you can end up with conclusions that are the exact opposite of the truth. In the case of the Israeli data, reanalyzing numbers available from the Israeli government data dashboard showed that it is likely that claims of waning vaccine efficacy are a byproduct of higher vaccination rates in the elderly, who have more underlying risk factors, compared to the highly unvaccinated group of teenagers and young adults. When you correct for the confounding, or hidden, variables, the vaccine efficacy looks great for all ages.

Please Don't Administer Vaccines Off-Label!

I received an email from the mother of a child enrolled in our COVID-19 vaccine trial at Children's National. He received his 2 injections (either vaccine or placebo) a couple months ago and is doing very well, but mother is concerned because he will start school soon and has very mild asthma. (She felt this was a risk factor for severe disease but plenty of studies show that this is not a risk.) She had a few great questions, but what concerned me most is that she had heard from others that some healthcare providers were planning to administer COVID-19 vaccines to children 9-11 years of age which of course is not an authorized age group. I don't know if this is true, but I do know this is one of the worst ideas I've heard, right up there with drinking bleach.

DO NOT make up your own vaccine schedules. How would you know what dose to give a 9 year old child? What if there is a severe adverse reaction? Remember also that you need to report all vaccine doses administered. If you want to get involved with giving COVID-19 vaccines to children under 12, join a clinical trial. 'Nuff said.