My long-suffering (try putting up with me for 40+ years) wife, a retired general pediatrician, thought my posting about the FDA VRBPAC meeting was a bit too detailed for front-line healthcare providers. Reading it again, she's right, but of course I'll push back a little bit given that FDA has backtracked on the committee's recommendation. First, a few other updates.
Wild-type Polio Update
Thankfully we haven't had any recent polio appearances in the US, but it's a bit discouraging to see what's going on worldwide. The Global Polio Eradication Initiative reported 1 case of wild-type polio in each of 2 countries, Afghanistan and Pakistan. Here's the complete list for the week which includes vaccine-derived cases and environmental sampling as well:
- Afghanistan: one WPV1 case
- Pakistan: one WPV1 case and 20 positive environmental samples
- Côte d’Ivoire: three cVDPV2-positive environmental samples
- DR Congo: one cVDPV1 case
- Ghana: one cVDPV2-positive environmental sample
- Liberia: four cVDPV2-positive environmental samples
- Niger: one cVDPV2-positive environmental sample
- Sierra Leone: six cVDPV2-positive environmental samples
- South Sudan: one cVDPV2 case
- Yemen: three cVDPV2 cases
Remember that paralytic polio cases represent only the tip of the iceberg for polio infections; the vast majority of infections are asymptomatic, with a few percent manifesting as nonspecific febrile illnesses. Paralytic polio cases comprise less than 1% of infections. So, the appearance of 1 case can imply that at least 100 more infections were present in an area.
Can Infants Spread C diff in Households?
Asymptomatic Clostridioides difficile carriage is common in infants, and the organism seldom causes illness under 2 years of age. That's why you don't want to test for C diff in younger children. An interesting new study suggests, but by no means proves, that these asymptomatic carriers might be the source of household spread which could include spread to more vulnerable individuals.
Thirty families were recruited at their child's 4-month checkup to participate in this longitudinal study where participants mailed soiled infant diapers every 2 weeks to the study site, until the infants reached 8-9 months of age. Rectal swabs from mothers, and sometimes from fathers, were mailed at the same time but in separately sealed containers. (Sorry, but I couldn't help but wonder if the mail carriers had to put up with some unusual package odors!) The specimens were tested for C diff and positive samples were strain-typed and tested for toxin production.
Probably the researchers would have needed to perform more frequent sampling to prove the directionality of transmission, i.e. infant to adult or vice-versa, but they did note that the adults, compared to the infants, seldom were the initial positive carriers in these families. Sharing of C diff strains in the families was common and more often implicated infant to adult directionality. C diff prevalence in infants was 50.0-71.4%, including nontoxigenic strains, while maternal positivity was 20.0-40.0% and fathers were 20.0-37.5% positive. None of the infants or parents developed symptoms.
Something to keep in mind, but please don't start testing infants for C diff.
Misinformation Tracking
A recent article tried to look at patterns of misinformation and flagged content on Facebook. Unfortunately subscription is required for full article access, but the abstract is an accurate summary. Based on numbers of pageviews, the authors felt that unflagged content on Facebook was more likely to be influential.
The authors used a complicated (to me) combination of crowd-sourcing and machine learning to derive estimates of vaccine hesitancy and matching to pageviews. They concluded "...We estimate that the impact of unflagged content that nonetheless encouraged vaccine skepticism was 46-fold greater than that of misinformation flagged by fact-checkers." This unflagged content predominantly consisted of real facts, e.g. rare deaths following vaccination, that then were misinterpreted by viewers as the vaccine causing the death rather than within the expected death rate based on the general population, regardless of vaccination status. A classic misinterpretation due to lack of a control group!
Along the same lines, I noted with sadness that the Stanford Internet Observatory that tracks misinformation is shutting down, in large part due to lawsuits and online attacks received by staffers. Enough said.
The Flu Front
Not to be overly concerned, but a brief CDC report demonstrated spread of neuraminidase mutations in influenza A H1N1 strains in the US, showing reduced susceptibility to oseltamivir. We'll hope these don't become more common.
On the positive side, a NEJM article demonstrated that heat readily inactivates the current influenza A H5N1 strains showing up in cows' milk. I didn't see any earth-shattering news on H5N1 in the past week, but I happened upon the AAP Red Book's outbreak pages for the topic and really liked their bullet summary for current status in different populations:
People: 3 cases (in 2024)
Person-to-person spread: None
Current public health risk: Low
Dairy Cows: Ongoing multi-state outbreak
Wild Birds: Widespread
Poultry Flocks: Sporadic outbreaks
Mammals: Sporadic infections
Apparently these outbreak pages are free to the general public, so check it out. (IMHO every pediatric healthcare provider should have a Red Book subscription, included in AAP membership.)
Covid Vaccine Backtracking
Initially following the FDA VRBPAC meeting on June 5, FDA recommended that the JN.1 covid strain be utilized in the next iteration of covid vaccines. On June 13 they amended this, stating that "if feasible" the KP.2 subvariant should be used instead. What's going on?
After a day-long discussion, the VRBPAC members were asked to vote on whether or not to choose a monovalent JN.1-lineage strain to use in the next vaccine. The vote was unanimous in favor of this. As seen in the section of the lineage chart below, the JN.1 lineage includes that purple JN.1 strain at the left, as well as all the subvariants derived from it.
The VRBPAC vote didn't specify which of all those strains to pick. The discussion following the vote did address that, with the important concerns being whether newer subvariants like KP.2 and KP.3 might be dominant this fall and could evolve further to be more antigenically distinct than JN.1. In that scenario, a JN.1-based vaccine might be less effective. On the other hand, KP.2 might prove to be a worse choice if its derivatives became more antigenically distinct from other subvariants that might predominate in the fall. This is too tough to predict now. (I was also interested that CDC stated they aren't emphasizing research on using generative artificial intelligence as a predictive tool.)
I was monitoring the VRBPAC meeting in real time, and it was clear that Dr. Peter Marks, director of the FDA's Center for Biologics Evaluation and Research, leaned more towards using the KP.2 subvariant, and I guess eventually this view prevailed. I'm not quite clear why the initial guidance didn't say that since I don't see any startling new information this past week. Here's the latest variant picture from the same link as above.
As always, the last 2 fortnight periods are only estimates; in the past, these estimates have been fairly accurate predictors. The 6/8/24 bar isn't that different from 5/25/24 in that KP.2, KP.3 and LB.1 seem to be trending towards dominance while JN.1 itself fades. It's important to remember that in general across the US covid activity is pretty low, with a few spots of minor uptick but nothing approaching a big surge so far.
Shifting towards the KP.2 variant as the vaccine component won't affect the Moderna and Pfizer planning, they were already working on both JN.1- and KP.2-based vaccines. However, Novavax won't be able to supply a KP.2 vaccine for another 6 months. They would however be able to have a JN.1 vaccine by fall. That "if feasible" phrase in the FDA announcement seems to leave the door open to allow Novavax to continue with JN.1 vaccine production; the company submitted an FDA application for authorization of this vaccine on June 14, the day after the new FDA announcement. I'll be interested to see how the discussion goes at the CDC ACIP meeting in a couple weeks.
The Eight Queens Puzzle
I was looking up the term backtracking to see where it came from (apparently it showed up in 1870), but instead found another use of the term in computer science, specifically computer algorithms. That led me to the eight queens puzzle.
Apparently there are 92 separate solutions to this puzzle, first published in 1848. Subsequently the puzzle was expanded to all natural numbers, the n queens puzzle, for which solutions exist for all of them except the numbers 2 and 3. The chessboard then has n rows and columns. This is a dangerous rabbit hole into which I hope I never fall.
Happy Father's Day to all you dads, granddads, step-dads, and every other iteration!