On Tuesday, Pennsylvania Governor Josh Shapiro announced devastating news: two Pennsylvania residents had died after contracting measles, the first measles-related deaths reported in the United States this year.
Both were unvaccinated. The deaths come during the worst U.S. measles resurgence in decades. Two doses of the MMR vaccine are approximately 97% effective at preventing measles.
Hours later, Health and Human Services Secretary Robert F. Kennedy Jr. used his personal X account (not the official HHS account) to accuse Shapiro of “gaslighting” the public about vaccines.
Kennedy said he had told Shapiro during a phone call that some Americans object to MMR vaccination on religious grounds because the vaccine “contained fetal tissue.”
According to Kennedy, Shapiro responded:
“There is no fetal tissue in the MMR.”
Kennedy then posted an old video of vaccine scientist Stanley Plotkin being deposed by attorney Aaron Siri and argued that Shapiro was wrong.
Siri subsequently joined the argument, declaring:
“RFK Jr. is correct.”
But is he?
There is an important piece of vaccine history underneath Kennedy’s post. There are also residual materials from the vaccine manufacturing process that deserve to be described accurately.
However, Kennedy’s central characterization, that MMR contains fetal tissue, is misleading.
Understanding why requires separating four things that Kennedy and Siri repeatedly blur together:
Fetal tissue.
A laboratory cell strain originally derived from fetal tissue.
Residual cellular material.
Fragments of DNA.
These are not interchangeable terms, and if we’re going to talk about “truth” and “misinformation,” we should get those distinctions right.
There are currently two MMR vaccines licensed in the United States: Merck’s M-M-R II and GSK’s PRIORIX.
Kennedy’s post specifically discusses M-M-R II.
The FDA-approved prescribing information tells us exactly how it is manufactured.
The measles and mumps viruses are propagated in chick embryo cell cultures. The rubella component, the RA 27/3 strain, is different. It is propagated in a laboratory cell strain called WI-38 human diploid lung fibroblasts.
That part is true, but read what the manufacturer lists in each final approximately 0.5 mL dose:
sorbitol
sucrose
hydrolyzed gelatin
recombinant human albumin
less than 1 part per million fetal bovine serum
approximately 25 micrograms of neomycin
other buffer and media ingredients
The prescribing information does not say that a dose contains fetal tissue. What it says is that the rubella virus is propagated in WI-38 cells. Manufacturing substrate and final vaccine are not the same thing.
That distinction is the heart of this entire controversy.
WI-38 is a human diploid fibroblast cell strain developed in the early 1960s.
The original cells came from lung tissue obtained following an elective abortion in Sweden. The fetus was female and approximately three months’ gestation. The cell strain was subsequently cultured, banked and propagated in laboratories.
There is no reason to conceal this history. It is well documented. It is also precisely why saying “the vaccine contains aborted fetal tissue” creates the wrong mental picture.
Manufacturers are not obtaining a new fetus every time they make a batch of MMR. They aren’t putting pieces of fetal lung into vaccine vials. They are using descendants of cells cultured from tissue obtained more than six decades ago.
This technology is one reason cell strains are so useful: cells can be frozen at early passages, expanded when needed, and used repeatedly rather than continually obtaining new tissue. Plotkin himself explained this in the deposition Kennedy is circulating.
So:
Original fetal tissue → laboratory cell strain → generations of cultured cells → vaccine manufacturing substrate.
Those concepts should not be collapsed into:
“There is fetal tissue in your child’s shot.”
This is where Kennedy and Siri make their strongest argument, and where precision becomes even more important. Can residual material from the cell substrate remain after vaccine manufacturing?
Yes.
This is not a secret. In fact, the FDA has extensive guidance addressing residual DNA from the cells used to manufacture biological products.
For widely used human diploid cell strains specifically including WI-38 and MRC-5, FDA guidance says residual DNA is not considered a safety issue, based on extensive experience with these substrates.
That doesn’t mean manufacturers ignore residual DNA. Regulators consider the cell type, amount, biological characteristics and, in circumstances where it matters, the size of DNA fragments, but this is where the rhetoric becomes particularly deceptive.
Residual DNA from a fetal-derived laboratory cell strain is not fetal tissue.
A DNA molecule isn’t a cell. A cell isn’t a tissue, and a cell strain maintained in laboratories for decades is not an aborted fetus. Those aren’t semantic technicalities. They are fundamentally different biological things.
Kennedy quotes Siri asking Stanley Plotkin during a 2018 deposition:
“Isn’t it true that MMR II contains approximately 150 nanograms cells substrate double-strand DNA and single-strand DNA per dose purposefully fragmented to approximately 215 base pairs in length?”
Plotkin replied:
“Yeah, that’s probably correct, yes.”
That exchange really happened. It appears on page 328 of the deposition transcript, but there is some important context missing. The approximately 215-base-pair measurement Kennedy and Siri are discussing comes from a 2015 paper that tested vaccine samples and reported an average of roughly 142 ng of single-stranded DNA and 35 ng of double-stranded DNA in the rubella vaccine samples examined.
That’s a legitimate observation to discuss. It does not establish the proposition:
“MMR contains fetal tissue.”
It establishes that measurable residual human DNA can remain from the manufacturing substrate. Those are different claims.
This may be the most rhetorically effective part of Kennedy’s post.
It is also an excellent example of how a number can sound terrifying while providing almost no useful information about risk. Kennedy writes that approximately 150 nanograms of residual DNA translates into roughly 646 billion pieces of human DNA.
Siri similarly calculates hundreds of billions of DNA fragments.
Hundreds.
Of.
Billions.
That sounds enormous, but DNA molecules are extraordinarily small.
If you take a microscopic quantity of DNA, chop it into microscopic fragments, and then count every molecular fragment individually, of course the number becomes gigantic. The relevant quantity Kennedy himself cites is on the scale of nanograms.
One nanogram is one-billionth of a gram, so 150 nanograms is:
0.00000015 grams.
Calling that “hundreds of billions of pieces” doesn’t make the mass larger.
It’s a little like describing a pinch of salt by calculating the astronomical number of individual sodium and chloride ions inside it. The arithmetic can be correct while the framing is designed to make a tiny quantity sound enormous.
More importantly, molecule count by itself does not establish toxicity.
Risk depends on biological properties, dose, route of exposure and what the material can actually do, not whether you can convert a tiny mass into an intimidating number of molecules.
FDA’s own guidance is especially important here. It explicitly discusses potential theoretical concerns from residual cellular DNA and ways to mitigate those concerns. Yet for extensively characterized human diploid strains such as WI-38 and MRC-5, FDA states that it does not consider residual DNA from these cells a safety issue. That is a rather important fact to omit while announcing “646 billion pieces of human DNA.”
This is another source of recurring confusion.
FDA guidance does discuss a 10 ng per dose recommendation for certain cell substrates but read the actual guidance. That recommendation pertains to continuous non-tumorigenic cell lines, such as low-passage Vero cells.
Immediately before that, the FDA specifically distinguishes extensively characterized human diploid cell strains like MRC-5 and WI-38, stating that measurement of residual DNA from those strains might not even be necessary because FDA does not consider their residual DNA a safety issue, so taking the 10-ng recommendation from one category of manufacturing cells and applying it indiscriminately to WI-38 is not an accurate reading of the document.
Siri’s follow-up post says:
“There are billions of fragments of human cellular debris and DNA…”
There can indeed be residual cellular components following biological manufacturing.
For some vaccines, the package labeling says so explicitly.
For example, the current Varivax prescribing information says the chickenpox vaccine contains residual components of MRC-5 cells, including DNA and protein.
The current Havrix label states that the hepatitis A virus is propagated in MRC-5 human diploid cells and that the final vaccine contains no more than 5 micrograms/mL of residual MRC-5 cellular proteins.
Again: none of this is hidden, but notice how rapidly the terminology changes:
Residual DNA and proteins become “cellular debris.”
“Cellular debris” becomes “fetal tissue.”
Suddenly, the public is picturing something entirely different from what’s actually in a vaccine vial. Scientific communication shouldn’t work that way.
This is the most emotionally charged claim in Kennedy’s thread.
Kennedy writes:
“Plotkin admits to butchering 76 aborted normally developing fetuses, all over three months old, in just one of the many studies that led to the development of his vaccine.”
There really was an early research study involving tissue from 76 fetuses. Plotkin really did acknowledge that during his deposition, but Kennedy’s wording leaves out crucial context from the very same deposition. When Siri asked Plotkin how many fetuses were involved specifically in making vaccines, Plotkin answered:
“There were only two fetuses involved in making vaccines.”
He then explained that the study involving the larger number of fetuses was research intended to characterize fetal cell strains and determine whether they could potentially be used for vaccine production.
Siri asked whether that work was related to vaccines. Plotkin agreed that it was “preparatory.”
That’s important. It is fair to say:
A historical research program studying whether fetal-derived cells could serve as vaccine substrates examined tissues from dozens of aborted fetuses.
It is not accurate to leave readers believing that today’s MMR vaccine is manufactured from 76 fetuses, or that 76 fetuses were somehow incorporated into MMR.
They weren’t. In fact, the cell substrate at issue in Kennedy’s own post, WI-38, traces to a single original fetal donor.
There is another problem with Kennedy’s description.
He says Plotkin “admits to butchering 76 aborted fetuses.”
That’s Kennedy’s word.
In the deposition, when Siri asked Plotkin what organs he harvested, Plotkin explicitly responded that he did not personally harvest them; coworkers had harvested a range of tissues that were then cultured.
One can have strong ethical objections to fetal-tissue research without rewriting what a witness actually testified to.
Kennedy calls Plotkin:
“the inventor of the MMR vaccine.”
That’s another oversimplification.
Plotkin was one of the most important vaccine scientists of the 20th century, and he developed the RA 27/3 rubella vaccine strain that eventually became the rubella component used in MMR, but the combined MMR vaccine was a separate development. Maurice Hilleman and his team at Merck developed the combined measles-mumps-rubella vaccine, and Merck later replaced its earlier rubella strain with Plotkin’s superior RA 27/3 strain. Plotkin himself has described that history.
That may sound like a small historical correction, but in a post explicitly accusing other people of spreading misinformation, details matter.
I want to be very clear about something. People can have sincere religious or ethical objections to the historical use of fetal-derived cell strains. That conversation shouldn’t be mocked.
It also doesn’t require anyone to distort the biology. Even the Catholic Church, perhaps the most prominent religious institution with a categorical moral objection to abortion, has extensively considered this question.
The Vatican has distinguished the abortion that originally produced a cell strain from the actions of someone vaccinated decades later. Its Congregation for the Doctrine of the Faith concluded that, when alternatives aren’t available and serious infectious disease is at stake, receiving vaccines developed using these cell strains can be morally acceptable and does not constitute endorsement of the abortion from which the original cells were obtained.
You can disagree with that ethical analysis, but the fact that a religious body can distinguish between an abortion, a cell strain derived from fetal tissue, and someone receiving a vaccine decades later demonstrates exactly why precision matters.
There is another part of the rubella story worth remembering. Rubella can be relatively mild in a child, but infection during pregnancy can be catastrophic.
The massive U.S. rubella epidemic of 1964–65 caused approximately 12.5 million infections and was associated with thousands of pregnancy losses, infant deaths and roughly 20,000 infants born with congenital rubella syndrome, including deafness, blindness, heart defects and other severe disabilities.
The RA 27/3 vaccine that Plotkin helped develop became extraordinarily effective at preventing rubella and congenital rubella syndrome, so a vaccine Kennedy is now invoking through the language of abortion helped prevent countless pregnancies from ending in miscarriage, stillbirth or devastating congenital infection.
That history deserves to be part of the discussion too.
This isn’t an abstract debate about vaccine manufacturing history. Pennsylvania just reported two measles-related deaths. The United States is experiencing its largest measles outbreak in decades. Measles is one of the most contagious infectious diseases humans face, and two doses of MMR are approximately 97% effective at preventing it. That makes the communication choices of the country’s top health official particularly consequential.
If Kennedy wanted to give Americans the full, technically accurate story, he could say:
The rubella component of M-M-R II is manufactured using WI-38, a laboratory human diploid cell strain originally derived from fetal lung tissue following an elective abortion in the 1960s. The cells have been propagated in laboratories for decades; new abortions are not required to manufacture each batch. The finished vaccine does not contain intact fetal tissue or an aborted fetus. Trace residual material from the manufacturing substrate, including DNA fragments, may remain. FDA has specifically evaluated residual DNA from extensively characterized human diploid strains such as WI-38 and MRC-5 and does not consider it a safety issue.
That’s the history. Nothing needs to be hidden. Nothing needs to be sanitized, and nothing needs to be sensationalized.
The most effective misinformation isn’t always fabricated from scratch. Sometimes, virtually every individual fact can have a kernel of truth. WI-38 really did originate from fetal tissue. The abortion really did happen. Human diploid cell strains really are used to manufacture some vaccines. Residual DNA really can remain. A study really did involve tissues from 76 fetuses. Plotkin really did answer Siri’s questions under oath, but now watch what happens when the distinctions disappear:
A cell strain originally derived from fetal tissue becomes fetal tissue.
Trace DNA fragments become hundreds of billions of pieces of an aborted fetus.
A historical study of dozens of fetal specimens becomes 76 fetuses used to make your vaccine.
Stanley Plotkin’s development of the rubella vaccine becomes “the inventor of the MMR vaccine.”
And a legitimate ethical question gets transformed into an image of fetal tissue somehow being injected into children. That is not greater transparency. It is less accuracy, and that’s the frustrating part of this entire situation.
Governor Shapiro’s sentence, “There is no fetal tissue in the MMR” is a reasonable description of what is in the finished vaccine. If you want to make it maximally precise, add another sentence:
The rubella component is manufactured using a laboratory cell strain originally derived from fetal tissue more than 60 years ago, and trace residual molecules from that manufacturing process can remain.
Both statements can be true at the same time. Because:
Fetal tissue is not a fetal-derived cell strain.
A fetal-derived cell strain is not residual protein.
Residual protein is not fragmented DNA.
And fragmented DNA is not an aborted fetus.
Those distinctions shouldn’t be difficult for the country’s top health officials to communicate. At a moment when Americans are dying from a vaccine-preventable disease, they are distinctions we desperately need them to get right.
And this controversy took another turn today.
The Pennsylvania Department of Health had already publicly confirmed two measles-associated deaths in Lancaster County. Both people were unvaccinated. They were the first measles-associated deaths reported in Pennsylvania in 35 years, but rather than simply accepting the state’s announcement, Kennedy publicly suggested that the circumstances surrounding those deaths still needed to be investigated.
Speaking Wednesday in Tampa, Kennedy said the CDC was looking into the deaths and questioned whether Pennsylvania had provided sufficient information to federal officials. STAT reported that Kennedy said the CDC was examining the circumstances surrounding the fatalities.
Separate reporting quoted Kennedy accusing Pennsylvania of “refusing to cooperate” with the CDC regarding the outbreak and the deaths. There is an important distinction here.
It is perfectly reasonable for epidemiologists and public-health agencies to review medical records and determine exactly how a death should be classified. In infectious-disease surveillance, terms like “measles-associated death” are used deliberately. A person can die from complications triggered by measles such as pneumonia or encephalitis or measles can contribute to a death alongside other medical conditions.
However, that is very different from publicly casting suspicion on whether the deaths were really related to measles after the state health department has already investigated them and formally announced them as measles-associated deaths.
Pennsylvania’s statement was not ambiguous:
The Department of Health confirmed two measles-associated deaths. Both individuals were unvaccinated.
The state has declined to release the victims’ ages, identities, or additional medical information because of patient and family privacy.
That lack of publicly available medical detail should not be confused with an absence of evidence. Health departments routinely possess confidential medical records that cannot simply be released to satisfy demands on social media, and this matters because we have seen this pattern before.
When measles deaths occur, the response from the country’s top health official should be extraordinarily straightforward:
Measles can kill. These deaths are tragic. The MMR vaccine is the most effective way to prevent measles. Please get vaccinated.
Kennedy did eventually encourage vaccination Wednesday, but at almost the same time, he was publicly challenging Pennsylvania over the circumstances of the deaths and accusing the state of withholding information.
That creates two competing messages:
“Get vaccinated because measles can be dangerous.”
And:
“But perhaps these deaths aren’t what Pennsylvania says they are.”
Those messages do not reinforce one another.
They undermine one another, and that is especially concerning because these deaths did not occur in isolation. As of August 20, the CDC had recorded 2,777 confirmed measles cases in the United States in 2026, already making this an extraordinary resurgence of a disease that the United States declared eliminated in 2000.
Pennsylvania alone has reported hundreds of cases, with Lancaster County at the center of the outbreak. The two people who died were unvaccinated. There is nothing scientifically inappropriate about reviewing a death.
There is something deeply problematic about using the existence of an ongoing review, or the fact that private medical information has not been released publicly, to insinuate that a state health department may be falsely attributing deaths to measles.
Absence of public medical records is not evidence that health officials don’t have them.
Once again, the issue isn’t merely what Kennedy technically said. It’s the impression his words leave behind.
First, the public is told that MMR contains “fetal tissue.”
Then, when two unvaccinated people die during a major measles outbreak, uncertainty is injected into whether those deaths should really be attributed to measles.
That is not how you rebuild confidence in public health.
It is how you manufacture doubt.
Although not required, any support is greatly appreciated, including preordering my upcoming book Think Like a Doctor.
No posts
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.