The ‘10ng’ DNA safety limit for vaccines is a fallacy

The 10 ng safety limit for residual DNA in vaccines has jumped 1,000-fold over the decades. An investigation reveals the global standard is built on assumptions—not robust evidence.

International drug regulators have set a safety threshold that allows up to 10 nanograms (ng) of residual DNA per vaccine dose—a trace amount they claim poses no risk to human health.

It was originally 1,000 times lower—at just 10 picograms (pg) but over the decades, that threshold quietly increased without fanfare or public scrutiny.

Regulators insist that the 10 ng limit is supported by sound science. But is it?

I set out to investigate what evidence—if any—supported this dramatic shift. When I asked the American, European and Australian drug regulators for human data to justify the 10 ng threshold, none could provide any.

The European Medicines Agency (EMA) responded that there is “no fixed limit” of residual DNA and that “higher amounts of DNA can be accepted if the DNA is sufficiently fragmented.”

The US Food and Drug Administration (FDA) initially agreed to respond but, after multiple delays, stated: “We have nothing further to share.”

Australia’s Therapeutic Goods Administration (TGA) also failed to produce any human data supporting the limit.

Rather than citing clinical studies, regulators pointed to the “millions of doses” administered without obvious harm.

So, despite the glaring absence of human safety data, the 10 ng threshold has become the global standard.

Residual DNA in vaccines

Components of many modern vaccines are made with recombinant DNA technology—a method that relies on plasmids, small circular pieces of DNA engineered in the lab to carry genetic instructions into cells.

Gardasil, for instance, uses plasmid DNA grown in yeast cells to manufacture the L1 protein, a component of the human papillomavirus (HPV) outer shell.

Covid-19 vaccines on the other hand, use plasmid DNA to produce messenger RNA, which instructs the body’s own cells to make the spike protein.

Manufacturers are supposed to eliminate or degrade this DNA during the production process—but it’s an imperfect system. Residual DNA fragments often remain in the final product.

These fragments are formally referred to as “process-related impurities,” but in plain terms, they are contaminants.

This raises the crucial question: how much of this DNA contamination is actually safe to inject into the human body?

The origins of the 10 ng limit

In the 1970s, as biotech bloomed with drugs like interferon grown in human cells, there were fears that residual DNA could carry oncogenes—genes that might trigger cancer—or viral sequences that could infect recipients.

In 1984, with little to no human data available, regulators adopted a conservative cap of 10 pg per dose, loosely based on an earlier polio vaccine grown in monkey cells that had not been associated with known harm.

By 1986, the World Health Organization revised the safety limit to 100 pg, reasoning that the cancer risk remained low.

But these were educated guesses—nothing more. There were no long-term safety trials, and no direct evidence that 100 pg was a biologically meaningful threshold.

By the 2000s, recombinant vaccines—like Gardasil, which use plasmids in yeast cells to produce HPV proteins—pushed the debate forward.

In 2008, a study by Sheng et al. showed that extraordinarily high doses—9,400 ng—of plasmid DNA could cause sarcomas in mice. These data were used to infer that lower doses were probably safe.

Then in 2013, statistician Dr Harry Yang proposed a new safety limit of 10 ng per vaccine dose.

His argument was not based on any human data, but a mathematical model suggesting that fragmented DNA at less than 200 base pairs would be unlikely to contain a full oncogene, and therefore pose little risk.

Yang reasoned that the manufacturing process would sufficiently fragment and inactivate the DNA, creating a “large safety margin.”

A flawed foundation

Despite its global acceptance, the 10 ng limit rests on a foundation of assumptions, not evidence.

Yang’s model was never validated in humans. It drew on animal studies and cell lines and made broad assumptions about how DNA behaves inside the human body.

Moreover, the notion that smaller fragments are safer is increasingly contested by genomic experts.

Cancer researcher Professor Philip Buckhaults warns that breaking DNA into smaller pieces may actually make matters worse.

“Instead of one big piece, you’ve got a bazillion little ones,” he said. “They’ve turned a shotgun slug into buckshot pellets.”

Extensive lab evidence backs him up.

A study in Nature Scientific Reports showed that roughly 7% of transfected cells could integrate foreign DNA within hours.

Buckhaults’ own lab showed that plasmid DNA from Pfizer’s Covid-19 vaccine integrated into the genome of human colon organoids and persisted for over a month.

Genomic researcher Kevin McKernan has also found that DNA fragments in Pfizer’s vaccine were capable of integrating into the genome of ovarian cancer cells in culture.

And integration is only one of several concerns.

More than just genomic integration

Residual DNA in the cytoplasm of human cells can trigger innate immune responses with potentially damaging consequences.

For instance, residual DNA in cells can activate the cGAS-STING pathway - when overstimulated, this pathway can lead to cancer development.

Dr Sin Hang Lee, a molecular biologist, found that DNA fragments in Gardasil can trigger Toll-like Receptor 9 (TLR9). When chronically activated, it can lead to autoimmune disorders like POTS (postural orthostatic tachycardia syndrome).

There’s also growing concern in the field of epigenetics, where DNA fragments may reprogram gene expression without altering the underlying genetic code.

The danger may not lie in a single vaccine dose—but in cumulative exposure over time.

Typically, a child receives up to 68 vaccine doses before reaching adulthood. Yet no regulator has evaluated the total DNA burden created by repeated injections over a lifetime.

Importantly, earlier safety limits were based on “naked” DNA, which they assume degrades once injected. But aluminium adjuvants or mRNA vaccines with lipid nanoparticles, are able to transport genetic material directly into cells.

Hence, these modern vaccines with DNA contamination could have far greater biological impact than regulators have previously accounted for.

DNA contamination in vaccines

In 2023, genomics researcher Kevin McKernan ignited global awareness after discovering excessive plasmid DNA contamination in vials of Covid-19 mRNA vaccines.

Independent laboratories soon confirmed his findings, reporting DNA levels hundreds of times above the 10 ng safety threshold.

Even student researchers inside the FDA’s own laboratory detected contamination levels in Pfizer’s vaccine that exceeded the limit by as much as 470-fold.

Excessive levels of HPV DNA have been detected in the Gardasil vaccine, which the FDA summarily ignored.

Recently, court documents from litigation against Merck revealed the company intentionally retained HPV-18 DNA fragments in Gardasil because it boosted the immune response, effectively functioning as an additional, unlabelled adjuvant.

The issue of residual DNA extends beyond Covid-19 and Gardasil vaccines. Other recombinant vaccines, like those for hepatitis B or shingles—could carry similar risks if regulators don’t rigorously assess its safety in humans.

Time for science—not assumptions

The assumption that 10 ng of residual DNA per dose is safe has never been clinically verified.

Moreover, genotoxicity and carcinogenicity studies are often bypassed in favour of faster approvals.

Regulators have quietly increased the safety threshold by 1,000-fold—without producing safety data in humans.

It’s time for science to replace assumptions. That means mandatory batch testing for residual DNA in all recombinant vaccines, and long-term studies to assess its biological effects in humans.

Until then, the 10 ng threshold remains what it always has been -- a theoretical construct, untested in humans and passed off as sound science.

---