First Clinical Outcomes from Mitochondrial Donation: 8 Babies Born via Three-Parent IVF

Last updated: March 2026

Eight healthy babies. Four girls, four boys, including a set of monozygotic twins. All born to women carrying mitochondrial DNA mutations that can cause devastating neurological, cardiac, and metabolic disease in their children. All born using a technique that Parliament spent two days debating in 2015, that the tabloids call “three-parent IVF,” and that took another decade to produce definitive clinical data.

The results, published in the New England Journal of Medicine in July 2025 by McFarland, Hyslop, and colleagues at Newcastle Fertility Centre, represent the first clinical outcomes from mitochondrial donation via pronuclear transfer (McFarland et al., NEJM 2025; 393:461-468). The numbers are small. The follow-up is short. The oldest child is five. But these are the first babies born from a technology that exists for one reason: to give women with mitochondrial disease a chance at having genetically related children who don’t inherit the condition.

What Mitochondrial Donation Is (and Who Needs It)

Roughly 1 in 5,000 children are born each year with pathogenic mitochondrial DNA mutations (McFarland et al., NEJM 2025). Mitochondria, the cell’s energy-producing structures, carry their own small genome inherited exclusively from the mother. When that genome is faulty, the consequences range from mild muscle weakness to fatal multi-organ failure in infancy. There is no cure.

The standard options for women who carry these mutations are limited. Preimplantation genetic testing (PGT) can screen embryos during IVF to select those with lower mutation levels. Donor eggs eliminate the risk entirely but mean the child shares no genetic connection to the mother. Mitochondrial donation offers a third path.

Pronuclear transfer works like this: the mother’s egg is fertilized normally via IVF. Then the pronuclei (the nuclear DNA from both parents, containing 20,000+ genes that determine virtually everything about who the child will be) are extracted and placed into a donor egg that has had its own nuclear DNA removed but retains healthy mitochondria. The resulting embryo carries nuclear DNA from the mother and father, and mitochondrial DNA from the donor.

That donor contribution is 37 genes. Out of roughly 20,000. It codes for energy production, not eye colour or personality or anything the child will recognise as identity. The “three-parent” label makes for good headlines. As a description of the biology, it’s doing a lot of heavy lifting for 0.1% of the genome.

The Newcastle Results

The pathway at Newcastle, the only clinic in the world licensed for the procedure by the HFEA, received 196 referrals (McFarland et al., NEJM 2025). Of those, 163 women were evaluated, 133 entered the care pathway, and 70 proceeded with some form of reproductive intervention. 32 women applied specifically for mitochondrial donation via pronuclear transfer. All 32 were approved.

22 women underwent the procedure. The outcomes:

(McFarland et al., NEJM 2025)

The critical measurement is heteroplasmy: the percentage of the mother’s mutant mitochondrial DNA that carried over into the child despite the transfer. Five of the eight children had undetectable heteroplasmy at birth. One child (carrying the m.4300A>G variant) showed 5% in blood and 9% in urine at birth, dropping to undetectable in blood by 18 months. Two children showed higher levels: 16% blood/20% urine (m.3260A>G) and 12% blood/13% urine (m.11778G>A). All remain well below the disease threshold, which sits around 80% for most variants (McFarland et al., NEJM 2025).

A companion laboratory paper reported parallel results from the PGT pathway: 36 women offered PGT, 16 of 39 achieving clinical pregnancy (41%), 18 live births, with infant heteroplasmy ranging from 0% to 7% (Hyslop et al., NEJM 2025; 393:438-449).

Complications, Stated Plainly

Three of the eight children experienced medical issues. Five had none.

One infant (carrying m.3460G>A) developed myoclonic epilepsy at seven months. It self-resolved after three months. Development is normal (McFarland et al., NEJM 2025).

A second child had a urinary tract infection. Antibiotics resolved it (McFarland et al., NEJM 2025).

The most complex case involved an infant (m.4300A>G, breastfed by a mother with pre-existing cardiomyopathy) who presented with hepatic steatosis, hyperlipidaemia, a Wolff-Parkinson-White pattern, and atrial tachycardia. Anti-arrhythmic medication normalised the cardiac issues. A Bayley-III developmental assessment at 18 months was normal. Medication is being weaned (McFarland et al., NEJM 2025). The mother herself experienced paroxysmal atrial fibrillation and extreme hypertriglyceridaemia (77.17 mmol/L) during pregnancy, both resolving postpartum.

Whether these complications relate to the pronuclear transfer itself, to the underlying mitochondrial mutations the families carry, or to ordinary bad luck is not yet clear. The sample is eight children. That is not a number that supports causal conclusions.

The Debate

The reaction split along a predictable fault line: cautious celebration from the technique’s supporters, and specific, substantive criticism from scientists who think the risk-benefit calculation doesn’t yet add up.

The supportive voices were measured. Robin Lovell-Badge at the Crick Institute called it a “cautiously good outcome and well worth the wait” (Nature, July 2025). Paula Amato at OHSU described the work as “super influential” (Nature, July 2025). Peter Thompson, CEO of the HFEA, called it “wonderful news” (Nature, July 2025). Dusko Ilic at King’s College London used the word “remarkable” (Nature, July 2025). Appropriate language for a first-in-human study with eight births and five years of follow-up: good, but careful.

Joanna Poulton at Oxford raised the comparison that matters most. She pointed out that it is “not clear that MD has any advantage over PGT for heteroplasmic disorders” (Nature, July 2025). The take-home-baby rates are similar. PGT produced 18 live births in the companion paper versus 8 from pronuclear transfer, from roughly comparable patient numbers, without the added complexity of a donor egg and nuclear transfer.

Nuno Costa-Borges in Barcelona flagged the reversal phenomenon: 3 of 8 infants showed measurable heteroplasmy (5-16%) despite the transfer being designed to eliminate it (Nature, July 2025). Maternal mtDNA carryover can expand. Marcus Deschauer at TU Munich echoed this: “Cannot be ruled out that proportion of mutated mtDNA will continue to increase over a lifetime” (Nature, July 2025). The oldest child is five. We don’t know what happens at fifteen, or thirty-five.

Heidi Mertes, a medical ethicist at Ghent, made the most pointed argument. Standard donor egg conception eliminates the risk of mitochondrial disease entirely. No nuclear transfer required. No heteroplasmy risk. No reversal phenomenon. If women who would have chosen donor conception now opt for mitochondrial donation instead, she argued, “this is actually a risk-increasing technology” (Nature, July 2025).

That last point is worth sitting with. Mitochondrial donation exists to preserve genetic relatedness between mother and child. The 37 mitochondrial genes come from a donor; the 20,000+ nuclear genes are the mother’s own. Whether that distinction justifies the added complexity and residual risk is not a scientific question. It’s a personal one, and different families will answer it differently.

Bert Smeets at Clinical Genomics added a practical observation: 8 births over 7 years, against a projection of 150 per year (Nature, July 2025). One licensed clinic in the world, processing 196 referrals to produce 8 babies. If this technique is going to matter at population scale, either more clinics need licences or the conversion rates need to improve substantially. Ideally both.

What This Means

Mitochondrial donation via pronuclear transfer is legal in two countries: the UK and Australia. The United States prohibits all heritable embryo modifications. The rest of Europe has no specific regulatory framework for the procedure.

For patients, the practical reality is narrow. Newcastle is the only licensed clinic. The pathway took 196 referrals to produce 8 live births. The pregnancy rate (36%) is within the range of standard IVF, which means the technique doesn’t appear to harm embryo viability, but it also doesn’t improve it. You are going through a more complex version of IVF to solve a specific genetic problem, not to increase your odds of pregnancy.

The open questions are real. Five years of follow-up is not a lifetime. The reversal phenomenon (maternal mtDNA expanding in the child over time) has been observed in 3 of 8 children, all currently at safe levels, all currently healthy. “Currently” is doing meaningful work in that sentence.

And then there is the Mertes argument, the one that will sit differently depending on who you are. If you carry a mitochondrial mutation and want a genetically related child, donor eggs mean giving that up. Mitochondrial donation preserves that connection at the cost of added procedural complexity and a small, not-yet-fully-characterised residual risk. If you were already considering donor eggs, mitochondrial donation adds risk to a problem that donor conception already solves. Two reasonable positions. The right answer depends on what matters most to you, and that’s a conversation for your clinical team, not a website.

For the broader field, this is proof of concept. A technique debated in Parliament a decade ago now has clinical data. Small-sample, short-follow-up, single-centre clinical data, but real outcomes from real patients making real decisions under the most regulated fertility framework in the world. Nils-Goran Larsson at Karolinska is right that the results need independent confirmation (Nature, July 2025). They also need time. The children born from this programme will be studied for years, and the answers that matter most are the ones we can’t have yet.

If you or your family are affected by mitochondrial disease and want to understand your reproductive options, send us your situation. We connect patients with specialist clinics across Europe.

Browse all research summaries or compare clinics by country.