Childhood cancer risk varied by fertility treatment in a nationwide Danish study published on 21 October 2026, but cancer remained rare in every group. Independent specialists stress that the findings show an association, not proof that any fertility procedure causes cancer.
The study published in JAMA examined 1,086,136 live births in Denmark from 1996 to 2012, including 37,228 children conceived through assisted reproductive technology (ART). Researchers used national records to follow the children through 2023. They recorded 3,736 cancers during a median follow-up of 20 years.
The absolute childhood cancer risk remains low
The overall incidence was 23.6 cancer cases per 100,000 person-years among children conceived using ART, compared with 18.3 among other children. That is a difference of about five cases for every 100,000 years of observation. A person-year means one child followed for one year: these are rates of diagnosis over time, not the percentage of children who eventually develop cancer.
Researchers estimated a 26% higher relative hazard in the ART group (hazard ratio 1.26). That does not mean the absolute chance of developing cancer increased by 26 percentage points. The likelihood remained low in both groups.
Rates differed by procedure. For frozen embryo transfer (FET), which involves thawing and transferring an embryo previously frozen, the rate was 41.1 cases per 100,000 person-years. For intracytoplasmic sperm injection (ICSI), in which a single sperm is injected into an egg, it was 27.9. For conventional in vitro fertilisation (IVF), the rate was 18.5, almost the same as the 18.3 recorded among children conceived without ART. The study did not detect an increased risk with conventional IVF.
An association is not proof of a cause
The researchers also reported associations for certain individual cancers: leukaemia, lymphoma and soft tissue sarcoma after frozen embryo transfer, and lymphoma, soft tissue sarcoma and central nervous system tumours after ICSI. But these results require particular caution. There were only 23 cancer cases in the frozen embryo transfer group and 66 in the ICSI group. The reported association between frozen embryo transfer and leukaemia rested on just six cases, according to independent experts consulted by the Science Media Centre.
Alastair Sutcliffe, professor of general paediatrics at University College London, praised the quality and duration of the Danish register data but warned that small numbers and numerous comparisons make the results for individual cancers uncertain. He said:
“The most important point for parents is that the absolute risks are very small.”
Observational studies can identify differences between groups, but not establish what caused them. ICSI is often used in cases of severe male-factor infertility, and the study could not fully account for paternal factors. The records also did not indicate whether embryos later transferred after freezing had originally been fertilised through IVF or ICSI, Jason Kasraie, president of the Association of Reproductive and Clinical Scientists, noted in his assessment.
The Danish researchers tested alternative comparison groups, including children born to women with a history of infertility and children conceived through conventional IVF. They found broadly similar results and suggested that factors related to the procedures deserve further investigation. These comparisons, however, do not eliminate the possibility that differences between parents or pregnancies help explain the findings.
What Denmark’s health registers can reveal
Denmark’s nationwide health registers allowed researchers to examine a rare outcome across more than a million births and a long period of follow-up. This is a major strength of the study. Yet even a population of that size leaves very few cases when the results are divided into specific treatments and cancer types.
There is also a question of timing. The children were conceived before or during 1996 to 2012, and fertility treatments have since changed. Embryo freezing has increasingly moved from slower methods to vitrification, a rapid-freezing technique. The results therefore cannot automatically be applied to procedures used today.
In a statement published by the Danish Cancer Society, study co-author Marie Hargreave said it remains uncertain whether the observed differences reflect underlying parental infertility or elements of the treatment itself. She argued that prospective parents should receive clear information about possible risks, while emphasising that the risk for an individual child is small.
Independent confirmation in other large registries, together with research into possible biological explanations, will be needed to understand the findings. For now, the study identifies questions about particular fertility procedures rather than offering a prediction about any individual child’s health.
