October 21, 2026

expert reaction to cohort study looking at childhood cancer risk after assisted reproductive technology

Scientists comment on a cohort study, published in JAMA, investigating childhood cancer risk after assisted reproductive technology.

Prof Jason Kasraie, President of the Association of Reproductive and Clinical Scientists (ARCS), said:

“This interesting national cohort study, which looked at over a million children born between 1996 and 2012 included 37,228 children conceived by assisted conception. The authors conclude that two assisted conception procedures are associated with an increased childhood cancer risk, although the overall risk was still very low. Whilst large population-based studies are useful, and there is nothing major inherently wrong with the design of this study, the conclusions should be viewed with caution for a number of reasons:-

1. The relative risk of childhood cancer for all patients, whether they had assisted conception or not, is still very low.
2. Whilst the study identifies two assisted conception procedures that may be associated with an increased risk of childhood cancer (Intra-Cytoplasmic Sperm Injection (ICSI) and Frozen Embryo Transfer (FET)). It is very reassuring to see that there was no increased risk identified for children conceived by standard In-Vitro Fertilisation (IVF).
3. The absolute number of cancer cases is very low, for example there were only 23 cases overall in the FET group and 66 in the ICSI group and the number of cases for individual cancers is much lower, with less than five for most of the childhood cancers reviewed in FET cases.
4. Severe male factor infertility has not been controlled for, and as this is the main indication for ICSI treatment, we cannot rule out the effect of male factor infertility, which is associated with higher rates of sperm DNA damage, on cancer risk (i.e. any increased risk could be related to male partner fertility issues rather than the ICSI procedure itself).
5. The study relates to children born between 1996 and 2012, which is so long ago that the majority of assisted conception procedures have changed significantly. During this time embryos were predominantly frozen using a technique known as ‘slow freezing’, often after three days of development whereas now embryos are frozen using a technique known as ‘Vitrification’, usually after five or six days of development.
6. The study does not identify whether embryos utilised in FET were originally fertilised using IVF or ICSI treatment, meaning that any increased risk could be inherently associated with ICSI, and thus severe male infertility, rather than the FET procedure itself.

“The study does serve as a reminder that procedures should only be performed in line with national guidelines and when absolutely necessary, with ICSI usually only being performed for male factor infertility and FET only being performed when there is no option to transfer the embryos in a fresh treatment cycle.

“Parents of children born through ICSI or FET should be reassured that the overall risk of childhood cancer remains low and that further work is needed to determine if the observations in this study truly represent an increased risk of childhood cancers when these procedures are undertaken.”

Prof Yacoub Khalaf, Consultant in Reproductive Medicine and Surgery, Guys and St Thomas’ NHS Foundation Trust, Professor of Reproducible Medicine at King’s College London, said:

“The major is strength of the study is the large cohort and the dataset that shows some consistency of the association between ICSI and frozen embryo transfer on one hand and childhood cancer of certain types on the other. This is best available evidence. 

“Having said that, this association is not adequate to establish causation therefore can’t be interpreted as such. The action that should follow this study, is to try to replicate this observation by analysis of data that may be collected by other registries if replicated, it would lend presence to the findings of this study. Additionally, some scientific robust scientific exploration of the potential mechanism by which these interventions may be implicated in the genesis of such serious consequences. This work is of a fundamental importance as without which, biological plausibly would remain questionable given the very low, albeit significant, increase in the incidence of these serious cancers.”

Prof Dusko Ilic, Professor of Stem Cell Sciences, King’s College London (KCL), said:

“Statistical significance should not be confused with biological significance, and certainly not with evidence of causation. Large datasets and sophisticated statistical analyses can identify associations, but they cannot compensate for a lack of biological understanding. This study links broadly defined fertility treatments to childhood cancer without identifying which, if any, aspects of those treatments might be responsible. Such findings risk generating unnecessary anxiety among patients and families, particularly when the absolute risks are very low and the treatments studied largely reflect clinical practice from decades ago.”

Dr Rubina Ahmed, Director of Research, Policy and Services at Blood Cancer UK, said:

“This is a large and carefully conducted study, looking at more than a million children in Denmark over several decades. The researchers have tried to understand whether the increased risk might be linked to the treatments themselves rather than underlying fertility problems. But while the findings are worth taking seriously, they cannot tell us that fertility treatment causes childhood cancers.

“The researchers found higher rates of childhood cancer following frozen embryo transfer and a technique known as ICSI, but not conventional IVF. However, even where an increased risk was identified, the likelihood of developing cancer remained very low. Looking specifically at blood cancers, the possible increased risks were only seen for certain types, including leukaemia following frozen embryo transfer and lymphoma following both treatments. The link between frozen embryo transfer and childhood leukaemia was based on just six cases, so these findings need to be treated with caution. As fertility treatments have changed since these children were conceived, we also cannot assume the same results would be seen today.

“Blood cancer is the most common type of childhood cancer, and understanding what causes it remains incredibly important. When Blood Cancer UK was founded in 1960, just one in ten children with leukaemia survived for five years. Today, nearly nine in ten do, thanks to decades of research and advances in treatment. Studies like this can help us explore why childhood cancers develop, but for families, the important message is that these findings do not establish cause and effect, and the risk of childhood cancer in children conceived through fertility treatment or otherwise, remains low.”

Prof Alastair Sutcliffe, Professor of General Paediatrics, University College London (UCL), said:

“This is a well-conducted analysis of high-quality Danish national registers, with long follow-up and near-complete cancer ascertainment, and those are genuine strengths.

“However, it is an observational study, and it can show only an association between some fertility treatments and childhood cancer, not that the treatments cause cancer.

“The most important point for parents is that the absolute risks are very small. Even in the highest-risk group, frozen embryo transfer, the study estimates around 41 cancers per 100,000 children per year, compared with around 18 in naturally conceived children. That means well over 99% of children conceived this way will not develop cancer in childhood.

“The cohort is large, but the number of affected children in each treatment group is small. The overall frozen embryo transfer finding is based on 23 cancers, and several of the specific cancer-type findings rest on fewer than five cases each. This explains the very wide confidence intervals.

“The authors examined around 30 combinations of treatment type and cancer type without correcting for multiple testing. With this many comparisons, some ‘significant’ results would be expected by chance alone. The authors rightly describe these analyses as exploratory, and they should be treated as hypotheses, not findings.

“The study didn’t account for some well-known influences on children’s health, such as the mother’s age, education, smoking and family size, so the differences it found might be partly or wholly explained by those factors rather than by what was being studied. The study also has no information on fathers, which matters greatly because ICSI is mainly used for male-factor infertility. The ICSI results therefore cannot separate any effect of the procedure from the characteristics of the fathers who need it.

“Children conceived by ART differ substantially from other children. A third were twins or triplets, and they had more preterm births, low birth weight and congenital malformations, all of which are linked to childhood cancer. The paper does not examine whether these factors explain the associations, and it does not report accounting for twins from the same pregnancy, which could make the results look more precise than they are.

“The comparison with women who had a history of infertility but conceived without treatment does not fully address confounding. Those women conceived naturally, so their infertility was likely less severe than that of couples needing IVF, ICSI or frozen transfer.

“The suggestion that the procedures themselves are responsible goes further than the data can support. Apart from effects of freezing and thawing on birth weight, consistent differences in child health between IVF, ICSI and frozen embryo transfer have not been established, and residual confounding is a more likely explanation for much of what is reported here.

“This is largely an extension of the same Danish cohort that produced a similar 2019 paper, so it is not independent confirmation. The wider international evidence remains mixed, and several large studies, including from the UK, have found no overall increase in childhood cancer after assisted conception.

“The treatments studied date from 1996 to 2012. Embryo freezing methods have changed considerably since then, so the results may not apply to children conceived with current techniques.

“Couples considering or who have had fertility treatment should not be alarmed by this study. Any decisions about treatment type should continue to be guided by clinical need, in discussion with their fertility team.”

‘Childhood Cancer Risk After Assisted Reproductive Technology’ by Katrine Mose, et al. was published in JAMA at 16:00 UK time on Wednesday 21 October 2026.

DOI: 10.1001/jama.2026.18862

Declared interests

Prof Jason Kasraie: “I am currently President of the Association of Reproductive and Clinical Scientists (ARCS). I am a past Chair of ARCS. I am a past Chair of the Association of Clinical Embryologists (the predecessor to ARCS). I am a past board member of the Human Fertilisation and Embryology Authority (HFEA). I sat on the specialist advisory committee for the NICE Fertility guideline, which was published earlier this year. I received funding from IBSA to attend the European Society of Embryology and Human Reproduction (ESHRE) meeting this year. I am on the scientific steering committee for the Merck ‘Masteclass’ Meeting which will take place in November and will receive remuneration for Chairing a session at the meeting.”

Prof Yacoub Khalaf: “I have no conflict of interest to declare”

Prof Dusko Ilic: “I declare no conflict of interest”

Dr Rubina Ahmed: “Rubina has no conflicts to declare.”

Prof Alastair Sutcliffe: “Professor Sutcliffe is a collaborator on a large international study led from Australia addressing the issues in this paper”