That scary “IVF and cancer” headline, and how to actually read it

TLDR: A headline linking IVF to hormone cancers sounds frightening. Read it closely and the extra risk looks small, and much of it traces back to the reasons women need treatment rather than the treatment itself.

Why does that “IVF and cancer” headline feel so scary?

Picture the moment. You are already carrying a lot. Maybe you are mid-cycle, counting days, tracking follicles, holding your breath before the next scan. Then a headline slides across your phone: something about fertility treatment and a raised risk of cancer. Your stomach drops.

Take a slow breath. That feeling is completely understandable, and you deserve a calmer, fuller picture than a headline can give. So let us sit down together and read it properly, the way a good friend who happens to love data would read it with you.

The headline traces back to a large study published in a peer-reviewed journal. Researchers followed nearly 1.75 million Australian women over almost three decades, comparing those who had medically assisted reproduction (MAR, the umbrella term for IVF and related fertility treatments) with those who conceived without it. IVF stands for in vitro fertilisation, where an egg meets sperm in a laboratory before an embryo returns to the womb. The study reported a raised risk of most hormone-related cancers after treatment, with hazard ratios landing between 1.09 and 1.64.

Those numbers are the whole story behind the scare. So the kindest thing we can do is understand what they actually say.

What does a hazard ratio of 1.09 to 1.64 really mean?

A hazard ratio (HR) is simply a way of comparing two groups. An HR of 1.0 means the two groups share the same risk. An HR of 1.5 means one group shows about 50 per cent more of something across the study. It sounds enormous, and here is the gentle catch: a big-sounding percentage sits on top of a starting number that may be very small.

This is the difference between relative risk and absolute risk, and it is the single most useful idea to carry with you. Relative risk is the headline percentage. Absolute risk is how many real women it touches.

The study spelled the absolute figure out plainly. Comparing treated women with the comparison group, the researchers estimated fewer than 20 extra cancers per 100,000 women each year. Read that again slowly. Out of 100,000 women, the signal amounts to a handful.

Imagine a woman named Sofia, 34, about to start IVF. A phrase like “up to 64 per cent higher risk” feels like a wall falling on her. Yet translated into her real life, the study’s own arithmetic points to a very small addition to a baseline risk that was already low for a woman her age. The relative number shouts. The absolute number whispers. Both are true, and the whisper is the one that describes your day.

Is it the treatment, or the reason for the treatment?

Here is the second idea, and it is the one most headlines leave out entirely.

Women rarely arrive at IVF at random. Many arrive because of conditions such as endometriosis or polycystic ovary syndrome (PCOS, a common hormone condition that disrupts ovulation), and many have yet to carry a pregnancy to term. Researchers call this confounding by indication: the reason for the treatment carries its own risk, separately from anything the treatment does.

That matters, because those very conditions are known to nudge hormone-cancer risk on their own. The National Cancer Institute notes that infertility itself links to higher risks of some of these cancers, even for women who take no fertility drugs at all. Carrying a full-term pregnancy tends to lower the baseline risk of ovarian and endometrial cancer, so women who have yet to do so sit at a naturally higher starting point.

There is a second, quieter effect too. Women in fertility care receive far more scans, blood tests and ultrasounds than the general population. More looking means more finding. Researchers call this detection bias, and the study saw its fingerprint in the early years after treatment, when extra monitoring surfaces conditions that were already quietly present. The research team also built in careful cross-checks to test whether the signal was real biology, and those checks are worth a closer look.

Which cancers did the study actually flag?

It helps to see the pattern cancer by cancer, because the study looked at each one separately. For ovarian, uterine and thyroid cancer, the raised signal was real in the data, yet the researchers’ formal analysis suggested that the underlying conditions and closer monitoring could account for the whole of it. In plain terms, the arrow points back toward endometriosis, PCOS and the extra scans rather than toward the treatment.

Breast cancer showed a gentle rise that clustered in the early years after treatment, which is the classic signature of detection bias rather than a slow-building biological effect. The thyroid signal deserves the most caution of all, since the thyroid tends to be examined so often during fertility care that extra findings are almost expected.

Then come the two quiet tells that reassure the experts most. Blood cancers, which have no sensible link to fertility hormones, rose by a similar amount, which points to a background factor the study could not fully measure. Lung cancer, meanwhile, actually fell. A treatment that truly drove cancer would be unlikely to lower one type while raising another at random, so the uneven picture reads far more like statistical noise and confounding than like cause and effect. The authors’ own read was measured: the risk may be associated with treatment, yet it may reflect the women being treated rather than the treatment itself.

The condition often carries the risk the treatment gets blamed for

When you line up what research says about the underlying conditions beside what it says about the treatment, the picture softens considerably. The exhibit below gathers the well-anchored figures in one place.

What do the longer-term studies say?

The reassuring part is that this fresh study joins a shelf of earlier work that already looked hard for a treatment effect and mostly found reasons for calm.

On breast cancer, a Dutch cohort followed women for a median of 21 years and found that IVF carried a breast cancer rate in line with the general population, with a standardised incidence ratio of 1.01. That is about as close to “the same” as data gets.

On ovarian cancer, a large British analysis found a raised figure at first glance, yet the researchers showed that the excess belonged to women with endometriosis or fewer pregnancies, tracing it to patient characteristics rather than assisted reproduction. Cancer Research UK reaches a similar view, explaining that the ovarian pattern reflects factors such as number of children and breastfeeding rather than the IVF procedure.

The professional bodies land in the same calm place. The American Society for Reproductive Medicine (ASRM) reviewed the evidence and concluded that fertility drugs show no apparent rise in breast cancer risk, that any small ovarian signal likely stems from underlying endometriosis and infertility, and that confounders such as never having carried a pregnancy probably explain the associations rather than the medicines. In Europe, the European Society of Human Reproduction and Embryology (ESHRE) maintains the current guidance on how ovarian stimulation is carried out, keeping the practice under continual professional review.

None of this hands you a guarantee, and honest reassurance keeps its nuance. Estrogen is a recognised human carcinogen, and long or high hormone exposure is genuinely part of the biology of several female cancers. The fair summary is this: the absolute signal is small, and much of it appears to belong to who is being treated and how closely they are watched.

Why closer monitoring can work in your favour

The idea of detection bias can sound like a technicality, yet it carries a genuinely comforting message. Women in fertility care are seen often. They have scans, blood tests and honest conversations with specialists who know their history. That closeness is part of why a few extra findings surface early, and finding something early is almost always the kinder outcome.

So the same attention that nudges the numbers upward on paper is the attention that tends to catch things while they are small and treatable. Many women find it easier to read that monitoring as a quiet layer of care wrapped around a hopeful chapter, rather than as a source of worry.

It still leaves room for the sensible basics. The scans and check-ins that come with treatment are working for you, and they sit comfortably alongside the everyday habits that genuinely shape long-term risk.

So what can you actually do with all this?

The most powerful thing you can carry into your next appointment is your own story. The risk picture bends around your personal history, so knowing it turns a scary headline into a real conversation.

If a baby is on your mind right now, you might find a calm next read in our gentle guides to knowing your fertile window and giving yourself a three-month head start before trying.

You started this article with your stomach in a knot over a headline. You get to walk away from it holding two calm, portable ideas instead. First, a big percentage can sit on a very small real-world number, so always ask for the absolute figure. Second, the reason a woman reaches for treatment often carries risk of its own, which means a signal after IVF may be describing her history rather than her fertility care.

Another future is possible for you, and it can be a well-informed one. Be gentle with yourself, know your own history, and let your clinician walk the rest of the road beside you.

References

1. JAMA Network Open, 2026. Medically Assisted Reproduction and Hormone-Related Cancers, Australian cohort. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2851593
2. National Cancer Institute. Hormones and Cancer Risk. https://www.cancer.gov/about-cancer/causes-prevention/risk/hormones
3. National Cancer Institute. Reproductive History and Cancer Risk. https://www.cancer.gov/about-cancer/causes-prevention/risk/hormones/reproductive-history-fact-sheet
4. van den Belt-Dusebout AW, et al. Ovarian Stimulation for IVF and Long-Term Breast Cancer Risk. JAMA, 2016 (OMEGA cohort). https://jamanetwork.com/journals/jama/fullarticle/2533505
5. Williams CL, et al. Risks of ovarian, breast and corpus uteri cancer after assisted reproduction, HFEA data. BMJ, 2018. https://pmc.ncbi.nlm.nih.gov/articles/PMC6039832/
6. Cancer Research UK. Ovarian cancer risk factors. https://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/ovarian-cancer/risk-factors
7. Barry JA, et al. Risk of endometrial, ovarian and breast cancer in women with PCOS. Human Reproduction Update, 2014. https://academic.oup.com/humupd/article/20/5/748/2952619
8. Cancer Research UK. Womb cancer risks and causes. https://www.cancerresearchuk.org/about-cancer/womb-cancer/risks-causes
9. American Society for Reproductive Medicine. Fertility drugs and cancer: a guideline, 2024. https://www.asrm.org/practice-guidance/practice-committee-documents/fertility-drugs-and-cancer-a-guideline-2024/
10. European Society of Human Reproduction and Embryology. Ovarian stimulation guidance, 2025 update. Human Reproduction. https://academic.oup.com/humrep/article/41/4/498/8495097