Does fertility treatment raise cancer risk? What 1.7 million women show

It is one of the most common questions asked before starting treatment, and one of the hardest to get a straight answer to: does the hormonal stimulation used in IVF increase the risk of cancer?

A study published in JAMA Network Open in July offers the largest and most recent answer available.

What was measured

Walker, Venetis, Opdahl and colleagues linked health records for 1,748,927 Australian women aged 18 to 55, followed from 1991 to 2018. Of those, 396,661 had undergone medically assisted reproduction — a group large enough, followed long enough, to detect effects that smaller studies would miss.

They looked at hormone-related cancers: breast, ovarian, endometrial and thyroid.

Hazard ratios ranged from 1.09 to 1.64. The largest was for non-serous ovarian cancer after assisted reproduction, at 1.64 (95% CI 1.31–2.05) — that is, a 64% higher relative risk.

Why the relative number is misleading on its own

A 64% increase sounds alarming. Here is what it means in absolute terms: for most of the treatments examined, the elevated risk translated to fewer than 20 additional cancers per 100,000 women per year.

That is the number that actually matters for a decision. Hormone-related cancers are uncommon in women of reproductive age, so a large proportional increase on a small baseline is still a small absolute number. Reporting the first figure without the second is how genuinely reassuring data ends up sounding frightening.

The detection-bias explanation

The authors highlight a pattern that shapes how the whole finding should be read: the elevated risk clustered immediately after treatment and diminished over time.

If fertility drugs were causing cancer through a biological mechanism, you would expect risk to appear years later, not immediately. A cluster right after treatment fits a different explanation: women going through fertility care undergo intensive medical surveillance — repeated ultrasounds, blood work, physical examinations. That finds existing cancers that would otherwise have been diagnosed later, or not at all.

The authors interpret their results as consistent with detection bias rather than a direct causal effect of treatment. That is not a way of dismissing the finding; it is the most likely explanation for its timing.

The limitations, stated plainly

This is observational registry linkage, not a randomised trial, and it carries real constraints:

• No data on the underlying cause of infertility. This matters more than it might seem. Some conditions that cause infertility — endometriosis, PCOS, never having been pregnant — independently affect cancer risk. Untreated infertility is itself associated with higher rates of some of these cancers, so separating the treatment from the condition is genuinely difficult.
• No data on race, ethnicity, smoking or individual socioeconomic status, all of which influence cancer risk.
• Industry ties disclosed: two co-authors reported pharmaceutical grants or fees, and one reported ownership of stock in a fertility clinic group.
• Australian population, so it may not transfer perfectly elsewhere.

What to take from it

For someone deciding whether to proceed with treatment, the practical summary is this: the best available evidence, from nearly 1.75 million women followed for up to 27 years, finds small absolute increases in hormone-related cancers after fertility treatment, concentrated in the period right after care, and most consistent with more thorough medical attention rather than harm from the drugs themselves.

That is a reasonable basis for reassurance. It is not a basis for skipping screening. Whatever route you take to parenthood, keeping up with routine cancer screening for your age group remains sensible — and if you had fertility treatment, your regular doctor should know.

If you have a personal or family history of breast, ovarian or endometrial cancer, or a known genetic predisposition, this is a conversation to have with a specialist before starting, not a question to settle with a study average.

This article is general information, not medical advice. To see how a treatment plan is built around your own history, read the patient journey.