However, while these relative risks sound alarming and large, concerned women should bear in mind that in absolute terms they are very small indeed and this is only a preliminary study. The researchers concluded that further studies with much larger numbers of participants are now needed to investigate this further.
Flora van Leeuwen, a professor at the Netherlands Cancer Institute in Amsterdam, and colleagues, write about their investigation in the 26 October online issue of Human Reproduction.
In vitro fertilization (IVF) is used to help couples having difficulty conceiving. The woman is given hormones to stimulate her ovaries to make eggs which are harvested and fertilized with the man's sperm in a test tube. The fertilized egg is then inserted into the woman's uterus where it hopefully results in a successful pregnancy.
The world's first "test tube" baby was Louise Brown, born in 1978 in Oldham, Greater Manchester, UK. The physiologist who helped her parents conceive through IVF was Robert G. Edwards who won the Nobel Prize in Physiology or Medicine in 2010 for developing the treatment.
However, although the treatment has been around for three decades, we know don't know if and to what extent the hormone stimulation may affect the risk of developing ovarian cancer.
So, for their preliminary study, Van Leeuwen and colleagues followed up a nationwide cohort of 19,146 women who received IVF treatment in the Netherlands between 1983 and 1995. For comparison, they also had data on 6,006 counterparts with similar fertility problems who did not undergo IVF. By cross-referring with disease registers and other sources, they managed to find enough complete data to do a risk analysis on 65% of the women.
They compared risks of developing ovarian malignancies in the IVF group with risks in the general population and the non-IVF comparison group.
The results showed that:
- After a median follow-up of 14.7 years, the risk of borderline ovarian tumors was higher in the IVF group than in the general population (Standardized Incidence Ratio, SIR, was 1.76, with Confidence Interval CI ranging from 1.16 to 2.56).
- The risk for invasive ovarian cancer was not significantly higher, but this did go up with time, such that the longer the follow up period since first IVF, the higher the risk (after 15 years the SIR was 3.54, with 95% CI of 1.62-6.72).
- After taking into account age, parity and cause of subfertility, the risk of borderline ovarian tumors was signficiantly over 4 times higher in the IVF group than the comparison group of similarly subfertile women who did not undergo IVF (hazard ratio HR 4.23, with 95% CI 1.25-14.33).
- After similar adjustments, the risk of cancerous ovarian tumors was also significantly over 2 times higher in the IVF group than the comparison group of similarly subfertile women who did not undergo IVF (HR 2.14, 95% CI 1.07-4.25).
"Ovarian stimulation for IVF may increase the risk of ovarian malignancies, especially borderline ovarian tumours."
They call for "more large cohort studies ... to confirm these findings and to examine the effect of IVF treatment characteristics".
Van Leeuwen told the BBC that "the absolute risk of these tumours is very low. But there is an increased risk of a borderline malignant tumour that needs surgery."
She said doctors should tell women about this risk but not overstate it.