Fertility Treatment and Ovarian Tumor Risk
Recent headlines have reported that IVF treatment is linked to ovarian tumors based on a recently published study. The findings are not different than was was reported over 10 years ago, and media coverage often misleads the public into making assumptions that may not be true. Many studies in the past have reported positive or no linkages with fertility treatment and ovarian stimulation and future risk of ovarian tumors. This study from the Netherlands was retrospective in nature, meaning that they started with women who had tumors, then looked to see what type of treatments they had. This type of study can possibly find causal linkages between a treatment and an outcome, but are in no way definitive as there are many biases that are introduced. You can breathe a sigh of relief that the study did find no link between IVF and ovarian cancer in a follow up of 15 years time. The only relationship was with possible “borderline” tumors, often called “low malignant potential” tumors. These are not invasive cancer and are not considered pre-cancer tumors, though they can grow like a tumor over time. The prognosis with this tumor is usually very good and tremendously different than with invasive ovarian cancer. Below is a blog from the American Fertility Association that helps to clarify a lot of this information.
Tumors of low malignant potential
Posted by Corey Whelan on Oct 27, 2011 with 1 Comments
A new study out of the Netherlands and published in Human Reproduction is making the news today. It is entitled “Risk of borderline and invasive ovarian tumours after ovarian stimulation for in vitro fertilization in a large Dutch cohort”. Reuters picked it up as “Fertility treatment raises tumor risk in study”.
It is of course understandable that The AFA’s phones have been ringing off the hook. Women are alarmed by this report. Briefly, I wanted to share a few facts about the study and urge you to please relax while you read this.
This long term study compared women who had done in vitro fertilization (IVF) with two other groups of women; 1)the general population, and 2)sub fertile women who had not done IVF. The researchers reported an increased risk level for the IVF group of developing borderline ovarian tumors. The research also showed that the overall incidence of invasive ovarian cancer was not significantly elevated, but increased with longer follow up.
The important thing to note here is that borderline ovarian tumors are actually cysts of low malignant potential. I am not minimizing the value of this study However, the term as is being reported in Reuters and picked up by media outlets, that women are twice as likely to develop ovarian cancer after IVF, is misleading. Patient’s individual risk of getting ovarian cancer if they do IVF is not 50% higher than it would be were they not undergoing treatment. The increased risk of acquiring invasive ovarian cancer for this group is around 1.76% greater after 15 years. If you however, put together the statistics for borderline ovarian tumors along with the statistics for ovarian malignancies, the rate goes up to 4.23%. That being said, we don’t want women to wind up in either group if they can help it.
I reached out to Dan Potter, M.D. to get his opinion on the study. His comments, verbatim, were:
“This is no different than the study that was published here in the U.S. about 10 years ago. The lead author is an epidemiologist. Oftimes in studies such as this, the researcher starts with the conclusion they wish to arrive at and then they work backward to prove it. I’m not saying that’s what happened here, but these are the most problematic issues that I see in this study.
- They linked all of the ‘malignancies’ in the registry to patients in the two study groups (IVF and non-IVF) regardless of whether they returned the surveys or not. This creates a falsely high incidence. They compared them to ‘general population rates’ who basically have the equivalent of a survey return rate of 100%. See how they appeared to start at the conclusion and work backwards?
- Borderline tumors, also known as tumors of ‘low malignant potential’ or LMP are not malignancies. There is no such thing as ‘invasive cancer’ and ‘non-invasive cancer’. Cancer is invasive. Borderline tumors are not cancer and while not benign have low malignant potential. The 5 year survival rate of borderline serous tumors (the most common type) is ….…100%.
- They seem to speak of ‘invasive cancer’ and consider the borderline tumors ‘non-invasive’ cancer. The definition of malignancy is invasion so this simply does not make sense from a medical perspective.
- Their control group is women diagnosed with infertility that did not do IVF. They did not stratify these patients as to whether they became pregnant or whether they were on the birth control pill, both known protective factors. In fact, there is not control for whether the patient even had a hysterectomy or oophorectomy for endometriosis. It is hard to get ovarian cancer when you do not have ovaries. Prior use of fertility drugs in the controls may have led to pregnancy so that they did not pursue IVF. Pregnancy, BCPs, oophorectomy are all protective.
- There were 19,146 patients in the IVF group and 28 ‘invasive cancers’. Using their methodology they would have expected 21. Both of these number are very high compared to what we would expect in the US (about 10x higher). Not sure why they diagnose it more frequently there.”
So, to sum it up. You should absolutely discuss this study with your physician and weigh your own personal risk, based on medical and family history and genetic background. And, if you want to talk about it, please feel free to call.
Great news about fertility treatment from the Netherlands!
The study, a large cohort of Dutch women with infertility, showed that
the incidence of invasive ovarian cancer was NOT increased in either IVF
treated women or Non-IVF subfertile women followed for up to 15 years
The study of 19,146 women who underwent IVF and 6,006 women treated with
lesser forms of fertility therapy over a period of 13 years, found 28
and 9 women respectively developed invasive ovarian cancer, (37 cases /
25,152 women) a rate that was the same as the general population in the
9 of the 28 cases of invasive ovarian cancer found in IVF treated women
occurred beyond the initial 15 year observation period when the authors
would have expected to find approximately 3. No medical information
about other exposures that these women may have had during those
intervening years was accessible.
Borderline ovarian tumors, a non-lethal ovarian growth that may never
become cancer but may require surgery, were seen more frequently in
women who underwent IVF compared to Non-IVF subfertile women.
There are many risk factors for ovarian cancer and non-lethal borderline
ovarian tumors independent of treatment including infertility itself,
use of exogenous hormones; lifestyle factors; and family history of
cancer. The current study obtained this critical information by
A limitation of the study is that while 71% of the subjects who
underwent IVF returned the 23 page questionnaire, less than half (48%)
of those who received lesser treatment responded.
Factors known to decrease the incidence of invasive ovarian cancer
including pregnancy and the use of oral contraceptives, now commonly
used in conjunction with IVF treatment, were not independently assessed.
Alan Penzias, MD