The genetics of ovarian performance and reserve

IVF is a complex treatment for infertility requiring costly drugs and carrying significant risk of complications. Part of the procedure aims to stimulate the ovaries to produce more eggs, and conventional methods include a combination of hormones to induce follicle growth, from which eggs are collected.

Before beginning the first IVF cycle, knowledge of individual ovarian activity may help predict the response to stimulation and allow clinicians to tailor-make treatments. Ovarian reserve tests, which assess the function and quality of the ovaries, have become a routine part of pre-IVF screening, but currently their accuracy is only modest.

Clinically speaking, we need to individualise and optimise these methods to reduce the risks of extremes of ovarian response while maximising the probability of live birth. Over the last 50 years it has become evident that studying the genetic mechanisms behind variations in drug response in different individuals can help control and minimise adverse effects of drugs and their related actions.

The interaction between follicle stimulating hormone (FSH) and its receptor is necessary for normal ovarian activity, and studies have shown that variations in the (FSH) receptor gene may be associated with abnormal reproductive activities (1).

Hypothetically, a genetic test looking at the variations in this FSH-receptor gene may help identify the dosage required for a satisfactory ovarian response.

Screening for FSH-receptor mutations in hundreds of subjects in different populations worldwide has revealed two frequent genetic variations (each altering just one DNA letter)  in one part of the FSH-receptor gene, called exon 10, that codes for its protein (2).

One of these variants (at position 680 of exon 10) has been linked to differences in FSH action during ovarian stimulation. It was thought that those with a recessive form of the variation had a poor ovarian response, but other studies failed to confirm this (3).

On the whole, there is lack of consistency between the various studies investigating the relationship between the FSH-receptor gene variants, controlled ovarian stimulation and the outcome of treatment (4). 

Indeed, we undertook the largest prospective study in the available literature, and did not see a relationship between FSH-receptor gene variations and the outcome of ovarian response, and certainly could not use this as a method of prediction (5).

We also studied the relationship between other currently used markers of ovarian reserve, including hormone levels (FSH and anti-Mullerian hormone) and the ultrasound marker, antral follicle count. Again, we could not see any association between markers and the FSH-receptor gene variants (6).

We surmise that the outcome of ovarian stimulation could be dependent on several factors - both genetic and non-genetic - in a complex multi factorial model. In the future, genome wide screening strategies may allow the discovery of further genetic variants associated with ovarian response to stimulation.

However, there is currently a lack of evidence to suggest that genotyping women scheduled for ovarian stimulation would be a useful test in tailoring the drug dose. Therefore, FSH-receptor gene variant assessment should not be used to direct clinical management at present.