Unlike many expensive medical treatments that are covered by health insurance, infertility treatment is usually paid for out of pocket. A recent publication estimated the median cost of an IVF cycle in the United States at $24,373, with the cost of an unsuccessful IVF treatment even higher ($25,921) . A significant portion of the total cost of IVF is spent on gonadotropin injections. This observation is especially true in poor responders who are usually treated with high daily doses of FSH, at times exceeding 600 IU a day. Unfortunately, after several days of injections and a few thousand dollars spent on fertility drugs, the cycle may be cancelled for lack of adequate response, or oocyte retrieval will be pursued with less than the desired yield of oocytes.
In the present study, we show that a significantly higher concentration of serum FSH following administration of exogenous FSH did not result in a higher number of oocytes and embryos. The administration of similar doses of FSH resulted in a wide range of serum FSH concentrations (6.8-36 IU/L) on CD 7 that negatively correlated with oocyte yields. The steady state of any substance in the serum represents its rate of intake and clearance. In our clinic we see the patients on CD3 when they start gonadotropin administration and again on day 7. In between these days the patients are not monitored and no changes are made to the dose of gonadotropins. These five days between the start of gonadotropin injections and the next measurement of its serum concentration is the time in which injected gonadotropins were shown to reach a steady state. Despite the lack of significance, it is clear that there is a correlation between the injected dose of FSH and its serum concentration. However, the narrow range of daily FSH dose,(200–300 IU a day) was associated with a much wider range of serum FSH concentrations, with the highest concentrations being 6 fold greater than the lowest. This observation suggests that the rate of FSH clearance plays an important role in its serum concentration. This study is retrospective and, therefore, possibly subject to unrecognized biases. In order to reduce the impact of age on endogenous FSH concentrations, we only included cycles with a long mid luteal GnRH agonist protocol in order to reduce the contribution of endogenous FSH as much as possible. As seen in Figure 1, the cycle day 3 FSH concentration was similar in all 4 age groups (average of 5 IU/L) after pituitary downregulation by GnRH agonist. Moreover, only cycles with recFSH were included in order to prevent differences in the rate of clearance that are attributed to difference in acidic isoforms between recFSH and urinary gonadotropins . Although long luteal protocols are considered to be overly suppressive for some older patients, we felt that the use of this protocol was the only way to eliminate the endogenous FSH differences between groups that would preclude any meaningful interpretation of the data.
In our study, serum concentration of FSH on CD7 showed a significant negative correlation with oocyte and embryo yield, as well as pregnancy rates despite the administration of a similar dose of FSH. The changes in serum FSH concentration on CD7 cannot be attributed to the baseline FSH or to the patient’s age since both did not show a significant correlation.
At least part of the process of clearance of FSH from the circulation involves binding of FSH to its receptor and internalization of the hormone receptor complex. Therefore, a low number of follicles containing FSH receptors on granulosa cells will inevitably result in the accumulation of FSH in circulation as a result of less binding of FSH to FSHR .
We observed that the serum FSH concentration on CD7 predicted response to FSH administration with a dramatic change in oocyte retrieval numbers and pregnancy outcome above a serum FSH concentration of 22 IU/L. Our results demonstrate that there was excessive circulating FSH in poor responders and suggest that increasing the dose of FSH would not be helpful since the FSH administered already was not being completely used. This observation is consistent with our clinical experience that doses of FSH above 300 IU per day are unlikely to increase follicular response in older women or previous poor responders, and only adjunctive measures are likely to be of help in increasing oocyte yield. These include sensitizing the follicles using androgens prior to ovarian stimulation  or by adding LH or hCG to recFSH during ovarian stimulation . Alternatively, if the FSH serum concentration on CD7 is less than 22 IU/L, there may be a place for increasing FSH dose if follicular response is less than desired.