Vaginal progesterone for luteal phase support in frozen embryo transfer cycles: does it deliver?

The window of implantation in the luteal phase is the critical period in the female reproductive process and enhancing it should be our central task. To optimize the endocrine environment during this period, vaginal preparations have been used for many years in both “fresh” cycles and frozen embryo transfer (FET) cycles. In the latter, it remains uncertain if it achieves its objective. The first doubt arises when one examines the vagina on the day of embryo transfer in patients who use the same vaginal progesterone (P) product and the same dose. In some patients, the vagina is clear, and in others, a copious amount of discharge is observed. It is known that mechanical alterations such as coitus, or chemical changes such as pH and viscosity, modify the release of drugs. In addition, changes in vaginal microbiota can influence the entire vaginal microenvironment, thus determining the disposition of drugs in the vaginal cavity and decreasing their therapeutic efficacy. ¹ These facts raise the question whether product absorption is consistent across different patients. On a large scale, this does not seem to be the case.

The published data add more to the puzzlement. A notable example is the study by Vuong et al. in which 1355 patients were treated with vaginal P, 400 mg twice daily for FET. ² Four days after starting vaginal P, a blood sample was collected 2–3 h after the morning dose. Serum P ranged between 0.873 and 207 ng/ml (table 2 in that publication). This marked difference demonstrates that vaginal P absorption is erratic, unpredictable, and may be unreliable. Moreover, when comparing the pharmacokinetic parameters and bioavailability of two vaginal P formulations—a gelatin capsule (Utrogestan; Basins–Iscovesco, Paris, France) and a vaginal tablet (Endometrin; Floris Company Ltd., Gush Segev, Israel)—notable differences were observed. ³ Following administration of the gelatin capsule, mean P concentrations reached a plateau at approximately 12 h, and remained relatively stable thereafter. In contrast, P levels following the vaginal tablet gradually declined over time. Additionally, the time to reach individual peak plasma concentration after administration of the gelatin capsule showed significant variability, as indicated by a large standard deviation. These findings highlight the challenges associated with achieving consistent pharmacokinetics across different vaginal P preparations.

Furthermore, Alvarez et al. described individualized luteal phase support in hormonal FET. ⁴ Patients received vaginal micronized P treatment at a dose of 200 mg every 8 h. On the day before embryo transfer, P was measured, and if found to be less than 10.6 ng/ml, it was concluded that the luteal phase P is insufficient. In those patients, daily injections of subcutaneous P were added to correct the lack of adequate luteal P. Surprisingly, 38% of the patients required the added P injections. All the above may lead to the fact that the routine vaginal P treatment may fail to deliver and should be replaced with a more reliable treatment.

It can be posited that the concentration of P within the endometrial tissue (target organ) holds greater significance than serum P levels. Studies have indicated that P vaginal administration leads to rapid absorption, resulting in higher concentrations within the endometrial tissue and lower systemic exposure due to the “first uterine pass effect.”^5–7 Nonetheless, serum P levels are frequently cited as an indicator of adequate P support. Labarta et al. established a cut-off serum P level at 8.8 ng/ml. ⁸ Additional research has reported serum P threshold concentrations ranging from 8 to 11 ng/ml during FET cycles; levels below this range were associated with poorer outcomes: 9.2 ng/ml, ⁹ 35 nmol/l, ¹⁰ 10 ng/ml, ¹¹ and 9 ng/ml. ¹² Importantly, consideration must be given to the source of P in the serum: either from exogenous drug source, or from a functioning corpus luteum (CL). Since the CL secretes not only sex steroids but also numerous peptides, most notably relaxin, oxytocin, and inhibin A, which may have certain protective or immunomodulatory effects. At any rate, the principal concern remains that adequate luteal support must encompass a considerable safety margin, given its critical role in achieving our primary objective: the establishment of a viable pregnancy.

To further complicate the challenge, P-driven immunologic modulation plays a critical role in maintaining maternal-fetal tolerance and promoting a homeostatic environment necessary for a successful pregnancy. These immunologic effects, which include suppression of pro-inflammatory cytokines and promotion of regulatory T-cell activity, are largely mediated systemically rather than locally within the endometrium. As such, they are primarily dependent on circulating serum P levels rather than local endometrial tissue concentrations. This distinction underscores the importance of achieving adequate systemic absorption in P supplementation, where immune factors can significantly impact implantation success and pregnancy maintenance. ¹³

The erratic pharmacokinetics of natural P vaginal suppository Pharmacokinetics was previously described as different doses of vaginal P (25, 50, or 100 mg) resulted in comparable serum concentration. ¹⁴ A similar pharmacokinetic study showed a lack of dose–response effect of a vaginally administered formulation of P, suggesting that the vaginal mucosa or the total surface area of the vagina may limit the absorption of P from the vagina. ¹⁵ In contrast, pharmacokinetics of a subcutaneous P formulation was linear over the range of doses studied. ¹⁶ There is limited direct evidence that exercise, body temperature, or co-medications significantly alter P absorption in clinical practice. However, as mentioned above, sexual intercourse may reduce P vaginal absorption. ¹⁷

It appears that there is likely a clinically significant minimum concentration of luteal serum P necessary to maintain an optimal endocrine environment during embryo implantation and early pregnancy following FET treatment. Vaginal P may not meet this requirement.

In summary

Although used routinely for many years for luteal phase support, vaginal P can be a broken reed in FET cycles especially those with no CL, and therefore should not be used alone, if at all.