Case: A 43-year-old woman with normal BMI presents with desire to conceive. She has previously failed 5 IVF cycles, including 3 involving preimplantation genetic screening (PGS) attempts. None of the 3 PGS cycles reached embryo transfer. Her FSH is 11.4mIU/mL with normal range estradiol; her AMH is 2.9ng/mL; her total and free testosterone are mildly below mid-point of normal range; her DHEA is 246 ng/mL, her DHEAS is, however, 64ug/dL; and her sex hormone binding globulin (SHBG) is 134nmol/L. The patient had a history of Hashimoto’s thyroiditis, had abnormal elevations of IgM and IgE immunoglobulins and, based on elevated IL-6, evidence of inflammation.
Learn More about PCOS
Analysis: This is a woman of advanced age with normal BMI who presents with repeated IVF failures. Her FSH is within expected range for age, though somewhat elevated. Her AMH is, however, unusually high for her age. This discrepancy between FSH and AMH, both reflecting functional ovarian reserve (FOR), is one key point in resolution of this case.
Total and free testosterone as well as DHEA appear in normal range; DHEAS is, however, abnormally low, while SHBG is abnormally elevated. This is a fairly typical presentation of what we, here at CHR, have come to call the hypoandrogenic polycystic ovary syndrome (haPCOS).
Background: The discovery of this new haPCOS phenotype was initiated at CHR with a study published in 2013 in the prestigious Journal of Clinical Endocrinology and Metabolism (Gleicher et al., Clinical relevance of combined FSH and AMH observation in infertile women, 2013;98:2136-45). This study revealed that, amongst all possible FSH/AMH combinations, the combination of high/high overall resulted in best IVF outcomes. Yet, at the same time, a combination of abnormally high FSH and abnormally high AMH was contradictory because, as FOR declines, both hormones are expected to go into opposite directions (FSH up, and AMH down). This unexpected finding initiated further research at CHR, which, ultimately, led to the discovery of the new haPCOS phenotype.
It is characterized by exactly the hormone profile above described patient demonstrates: Her FSH as well as AMH are abnormally high. While her testosterone and DHEA appear in normal range, they actually are not. This can be assumed because her DHEAS is very low, and especially because her SHBG is very high (it usually goes into the opposite direction to testosterone). Combined, these data suggest that the patient’s seemingly normal testosterone and DHEA levels are actually, likely, relatively hypoandrogenic.
This conclusion should not surprise: Her still very high AMH of 2.9ng/mL at age 43 strongly suggests that, when this patient was younger, her AMH probably was at PCOS levels. Since she has a normal BMI, and never was obese, she at younger years, likely, was a non-obese (“skinny”) PCOS phenotype, which in the literature has been differentiated from the “classical” PCOS phenotype with truncal obesity and obvious signs of hirsutism.
In continuing the investigation, CHR investigators then discovered that PCOS patients at CHR were almost exclusively of the “skinny” phenotype. Since over 90 percent of CHR patients failed prior fertility treatments elsewhere, this observation suggested that “classical” PCOS patients, likely, conceive more easily and earlier during fertility treatments than “skinny” PCOS patients (For further details, please see Kushnir et al., Relative importance of AMH and androgen changes with aging among non-obese women with polycystic ovary syndrome, J Ovarian Res 2015;8:54).
Like women with “classical” PCOS, “skinny” PCOS patients at young ages, however, are also hyperandrogenic; yet, older women with likely haPCOS, as ours studies revealed (same Kushnir et al reference), usually had become severely hypoandrogenic and, at best, barely reached mid-range normal levels, as here presented case. However, even haPCOS patients with seemingly normal testosterone levels, have to be considered functionally hypoandrogenic because their ovaries’ tissue memory is, likely, expecting higher testosterone levels of their younger years. Therefore, we consider this PCOS phenotype to be hypoandrogenic and coined the term haPCOS.
We, since, have had the opportunity to see quite a number of relatively younger haPCOS patients and were able to add additional clarifications. For example, we now know that the hypoandrogenism of haPCOS is not the classical age-dependent decline in androgens seen in women after age 40. haPCOS patients suddenly drop their androgen levels already at relatively young ages (late 20s to mid-30s).
Why that happens is still unknown, though CHR’s strong suspicion is an autoimmune attack on the zona reticularis of the adrenals, where adrenal androgen production takes place. The principal reason for our suspicion is, like in the here reported case, low DHEAS, which is almost exclusively an adrenal product. Autoimmunity is suspected because, as in the here reported case, haPCOS phenotypes almost always demonstrate significant evidence for autoimmunity and/or inflammation. Moreover, in a small number of haPCOS patients, we actually diagnosed previously unknown autoimmune-induced classical adrenal insufficiency. Confirmation of an autoimmune anti-adrenal immune response causing the observed hypoandrogenism in haPCOS patients, however, awaits experimental confirmation.
Treatment: Low intra-ovarian testosterone levels adversely affect early stages of follicle maturation. haPCOS patients, therefore, require androgen supplementation, which we in most cases provide with dehydroepiandrosterone (DHEA), though sometimes with testosterone directly, following in principle the same protocol as in women with low FOR. Since in haPCOS patients higher levels of “normal” testosterone are likely required than in non-PCOS patients with low FOR, the one difference is that we carefully watch testosterone as well as SHBG levels, and will not initiate IVF cycles until SHBG drops below 80.0nmol/L.
Conclusions: Once the features of the new haPCOS phenotype were identified, it was quite surprising how often patients with haPCOS are seen in daily practice at CHR. Here described case report represents the fairly typical clinical presentation of these patients. The patient conceived in her second IVF attempt at CHR, and delivered, after receiving immunosuppressive therapy during her IVF cycle and into early gestation.
Since CHR for over 10 years has been routinely preparing women above age 40 for IVF with DHEA, we, for all of this time, unknowingly, have been treating women with haPCOS correctly. When we reported in 2013 CHR’s outcome experience in IVF based on FSH and AMH combinations, we were unable to explain our findings (see above reference Gleicher et al). It took CHR’s investigators over three years to figure out the mystery. We now fully understand why women with elevated FSH and high AMH do so well at CHR, and are in the process of preparing a manuscript for publication, describing statistical outcome data for a few hundred haPCOS patients.