REVIEW AND COMMENTARY ON RECENT LITERATURE IN REPRODUCTIVE MEDICINE
The staff of THE REPRODUCTIVE TIMES here offers commentaries on recently published articles, primarily chosen for educational values—in the positive but also in the negative—for clinical purposes, and for their potential translational values to clinical medicine when addressing basic science issues.
September 6, 2024. Revised with permission from the September CHR VOICE.
SUBJECT: Endocrinology
The persistent controversy regarding supplementation of androgens in female infertility
It has been nearly 20 years since the first paper appeared in 2005, reporting increased oocyte yield in certain infertile women following androgen supplementation with DHEA. Yet androgen supplementation in female infertility has remained controversial. The principal reason for this persistent controversy has been the absence of properly powered prospectively randomized studies. Though many have been attempted, most failed because patients were – understandably – unwilling to be randomized to a placebo. As women who qualify for androgen supplementation usually have low functional ovarian reserve, once cannot blame them for not wanting to lose potentially valuable time on a placebo.
When prospectively randomized studies for whatever reasons cannot be performed, medicine usually turns to study formats with lower evidentiary levels, such as prospectively matched studies, controlled cohort studies, etc. This also happened here but results once again were not consistent, thus extending the uncertainty.
This inconsistency of results had one obvious reason, and that was the inappropriate selection of patients. Practically without exception, androgen supplementation was based on a female infertility diagnosis and not, as it should be, on hypo-androgenism in a female infertile patient population. In other words, infertile women were treated with androgen supplementation whether they were hypo-androgenic or not. Except for studies by one group of researchers, not even a single other study in the literature ever tested androgen levels in infertile study patients before supplementation. And, just as aspirin will cure a headache only if the patient has a headache, so can androgen supplementation not be expected to show effects if women are not hypo-androgenic.
Evidence can, however, also be derived from other research resources, especially animal experiments. And, while when androgen supplementation was first reported to improve female infertility,1 the reasons for these effects were unclear, collaborations between clinicians and basic scientists in subsequent years established in a mouse model why normal androgen levels are essential for normal ovarian function especially during the small growing follicle stages of folliculogenesis, when androgens act synergistically in promoting follicular growth and maturation.2 These mouse studies – and later in larger animals from sheep to apes – beyond any reasonable doubt established why and how hypo-androgenism can lead to female infertility. Why androgen supplementation in hypo-androgenic infertile women is still considered controversial, is not very clear.
Androgen supplementation can be done with testosterone directly or with its precursor, dehydroepi-androsterone (DHEA). Most clinicians prefer the latter, since it practically makes overdosing impossible, while with direct transdermal testosterone such overdosing can happen very easily. And too much androgen may be worse for ovaries than too little. Even if using DHEA or testosterone for androgen supplementation, most fertility providers to this day do not test androgen levels in infertile women.
Another major reasons for the persistent controversy regarding this subject is the lack of understanding by clinical practitioners as to why low androgen levels are harmful for female fertility. This lack of understanding then leads to all kinds of supplementation protocols, which often negate the efficacy of androgen supplementation. As already noted, normal androgen levels are especially important at small-growing stages between secondary and small preantral follicle stages. These follicles, however, still require at least 6-8 weeks to reach gonadotropin-dependency, when they become sensitive to external gonadotropins use in ovarian stimulation cycles. Androgen supplementation, therefore, must be started at least 6-8 weeks before start of ovarian stimulation. If supplementation, for example, is started only two weeks before cycle start (as one frequently can find in the literature) the follicle cohort that would have the most benefit would not be the cohort in the treated cycle but the cohort of next month’s cycle.
This lengthy introduction is necessary in preparation for a here discussed paper, which colleagues from New Zealand and Australia recently published. Based on a so-called systematic review of the literature, they concluded that pre-treatment with testosterone likely improves, and pre-treatment with DHEA likely results in little to no difference, in live birth and clinical pregnancy rates in women undergoing IVF who have been identified as poor responders.3
A systemic review of the literature does not produce new data; such a review selects what are considered well executed studies – often prospectively randomized clinical trials and at times also including studies of lower evidentiary power. Unsurprising, considering our comments above about prior androgen supplementation studies, none of the here-included studies, indeed, checked androgen levels of patients or controls. Undoubtedly, therefore, some study patients who were supplemented with androgens didn’t need it and, equally likely, some control patients would probably have done better had they been supplemented considering their low androgens. In other words, even though it appeared in the prestigious Cochrane Database System Review, it is yet another completely worthless study on this very important subject. Unsurprisingly, it was awarded a “Worst Paper Award” of the month by the CHR VOICE.
One final word about androgen supplementation in selected infertile women appears of importance. Androgen supplementation is likely the least costly infertility treatment that exists. Therefore, considering all the evidence that has been accumulating over almost 20 years, once cannot but wonder about the resistance to androgen supplementation some reproductive endocrinologists still demonstrate. And those with most profound resistance, often have absolutely no hesitance in imposing extremely costly treatments like, for example, preimplantation genetic testing for aneuploidy (PGT-A) on many – if not all – of their IVF patients, which even ASRM and SART no longer can defend as improving IVF outcomes.4
REFERENCES
1. Barad DH, Gleicher N. Fertil Steril 2005;84(3):756
2. Prizant et al., J Endocrinol 2014;222(3):R141-151
3. Nail et al., Cochrane Database Syste Rev 2024;6(6):CD009749
4. Practice Committees of ASRM and SART. 2024; Fertil Steril. May 18: S0015-0282(24)00241-3. doi: 10.1016/j.fertnstert.2024.04.013. Online ahead of print.
News on FSH effects on ex-vivo (mouse) folliculogenesis
Though based on mouse work, a recent paper on FSH effects in ex-vivo folliculogenesis by investigators from Rutgers University in New Jersey1 recently attracted attention: At least in the mouse, a minimum FSH threshold appears required for follicle maturation into the high estradiol-secreting preovulatory stage, inducing distinct expression patterns of genes related to follicle maturation, follicular transcriptomics, and follicular cAMP.
The most interesting finding was, however, that unusual high FSH levels (20-30 mIU/mL) resulted in premature luteinization, high androgen production and production of proinflammatory factors, as well as reduced expression of genes related to energy metabolism in oocytes.
These findings are of considerable interest in view of several recent reports that suggested, especially in older women, gonadotropin toxicity with increasing dosages of gonadotropins.
REFERENCES
1. Zhan et al., Endocrinology 2024;165(7):bqae054
2. Ni et al., JAMA Netw Open 2024;7(4-:e244438
Avoiding Clomiphene Citrate periconceptionally?
Again in mice – and, yes, mice are not men – Australian investigators fed mice within 36 hours after mating Clomiphene Citrate at body-weight-adjusted dosage similar to human use and recorded a dose-dependent adverse effect on pubs at a dose of 0.75 mg/kg: Viable pregnancies were reduced by 30%, fetal weight was reduced by 16%, and approximately 30% of fetuses demonstrated delayed development and congenital abnormalities in multiple organs not seen in controls. The investigators also observed a 30-hour delay in birth, together with increased early postnatal death of pups.
These are surprisingly severe consequences of periconceptional Clomid exposure and, of course, raise concern about periconception exposure to Clomiphene Citrate in humans. Since this medication is in principle used in anovulatory females with oligo-amenorrhea, early pregnancies may be missed. These findings suggest that a pregnancy before treatment start with Clomiphene Citrate may not be a bad idea.
REFERENCE
1. Chin et al., Endocrinology 2024;165(7):bqae047
Hypopituitarism
It is rare to see women with complete deficiency of anterior or posterior pituitary hormone production in infertility practice, but for anybody interested to read an excellent comprehensive (16 pages) and up-to-date summary on the subject we recommend a recent “Seminar” by several experts in The Lancet.1
Quoting some of their conclusions, treatment aims at replacement of missing hormones. Increased mortality may, however, persist despite treatment, particularly in younger patients, especially in females, and in those with arginine vasopressin deficiency. Patients with complex diagnoses, pregnant patients, and adolescent patients transitioning to adulthood should ideally be managed at a pituitary tumor center of excellence.
REFERENCE
1. Chin et al., Endocrinology 2024;165(7):bqae047
SUBJECT: Gynecology
Polycystic Ovary Syndrome (PCOS)
Here is an interesting PCOS story in a recent publication by Chinese investigators in Science magazine: They reported that compounds of artemisinins, derived from Artemesia plants which are routinely used to treat malaria, apparently successfully targeted all clinical aspects of PCOS, and they did so by suppressing ovarian androgen production.1 Specifically, the paper reported that these compounds decreased testosterone levels in blood and eliminated the ovarian phenotype in a PCOS mouse model. In a rat model they improved implantation rates and litter size (one, of course, wonders why two different small animal models were used to demonstrate different effects).
Most importantly, however, dihydroartemisinin treatment ameliorated hyperandrogenemia, reduced AMH (anti-Müllerian hormone) levels, improved ovarian polycystic morphology, and normalized menstrual patterns in 19 PCOS patients.
If confirmed by more data (19 patients are really not enough and neither mice nor rats are men), this may turn out to be a very promising new treatment approach to hyperandrogenic PCOS and that is, of course, potentially great news. After all, PCOS is widely believed to affect at least 10-13% of all women in reproductive years and this may, indeed, be an underestimate. The potential importance of this paper was further reflected in an accompanying commentary in Science2 and a brief summary in the News Section of JAMA.3
But here is an additional point of importance: All three here cited papers/commentaries started by describing PCOS with an almost identical cliché of PCOS, as “characterized by hyperandrogenemia,”1 by “elevated androgen levels,”2 and by “elevated levels of male sex hormones known as androgens.”3 And all three introductions were, of course, incorrect because they ignored what, under so-called Rotterdam criteria, is known as the D-phenotype (also called “lean” or “ovulatory” phenotype).
Though this phenotype after menarche – and until approximately age 25 – is also hyperandrogenic, it after that age develops progressive insufficiency of adrenal androgen production (adrenals and ovaries produce, each, approximately half of all androgens), resulting in steadily declining peripheral androgen levels. For approximately 10 years, between ages 25 and 35, these women remain in what generally are considered age-appropriate androgen levels, but after age 35 they enter into a hypo-androgenic range and become increasingly hypo-androgenic with advancing age.4-6 The roughly 10-year period of being normo-androgenic is also the age range when most PCOS cases are diagnosed. It is therefore un- surprising that the original Rotterdam criteria erroneously described phenotype D PCOS as “normo-androgenic.”
We are reviewing all of this in conjunction with here discussed recently published paper1 because by describing PCOS as a hyper-androgenic condition–and this is, indeed, how PCOS is almost universally described in the medical literature–three important negative consequences follow:
(i) PCOS phenotype D diagnoses will often be missed; PCOS is therefore even more prevalent in the general population than is widely appreciated, with especially the prevalence of the D phenotype being widely underestimated.
(ii) While lowering androgens in the other three phenotypes makes sense (whether with myoinositol, as these days widely practiced, or in the future, possibly, with artemisinins), in D phenotype lowering androgens will after ca. age 35 only further aggravate already existing hypo-androgenism.
(iii) Though these women, while still normo-androgenic, still often spontaneously conceive, once hypo-androgenic, they not only become infertile (ovaries need normal androgen levels to produce good quality eggs in good numbers), but often, indeed, become IVF treatment-resistant until their hypo-androgenism is corrected.6
While the other three PCOS phenotypes later in life are associated with metabolic syndrome (diabetes, hypertension, and cardiac disease), this does not apply to the lean PCOS phenotype. This phenotype has, however, another associated risk, and that is a hyperactive immune system (autoimmunity, inflammation, allergies). And a hyperactive immune system, if not appropriately treated, predisposes to miscarriages. These women, therefore, often not only need help in getting pregnant but also in staying pregnant.
And as a final word regarding PCOS in general: It is encouraging to see that voices are starting to speak out about PCOS representing a broad spectrum of disease, a point often widely ignored and, likely, a principal reason why progress in PCOS research has been so inadequately slow. A recent commentary in Fertility and Sterility made this point quite bluntly in discussing two other articles on PCOS in the same issue of the journal7 by noting that – though both of the articles they addressed represented potentially important contributions – neither adequately addressed this fact.7 Their advice of investigating individual sub-groups of PCOS within the broad PCOS family of conditions independently and stop throwing all subgroups into one basket, must be heeded. Otherwise, PCOS will never be really understood.
REFERENCES
1. Liu et al., Science 2024;384(6701):1187 70
2. .Stener-Victorin E. Science 2024;384(6701(:1174-1175
3. Harris E, JAMA 2024;332(4):274
4. Gleicher et al., J Steroid Biochem Mol Biol 2017;167:144-152
5. Gleicher et al., Endocrine 2018; 59(3):661- 676
6. Gleicher et al., Biomedicines 2022;10(7):1505
7. Rebar RW, Keator CS. Fertil Steril 2024;121(6):934-936
Endometriosis
ENDOMETRIOSIS AND PCOS: Endometriosis has much in common with PCOS. Like PCOS, endometriosis is one of the most common diagnoses encountered in fertility practice; like in PCOS, progress in the understanding of this condition has been painfully slow; and like in PCOS, one of the main reasons has been the inability to reach accurate diagnoses. Though textbooks still tell us that through laparoscopy a diagnosis of endometriosis is accurate, this does not consider how frequently the condition is only microscopic and not visible to the naked eye. And whenever we cannot rule out a possible diagnosis, one has to automatically assume that control populations in studies will be contaminated and, therefore, biased.
ENDOMETRIOSIS AND “UNEXPLAINED INFERTILITY.” The above preamble on endometriosis and PCOS leads straight to a recent study in the format of a systematic review by Belgian colleagues (an increasingly fashionable study format in medical journals) which claimed to determine “the prevalence of endometriosis in so-called “unexplained infertility.”1 The title alone already suggested a possible oxymoron rather than a valuable paper, because once diagnosed with “unexplained infertility,” how can the patient in parallel also have a diagnosis of endometriosis (and vice-versa)? Moreover, the diagnosis of “unexplained infertility,” in itself, is an absurdity because how long something remains “unexplained,” of course, depend on how hard one searched to find an explanation.2
This is how deep they looked in this case, quoting the authors themselves: In performing a systematic search of the literature, they attempted “to identify all studies reported on pelvic pathologies found by laparoscopy in couples diagnosed with unexplained infertility.” Based on the methodology of their literature search, not much more therefore, needs to be said; they pursued a phantom diagnosis (unexplained infertility) with a diagnostic procedure which is known to miss a significant portion of cases (laparoscopy). The result that 44% of all patients by laparoscopic diagnosed with pelvic pathology being endometriosis patients, therefore, cannot be taken seriously.
That a significant portion of women with diagnosis of “unexplained infertility” in reality are endometriosis cases has been known for decades2 and is, of course, a logical conclusion if one considers that a significant portion of women with microscopic disease go undiagnosed even after laparoscopy. This is, however, not the reason why this paper was here selected for presentation. The principal purpose of presenting intermittently also really bad papers, is to demonstrate that decent medical journals – and at times even very good journals – on occasion publish bad papers. Like everything else in the world, the peer review process at medical and basic science journals is anything but perfect. Before accepting the conclusions of a paper – which in practical terms means to cognitively integrate its information and conclusions into one’s understanding of an (in this case) medical circumstance, it is of crucial importance to learn to assess the value of the offered information which in this case, of course, was basically completely lacking. In short, a paper to forget as quickly as possible!
ENDOMETRIOSIS AND THE IMMUNE SYSTEM: Another overlooked diagnosis, often misdiagnosed as “unexplained infertility,” is a hyperactive immune system. Endometriosis in itself has, indeed, for decades, been associated with a hyperactive immune system due to autoimmunity and inflammation.2-4
Now Bruce A Lessey, MD and colleagues published an interesting relevant paper which supported a role for endometriosis in unexplained euploid embryo transfer failure.5 More specifically, the authors concluded that suppression of endometriosis with a GnRH antagonist prior to embryo transfer may improve IVF success by addressing underlying inflammatory and epigenetic changes found in so-affected patients.
This is not a new idea; already in 1988 investigators suggested that suppression of endometriosis may have beneficial immunological effects on women with endometriosis. They, however, already then also demonstrated that the effects of a GnRH -agonist were significantly inferior to the effects of the androgen, danazol, which in those days was still a preferred drug in suppressing endometriosis.6 Danazol was then, several years later, demonstrated to have an immunomodulatory effect on the autoimmune response in endometrium associated with endometriosis.7
ENDOMETRIOSIS AND CANCER RISK: That endometriosis has an association with ovarian cancer risk has been known for some time. A recent study in JAMA now not only reemphasizes this risk but defined it in further detail:8 Women with endometriosis had a 4.2-fold higher cancer risk than women without (known) endometriosis. This risk increased to 9.7-fold in women with ovarian endometriomas and/or deep infiltrating endometriosis. Moreover, associations were higher with type I (endometrioid, clear cell mucinous, and low grade serous) than type II (high grade serous) ovarian cancers. An accompanying editorial – like the manuscript – reemphasized that these data support the importance of counseling especially women with deep infiltrating and/or ovarian endometriosis about their increased ovarian cancer risk.9 The editorial raised two additional important points, especially in more high-risk situations with deep endometriosis and/or ovarian endometriosis: the possibility of prophylactic surgery, and – as a potentially important hypothesis – the question whether endometriosis in certain cases can be considered a precursor condition of ovarian cancers.
ENDOMETRIOSIS AND ADENOMYOSIS: Endometriosis and adenomyosis are distinct clinical entities with almost identical pathophysiological features. In a commentary, several European colleagues recently argued that both are “hormonally active diseases” which need to be treated by controlling local hyper-estrogenism and progesterone resistance. The suggested purpose for this commentary according to the authors was to stimulate discussion and spark interest in research of these two conditions.10
This, of course, appears like a correct sentiment, though, unfortunately, the paper did not offer much help. Whether the basic underlying premise that adenomyosis is nothing more than endometriosis of the uterus is correct remains to be determined. It appears to us that proliferation of endometrium-like cells within muscular tissue must have very different physiological characteristics from proliferation of endometrium-like cells within pelvic and abdominal cavities. Related, JAMA Surgery published an editorial about the intersection of endometriosis and ovarian cancer prevention.11 It’s worth a read!
REFERENCES
1. Van Gestel, et al., Reprod Biomed On- line 2024; 49(3):103848.
2. Gleicher N, Barad DH. Hum Reprod 2006;21(8):1951-1955
3. Gleicher N, Pratt D. Int J Obstet Gynecol 1993; 40 Suppl S21-S27
4. Gleicher et al., Obstet Gynecol 1987;70(1):115-122
5. Lessey et al., Int J Mol Sci 2024;25(13):6852
6. el-Roeiy, et al., Fertil Steril 1988;50(6):864- 871
7. Ota et al., Fertil Steril 1996;65(3):545-551
8. Barnard et al., JAMA 2024; 332(6):482-489
9. McHale MT. JAMA 2024;332(6):460-461
10. Cozzolino et al., Reprod Biomed Online 2024;48(4):103737
11. Ring KL. JAMA Surg 2024; doi:10.1001/jamasurg.2024.2974