How to individualize fertility treatments in women with low functional ovarian reserve (LFOR)

We here continue the series on how to individualize fertility treatments in women with low functional ovarian reserved (LFOR), often also called diminished ovarian reserve (DOR). In the May issue of the VOICE, we offered a general introduction to the subject and explained some of the terminology used. We then described how follicle mature after being recruited out of their resting phase as very primitive so-called primordial follicles into weeks to months of follicle growth and maturation, called the folliculogenesis. We finished the 1st Volume by explaining how CHR investigators modified the hypothesis of ovarian aging in a way that suggests that during growing stages of follicles, therapeutic interventions into the microenvironment in which these follicles develop, can beneficially affect resulting egg and, therefore, embryo quality. In that regard, we also pointed out how “revolutionary” this concept was because in 60 years of “modern” infertility care, all treatment interventions, without exception, had centered only on the last two weeks of folliculogenesis, the so-called gonadotropin-dependent stage of follicle maturation, when follicles become sensitive to, and respond to gonadotropins and other fertility drugs.

Here in the 2nd Volume, we will address how to select patients for individualized treatments, what represents such individualized treatments, how ovaries in many patients must, first, be prepared before they can be stimulated in an IVF cycle and, finally, how ovarian stimulation must be individualized. Volume 3, the last installment, will follow either in July, should there be a July issue or in the first issue of the VOICE in the new academic year in September. (The VOICE, when possible, goes on a well-earned hiatus during July and August but did so in 2019 for the first time in many years. Considering the COVID-19 pandemic, such a hiatus appears, as of this point, quite unlikely.)

Identifying patients in need of individualized fertility treatments 

For decades, most IVF centers have followed treatment protocols that rarely diverge from established baselines. To maintain consistency of results, such an approach made sense, as long as patient populations were relatively cohesive. However, with IVF outcomes in the U.S. till ca. 2010 steadily improving, initially homogeneous patient populations started to diverge into better- and poorer-prognosis patients. Better-prognosis patients conceived more quickly and rapidly exited fertility treatments. Poorer-prognosis patients, however, lingered at IVF centers and, over time, increased in absolute numbers as well as in proportions of total patients in fertility treatment.

Poorer-prognosis patients were mostly older women, desirous of conceptions at ever-increasing ages, but also younger women with POA/oPOI and almost equally poor pregnancy and live birth chances. Both of these patient populations were in obvious need of distinct treatment algorithms and more individualized infertility treatments. In many treatment aspects, both groups also overlapped and shared the fate of, often, being given no choice but third-party egg donation cycles rather than individualized IVF cycles with autologous oocytes (patient’s own eggs). Thus, the IVF field largely missed out on the rapidly evolving concept of “personalized medicine” that swept medical practice in most specialty areas over the last decade. This kind of individualized, personalized medicine is, however, exactly what is required if the dramatic declines in live birth rates observed all over the world since 2010 are to be reversed.

Though these developments seemed too obvious to be overlooked, quite surprisingly, they did not lead to reassessments of mostly uniform protocols in favor of more individualized approaches in poorer-prognosis patients. What happened instead was a rush toward third-party egg donation. As a consequence, to this date, the IVF field has made no serious efforts to improve treatment outcomes in poor- prognosis patients and really never even learned to apply currently available individualized treatment protocols to such patients. Egg donation, simply, appeared like a so much better treatment option. because of much better pregnancy and live birth chances.

Based on age-specific FOR, women at all ages can be classified into three diagnostic categoriesgood-average- and poor-prognosis. The basis for this classification is almost universal interdependencebetween quantity and quality of oocyte yields. As CHR investigators clearly demonstrated, embryo numbers available for transfer ultimately determines pregnancy chances and live birth chances at all ages. Since FOR is age-dependent and declines as women age, what represents a favorable FOR associated with good prognosis, therefore, changes over time, as do determinations of average- and poor-prognoses.

Traditionally, FOR is clinically determined through FSH, AMHand in some centers via antral follicle counts (AFCs). Theoretically, these indicators should correlate, but some distinct differences exist. Many centers consider AFCs a more subjective FOR parameter than FSH and AMH levels and, therefore, do not use them as objective parameters in daily patient assessments. Others use AFCs as primary assessment tools in order to avoid blood draws. Also, at extremely low and high levels, AMH loses clinical predictability, while evaluations of FSH are, of course, cycle-day-dependent (should be drawn on days 2/3 of cycle) and easily influenced by estradiol levels (FSH, therefore, must be drawn in association with estradiol).

A study performed by CHR investigators a number of years ago investigated various FSH/AMH combinations and their respective outcomes in clinical IVF practice. Surprisingly, best pregnancy rates were achieved in women with high-FSH/high-AMH patterns. When originally reported, the authors had no good explanation for these findings. Since then, they were able to explain the findings after discovering a new diagnostic entity in infertile women (named H-PCOS; see below for further detail). Though FSH and AMH are inversely correlated in general, there are exceptions with potentially significant diagnostic relevance.

Since FOR changes with age, FSH, AMH and AFC must be evaluated in age-specific ways. Remarkably, this, however, happens only in a small minority of IVF centers. IVF practitioners who do not consider FSH, AMH and/or AFCs in age-specific ways will fail to diagnose LFOR (in POA/oPOI and older patients) and/or excessively high FOR in association with PCOS. Since many laboratories consider FSH levels up to 10.0 or even 12.0 mIU/mL as “normal range,” many women with POA/oPOI do not necessarily present with FSH levels they consider to be “abnormal.” An FSH of 9.9 mIU/mL in a 25-year-old should, of course, be considered abnormal, even if it is still in a given laboratory’s “normal” range. (As normal lab ranges include individuals at all ages, they skew higher.). The same FSH level in a 46-year-old, on the other hand, may actually denote unusually good FOR considering her age. In reverse, an AMH of 1.0 ng/mL at age 25 is strongly suggestive of POA/oPOI but at age 45 would actually denote a good FOR.

Table 1 summarizes diagnoses mandating special attention. These diagnoses should signal the need for individualized IVF protocols. Because age 38 represented the upper age limit for IVF treatments in the early days of IVF, advanced female age has historically been defined as age above 38 years. With improved abilities to achieve pregnancies in older women, the definition of advanced female age moved to age 40. As national registries demonstrate, contemporary IVF practice refers women above age 42 almost automatically into third-party egg donation. Advance female age, under current practice patterns, therefore, can be considered 42 and above. However, a more appropriate terminology would be “advanced ovarian age” because younger women can have “older-behaving” ovaries (POA/oPOI) and, when this is the case, are just as frequently referred into third-party egg donation. Published data, however, very clearly demonstrate that, assuming the same degree of LFOR (measured by FSH and AMH), younger women have better pregnancy chances with own eggs than older women.

Though in the U.S. IVF with autologous oocytes is rare above age 42-43, we have been advocating for treatments of older women with autologous oocytes with the argument that outcomes can be better than has been widely perceived. Outcomes with patients’ own eggs, of course, cannot approach treatment successes achieved with young third-party donor eggs. Genetic maternity does, however, represent a principal desire and goal of most female infertility patients, and must be respected under a patient’s right to self-determination.

Young women with POA/oPOI mimic older women in many physiologic as well as clinical observations. Consequently, treatments are often similar. Because diagnostic terminologies can at times be confusing, a clear understanding of which patient populations should be given specific treatments is of utmost importance. So, for example, a recent systematic review in a reputable medical journal claimed to address pregnancy following premature ovarian insufficiency. It further defined this condition by criteria of the European Society of Human Reproduction and Embryology (ESHRE) as primary or secondary amenorrhea for more than 4 months, onset before age 40 and FSH over 25 mIU/mL. One of the two references the authors cited, however, referred to a diagnosis of primary rather than premature ovarian insufficiency, a more commonly used terminology. Though both terminologies can be abbreviated as POI and are defined by similar criteria, they are substantially different because primary ovarian insufficiency is defined by FSH levels above 40.0 mIU/mL rather than above 25 mIU/mL.

Box 2 defines terminologies, and with it, patient populations, as they are described in this book chapter.

Individualization of treatments is obviously dependent on accurate and consistent differential diagnoses between precursor and full ovarian insufficiency. A clear understanding of here used terminology is, therefore, of importance. Women with POF/POI (menopausal FOR before age 40), or with simply early physiologic menopause between ages 40 and 51-52 years, are obviously distinct in their respective therapeutic interventions from POA/oPOI patients, even though they may share underlying etiologies. 

LFOR usually produces abnormally low oocyte yields in IVF (i.e., affected women exhibit ovarian resistance to stimulation). Conditions with abnormally high FOR, in contrast, result in unusually large egg numbers for age and, therefore, often create risks of ovarian hyperstimulation. In an overwhelming number of cases, they represent PCOS patients, with classical and lean phenotypes (phenotypes A and D, respectively, under Rotterdam Criteria), combined, representing somewhere between 80-90% of cases. In contrast to the classical phenotype, the lean phenotype often goes undiagnosed, a subject we will return to later.

Extremes of ovarian response and ovarian function are, therefore, defined by POA/oPOI and POF/POI at low end and PCOS at high end. Neither extreme represents, however, homogeneous patient populations. Both, therefore, require highly individualized care, even though reasons differ. Especially at younger ages, large egg numbers, for example, automatically denote increased risk for ovarian hyperstimulation and ovarian hyperstimulation syndrome (OHSS), a potentially life-threatening complication of IVF.

PCOS has remained controversial since it does not represent one distinct diagnosis but, likely, a basket of conditions. A detailed review of PCOS would exceed this framework; therefore, only so much: 

When contemplating a PCOS diagnosis, even most physicians only consider the so-called classical PCOS phenotype (phenotype A) because of its very typical and very obvious appearance: It is characterized by hyper-androgenismtruncal obesityanovulation and oligo-amenorrheahirsutismacne, and development of the metabolic syndrome later in life. The lean phenotype (phenotype D), in contrast, are women who look entirely normal and have none of the above-noted stigmata of the “classical” phenotype, are usually ovulatory and have regular menstrual patterns (Table 2). Like classical phenotypes, they, however, in their teens, are hyper-androgenic and may encounter rare short-term menstrual irregularities. Investigating the ontogeny of this phenotype, CHR investigators raised the possibility that this phenotype represents a continuum of advancing age, starting with nodular adrenal hyperplasia described by the Stratakis’ laboratory at the National institutes of Health (NIH) in teenagers, in early 20s becoming phenotype D of PCOS and sometimes between mid-20s to mid-30s turning into the so-called hypo-androgenic PCOS-like phenotype (H-PCOS), further discussed below. Because the lean PCOS phenotype is rather unremarkable in its clinical presentation, it is frequently overlooked, and patients are often classified as “unexplained infertility.”

The only characteristics classical and lean phenotypes share are high androgen levels at younger years (teens to mid-20s) and high AMH persistently into advanced ages. Since classical and lean PCOS patients quite often present with mildly elevated FSH levels, they often represent previously noted high FSH/high AMH pattern. In contrast to classical PCOS, the lean PCOS phenotype, however, loses its hyper-androgenism in the age-range of 25-35 yearswhile high AMH persists. By mid-30s, phenotype D PCOS patient usually have become hypo-androgenic (and, therefore, often demonstrate high SHBG). Because international bodies so-far have not been willing to consider elevated AMH levels in the diagnosis of PCOS, investigators at CHR, who first discovered the unusual ontogeny of lean (phenotype D) PCOS women, have been given patients who have reached the stage of hypo-androgenism the diagnostic acronym hypo-androgenic PCOS-like phenotype (H-PCOS).

Professional consensus for the longest time has been that, clinically, the classical PCOS was the more “unfavorable” phenotype to have. To a degree, this argument still holds, but this is the case only because patients are phenotypically severely affected and later in life suffer from metabolic syndrome. Regarding infertility, lean PCOS has, however, counterintuitively, been found to be the more resistant to traditional infertility treatments. This was discovered as investigators at CHR attempted to research the ontogeny of PCOS with advancing female age. To everybody’s surprise, in extracting PCOS patients from the center’s anonymized electronic patient database, the researchers discovered that the center almost exclusively served “lean,” phenotype-D PCOS patients. As a center of last resort for most patients who previously underwent unsuccessful IVF treatments elsewhere, the only explanation for this discovery was that classical PCOS patients conceived before needing to reach out to CHR, while lean PCOS patients apparently did not.

Investigating the “lean” PCOS phenotype further, investigators were able to define its characteristics, its above-described ontogeny and even hypothesize about its etiology. It, indeed, in principle, encompasses what has been described as the lean PCOS phenotype D under Rotterdam Criteria. Table 2 summarizes its characteristics. Once well defined, its prevalence among CHR’s patients was found to be unexpectedly high, demonstrating well this phenotype’s resistance to traditional “routine” infertility treatments and the need for individualization of care.

As H-PCOS patients became better understood, it became apparent that older H-PCOS patients who already developed hypo-androgenism, in analogy to POA/oPOI patients, required androgen supplementation before the start of IVF cycles if their egg and embryo quality was to be improved and, with it, their pregnancy chances. At CHR, patients with H-PCOS, therefore, receive supplementation with DHEA in an effort to raise their testosterone.

These observations also, finally, offered CHR investigators an explanation for the previously noted study of FSH/AMH pairings that discovered high-FSH/high-AMH combinations to produce best pregnancy chances in association with IVF, an at that time unexplained finding. The center then already routinely supplemented hypo-androgenic infertility patients with DHEA. That included women with, at that point still unknown H-PCOS, who typically presented with discrepant FSH and AMH levels, the unusual combination of high FSH and high AMH, low androgens and elevated SHBG (Table 2). Up to that point treatment-resistant at other IVF centers, supplemented with DHEA, these patients experience the center’s highest IVF pregnancy rates, explaining previously perplexing findings in the original publication.

Individualized treatments

The preceding section described which patients represent the most important potential targets for individualized fertility treatments. In this section, we describe what those individualized treatments entail. In recognition of almost all fertility patients preferring to conceive with autologous oocytes, the context remains a treatment philosophy that considers third-party egg donation strictly the last-resort treatments, though fully recognizing and communicating to patients that, in practically all cases, pregnancy and live birth chances with autologous oocytes will be significantly inferior to donor eggs. Ultimate treatment choices are, however, always the patients’, with clear understanding that physicians have the absolute obligation to present all available treatment options to patients with unvarnished outcome prospects but are not qualified to tell patients how to live their lives. 

In practical terms, this philosophy means that treatment choices are always left to patients, even if their choices contradict recommendations made by physicians. Recognizing that the decision to turn toward third-party egg donation represents a highly complex and very personal decision and that patients, therefore, should not be rushed, or even forced, into such a decision, a patient’s treatment choice should never be refused, unless it carries with it significant risks to the patient’s health. This philosophy is also based on the recognition of how important certainty is for many women (and often also to their partners) that they, indeed, are unable to conceive with use of their own eggs before choosing third-party egg donation.

Preparation of ovaries

Not yet being able to produce “new” oocytes for women with either LFOR and/or with poor-quality oocytes, the concept of pre-treating ovaries prior to IVF cycle start has been gaining. Though many colleagues still believe that oocyte quality is fixed and unassailable, as previously explained, evidence suggests otherwise: With the microenvironment of ovaries aging, crucially important components for follicle maturation within this microenvironment are becoming deficient. Androgens and IGF-1/HGH are the only such components identified so far. It seems logical that essential components of the ovarian microenvironment, lacking in appropriate concentrations in women with LFOR, should be replenished. Such replenishments, moreover, must occur in a timely fashion.

When it comes to supplementation of androgens and IGF-1/HGH, that time point is the small-growing follicle stage immediately after recruitment, which means that supplementation must be initiated at least 6-8 weeks before the start of an IVF cycle. Such pre-supplementation appears essential because of the time it takes the so-treated follicles to reach gonadotropin-dependency and become sensitive to the effects of exogenous gonadotropin stimulation used in IVF cycles. Older women above age 40, POA/oPOI, POF/POI, and more recently as reported above, H-PCOS patients, have all been identified as women in need of androgen pre-supplementation.

Since the human body produces testosterone from DHEA, such supplementation via DHEA (25mg TID, p.o. Fertility Nutraceuticals, LLC, New York, N.Y. 10021 appears advisable; please note the authors’ conflict statement at the end). This way, every organ in the body capable of producing testosterone will take up only as much DHEA as it needs to achieve desired testosterone levels (and testosterone levels vary between human organs). Though testosterone can also be administered directly, its administration via patch or transdermal gel floods the whole body, with every organ getting equal amounts. Side effects of direct testosterone administration are, therefore, somewhat more pronounced. Moreover, risk of reaching toxic testosterone levels for reproductive success, with DHEA supplementation minimal, are significantly higher with testosterone.

Because of a large number of animal models that not only demonstrated the importance of androgens for normal follicle maturation but also demonstrated beneficial effects from supplementation of low androgen levels, even in the absence of adequately powered prospectively randomized studies, androgen supplementation can be considered an adequately established treatment modality. The same, however, cannot be said about HGH pre-supplementation, which we still consider experimental and, currently, investigate in a prospectively randomized study in which patients receive 3-6 IU of HGH (various manufacturers) daily for 8 weeks prior to IVF cycle start.

In addition to prenatal vitamins, poor-prognosis patients also receive supplementation with CoQ10, depending on absorption characteristics at 600-1,000 mg daily. In mice, this antioxidant has been found effective in improving egg quality and quality of cumulus cells . Though human data for such effectiveness are lacking, considering its relatively low cost and lack of side effects, we consider supplementation with CoQ10 in poor- prognosis patients advisable. From research in other medical specialty areas, this antioxidant is known to improve mitochondrial function in human cells. With the mature egg being the largest cell in the female body, also carrying the largest mitochondrial load, supplementation prior to IVF cycle start appears to make sense. 

Ovarian stimulation

How to stimulate poor-prognosis patients with LFOR is more an issue of what not to do in ovarian stimulation, rather than how to stimulate correctly. So, for example, since after age, the number of good-quality embryos available for embryo transfer is the second most important predictor of pregnancy and live birth chances, we do not understand the arguments made in the literature for mild-stimulation cycles or even natural cycles. We, therefore, stimulate such patients aggressively, which uniformly means stimulation with a mixture of FSH (350-450 I.U.) and hMG (150 I.U.) per day.

Another stimulation feature that must be avoided is ovarian suppression. In women of advanced age and in other poor-prognosis patients with LFOR, we avoid hormonal contraceptives, full-dose agonists and antagonists and, if prevention of premature spontaneous ovulation is required, that is achieved with the microdose agonist protocol initially reported by Surrey et al (Colorado).

Recognizing that premature luteinization of the follicular microenvironment accelerates with advancing female age as well as in women with POA/oPOI, the need for such microdose agonist protocols has, however, at our center in the last five years greatly diminished due to increasing utilization of Highly Individualized Egg Retrieval (HIER), described in the third and final volume of this series.

Conflict Statement

CHR and some of CHR’s scientists are listed as co-owners of a number of already awarded and still pending U.S. patents, some claiming benefits from androgen supplementation in women with low ovarian reserve, a topic addressed in this series of manuscripts. They also are co-owners of a number of U.S. patents relating to other fertility issues, unrelated to here discussed topics, including diagnostic importance of the FMR1gene and potential diagnostic and/or therapeutic functions of AMH. CHR and some of its scientists are shareholder in a number of start-ups, including Fertility Nutraceuticals, LLC, which produces a DHEA product and pays CHR and some of its scientists royalties. CHR and individual scientists have in the past received research support, travel funds and/or speaker honoraria from different Pharma companies as well as medical device corporation – none, however, in any way related to here discussed subjects. 

This is a part of the June 2020 CHR VOICE.

Norbert Gleicher, MD, leads CHR’s clinical and research efforts as Medical Director and Chief Scientist. A world-renowned specialist in reproductive endocrinology, Dr. Gleicher has published hundreds of peer-reviewed papers and lectured globally while keeping an active clinical career focused on ovarian aging, immunological issues and other difficult cases of infertility.