BASICS OF INFERTILITY TREATMENTS
Medicine is a complicated enterprise. It at least pretends to always strive for best and most efficient medical practice but, likely, more often fails than succeeds and the reasons are manyfold. In today’s postings we at least attempt to present a few examples why that is and – in order to not appear too pessimistic – at the same time, where recently offered in the literature – suggest solutions or at least new thinking and/or approaches to improve the situation.
In our next posting in the coming week we then intend to return again to more specific clinical subjects addressed in the recent medical infertility literature, - even though the today discussed issues very obviously have considerable relevance to the clinical practice of infertility.
The CHR’s Editorial Staff
LOW FUNCTIONAL OVARIAN RESERVE (LFOR) – What it Is and What it Is Not!
By David H. Barad, MD, MS, one of the CHR’s REI physicians, Associate Editor of CHR Publications, Director Clinical IVF, Director of CHR-Research and a Senior Scientist.
Dr. Barad here addresses an important issue in explaining the term ovarian reserve (OR) in its various applications and implications. The terminology can be confusing, starting with the fact that OR reflects all remaining follicles in ovaries and the eggs they contain. A large majority of follicles and eggs exist, however, as very primitive and very small, so-called primordial or resting follicles, - and only a small minority at any given time, - after having been recruited out of this primitive resting stage – become visible on ultrasound as they grow in the ovary toward maturity. It is this part of the OR - called functional ovarian reserve (FOR) – that is important in fertility treatments and is evaluated during IVF cycle monitoring in every patient.
This afternoon, I met a patient who came to our consultation carrying more than lab results, - she carried the weight of having been told that her chances of pregnancy with use of her own eggs were essentially gone!
Based on hormone tests showing low ovarian reserve (OR), she had been advised that in vitro fertilization (IVF) with her own eggs was not an option and that – if she really insisted on continuing to try with autologous eggs – only lower-cost intrauterine inseminations (IUIs) made sense going forward. In other words, - she was refused the chance of an IVF cycle because of her alleged low functional ovarian reserve (LFOR). In two earlier such IUI attempts, one had resulted in an ectopic pregnancy that not only required treatment but – as an undesirable consequence of her LFOR - forced a significant pause in her once more trying to conceive. Even after the ectopic pregnancy experience – potentially suggesting tubal disease - she, therefore rather paradoxically, was still because of LFOR refused IVF by her treating physician(s), even though IVF in women with LFOR in most cases is a much more effective treatment than IUIs, and – in addition - was urged to continue relying on her likely diseased fallopian tubes.
As she told her story, what was bothering her and why she had come to the CHR for a second opinion became quickly obvious: what bothered her was not only her tubal pregnancy; but – after asking several A.I. platforms a few relevant questions, she had started to wonder whether she had allowed a set of numbers assessing her so-called FOR to define her future fertility.
So, what are the tests doctors use to assess FOR, and what do they actually tell us? The most commonly used measures are FSH (follicle-stimulating hormone), AFC (antral follicle count), and AMH (anti-Müllerian hormone). Each of these three tests provides information on FOR but does so from different viewpoints, and none should, therefore, be interpreted in isolation.
FSH is a hormone produced of the pituitary gland that signals the ovaries to begin growing out of their dormant state (as primordial follicles) freshly recruited follicles. As the ovaries’ general ovarian reserve declines with advancing female age and fewer follicles are available for recruitment out of resting stage, the body tries to compensate by producing higher levels of FSH in an effort to stimulate the ovaries harder. Rising FSH is, therefore, not a sign of the ovaries “failing,” but rather of the body pushing harder to get an FSH response.
The AFC., in contrast, is assessed by ultrasound at the beginning of a menstrual cycle and represents the number of so-called small antral follicles visible in the ovaries at that moment in time. Each of these follicles is assumed to contain an at that point an extremely immature egg and, together, they offer a snapshot of how many follicles may be available to respond during the coming cycle. Importantly, AFC can vary from month to month, between ovaries, and even between individuals assessing the ultrasound exam. The AFC, therefore, just reflects what is visible, and not necessarily what is possible. A lower count , therefore, may suggest fewer follicles to work with, - but it does not necessarily predict whether one of those follicles can and/or will develop well enough to mature a healthy egg that after fertilization can produce a healthy embryo, and/or, ultimately, will result in a successful pregnancy.
Finally, AMH is produced by the cells surrounding developing follicles, including indeed the earliest small antral follicles. Because AMH therefore reflects the cumulative activity from such a broad assembly of follicles (many not even yet visible on ultrasound) AMH is often used as a general marker of FOR. One way to think about the meaning of AMH values is to consider what the sound produced in gently shaking a closed box of candy will tell about f how much candy may be inside. But it will not tell you which piece you may pick up next, or how much you will like it.
Because FSH, AFC, and AMH are each looking at ovarian function from different angles, it is not unusual for them to tell slightly different stories. One test may appear more concerning while another more reassuring, and that can be confusing for patients (and at times also for some of our colleagues). These differences do not mean that one test is “right” and the others one is “wrong.” Rather, they reflect the fact that ovarian function is dynamic and varies from cycle to cycle and - especially at more advanced ages – can quickly change.
Hormonal signals and what is visible on ultrasound in a given month, and the activity within smaller follicles do not always move in perfect synchrony. For this reason, experienced clinicians interpret these tests together, and in the context of a patient’s age, history, and prior responses, - rather than relying on any single number (including, of course, age) to define prognosis or guide care.
When these tests are placed into context, they stop being verdicts and start becoming tools that help shape a thoughtful treatment plan, - rather than defining what is or is not possible.
A LFOR in a patient, therefore, should consider how treatment should be approached, - and not whether treatment should be pursued at all!
LFOR usually means that fewer eggs may be available in any given cycle. This, of course, influences how medications are selected, how closely cycles must be monitored, and how expectations should be set appropriately from the outset. It also usually means that care in many of patients with LFOR – whether because of advanced age or because of premature ovarian aging, POA – must be more individualized and, at times, more iteractive, as each IVF cycle provides more information in refining the next cycle. In this context, the FOR guides strategy and planning but not outcomes.
LFOR however almost never automatically means that pregnancy is impossible, that treatment is futile, or that a negative outcome is predetermined. It, therefore, is difficult to understand how many of our colleagues in a case like here addressed can reach the conclusion that IVF is not even worth attempting. The patient, after all, did spontaneously conceive despite her LFOR, - even if it was a tubal pregnancy!
None of here discussed tests of FOR can predict egg quality in a given cycle, the potential of an individual embryo, or whether pregnancy will occur or not. They also cannot account for the natural variability that exists from month to month. While LFOR affects overall probabilities, it does not eliminate them, and it should never be used to close doors prematurely, - as, unfortunately, only too often happens in many IVF clinics.
In routine clinical IVF practice, patients are only too often told that a diagnosis of LFOR automatically excludes them from IVF treatments with use of their own eggs. In the case I described earlier, laboratory values were used to narrow options rather than to guide a broader conversation about strategy and risk.
Decisions like these are often driven by rigid thresholds applied to complex biological circumstances which simply don’t make sense, because probability in these cases is mistaken by the treating physician for certainty. While FOR testing is invaluable for planning care, using these measures as absolute gatekeepers can lead to options being closed prematurely, sometimes without fully accounting for individual circumstances, prior outcomes, or alternative approaches.
If – as noted before – OR reflects how the ovaries are functioning rather than final outcome, then LFOR also shapes how treatment can be thoughtfully individualized.
At the CHR, we therefor use the term FOR to describe not just the number of follicles suggested by testing, but the ovary’s current ability to access and recruit follicles in a given cycle. In other words, a LFOR reflects how the ovary is functioning at that moment in time, - not its fixed or permanent state. The concept of LFOR is that with individualized treatment we have the potential to change the ovarian response.
Because ovarian function is often influenced by the hormonal environment, treatment planning at the CHR includes thoughtful preparation before starting stimulation. In selected patients, this can involve pre-cycle therapies such as dehydroepiandrosterone (DHEA), other androgens, or human growth hormone supplementation (both synergistically with the hormone FSH enhancing follicular growth and maturation) , with the goal of optimizing follicle recruitment and response with the goal of improving LFOR.
We are also actively studying additional approaches aimed at improving how follicles are accessed and how individual follicles function, including carefully designed pre-treatment strategies and metabolic support. What remains unknown is which patients are most likely to benefit from these approaches, how durable any effects may be, and how best to individualize their use. But we are also hard at work to determine these parameters prospectively. The next guest article by Eriona Hysolli, PhD, offers some potential directions.
While none of these strategies can guarantee outcomes, they reflect a broader philosophy: FOR testing should guide how care is tailored, - not limit whether care is offered. At its best, fertility care is a partnership, in which medical expertise and patient goals are weighed together as decisions are made.
For patients facing fertility decisions, it is important to remember that test results are tools meant to guide conversations, not to shut them down. A finding of LFOR should prompt thoughtful planning and open dialogue, - not an automatic narrowing of options. Like fertility itself, - fertility treatments are rarely straightforward, and uncertainty is often part of the process. But uncertainty is not the same as impossibility and is – after all – in humans also integral to spontaneous conception attempts.
Identical treatment protocols for everybody, therefore, by definition cannot make sense and will end up hurting individual patients who deserve time, explanation, and a care planning that reflects their individual goals and values. When decisions are made collaboratively, with clarity and compassion, patients will move forward in their treatment journey feeling heard and truly supported. If treated with rigid protocols, they will perceive themselves – correctly – on an assembly line.
READING LIST
Gleicher N. Weghofer A. Barad DH. Improvement in diminished ovarian reserve after dehydroepiandrosterone supplementation.
Reprod Biomed Online 2010;21(3):360-365
Gleicher N, Weghofer A, Barad DH. Defining ovarian reserve to better understand ovarian aging. Reprod Biol Endocrinol 2011;9:23
GUEST COMMENTARY
A NEW METHOD OF GENETIC RISK ELIMINATION IN FEMALE INFERTILITY - The Possibility of Gene Corrections in Oocytes
By Eriona Hysolli, PhD, was a Co-founder of Manhattan Genomics, and is a Member of the CHR Publications’ Editorial Board.
With correction of gene defects in embryos already being a serious goal of research in several academic laboratories and start-ups, Dr Hysolli in this brief commentary on a recently published Chinese paper in Cell Genomics points out that gene defects – theoretically – can also be corrected in oocytes and, when affecting fertility, therefore in the future can play an important role in fertility treatments via in vitro fertilization (IVF).
Introduction
A recent paper published in Cell Genomics by Chen et al, tackled the genetic link of female infertility.1 While most attention goes to the power of prenatal genetic testing for monogenic disorders, aneuploidy, and polygenic scoring to choose the healthiest (and perhaps in the eyes of many - controversially - the best) embryo, not much attention is drawn to the genetic risk for early developmental defects of the oocyte and embryo that fail to produce a viable embryo in the first place.
This is now a key research area in reproductive medicine with several important questions left to answer: (i) How strong is the genetic link to female infertility, which blocks couples from producing viable
Embryos? (ii) Can we genetically screen oocytes, for infertility risks, - so that a clearer path to prevention and treatment can be charted earlier? And (iii) are there preventative correction pathways we can pursue?
Several genes have been correlated with female infertility in the past. In this study, a larger cohort of ~3600 women in China, who had failed two IVF/ICSI cycles with failures categorized as oocyte defects, abnormal fertilization, and embryo arrest, consented to whole exome sequencing of their blood samples. Sequence analysis revealed that approximately 13% of the cohort presented with - among a few other pathways - sequence mutations in key domains of genes involved in spindle assembly, cell cycle and check points, zona pellucida, maternal mRNA regulation, mitochondrial function, and homologous recombinations,.
As one example, TUBB8, encodes a primate-specific β-tubulin involved in human spindle assembly. Mutations in this gene have been characterized before, - but key was its overrepresentation in these datasets across the three categories of defects.
To capture the human phenotype in IVF/ICSI failure, the researchers conducted mouse functional studies in a few key targets involved in cell cycle regulation, - N-glycosylation and chromosome segregation - which showed embryo developmental arrest upon injection with mutated mRNA (CNTD2, SPDYC) or mRNA knockdown (DDOST, INCENP).
Infertility, of course, remains a complex issue. These findings are an important step toward clarifying the genetic contribution to female infertility and can be crucial for managing fertility treatments in the future.
While only roughly 13% of the cohort exome sequences had distinct mutations, hundreds of genes were
implicated, and more work on non-coding regions, epigenetic landscape, and paternal contributions is needed to comprehensively understand infertility overall; but a new path to a better understanding of human infertility has clearly emerged with this study.
Correction of Mutations in Oocytes and Embryos to Improve Fertility Outcomes -
The concept of germline gene correction has recently been making a powerful comeback in reproductive medicine conversations as well as in public discourse. The emergence of companies like Manhattan Genomics,, Preventive Bio and Bootstrap Bio, has created hope that avoidance of genetic diseases in future generations may become possible using powerful and precise editing technologies like base and
prime editors.
But the high mutational burden in the oocyte genome that contributes to developmental oocyte
as well as embryo defects can also be targeted using the same tools. IVF already can allegedly screen an embryo for PGT-M (monogenic defects), PGT-A (aneuploidy) and PGT-P (polygenic defects). But embryo screening, of course, requires substantial embryo numbers, which infertile women do not always produce in IVF cycles. The number of oocytes in an IVF cycle, however, almost universally exceeds the number of embryos. Corrections at oocyte stage, therefore, would produce obvious numerical advantages.
Gene editors or gene therapy to tackle infertility-producing genetic risks can make a significant impact when coupled with knowledge from omics analysis and better prediction algorithms. When used
thoughtfully and appropriately, gene correction or gene therapy for infertility risk at the oocyte and/or embryo stage ultimately will ensure more embryos for transfer into the patient’s uterus. And conception with such an embryos makes the genetic change also permanent for future generations, thus allowing them to avoid significant future infertility risks.
Such oocyte editing has already been established in multiple livestock species as well as non-human primates. It is time we move – at least in the research arena - towards integrating genomics into IVF cycle planning, and even bolder, open the path of exploration to gene correcting intervention of oocytes and embryos for infertility management and disease prevention in the human experience.
REFERENCE
Chen et al. Genetic Landscape of Human Oocyte/Embryo Defects. Cell Genomics 6, 101012
January 14, 2026. https://doi.org/10.1016/j.xgen.2025.101012
General Medical News with Implications for Infertility Practice
Even the AMA Now Recognizes the Need to Rebuild Trust in Evidence-based Health Information
It was nice to see at least one recent communication from the American Medical Association (AMA) that is not only driven by politics and/or ideology: In a press release on March 5, 2026, the AMA reported the results of a new public opinion survey by the Annenberg Public Policy Center which “high-lightened the widening trust gap in the nation’s health information landscape and underscore the importance of trusted medical voices grounded in science.”1
Considering the increasing politization of the AMA in recent years, this press release is almost an oxymoron, - but better late than never and better incomplete than not at all. A probably, however, more realistic inside into the AMA’s thinking comes likely from the press release’s last two sentences and we quote: “The AMA continues to advocate for science-driven health policy and clear communication grounded in the best available evidence in support of patients and physicians. The AMA recognizes the need for a strong health system that can foster and sustain a healthier future for everyone across our nation.”
Two comments regarding these two sentences: (i) Of course no “mea culpa” from the AMA for so many obviously – at least in retrospect – harmful AMA policies and practices in the recent past, - including at times quite disastrous comments and recommendations during the COVID-19 pandemic or the AMA’s quite aggressive initial support for medical gender transition for children and young adults. And (ii) Do we read the last sentence correctly by interpreting it as the AMA now supporting a national government “owned” health system?
If that is a yes, - many of the CHR’s Canadian patients may be the most unhappy, - no longer being able, after quickly crossing the border, to get medical services in the U.S. - including even simple blood tests - their own national health care system either does not offer at all or only with incredibly long delays. This is, indeed, how government – managed national health systems attempt to control costs, - constantly declining service quality.
REFERENCE
AMA. Press Release. March 5, 2026. Httpd:///www.ama-assn.org/press-center/ama-press-release/survey-shows-need-rebuild-trust-evidence-based-health-information
Why Is It So Difficult to Stop Pointless Medical Interventions?
This is, of course, a question we – here at the CHR – are constantly asking – too many times indeed – considering the many treatments in infertility practice which are routinely used, even though no evidence exists for any real clinical utility. The probably most consequential is, of course, preimplantation genetic testing for aneuploidy (PGT-A) where even the ASRM finally reached the conclusion that PGT-A does not confer any outcome benefit on IVF cycles.1
Yet over half of all U.S. IVF cycles – even often including donor egg recipient cycles – now routinely involve donor oocytes from very young donors – strong evidence that many colleagues now consider PGT-A a routine part of IVF. A good number of clinics by now, indeed, refuse treatment if patients object to the utilization of PGT-A.
But this problem does not only exist in the fertility arena (though it is, likely more prevalent than in most other medical specialties), - but, as a recent Opinion article in the BMJ pointed out, also exists elsewhere in medicine.2 And, based on his own behavior, the author – an intensivist - made a very interesting point in his commentary: “ You can (yourself) publish all the evidence and (may) still struggle to unlearn a habit.”
He described at least part of the problem as structural because the ability to charge for one’s activity rewards the activity, - as the feedback loop on harm is in general only very faint. The article also notes that in 2019 NHS in the UK published a list of 17 interventions judged inappropriate outside of defined circumstances, leaving the impression that this publication would result in decisive change. Yet nothing happened at all after the guidance was issued! Does this sound familiar to what is happening in infertility practice?
We could probably list 17 such interventions with ease in infertility practice alone! And more relevant information on this issue in the next commentary.
REFERENCE
Practice Committees of the ASRM and SART. Fertil Steril 2024;122(3):421-432
Morgan M. BMJ 2026;392:s374
Causal Inference in Medical Practice – How Certain is A the Cause of B?
Because of the preceding commentary, we here once more are bringing to attention two articles in the BMJ which disagree with each other to a degree on causal inference in medical research.1,2
So what is causal interference and why does it relate to the unavoidable uncertainty in medicine and science in general?
It is basically the process that determines whether an event A is caused by another event B or not; and that is, of course, a question asked in medicine and science all the time. It has, however, also been studied in philosophy, machine learning, psychology and, of course, statistics.3 It practically always boils down to the degree in which causal interference is based on controlled and uncontrolled observations.
You by now may be asking why would the BMJ publish two such contradictory essays regarding this subject, why we would spend valuable space on discussing , and what the two essays disagreed on?
To answer both of these two simple questions let us start with how Alex Broadbent, a Professor of Philosophy of Science at Durham University in South Africa explained the principal issue: Once a year he has been baking a Christmas cake, using always the same recipe. Yet the results always varied and that, in his mind, of course, raises the question, - why?
At least two hypotheses could explain the reason(s) for the observed variability,: A first he proposed was not a very serious one, - what he called “naughty elves” (a funny but not a very likely explanation). But a second hypothesis made more sense, - namely that ingredients he used and/or how he had handled them in the baking process had been inconsistent (likely the correct explanation).
He, therefore, first-of-all would have to figure out what exactly had caused these differences and, - unless he then perfectly succeeded in adjusting the process of baking uniformly every year (of course with great likelihood an unachievable goal for a human), one has to conclude that variability is unavoidable.
And this conclusion, of course, leads automatically to a second question: What are all those variabilities that arise? And the answer, of course, is that, simply from observation (even assuming we observe ourselves) we can never fully register and, therefore, consider and perfectly adjust – all of those variabilities. Broadbent, therefore, perfectly logically concluded that ”no method ever can take observational data and mechanically deliver causation.”
In other words, no observational study can ever with absolute certainty establish causality! And quoting him again: “Uncomfortable as it may be to admit, the nature of causation isn’t fully understood, and thus no satisfactory definition has been found. Perhaps the most useful thing philosophy (therefore) can offer doctors, confronted with huge numbers of complex studies, is the reminder that some questions really do remain open.”
The second essay by Timothy Feeney, MD, MPH, currently a postdoctoral research fellow at Boston University and Paul Zivich, PhD, assistant professor in the Department of Epidemiology at the University of North Carolina, in Chapel Hill followed a different line of thought but in reality really ended up not too far from Broadbent..
They in their essay concluded that researchers must “fully understand the underlying assumptions to uncover cause and effect” (so-far they make sense).
Though – at least on the surface this sounds more optimistic and doable than Broadbent’s “pessimism,” – and was presented as a contrarian opinion to the first paper, we – frankly – don’t see much of a difference between the two papers. Quoting this time verbatim from their paper, how they described the paper’s “key point” will explain why we reached this conclusion: ”Health researchers in medical care can’t solely rely on statistical analysis to uncover relations between cause and effect (we agree). To do this, they claim researchers need to have a clearly defined question and must articulate a so-called ‘estimand’ corresponding to this question.“
We, of course, never before even heard the term “estimand.” The authors, however, graciously defined the term as “the quantities a study aims to estimate” (note the word “estimate,” of course – once again – eliminates all certainty).
And to continue quoting them: “The researchers then must be able to translate from observed data to that causal ‘estimand’ and conceptualize an appropriate way of collecting the relevant data.” And one more quote: “The investigators (of, course) must critically understand the underlying assumptions and know how to design a study to obtain the estimates of interest. Only after the ‘estimands’ are formally defined can researchers consider how these ‘estimands’ might be learnt from observations of the world, a process often referred to as ‘dentification’ that entails mapping an ‘estimand’ to observed data.”
In short – while all of these efforts may come close to certainty, the authors’ own words once more reflect the fact that getting closer does not mean establishing causation beyond doubt (i.e., establishing proof). Sisyphus’s bolder is still rolling back down the hill before reaching the top of the mountain!
REFERENCES
Broadbent A. BMJ 2025;391:r2615
Feeney T, Zivich P. BMJ 2025;391:r2618