REPRODUCTIVE GENETICS
Considering that infertility practice produces human embryos and is heavily responsible for the creation of new human life, it should not come to a surprise that nothing in the field of reproductive medicine these days is more controversial than reproductive genetics. The recent articles we have chosen for presentation in today’s postings are meant to point out some of the most obvious flashpoints. They are also meant to demonstrate that much of the controversy is very obviously driven by economic incentives for the genetic testing industry, an obvious problem by no means only restricted to reproductive medicine.
It is time for medical journals in the infertility field to “clean up” the peer review process regarding PGT-A
New articles about preimplantation genetic testing for aneuploidy (PGT-A) still appear regularly in the literature. To a degree that is astonishing because - in an overwhelming majority of cases – they make it through peer review, even though they repeat the same mistakes in study design and statistical methodologies utilized, which now for decades have misled IVF practice into increasing utilization of PGT-A.
This is no longer only the lonely opinion of the CHR (Center for Human Reproduction) and a few other “dissidents’ in the IVF field, but at the end of last year was basically also – in our opinion in an overdue “opinion” - expressed by ASRM (American Society for Reproductive Medicine) and SART (Society for Assisted Reproductive Technology), its sister society, concluding that after decades of clinical utilization by the IVF field, PGT-A to this day has been unable to demonstrate any clinical benefit (1).
Without wanting to specifically blame only one journal (because pretty much all medical journals in the infertility field are guilty of this), the JARG recently published an Opinion article by French and Egyptian co-authors which – though overall more objectively critical of PGT-A than many other recently published opinions on PGT-A – nevertheless summarizes the subject with quite typical noncommittal language. The point we are trying to make here is that statements like, “PGT-A has not consistently proven its effectiveness. The clinical value of PGT-A, therefore, remains controversial,” (2) no longer appear appropriate.
After decades of PGT-A practice (including the time when the procedure was called preimplantation genetic screening, PGS), the above noted ASRM/SART statement was clear: PGT-A has not, as the JARG article suggested, only inconsistently proven its clinical utility but, per ASRM/SART statement has demonstrated no utility whatsoever in general populations. It seems high time to make it the accepted description of the effectiveness of routine PGT-A in association with IVF, especially since increasing numbers of IVF clinics have been making PGT-A mandatory in all IVF cycles.
Though based on the ASRM/SART Opinion document the medical literature - very slowly – appears to be shifting toward a more realistic assessment of PGT-A, the continuation of noncommittal language about PGT-A utility as used in above-noted JARG article is only hampering this very essential process. The need for a more correct understanding of the utility – or should we say lack of utility – of PGT-A in IVF may not even be the most important reason for the required changes in language used to describe the PGT-A procedure in the literature: That may be the additional recognition that, in certain patient sub-groups, PGT-A may, indeed, be harmful. And that at and additional expense for at least ca. $5,000 for every IVF cycle!
And then there is, of course, above noted ASRM/SART Committee Opinion on PGT-A, also published in Fertility and Sterility, the principal lead journal of the ASRM. The same journal then, however, nevertheless also published a (we have to call it an “alleged”) systematic review and meta-analysis by Turkish colleagues (one currently faculty at Yale University in Connecticut) (3) that concluded that “current low-quality evidence suggests that PGT-A enhances life birth rates per embryo transfer and per patient in unexplained repeated pregnancy loss.
This paper, however, not only again used rather wishy-washy language in describing an alleged outcome benefit for PGT-A but – by now shamefully – made again one of the quintessential mistakes in study design and data analysis responsible for years of false information on PGT-A – assessments of IVF cycle outcomes with reference point embryo transfer. By using embryo transfer as reference point (rather than cycle start as a reflection of intent to treat), outcomes are obviously biased toward good prognosis patients and, therefore, exaggerated for average populations.
And then there was of course, also a Reflection article attached meant as a commentary, something Fertility and Sterility, unfortunately, has made almost a routine. Reflections are usually written by one of the paper’s peer reviewers, in this case a well-known miscarriage expert, who used the opportunity to plug “new guidelines for recurrent pregnancy loss, integrating genetic testing of products of conception (which we, of course fully support) with PGT-A (which we, of course, do not support) (4).
One is once again left wondering what is going on with the peer review process in the infertility field, and where are the journal editors in the decision-making process about PGT-A articles (no journal in the field has as many editors as Fertility and Sterility). Who can be surprised that patients and IVF practitioners remain confused about the utilization of PGT-A if one of the most important medical journals in the field still allows such obviously incorrect conclusions to slip through peer review.
And since we are already talking about Fertility and Sterility, here is another paper published in this journal that is worth noting. This time it involved several very prominent U.S. investigators in the field who, based on a national data set, published a paper under the heading, “Success rates with PGT-A testing for aneuploidy in good prognosis patients are dependent on age” (5).
The dependence on age was however, not as one would have expected a decline in effectiveness of PGT-A with advancing female age, but exactly, the opposite: PGT-A was alleged to improve in effectiveness with increasing female patient age, an observation recently also suggested by some other papers in the literature.
But study results must also be compatible with basic logical consideration, in this case, for example, meaning that, (i) as an embryo selection method, PGT-A is trying to find “best embryos” and that means patients undergoing PGT-A must have a good number of embryos. Indeed, the more they have, the more likely will embryo selection make a difference. But this is where (ii) comes into play: as far as we know, oocyte and embryo numbers decline with advancing age. Basic logic, therefore, contradicts the alleged finding of better PGT-a utility with advancing age.
These authors at least acknowledge that they were reporting on good-prognosis patients; but what they – like others before them – overlooked is that PGT-A in every patient population investigators may choose for study will always select out best-prognosis patients because the decision to use PGT-A will always be made with significant consideration of how many embryos may be available for testing. If that number is even or smaller to the number one is willing to transfer, why would one even consider PGT-A at an additional IVF cycle cost of at least $5,000?
In other words, the recent message from some (unfortunately still existing) supporters of PGT-A utilization in IVF that- while not working in younger patients – PGT-a in older women does show outcome improvement (as this paper is also trying to demonstrate) is statistically malpractice. And once again one must ask, how does such a paper slip through peer review and editorial supervision?
References
1. ASRM Practice Committee Documents. https://www.asrm.org/practice-guidance/practice-committee-documents/the-use-of-preimplantation-genetic-testing-for-aneuploidy-a-committee-opinion-2024/
2. Viville A, Aboulghar M. J Assiste Reprod genet 2025;412:63-69
3. Mumusoglu et al., Fertil Steril 2024;123(1):121-136
4. Kutteh et al., Fertil Steril 2025;123:67-68
5. Harris et al., Fertil Steril 2025;123(3):428-438
Embryo editing of future diseases in offspring?
These pages have repeatedly expressed unhappiness about the concept of polygenic risk scoring/testing of human embryos, which some laboratories and IVF clinics in the U.S. now already offer in association with IVF as a routine commercial product. Two recently published articles in the same issue of Nature magazine, however, went even beyond this practice by raising the possibility of heritable polygenic editing of embryos.
In one of these papers the authors describe an “analysis” of the literature, based on which they concluded that polygenic genome editing, “could theoretically yield extreme reductions in disease susceptibility.” As examples they cited coronary heart disease, diabetes, schizophrenia and major depressive disorders, and Alzheimer’s disease. In addition, risk factors for diseases could be addressed, with examples cited being lipidemias and hypertension (1).
At the same time, the authors did recognize significant ethical issues in their models: For example, they concluded that this process – if successful – may deepen health inequities by being available only to the well off. In addition, by changing gene variants to reduce certain risks, other risks may be enhanced. The authors, however, nevertheless concluded that the technology to do this kind of embryo testing would in every individual be of such low risk and of such high effectiveness that deploying it on a large scale may be justified.
An accompanying commentary by three scientists from the U.S and Israel strongly disagreed, clearly stating that – currently - embryo editing is unsafe and unproven (2). They furthermore asked, “whether it is wise to distract stakeholders, including the public, with a technology that is still long way off at best, and might never actually be safe?” We couldn’t agree more!
References
1. Visscher et al., Nature 2025;637:637-645
2. Carmi et al., Nature 2025;637:554-556
Prenatal noninvasive cell-free DNA (cfDNA) testing
cfDNA testing is increasingly utilized beyond just routine noninvasive prenatal screening (NIPS). An example for other uses came about when accidental diagnoses of cancers during early prenatal testing were made. A recent review article in the Journal of the Endocrine Society was, therefore, timely.
The main purpose of the paper was to determine through NIPS established discrepancies between sex prediction and fetal sex after prenatal noninvasive cfDNA screening (1). This very clearly presented review article concluded that beyond screening for common autosomal trisomies (in chromosomes 13, 18, 21), NIPS offers an opportunity to discover also other fetal and parental sex chromosome variations, suspect other health-related conditions like for example cancers, and predict fetal sex noninvasively.
Physicians, however, must be aware of potential discordances between fetal sex predictions by prenatal cfDNA screening and/or prenatal ultrasound and actual phenotype and sex chromosomes at birth, as neither noninvasive cfDNA screening nor prenatal ultrasound in fetal sex predictions are 100% accurate. Most discrepancies are not due to fetal causes. Atypical sex development of infants, described as “difference in sex development (DSD)” represent ca. 30% of cases.
Reference
1. Witchel et al., J Endocrine Soc. 2025;9:bvaf007
Y-chromosome genes may regulate spermatogenesis even if they show no phenotypes when individually deleted
A consortium of European investigators published an interesting article in Science magazine adding important information regarding which genes on the Y chromosome regulate spermatogenesis (1). The findings using a mouse as well as a human model were quite surprising, because they suggested that that certain Y genes may regulate spermatogenesis even if they show no specific phenotypes when individually deleted.
The issue here is that the mammalian Y chromosome contains several gene families which have been known in some way to be involved in spermatogenesis and, therefore, have also been associated in male infertility. To study which genes specifically control spermatogenesis, the authors - using CRISPR-Cas9 - therefore generated 13 mouse models with specific deletions in Y chromosome genes and gene combinations previously implicated in cases of male infertility. What they found is that in murine spermatogenesis more Y genes are involved than previously estimated which are functionally active in more processes in spermatogenesis than was previously thought.
AZFa, AZFb, and ACFc gene deletions are believed to cause most Y-chromosome associated male infertility and automatically remove multiple Y genes when deleted. To determine which of these genes, therefore, is causing the fertility problem can be very complex. Here discussed paper now informs about the possibility that male infertility may at time be the consequence of cumulative loss of several Y genes, with each of them on their own having only minimal effects on spermatogenesis. The paper, moreover, also noted that some Y genes are also expressed outside of the testis, with such Y chromosome loss now having been linked to aging and related diseases, like cancer and heart disease
Reference
1. Subrini et al., Science 2025;387(6732):393-400
And the X chromosome impairs the development of male fetal germ cell
And since we don’t want to be discriminatory toward the X chromosome, there is also interesting news to report on X (unrelated to Elon Musk). Chinese investigators figured out how an extra X chromosome impairs the development of fetal male germ cells (1).
The development of fetal germ cells (FGCs) regulates the dosage of X-linked genes. However, in which way aberrant dosages of X-linked genes (like, for example in Klinefelter syndrome, XXY) impair FGCs was so-far not well understood, even though Klinefelter syndrome offers a natural model.
This paper demonstrated that in Klinefelter cases most FGCs are arrested in early stages, characterized by upregulation of genes related to pluripotency, along with downregulation of genes involved in FGC differentiation. Overall, the study for the first time demonstrated how the extra X chromosome in Klinefelter impairs the development of male FGCs.
Reference
1. Lu et al., Nature 2024;635:960-968
Paying attention to excessive family cancer history and testing for cancer genes
In an A Piece of My Mind article in a recent JAMA issue, a female physician reported her own personal medical experience to spread the message that an excessive family history of cancers should sound the alarm bells and result in the genetic evaluation of individual family members for cancer risk (1). And that does not only mean screening for BRACA 1 and BACA2 mutations which we all know are associated with significantly increased breast and ovarian cancer risk (they also are associated with increased pancreatic cancer, an increasingly frequently diagnosed cancer in women).
Concerned because of repeated cancers in her own family, the physician was seeking a way “to be watched.” By coincidence, her hospital had an excellent ovarian cancer surveillance program that was willing to accept her, even though she did not fulfill all of their usual criteria (it sometimes is after all, good to be a physician!). She consequently had annual ultrasounds and CA125 assessments. After 17 years of negative surveillance, another family member was diagnosed with cancer, and raising this fact in her medical community, a genetic counselor recommended she undergo a genetic evaluation for cancer risks.
Even though her mother’s cancer had been checked for Lynch syndrome (and was found to be negative), and while she had other gynecological cancers in the family but did not have a family history of ovarian cancer, she decided to ask for a 84-cancer gene panel. Everybody was shocked when this test revealed that she carried a mutation in the RAD51D gene, which denotes a 10-20% lifetime risk for ovarian, fallopian tube, or primary peritoneal cancer risk, and also increases breast cancer risk. As a consequence, she had prophylactic surgery, and a very early-stage ovarian cancer was discovered in the specimen.
Even as a physician, she acknowledged never to have heard about the RAD51D gene! It in principle protects against cancers and, when mutated, becomes permissive. And while the cancer risk from such a mutation is not a high a BRACA mutation, it is, still, significant and there are now quite a large number of cancer genes known, certainly making it worthwhile to test patients who have unusually high cancer rates in their families.
Reference
1. Di Maggio D. JAMA 2024;322(20):1703-1704