Fertility and Sterility On Air: Live from the ESHRE 41st Annual Meeting (Part 1)

Fertility & Sterility on Air is at the European Society of Human Reproduction and Embryology 41st Annual Meeting in Paris, France (Part 1)! In this episode, our hosts Kate Devine, Eve Feinberg, and Micah Hill cover:

outcomes from programmed ovulatory frozen embryo transfer cycles vs. modified natural frozen embryo transfer cycles with Philippos Edimiris (0:56),

non-invasive preimplantation genetic testing using metabolomics with Sofia Nunes (9:23),

optimizing timing of modified natural cycle frozen embryo transfer cycles with Amelie Johansen and Kristine Løssl (17:46),

mechanisms and consequences of ovarian aging with Francesca Duncan (27:18),

comparing pregnancy loss rates in natural cycle vs. artificial cycle endometrial preparation prior to euploid embryo transfer with Caroline Roelens (38:24),

and endometrial health and reproductive outcomes with Shari Mackens(49:44).

Welcome to Fertility and Sterility On Air, the podcast where you can stay current on the latest global research in the field of reproductive medicine. This podcast brings you an overview of this month's journal, in-depth discussions with authors, and other special features. F&S On Air is brought to you by the Fertility and Sterility family of journals, in conjunction with the American Society for Reproductive Medicine, and is hosted by Dr. Kurt Barnhart, Editor-in-Chief, Dr. Eve Feinberg, Editorial Editor, Dr. Micah Hill, Media Editor, Dr. Pietro Bortoletto, Interactive Associate-in-Chief, and Associate Editor, Dr. Kate Devine.

Welcome to Fertility and Sterility On Air. I'm Micah Hill, the Media Editor for F&S. I'm here with Eve Feinberg, one of our Editorial Editors.

Good morning, Eve. Good morning, Micah. So happy to see you in Paris.

Yes, this is wonderful. We are here live from Paris at ESHRE 2025, so we are here with Dr. Philippos Edimiris.

He is an REI in Berlin, Germany. His abstract caught our attention. It's titled, “Do Programmed Ovulatory Frozen Embryo Transfers and Modified Natural Cycle Frozen Embryo Transfer Protocols Yield Comparable Outcomes?” Welcome, Philippos.

I just wanted to say to our listeners, we are live from Paris, and what we did was we combed through the abstract books, and we combed through all the presentations. We'll try to select those that we thought would be most exciting for our listeners and most relevant to how we practice in the field today. So this one, you're one of my picks, and I was really excited to have the opportunity to talk to you, because I think what we spend a lot of time doing is trying to figure out what is the best protocol for frozen embryo transfer. So I was really intrigued by what you put together with your trial. So if you could tell the listeners a little bit—first of all, introduce yourself, where you're from, and tell us a little bit about why you chose to study this, and describe both of the protocols that you prepared.

Okay, yeah, thank you very much for the invitation. So my name is Philippos Edimiris, I'm from Germany. I just moved to Berlin, so before I worked almost 10 years at a university-based clinic in Düsseldorf, and now I'm in a private setting in Berlin.

And so, why did I choose this research? So we had a bigger multi-center study in which we studied the luteal phase of frozen embryo transfer cycles, and to study the luteal phase, it's important not to get progesterone, but dydrogesterone, because then it doesn't interfere with progesterone or estrogen levels. So that was why we had all these patients in this new protocol that we call P-O-F-E-T, programmed ovulatory frozen embryo transfer cycle. And it's called “programmed” because you just initiate progesterone in the late follicular phase, and then you schedule the embryo transfer. That's why we call it programmed; and “ovulatory” because you still have the ovulation, and you still have the corpus luteum in the luteal phase and in the pregnancy.

So I think, if I read it correctly, it sounds like you allow the patients to develop their main follicle, and you check their lining, you make sure their estradiol is over 180, and once they meet those criteria, then they start progesterone.

Exactly.

And then they still ovulate while they're on this protocol, but the timing of the transfer is really based on when you start progesterone, correct?

Yeah, exactly. When we start progesterone. So it's around 120 hours.

Right. So that's really interesting. So you're actually timing the transfer from whenever you want it, irrespective of when they ovulate, and then you compare that to what I would say is kind of a classical modified cycle, where you trigger ovulation and you time it based on the ovulation trigger.

Yes, correct.

So tell us a little bit more. How did you recruit? Was it retrospective? Was it prospective? And what did you find?

So we had these two groups. The POFET group—that were patients who were prospectively collected as part of this multi-center study—and we had the modified natural cycle patient group, and these were retrospective data from our clinical routine, because we used both protocols. The POFET protocol in this study, and the modified natural cycle was our clinical routine. So it was the idea to just compare them. What is the difference in the reproductive outcome?

How did you match the patients?

So at first the characteristics of both groups were very different, and also the pregnancy rate was quite different, and so we thought we have to match them somehow, because it's not a randomized controlled study. And with the matching, they were quite equal when it comes to the characteristics. So we matched for body mass index, maternal age, how many embryos were transferred, what about the embryo quality, the stage of the embryos—and so it was seven covariates that we matched for. So after the matching, the groups were quite similar when it comes to these characteristics, and what we also saw was that there is no difference when it comes to the reproductive outcome. So live birth rates, there was no significant difference.

And I know we talked about this a lot on the podcast—is our lab doesn't like the unpredictability of modified natural cycle.

Yes.

Do you struggle with something similar?

Yes, we do struggle with that, but for example, our clinic was quite big, so we transferred all days, so Monday, Tuesday, until Sunday. But nevertheless the lab wants the weekend to be free somehow, or just reduced in the workload.

Yeah. You know we are very effective in Germany. We try to be, and maybe this new protocol helps us to be a bit more effective, and on the other hand also more patient friendly, since the patients don't use hCG as an injection. So they only have to use dydrogesterone orally. They don't even have to use micronized progesterone, which is maybe, for me, also an advantage, I would say.

Yeah, I mean that's really nice. And I think that if we spend so much time doing monitoring and trying to chase the follicle—trying to figure out the exact best time to get that follicle—this is really nice.

Yeah.

I don't know—do you know, is dydrogesterone available in the United States? I don't know that it is. So Philip, do you know, are there any other oral progestins that work similarly, in that they can affect different changes on the endometrium, but yet not inhibit ovulation? Do you happen to know that off the top of your head?

No, unfortunately not.

Okay, no, I'd just be curious what similar oral progestins we might be able to try this with for people that don't have this particular one available. So maybe you could also even use vaginal progesterone like the same way you use dydrogesterone. I don't know, it might work, because I don't think that would inhibit ovulation.

So yeah, you're right, that probably would.

Not so—or Proluton, I mean, I'm sorry—not so that you allow ovulation to happen, and it just happens when it happens.

Yeah. But you schedule your treatments for an event or something.

Yeah. And that was also very interesting, because in the PPOS protocol you use that progesterone in Germany to prevent ovulation. Right, but then you start at the beginning of the cycle, but starting in the late follicular phase you don't prevent it.

Okay. So it's interesting.

Yeah. I don't know either, it's an interesting question. I really liked how you said that there were differences, but it's a retrospective cohort, so then you did the propensity score matching, and then you did the regression modeling for the confounders, and then when you normalized the two groups for the patient confounders they were similar outcomes. So I think even though it's a relatively small study, it's very novel, and you did a good job of methodologically controlling—so well done.

Thank you.

What do you think are the next steps? Like, are you going to carry on this research? Figure sample size? Or looking at downstream OB outcomes? Or what do you think are the next steps that you're interested in looking at?

So I think we have to confirm these findings, because as you said, the sample size is quite small, and we can only see big differences between the two groups. So probably a randomized controlled study would be the best to confirm these findings.

Yeah. Well, I want to say thank you so much for coming and speaking with us. I know you're a podcast listener, which is always really nice, and we will send you a copy of this once we're done, but it was great, and thank you so much.

Thank you very much for having me.

You're back. I am joined with Sofia Nunes. Welcome, Sofia.

Hi, thank you very much for this invitation.

Thank you so much for being here. I want to hear a little bit about—I know you spoke this morning on non-invasive PGT, and I was really intrigued to see what we were finding through abstracts, that you were looking at metabolomics.

Yes, which is a little bit different than DNA or non-invasive PGT.

Yes. Yes. Well, we are trying to combine, well, the use of metabolomics—a model—because our goal is to determine a non-invasive profile of embryo culture media as a goal to choose the best embryo to be transferred. So we wanted to check which metabolites would be good answers and to predict the ploidy of the embryos, you know.

So we were really intrigued to see your abstracts. It sounds like you did a paired study where you took euploid embryos and you did aneuploid embryos.

Yes. We did from the same couple, from the same cycle, and then we analyzed it through proton NMR spectroscopy, which is a very good technique because it's highly reproducible. So it can be used in the different spectrometers, or even different operators, and it's less costly than the PGT-A conventional, you know. So we could see there were some metabolites that could give us a good approach to choose the best embryo.

Yeah. What we did—we just, like, we cultured embryos in a time-lapse incubator, and on day three and on day five—so two different time points—we removed the culture medium, and then we sent the embryos to the PGT-A. After receiving the results, we were able to see which ones were euploid and which ones are aneuploid, and then we matched the pairs, and then we performed NMR spectroscopy—proton NMR spectroscopy—and then we did the statistical analysis, then we performed and developed the model, the predictive model.

That's really interesting. So you saw differences in the metabolomics of these embryos since all of these embryos achieved—differed whether or not the embryos were aneuploid or euploid?

Well, we saw—then we compared the different spectra of the metabolism of different embryos on day three—of embryos that were euploid and aneuploid and of the three medium. We didn't see difference in the composition of the metabolites presented. We saw difference in concentrations—that is also very interesting. So we checked on day three and then we checked on day five. We could see that on day five there were more metabolites present in the medium, and also they were present in higher concentrations comparing to day three. This is just a comparison between day three and day five. Then we also checked for the difference in aneuploids and euploid embryos, and yes, there were differences in some three metabolites on day three, and on day five we could find six metabolites—I don't know by heart—but we could find some differences, if you look at seven couples, seven pairs.

Yes.

Yes, they were.

Yes.

Yes, they were.

Yes.

What are the next steps for this research?

I mean, these were just—we were just presenting the results of five couples. We have more data to be presented. It's just thinking—we are analyzing it, you know, to have a—I mean, the model was very robust, was a very good model, but we have more data. We have to increase the predictive model.

So there are six different compounds on day five?

Yes.

Oh, you know—

Yes. But there are other compounds that are also looked at.

Yeah.

I mean, there were also other differences, but those were the compounds that gave statistical difference, you know. The other ones—we could see some difference, but were not statistical, you know.

What can you tell us about the embryos? What differences can you make from those compounds to say euploid embryos, or are they more metabolically active?

Yes. Let's see if I—I mean, we could see that it's like—you saw day five, we've seen day five aneuploid, and a different uptake of amino acids, you know. I cannot give you the right file—yeah, now I don't remember if I have—but we saw some differences, yes, according to the metabolites, yes.

So I guess it's great to see that. Do you think we can eliminate the day three part of the panels?

I mean, no. No, no. Day three gave us a good insight, yes, a good—the predictive model was very good on day three. So I think in the near future—I mean, if we have more data, and then to build a consistent model—I think it will be interesting if we could just, you know, do the test on day three of the samples, and then perhaps we could just, you know, even transfer the embryo without submitting it to the freezing procedure, you know.

Right. In the near future, if we could do that, you know.

How long does it take to run the NMR?

I mean, to run the NMR spectra—it could be in half an hour or one hour. It depends on the spectrometer you are using and the field you have, but yeah, in half an hour you have a result. Of course then you have to go to the statistics and everything, but if you know already which metabolites you have to look for, I think it will be faster.

In general, actually, we have—we use both media. We have sequential, we use sequential medias that we use for this project, and then we use unique media also. Well, we actually are now into difference, just to check whether it's a better option.

Yeah. Got it. Did you see any differences in how those embryos grew, relative to how the embryos normally grow in your lab, with changing their media and taking them out of the embryo scope?

I mean, the embryo scope—I think it's a very good machine. That's why, I mean, I use a unique media in the embryo scope, and outside I use the sequential medium. I think I prefer to use unique media in the embryo scope.

Yeah.

I think you get very good embryos.

Yeah. I guess my other question is, do you think we're going to get there? Do you really think we're going to get to a place where we can—we have to leave PGT-A—look at metabolomics to help answer those questions between embryos and their development?

So it's not better until it is deployed, and do you think that this is the wave of the future?

I think so, especially because, you know, PGT-A is not available—it's very costly—so it's not available to all couples. So whether it's going to substitute the PGT-A itself, or the analysis itself—no, but at least it's a good option to choose the best embryo to be transferred, you know, at least to see if you have a higher pregnancy rate.

Right. Yeah. No, I thought it was very compelling, and I really appreciate your taking the time to come on the podcast to share this with all of us here.

Thank you very much.

I am now joined by Amelie Johansen and Kristine Løssl. Hi and welcome.

Hi, thank you for having us. My name is Amelie Johansen, and I'm a PhD student at the fertility clinic at Copenhagen University Hospital, Rigshospitalet, in Denmark. And I'm Kristine Løssl, a senior consultant and associate professor at the same institution.

I think your pronunciation of your names is much more beautiful than I could do it, so I'm sorry. Anyhow, I thought your study was really interesting, which is why we invited you to come on. I'm wondering if you can share with our listeners what you looked at and why you chose to study that.

So we conducted a multicenter randomized control trial where we investigated the timing of a modified natural cycle frozen embryo transfer. So we randomized women to either undergo modified natural cycle in the cycle immediately following an ovarian stimulation and oocyte pickup, or to be randomized to at least one menstrual cycle of a pause or break, so in the subsequent cycle. And this specific study is a sub-study of that randomized control trial where we looked at the ovarian characteristics in the immediate cycle compared to the postponed cycle, and also the length of the follicular phase and how many ultrasound scans were needed in that particular cycle.

And the reason that the study was planned is that the data is definitely lacking, especially now on the modified natural cycle. In Denmark, at least, it has been the tradition through many years to postpone a frozen embryo transfer for at least one cycle after a fresh retrieval and a failed transfer, or total freeze. So we think it's important—before we offer this to our patients—to know if an immediate cycle, a natural FET in the natural cycle, is good.

Yeah. I think that's really relevant, and I will say that our patients in the U.S.—I think our trends have always been: finish your retrieval cycle, get a period, and then move on. I don't think many of our patients would do well if they were told that they had to rest, or they had to wait for a cycle. So I think that it's really timely, but I will say sometimes, when we bring those patients in for baseline and we do an ultrasound, we see lots of structures on their ovaries, and so I'm curious as to what exactly you looked at and what you found in your analysis.

We did ultrasound scans at baseline and then on the day of ovulation trigger. We followed them throughout the treatment cycle, and what we looked at was follicular structures above 10 millimeters at baseline and then on the day of ovulation trigger, and we categorized them into hypodense and hyperdense structures. And what we observed was that, at baseline, the women who underwent an immediate cycle had significantly more follicular structures above 10 millimeters, which was also an unexpected result. And then we also measured reproductive hormones throughout the treatment cycle, and also measured the length of the follicular phase—so time to ovulation trigger—and what we observed was that the time to ovulation trigger was also longer. So the follicular phase was approximately two days longer in the immediate cycle, which consequently led to an increased need for ultrasound scans in the immediate cycle, because we follow them with ultrasound scans until they're ready to get the ovulation trigger.

Yeah, we do the same. And so basically what you're saying is that, in going immediately from an egg retrieval to a natural cycle FET, it's delaying that follicular phase, leading to more ultrasound monitoring to precisely time your hCG trigger, and therefore your embryo transfer timing, and so you're seeing some alterations in that follicular phase immediately following the egg retrieval, right? And then what else did you look at?

Then we also looked at the cancellation rate, and what we observed was that in the immediate cycle there was a significantly higher cancellation rate—so we had a cancellation rate of 10.5 percent compared to 3.9 percent in the postponed cycle, so quite a huge difference, we would say. And the main reasons for cycle cancellation were spontaneous ovulation, but also difficulties in identifying the dominant follicle due to the follicular structures—multiple follicular structures above 10 millimeters—and then also some had, because they had a more prolonged cycle, a longer time for them to develop a dominant follicle, so we didn't observe that within the time frame that we usually observe.

Yeah. And so this was nested—you said this was nested as part of a larger trial. Did the larger trial show differences in pregnancy rates of those cycles that were not canceled? So of the 90 or so percent that went on to embryo transfer, that weren't canceled, were there differences in pregnancy rates—in live birth—in those two scenarios?

That is a great question. That is our primary outcome of the randomized control trial, and we're still awaiting the final result. So we're in the follow-up phase. We have completed patient inclusion, but we're still waiting for the last result.

And perhaps listeners will ask, “Why bother doing the natural cycle?” Why not perform artificial cycles, which have been investigated in randomized control trials showing no inferior live birth rate in immediate versus postponed cycle? But evidence is rising, and it is a hot topic, that perhaps the artificial cycle carries a higher risk of hypertensive disorders and may have a higher risk of miscarriages and a lower live birth rate.

Yeah, it's not solid, firm data yet, but until we have better evidence we definitely think that performing the natural cycle is the best option.

Yeah, and I will say we talk about that a lot on the podcast. It's a recurrent theme, and one of the other abstracts that we talked about earlier today looked at a modified natural cycle and looked at two different regimens. I think all over the world, one of the things that I love about getting to meet people from all over is I think we're all struggling with these same questions. But I think this specific question—on what are the risks in going from your retrieval cycle directly into your frozen embryo transfer—I think that's a really important counseling nugget, that if you don't wait a month you have a higher likelihood of not ovulating on time, or a higher risk of cycle cancellation. I think the bigger question, as you alluded to, is do we know whether or not, in those that proceeded to embryo transfer, that impacted outcomes? I guess we'll have to speak next year about that, but I think it is really interesting, and it may actually be worth it to encourage patients to wait a cycle. And I'm curious whether or not you are now more consistently telling patients to wait a cycle. I think you said that was your practice, but do these data strengthen that recommendation?

I think we will have to wait until we know the effects on clinical pregnancy rate and live birth rate until we can advise patients more safely, but I think it's important to take into account, both from a patient perspective and from a clinician's perspective, if it's worth it—if there's a higher risk of cycle cancellation and also more ultrasound scans. So it's also time-consuming for the patient to come back and forth more times, and also for the clinic, of course—unless we are able to tell which patient has the higher risk of cancellation. So we will look into associations between cancellation rate and patient phenotype, patient characteristics, but we haven't done that yet.

Yeah, no, that's really interesting. Well, I want to thank you both for coming on the podcast. I would strongly encourage you, once you have your data, to submit it to Fertility and Sterility so that we can all read it in print. And again, thank you, this was really interesting, and I loved meeting you and speaking with you.

Thank you for inviting us.

Thank you for inviting us.

Hi and welcome. I'm now joined by Francesca Duncan, who's very well known to me. Francesca's at Northwestern, and I'm going to have her introduce herself, and we're going to talk about her fantastic research on ovarian aging.

Thanks, Eve, so much for having me on this podcast. As you said, my name is Francesca Duncan, and I'm an associate professor at Northwestern University, where I'm also co-directing our Center for Reproductive Science. I am an egg and ovarian biologist by training, and our lab focuses really on mechanisms and consequences of female reproductive aging.

Thank you. So I think one of the most interesting facets of your research is some of your newer research on ovarian fibrosis, and I just want to highlight—we were talking a little bit offline before we started recording—Francesca was saying that at a previous ASRM people would stand up and talk about how difficult an egg retrieval can be in an older patient, and I think we all sort of know that tactile sensation when you're going into an ovary of a woman in her 40s—how sometimes it's just harder to retrieve than that nice, softer feeling of a woman in her 30s. And so, Francesca, I'll let you take it from here, but I want you to just tell a little bit about your story and a little bit about the work that you're doing surrounding aging and fibrosis, and how that observation really guided some of your really brilliant basic science work.

Yeah, thanks, Eve. So as I mentioned, I am an egg biologist by training, and as a basic scientist I use a mouse as a model system. The way that we study eggs and follicles is typically in isolation—so, in basic science, we take our ovaries from our mice and we use mechanical means, like with insulin needles, to poke the eggs or the follicles out of the ovary. And so we were studying reproductive aging, and we wanted to compare follicles from young and old mice, and so we had the ovaries from young and old mice and we did this poking strategy to get the eggs out. An observation that was made by many people in the lab, ranging from undergraduates to PhD students to technicians who were doing these procedures, was they would always say, like, “It's harder to get the follicles out of the old ovaries compared to the young ovaries.” Obviously that wasn't our primary focus—our primary focus was studying the eggs within the ovaries—but this was always such a robust observation that I felt like we had to look into this further.

At the time when we made this observation, I was at the University of Kansas, and there was a researcher in pharmacology/toxicology who was doing a lot of work on liver fibrosis, and, you know, fibrosis is where you have inflammation and you have excess accumulation of extracellular matrix in a tissue, which often makes it stiffer. And so I started to think, okay, well maybe fibrosis is happening in the ovary, and so I partnered with her and said, “What are the tools that you use in the liver to study inflammation and fibrosis, and let's just take a look and see if this is happening in the ovary with age.” And lo and behold, we used several different methods to look at inflammatory markers, but also stains and biochemical assays to look at collagen and fibrosis, and what we found was that there is, in fact, with the aging ovary in the mouse, an increase in this fibrosis and inflammation. This is common in aging tissues, so if you think about the liver, the lung, the heart, there's an increase in fibrosis and inflammation, but this was super interesting to us because the ovary is aging so much earlier relative to these somatic tissues, and so really it becomes an important model system to study, you know, how is fibrosis developing? Can we learn from the ovary and extrapolate or learn about these other tissues?

So this, which was sort of a side observation, really became a primary focus in the lab, and over the past 10 years we've tried to elucidate why this matters. So we've been able to show that a stiffer ovary can affect how follicles grow and the quality of the eggs that they produce. We've also been able to show—we and others—that a stiff ovary can block ovulation from happening with age. And also, in ovarian cancer, it can have a predisposition to ovarian cancer.

Yeah, that's fascinating. And so it sounds like ovarian stiffness could potentially be a factor in something like PCOS, where you have a high density of follicles. Are you seeing that?

Yeah, so PCOS is certainly associated with a lot of deposition of extracellular matrix and an increase in stiffness. Now, whether that's happening from the same mechanisms as aging or different mechanisms, we don't know, but certainly an aging ovary and a PCOS ovary are characterized by increased fibrosis and stiffness.

So hopefully, though—I know you've done some work using ultrasound to look at ovarian stiffness, and some of our fellows are doing research with you for their thesis projects, which is amazing—but can you talk a little bit about the use of ultrasound as a tool to measure ovarian stiffness, and some of those observations that you're seeing?

Yeah, so this actually was what stemmed out of what you referred to in the introduction, in terms of when I would go present this work at ASRM, several colleagues would remark that they knew exactly what I was talking about in the stiffness—what we saw in the stiffness of the mouse ovary—because in egg retrievals they would feel that when they would go aspirate the egg, that there would be this resistance that they would feel, and they were likening that to the stiffness that we observed in the mouse. And so this really got me thinking that we have to find a way to be able to, with minimal invasive mechanisms, measure the stiffness of the ovary. We think that this is important for two reasons. One is it could be a biomarker of itself—of ovarian aging or egg quality—but also, as we think about interventions to reduce the stiffness of the ovary, we have to have a mechanism to be able to see if our intervention is working.

And so that's where we started to look at what are the techniques that are available to measure stiffness of organs, and that's where we came across shear wave elastography, which again is what's used clinically in the setting of liver fibrosis to measure stiffness. The idea is that with the ultrasound you propagate a pulse through a tissue—the stiffer the tissue, the faster the waves are going to propagate; in a soft tissue, the slower those waves will propagate—and then that can be measured by an algorithm. So we essentially applied shear wave elastography to the ovary using transvaginal ultrasound, and so, as you alluded to, we have several ongoing studies at Northwestern where we're recruiting women who are coming in for assisted reproductive technologies. At their baseline or diagnostic scans we do shear wave elastography, and so we've been able to do this study in distinct cohorts of women and show that the stiffness parameters increase with age—both the maximum stiffness and elastic heterogeneity. These two markers increase significantly with age, in defined cohorts but also in a broad pilot cohort study.

Are there any anti-fibrotic therapies that are used in other organ systems that may be able to be used in the ovary?

Yeah, so that's an excellent question. We're really interested to see if we can actually target this therapeutically. We've done preclinical studies in mice using pirfenidone, which is an FDA-approved drug for idiopathic pulmonary fibrosis. Unfortunately—while it works well in mice to reduce fibrosis and extend reproductive longevity—unfortunately it has liver toxicity in humans, so we would never be able to move that forward. So what we're doing now is looking for specific anti-fibrotic targets that could be specific to the ovary. We're trying to understand what are the matrix components that are changing with age—are any of these specific to the ovary—and then can they be targeted with various drugs. So that's one angle. The other angle is that we do know there are drugs that are existing and have high safety profiles, like metformin, that have been shown in animal models and preclinical human models to, in fact, reduce ovarian fibrosis. I would love to be able to look at a trial with metformin use to see if we can actually modulate that stiffness parameter in an individual. So I definitely think this is a hot area with a lot of promise.

Yeah, I think it's a really interesting area too. I mean, I think we've been so focused on follicle numbers and follicle decline, and I think a lot of this is—what is the framework in which those follicles grow and develop? I think by extending to looking at fibrosis, you're probably learning a lot more about that follicle decline and how that happens. I really like this idea of ovarian environment as something to target, because the concept is—I always think about the environment as the nest of where the egg is developing. And so you could think about, you know, you can have the eggs, but if the nest where they're developing is compromised in any way, those eggs and their number or quality could be impacted—so either, you know, from direct effects from that nest or indirect effects. We're trying to figure out right now, you know, is it going to impact quality or quantity? What I can tell you—where the data is starting to emerge—I have a feeling that this is going to be more impacting the quality of the eggs versus the quantity. I think that will actually be really interesting, because we don't have great markers of quality. I mean, AMH is a wonderful marker of quantity but tells us nothing about quality, and so if we have something that differentiates, let's say, a young patient who has ovarian stiffness who may have worse quality, I think that's terrific—that we may soon have a marker.

So I could talk to you for hours, truly. In the spirit of the podcast, we're going to wrap up, but I just want to thank you for coming on. Your work is amazing. I love hearing you tell the story and the evolution of it. It is like—I always think it's like storytelling for science geeks like me. But seriously, amazing work, and thank you.

Thanks, Eve, and we'll definitely keep you posted on our successes.

Good afternoon, F&S listeners. We're back here at ESHRE, live in Paris. I'm here with Dr. Caroline Roelens, and she's going to be telling us about her randomized controlled trial that she conducted looking at frozen embryo transfer protocol. Welcome, Caroline.

Thank you very much for the invitation, and thank you for giving us the possibility to present our work and our RCT. So, as you mentioned, we did an RCT comparing natural cycle and artificial cycle endometrial preparation prior to euploid embryo transfer, and our main outcome was early pregnancy loss rates.

Okay, and what made you decide to look specifically at early pregnancy loss and comparing the two protocols?

Yes. It was a feeling that we had internally, that in these artificial cycles we felt institutionally that we had a higher risk of early pregnancy loss rates. For powering the sample size, we also looked into our data, and we really saw a difference already, in a retrospective way, in our own data comparing natural and artificial cycle preparation, and that was, in fact, the leading course to start the study.

Interesting. So tell me a little bit about the specifics of the protocols that you used, and then, of course, what did you find?

Yes. So we decided—because the study was set up back in 2019, and we all know that between 2019 and 2025 we really have seen a completely mind change, in every case in Europe, with regard to preparation for frozen embryo transfer cycles—so back in 2019 we decided to compare a complete true natural cycle, so with a spontaneous LH surge and without luteal-phase support, and we compared it to an artificial cycle using estradiol valerate and micronized vaginal progesterone 400 milligrams twice daily. This was the starting point at that time, and we discussed during the study whether, with the increased knowledge of measuring progesterone on the day of transfer, we should include that in the study, but we decided then to not do it and to wait for the results and see what happens.

Got it. So what did you find?

Well, we found indeed a higher early pregnancy loss rate in artificial cycles as compared to natural cycles. So we had, results, 21.6 percent early pregnancy loss rates in artificial cycles compared to less than 9 percent of pregnancy loss rates in natural cycles.

Got it. So about double. And then what about for secondary outcomes—were there any differences there?

Well, in live birth rates we also saw higher live birth rates in natural cycles as compared to artificial cycles, as well in clinical pregnancy rates at 12 weeks. Although—and that is important to mention—we halted the study prematurely because of increasing knowledge with regards to obstetrical complications, and we found it at a certain point unethical to still include ovulatory patients with regular menstrual cycles into the artificial cycles, and that's why—and that is important to mention—the results that we have are results from 356 patients included, although the calculated sample size was 261 in every group.

Yeah, I saw that you had stopped prematurely. So do you, at your clinic, now do modified natural cycles or true natural cycles for every single patient that's ovulatory?

Yes, yes. We try to. It's true that for ovulatory patients we do it, and we do not feel hampered by—because a lot of people think that the flexibility within the natural cycles is less—but we don't really feel that we are hampered, and I think, also in terms of patient friendliness, that patients prefer natural cycles, at least in our center.

And at your center, you're able to tolerate the strain that puts on workflow and irregularities in your schedules, etc.?

We are quite able to manage that, yes. We are implying also—so this study was using true natural cycles—we, of course, trigger ovulation with hCG, which already implements a certain amount of flexibility, and the NPP, the so-called NPP protocol by Samuel Santos-Ribeiro, we are also applying it. With all these different measurements we really can avoid weekend days—or try to avoid, as much as possible, weekend days. It's a little bit more of a puzzle than with artificial cycles, but we manage to do it.

Yes. Got it. And so you mentioned, as the reason for stopping the study prematurely, that you, as a group, felt it was unethical to continue offering artificial cycles in the setting of new knowledge about increased obstetrical risk in artificial cycles. Did you see any adverse pregnancy outcomes in your population with artificial cycles?

Well, of course—and that is something to mention—it's a very small group, so 356 patients. We all know that the incidence of preeclampsia is not very high, and that if you want to see real difference you need to include much more patients. But for the final manuscript we will indeed look at that, and we will communicate the results, of course always with the side mark that the study was not powered for these outcomes.

Yes, understood. I would challenge a little bit that we have definitive data. I know many in Europe—and I've had a lot of good conversations today—have kind of hook, line, and sinker gone towards natural cycles, and a lot of the reasons I think you gave are excellent reasons to offer it. That said, I would challenge that we still don't have level one evidence to show that there's an increase in the risk for preeclampsia or hypertensive disorders of pregnancy. But we shall see.

One more question I have for you is, what do you—how do you think your results would have been different, if at all, had you measured and supplemented those that had lower serum progesterone concentrations, and/or if you had used systemic progesterone in addition to just vaginal, or instead of just vaginal?

Yeah, that's indeed a very good question. To answer the first question, we have internal data—so we have my colleague, Shari Mackens, who has a publication in RBM Online about rescuing artificial cycles—and in that study we showed that 21 percent, approximately, of our patients showed low progesterone levels at the day of transfer. In natural cycles—I'm preparing also manuscripts—and there we see that 18 percent of patients experience low progesterone levels, and so these values are quite alike. They are around 20 percent of patients experiencing low progesterone levels. So I think that if you would have measured progesterone in both groups, maybe the results would even be higher than what we measured, but I don't know whether the difference would have changed so much if you do the same approach in both endometrial protocols.

And do you think that having a low serum progesterone concentration in a natural cycle matters? Do you think it has the same implications for the cycle outcomes as in artificial cycles with vaginal progesterone?

Well, if we look at the data that is out there, we have an RCT from Wanggren in true natural cycles where they really see a difference with or without luteal-phase support. On the other hand, we have also the trial of SOPSAT—the Danish group—who presented last year at ESHRE their study, and that was in modified natural cycles, and there it appears that luteal-phase support has less importance than in natural cycles, probably because of the hCG-induced support of the corpus luteum. But, of course, these are assumptions that we are making. I only think that with the results that we have now, performing again the same trial but by measuring progesterone—with the data that we have—I think it would be difficult to defend running the same trial. So I think that, with the data that we have now, we are prone to switch more towards these natural cycles, and, of course, we keep on investigating in these artificial cycles, because for POI patients or for really PCOS-resistant patients this is still the only possibility that they have in order to prepare the endometrium for a further transfer.

Yeah. And so, one final question, just as an expert wanting to get your opinion—I don't think we have data on this—just a couple of weeks ago we recorded the July edition of Fertility and Sterility, and Eve and I had a robust conversation around the appropriateness of natural cycles and modified natural cycles among older patients. So I noticed that in your study your patient inclusion criteria required them to be younger than 42 years old. Why did you do that, and what are your thoughts clinically on offering natural cycles to older patients?

Well, yeah, again, I think we made the same reflection, because what we see in older patients is, of course, that this follicular phase is getting shorter, and there you might think that maybe then the preparation of the endometrium would be less good for a transfer. Of course we included 42 as a cut-off level because also for the probability of having a euploid embryo, I think that if we would have gone higher maybe we would have had much more difficulties by including these patients—so they would maybe not even be into the study group. On the other hand, there is a paper coming out by the group of Santos-Ribeiro—I don't remember the first author anymore—but they looked into the difference between artificial and natural cycles in older patients undergoing oocyte donation, or after an oocyte donation program, and they did not see any difference when using natural cycles in older patients. So that is quite robust data. I think it must appear quite quickly—I think it was accepted in Human Reproduction.

So yeah, I'm looking forward to see the complete manuscript.

I'm looking forward to that too, because I know we're all afraid of it. That said, we think that it's probably the follicular phase that's getting shorter in these patients for the most part, and we don't know whether it actually is a clinical luteal-phase deficiency that occurs with age. So I'm glad to hear that there are reassuring data coming along. Well, thank you so much for joining us and for presenting your work today. We'll look forward to reading your manuscript as well.

Thank you very much for the invitation.

Here this afternoon in beautiful Paris, France, and I have Dr. Shari Mackens here with us. I just had the privilege of listening to Shari, along with Baris Ata and Nick Macklon, give a panel on endometrial health and reproductive outcomes, and Shari in particular gave a really great talk on microbiome and how the microbiome impacts reproductive outcomes—and particularly chronic endometritis. So, tell us, what are your thoughts? How do you practice today in 2025 in consideration of the microbiome, Shari?

Yeah, thank you, Kate, for inviting me, and to give you a small overview of the talk that I just gave: well, I think that the main message is that we are not quite ready to use nor microbiome testing nor chronic endometritis the way that we diagnose it today in daily practice. There are a lot of loose ends at the level of the diagnostics, but definitely also at the level of the therapeutics, where we do not really know how to handle this.

Yeah, I couldn't agree more. I also think that the diagnosis of chronic endometritis, in and of itself, is not even ready for prime time. We really don't know what it is, or if it is, in many ways. And you mentioned, or did a really nice review of the data that we have so far, looking at how helpful it is to determine plasma cell concentrations in endometrial biopsies—which is to say, unfortunately, not very helpful. So what do we say to our patients who are requesting this testing?

Well, I think that with the data that we have now, and definitely also with some new data that have been presented here at this ESHRE Congress—for the CERM trial, for example—I think we now really have arguments to also explain to our patients that they don't need the test. Before it was more difficult—it's kind of a historical diagnosis; it has been around for such a long time. An easy test: hysteroscopy, plasma cell count—not expensive—which probably led to the fact that it was there in our clinics for a long time without having the correct evidence. Now the evidence is pointing towards the other direction, which will probably give us the tools to hold back patients more from this non-evidence-based test.

Yeah, I'm very excited as well that we'll be interviewing, later in this meeting, the author of an RCT that showed no benefit from doxycycline treatment in the setting of a diagnosis of chronic endometritis. So yes, armed with negative studies—we don't only have to publish those positive studies—these are things that we can use to help counsel patients to do evidence-based treatments. You would expect that if you have plasma cell counts that are elevated, and if you would expect that an antibiotic treatment would cure these plasma cells, that you would also have different microbiome testings. But we actually see that these profiles are not having the potential nor the accuracy to really discriminate between the chronic endometritis patients and the ones that aren't—if this is a pathology, which is probably the very first question that we should also—

Yeah. Well, I really enjoyed the combination of the three talks today, and I really liked something that Baris Ata acknowledged, which is to say, you know, there's been a lot of criticism of studies showing or suggesting that recurrent implantation failure may not be a true phenomenon—right, that a true biologic resistance to implantation does not definitively exist. That said, we do see in studies like his and others that there's a very small decrease, with large enough sample, in the probability of a successful implantation with a transfer. So, small enough that I still believe we should reassure our patients to keep calm and transfer another embryo, but there probably is some endometrial barrier for some patients, and it may still lie in the microbiome for some of these patients. So how should we be studying this? If you could design the perfect study, what diagnostic test would you use, and how would you incorporate it? Tell me what you would do.

Yeah, that's a very difficult question, but I think what we should do is join forces, because, as you just explained, there is this small percentage of patients that will probably really have an insurmountable endometrial cause of not implanting. They are not passing at our consultations each and every day. So I think we first would need to join forces to have a sample size which is high enough, and then I would not immediately pick one or the other test nor one or the other therapeutic, but I would biobank for these patients. If we would be able to biobank blood, endometrial, microbiome samples, and then we follow up on these patients in a non-selective way—by preference with embryos—because it takes away so much noise for implantation, I think that this is the only way where we will learn, because probably this patient population is also very heterogeneous. Some will have an immune problem, others will have a microbiome problem, others might have a transcriptional problem, and it's a very, very small set of patients with probably very different reasons to not implant. So joining forces is essential, I think.

I love that idea. Our editor-in-chief, Kurt Barnhart, really dislikes the term “non-selective study,” but we don't have another better term—really “diagnostic accuracy study”—and I think that's a wonderful way to study this problem: to biobank. But it is such a huge undertaking if you think about taking vaginal samples, endometrial samples, also needing to do genotyping of these patients, and then also knowing when in the menstrual cycle to do this. So we have so much to learn, but I really appreciate all the work that you're doing to advance our knowledge base, and I really enjoyed your time today. Thank you so much.

Thank you.

This concludes our episode of Fertility and Sterility On Air, brought to you by Fertility and Sterility, in conjunction with the American Society for Reproductive Medicine. This podcast is produced by Dr. Molly Kornfield, Dr. Adriana Wong, Dr. Elena HogenEsch, Dr. Selena Park, Dr. Carissa Pekny, and Dr. Nicholas Raja. This podcast was developed by Fertility and Sterility and the American Society for Reproductive Medicine as an educational resource in service to its members and other practicing clinicians. While the podcast reflects the views of the authors and the host, it is not intended to be the only approved standard of practice, or to direct an exclusive course of treatment. The opinions expressed are those of the discussants and do not reflect Fertility and Sterility or the American Society for Reproductive Medicine.