Journal Club Global - The Association of Ovarian Reserve and Embryo Aneuploidy

Hello, everybody. I'm Kurt Barnhart, the Editor-in-Chief for Fertility and Sterility, and it is my pleasure to introduce you to A Journal Club Global, an annual event here at the American Society of Reproductive Medicine Annual Congress. So A Journal Club Global is live to the world, live screen, as well as to you in the audience.

So good morning, good evening, or good afternoon to wherever you are, might be in the world. Today we're going to talk about the Association of Ovarian Reserve and Embryoaneuploidy with a very distinguished panel of authors and discussants. It should be a very lively discussion.

It's my privilege to introduce Pietro Bortoletto, who is the Interactive Associate-in-Chief for Fertility and Sterility, and our apologies for Micah Hill not being able to join us today. Pietro will be our emcee and host, and we will, again, look forward to your participation during this event as well. Thank you.

Pietro. Hi, everyone. Welcome.

Thanks for making time out of your lunch break to come join us. We typically have these debates at ASRM every year, and they're typically in debate style, where we pit groups of authors against each other. We decided to do things a little bit differently this time.

Our goal is to highlight authors who have published papers in Fertility and Sterility that have kind of all coalesced around the same theme, which is the Association of Ovarian Reserve and Embryoaneuploidy, and have the authors here, highlighted, and being able to discuss what they did, how they did it, and what they found. But then we've also aggregated a group of experts to talk specifically about this topic and put this literature into context for us. How does it fit into the world? How does it fit into our understanding? I want to start by introducing our authors today, who are each going to be giving us a short three-minute presentation about their individual papers.

Our first author is Dr. Lenny Jaswa. She's an RAI at the University of California, San Francisco. We also have Dr. Yael Stavetsky, soon to be doctor, who's a fifth-year medical student at Weill Cornell, who's here presenting her research.

And then finally, we have Dr. Denny Sakas, who is here in place of Dr. Yuval Folks, who was supposed to be here to present his outstanding work, but instead is back home in Israel, serving as an Army reservist. And our thoughts are with him and all the men, women, and children, both in Israel and in Palestine, who are affected by this conflict. I want to start off by discussing these papers and really have the authors tell us about the research rather than us tell you about it.

So I'd like to invite Dr. Lenny Jaswa to begin and tell us about her paper. Thanks so much, Pietro and Kurt. I'll begin just by saying I really appreciate the opportunity to have this exciting conversation today.

I enjoy listening to this recorded Journal Club content during my commute from Silicon Valley to San Francisco, so it's a real privilege to be participating and sharing ideas today. So our paper, I published with Dr. Mitch Rosen at UCSF in 2000. When did we do this? 2020? And our hypothesis was that a reduction in ovarian reserve quantitatively would be associated with reduced egg quality.

And we used pre-implantation genetic testing as a proxy for egg quality. So what we did was review all of our PGTA cycles from 2010 to 2019, amassing over 8,000 trophectoderm biopsies generated from about 1,200 women over the course of 1,700 IVF cycles, and we really tried to focus on women with infertility. So we excluded PGT cycles done primarily for monogenic disease, PGTM.

We excluded those intended primarily for fertility preservation, and we excluded women with recurrent pregnancy loss. And what we found using generalized linear models with cluster robust standard errors for the modeling was a 23% reduction in the odds of a blast being euploid. I should mention these women were 19 to 42 years old.

And so we did indeed find evidence of a reduction in what we used as a proxy for egg quality in those women with diminished ovarian reserve, which was a clinician-assigned diagnosis. That was our predictor. We did a secondary analysis looking not at the clinician-assigned diagnosis of diminished ovarian reserve, but instead at those women generating the lowest quartile of mature eggs in any given IVF cycle.

We did this as an analog to poor ovarian response, and the findings in that analysis mimicked our primary findings, again, identifying a reduction in euploid rates overall. We then looked finally at whether, once you have a euploid embryo in your hands, when you transfer that to a woman, do live birth rates differ between women with DOR and without it? And at that point, we found no difference in live birth rates following embryo transfer of euploid embryos. So in summary, looking at women 19 to 42 years old with infertility, we found that clinician-assigned diagnoses of diminished ovarian reserve was associated with reduced euploid rates but not ultimate live birth potential of euploid embryos.

Thanks, Eleni. I remember reading this paper as a fellow and thinking, oh my goodness, this is going to open up a Pandora's box. We should study this.

And it's with great pleasure that we have one of our medical students who helped us with this project here to present her work, soon-to-be Dr. Stavetsky. Would you tell us a little bit about what you found? Yes, absolutely. Thank you so much for having me.

It's such a great honor to be here and to speak about our work. So we were motivated basically with the aim of understanding the correlation between AMH as a quantitative measure of ovarian reserve and blastocyst ploidy as the quantitative marker. We chose two populations of women.

One was the general population of women pursuing IVF with PGTA, and the second was a less-studied population of women who pursued IVF with PGTM and who were presumed to share the reproductive potential with the general population. We used all of our data from 2012 to 2020. And we decided to use AMH because we felt that some of the challenges of interpreting the quantity-quality question using available data was that there was a variety of quantitative markers used in the studies.

And there was also the use of clinical definitions of DOR, which we felt is interesting in its own way but may actually obscure the interaction. And so we chose to use AMH, and we used a cutoff of 1.1 to classify women as either low AMH or normal AMH. And we found overall no interaction in terms of euploidy rates as well as aneuploidy and mosaicism.

We continued to conduct sensitivity analyses using AMH quartiles, and we also conducted sensitivity analyses of age categories that were women less than 35 years old, women who were 35 to 37, and women who were 38 to 39. All of our women were less than 40 years of age. These sensitivity analyses actually added pretty important nuance to our study, and namely that was that the non-significant trend, which we have seen, whereby women with increasing AMH classified categorically as either less than 1.1 or not or as increasing AMH quartiles, this non-significant trend where the women with the increasing AMH actually had increasing euploidy rates was entirely driven by the age of the patient.

So when we looked at the age groups, we found that women who were 35 to 37 and 38 to 39 had that trend, whereby those who were less than 35 years old either had a negative trend or nothing whatsoever. So that finding really corroborated our conclusion that it was factors linked to age rather than just the pure effect of having a diminished ovarian reserve that compromises the quality of the egg. So for us, our take-home message from this study was that AMH at this point cannot predict euploidy rates and cannot be used in a clinical way to help patients who are pursuing IVF and would like to have a better idea of what their success rates might be.

Thank you. Dr. Sarkis, tell us a little bit about Dr. Foulkes' study from Boston IVF. Thanks, Pietro.

So firstly, I'd like to acknowledge Yuval who drove this study. He's an exceptional doctor and actually has incredible talents at statistical analysis which has led to quite a number of interesting papers that we've published with him. And what we did in our study, we used aneuploidy rates as a very good marker for embryo quality.

And we looked at the aneuploidy rates in young women who were less than 40 with diminished ovarian reserve which was assessed before treatment, before their IVF cycle. And we also looked at patients who had poor ovarian response post-cycle. So we actually created two groups.

We had the diminished ovarian reserve group and the poor ovarian response group and we used a statistical technique called propensity score matching to, you can do a pseudo-randomised trial if you want, so where you match statistically the patients in control to the patients of interest. So the two patients of interest we had was the diminished ovarian reserve patients and we actually matched them as a 1 to 2 ratio of other patients who had similar characteristics to them. And we also did the same for patients with poor ovarian reserve.

We actually matched them with a higher ratio of control groups. So allowing you to match these statistically gives you a very good way of actually comparing patients who have a slightly similar profile. And as I said, these were all young women who were less than 40, were doing autologous treatment, and we were looking at their first frozen transfer results.

To assess these patients, we actually did a comparison. We looked at the aneuploidy rates of all these patients and what we found was that young women diagnosed with diminished ovarian reserve or poor ovarian response, they exhibited equivalent aneuploidy rates and live birth rates per couple, per euploid embryo, before transfer and this was in the large match population that we had. And the matching was based largely on age, body mass index and IVF cycle initiation characteristics.

There was a lower percentage of cycles with no euploid embryos available for transfer in the diminished ovarian reserve and the poor ovarian response patients, but what we actually found was that the question for these patients is quantity, not quality. So all our euploidy rates, our pregnancy rates, once they had a blastocyst for transfer, a euploid blastocyst, were always exactly the same. They were not significantly different.

So the conclusion for our study was for this group of patients with diminished ovarian reserve or poor ovarian response, it's not the quality that actually is their limiting factor, it's actually the quantity of oocytes they have in the end. So if you can actually improve that quantity by maybe repeat cycles, you will actually have a good outcome for them. So we postulated that it's quantity, not quality, for these types of patients which hopefully will help diagnose these patients and treat them and counsel them in the future.

Thank you, Denny. It's really nice to have three excellent papers published in Fertility and Sterility over the course of the last three years that have looked at the same kind of topic but in slightly different ways. We're going to now introduce our discussants for today and have them tell you a little bit about how these papers fit into our understanding of diminished ovarian reserve and embryo aneuploidy.

Our first discussant today is Dr. Ann Steiner, an REI at the University of North Carolina and editor for FNS Reviews. We also have Dr. Scott Nelson who has my old job, the Muir Head Professor of OB-GYN at the University of Glasgow. I think that's the third time I've made that joke, Scott.

And then Dr. Enrique Shisterman who is the chair of the Department of Biostatistics at the University of Pennsylvania. I wanted to open up our discussion by talking a little bit about biologic plausibility. Does any of this even make sense? I think when we're all kind of considering hypothesis generation, why would this make sense to look at, Scott? How do you think that this fits as a question that's worth investigating? So I think there is some question about what is biologically plausible.

So this all started in 1999 with a research letter in The Lancet where it's associated with women with babies with Down syndrome had slightly higher baseline FSH compared to others. So that was the very first measure we had of ovarian reserve suggesting that there may be a very small and weak association between aneuploidy and ovarian reserve. And then people have tried to evaluate that.

But we know when they're thinking about ovarian reserve, across a lifetime there's essentially a normal distribution. At any given age there's a distribution. And people will potentially follow those trajectories across their lifetime.

And then there's also this accelerated ovarian decline where people are changing centiles. So they start on the 90th centile and then their AMH falls to the 50th centile and then their AMH falls to the 10th and then it falls through. And one of the questions is the mechanism of that accelerated ovarian decline is probably the one that we're thinking about here because that's a pathological process as compared to someone who's just following their standard genetic profile.

So when we're trying to think about chromosomal abnormalities versus ovarian reserve, it's trying to then tease out which of these two populations that we're looking at, those who are changing centiles and have got an accelerated pathological pathway and may end up with an early menopause or premature ovarian insufficiency versus the kind of control population who are just following their standard genetic trajectory. And I think that's one of the things we'll hopefully come on to in terms of discussing it. But you can see that there may be a biological possibility for that because of the DNA sequencing or the DNA mechanism, control mechanism, that are going to facilitate that accelerated decline because they've clearly got something wrong with their pathways that are underlying that.

And I think that's why we think there may be some biological plausibility in this. We know that AMH is a predictor of IVF success rates probably through egg yield predominantly. So any effect size is going to be absolutely small, absolutely tiny.

I think we just need to bear that in mind when we're thinking about how do we interpret some of these studies. What's the effect size we've got to be able to do that? Well, this is now a good question, I think, for Ann and Enrique. Scott mentioned AMH.

You've mentioned the effect size. Is AMH really the best marker we have for ovarian reserve if we're really trying to find an association? And do you think that there are other markers that we should be using that may arrive a little bit closer to the truth? Well, I think it's quite interesting. We can see kind of when we're using the impact of using various or different markers of ovarian reserve.

So in the presentation of the first paper, which is very nicely presented, we saw that it was physicians assigned using the Bologna criteria and which is defined by a combination, two out of the three, of low ovarian reserve by a marker or poor ovarian response or age over 40 years of age. That is very different from what was used in the other groups when we were talking about what is your marker of ovarian reserve. There we were specifically relying on