Journal Club Global LIVE at MRSi 2025: Sibling Oocyte Studies in ART

Experts discuss sibling oocyte trials, PIEZO-ICSI, and microfluidics in ART, evaluating outcomes, design limits, lab impact, and clinical implications.

Articles discussed:

• The impact of microfluidics sperm processing on blastocyst euploidy rates compared with density gradient centrifugation: a sibling oocyte double-blinded prospective randomized clinical trial
• Improved fertilization, degeneration, and embryo quality rates with PIEZO–intracytoplasmic sperm injection compared with conventional intracytoplasmic sperm injection: a sibling oocyte split multicenter trial

• Kate Devine, MD
• Josh Halpern, MD
• Richard Scott, MD
• Eric Forman. MD

• Micah Hill, DO, F&S Media
• Allison Eubanks, MD, F&S Editorial Fellow 

Thank you and welcome everyone to Fertility and Sterility Journal Club Global, live from MSRi in Chicago 2025. I'm Micah Hill, the media editor for Fertility and Sterility. We are delighted to be here once again to have a journal club live from this meeting, which we've been doing for about a decade now, and thank you MSRi, we really appreciate this partnership.

So we have a really interesting topic today on two ways that we can maybe improve fertilization and embryology outcomes. So the clinical aspect will be interesting, as we talk about PIEZO-ICSI and microfluidics, but we're also going to be talking about sibling oocyte trials. It's a really unique feature that reproductive medicine and ART allows us to leverage.

It's got some unique strengths, but also some limitations that we should be careful in our interpretation. So I'm excited to talk about that. The theme of this meeting is the stars of ART, and we have an all-star cast up here today to talk about this.

So I'm going to start at my left, we have Lieutenant Commander Allison Eubanks. She's a first-year fellow at the NIH. She's had two overseas deployments, and she is the very first-ever editorial fellow for FNS.

It's a program we started this year to get young people interested in how to be journal leaders. She's doing such a good job, Kurt extended her for a second year, so thank you Allison. We have Dr. Eric Forman.

So Eric and I have the distinction of being from the class of 2013 from fellowship. He is an HCLD and is the lab director medical director at Columbia. We're delighted to have his expertise.

We have Kate Devine. Kate, as you may or may not know, was a co-fellow of mine at the NIH. She's now the chief medical director and research officer for U.S. fertility.

So she's gone from being my co-fellow to my boss. We're glad to have her here. We have Dr. Richard Scott, who is prior Air Force.

So we have three of the panel here are either current or former military. He founded the program at Walter Reed that I led. He is a professor, an adjunct professor at Yale.

We all know him. We're delighted to have his expertise. He is also an HCLD, so we should have some MD HCLD expertise here.

And last but not least, we have Josh Halpern. So Josh, as you probably all know, is a urologist and an expert here in Chicago. And Josh is the chief medical officer at Posterity Health.

So what I like about this panel is we have people who are very well read. These people are also very critical and skeptical, yet are also patient-facing and deal with the realities of how we take care of patients. So I'm excited to have this discussion.

So Allison, we're going to start with you. Why don't you give us a brief presentation over the next seven minutes or so about these two articles so that everyone can understand what we're talking about. I will say that every article in fertility and sterility gets scored by the reviewers and the editors assigned to them.

And these were two of the top 10 scored articles from last year in the journal. So I'm excited to talk about the quality of these data and what we can learn from them. Allison.

Thank you everyone for joining us today. So I'm going to give a quick overview of the two studies we're going to talk about. I don't think my clicker is working.

There we go. All right. So let's start by framing the big question here.

So can embryology innovations actually improve outcomes in our clinical practice? So again, we're going to look at two sibling oocyte trials. One compares PIEZO-ICSI to conventional ICSI and the other compares microfluidics to density gradient centrifugation for sperm prep. Both use the sibling oocyte design to allow for intra-patient comparisons.

So what is a sibling oocyte design? In these trials, patients with a sufficient number of mature oocytes have them split into two groups. Each group is then exposed to different lab techniques allowing us to control for patient level confounders like age or stimulation response and egg quality. The first study looks at PIEZO-ICSI which uses piezoelectric pulses instead of aspiration to penetrate the oocyte.

The rationale is to reduce cytoplasmic disruption and mechanical damage during the injection. This study included 108 patients with at least six mature oocytes and each patient's oocytes were split 50-50 between conventional ICSI and PIEZO-ICSI and embryo development was then assessed. So these are the design details for this study.

Again, they had to have at least six M2s and oocytes were randomized. One embryologist performed both techniques per patient and extra oocytes were all put in the conventional ICSI category. So if the patient had nine, or sorry, 13, then the odd one was put in the conventional ICSI and this minimizes bias and ensures the technique is only the variable.

All injections were done on the same day, again using a consistent embryologist and culture protocol and some embryos underwent PGT-A and about half went on to frozen embryo transfer. And these are the results of that study. For the microfluidic sperm prep, now on to the this second study, we, in this study you can see that the trial compared the microfluidic sperm prep against the conventional and they, or their hypothesis was that microfluidics yield better sperm and that this can improve euploidy and embryo quality.

Here's the design for this study as well. So the eligibility for this one was at least eight M2s. It was, each of the oocytes were injected using the sperm prepared via microfluidics or DGC and all embryos were cultured under identical conditions and underwent PGT-A.

This again breaks down that trial. So random allocation of sibling oocytes, culture drops randomized, and embryologists were blinded at the sperm prep. Same embryologist performed ICSI for both arms.

And these are the results of that study. So you can see that there was non-significant results for euploidy, blastocysts, or pregnancy, but fertilization rate was significant. So very quickly, we're going to talk a little bit more about this later, but the benefits of the sibling oocyte design reduces confounding and increases power by using the same patient as their own control.

And it's ideal when testing lab techniques like injection or sperm prep, which don't require a whole new stimulation cycle. The limitations, certainly nothing is perfect. The design favors patients with a high oocyte yield.

So it's hard to generalize these results to poor responders. For example, 25 patients were excluded from the PIEZO study for not having enough oocytes ultimately. Again, in summary, both trials showed that the intervention did improve their early lab outcomes, and there was no clinical benefit observed in pregnancy or live births.

And so some questions for consideration. Thank you, Allison. So we're going to spend about 10 minutes doing more like a fellowship journal club style talk and just talk about the statistics, the methods, and think through these things.

So if you graduated 10, 20, 30 years ago, put on your fellow's hat and let's just think through this. So Allison, the first time I ever heard someone ask about what are the assumptions of a test was actually Richard when I was a fellow at the Frontiers course. He asked the fellows, what are the assumptions of a chi-squared test? So when we talk about assumptions, we're really just talking about when do you use a test? What are the statistical underpinnings that design the formulas? So do you think these authors use the right test? So for example, a paired t-test, when should we use that? So a paired t-test is going to be used with continuous paired data with a normal distribution.

You don't require a high sample size for this data because you can get higher power with lower sample sizes. So in this case, the data was paired and so it was appropriate to use. And what about for nonparametric data? If it's not normally distributed, do they use the right test for those? So non-normally distributed data is the Wilcoxon rank is one of the options for that one.

So it's paired nonparametric and that is, it doesn't require the normality. So we feel like they use the right tests when we did this. So that's great.

From a methodologic rigor standpoint, we don't think there's error there. You kind of got to it in your pros and cons. When we have a paired test, does it increase our power? Does it increase our sample size? What are the benefits from a methodologic standpoint of why we should utilize these study designs? So when you have a paired test, you can get a higher power with a lower sample size because of the intrapatient correlation.

So you can either increase your power or in this case, one of these studies kept the power at 80%, the other at 90%, but they massively reduced sample size that is needed to find this effect. So I want to go to Richard and Eric because you were two of the first that I really got exposed to sibling oocyte design. So we'll get to maybe some of the limitations and specifically what you did in one trial to overcome those limitations.

But I'd like to hear from both of you of why you've employed this design over the last 20 years and the research that you've both done. Yeah, I'll start. I mean, I think you hit on a lot of them that there's so many variables in ART that we do our best to control for with statistical models, but can never perfectly.

So at least when it comes to like laboratory outcomes, everything prior to that is controlled for the age, stimulation, laboratory, environmental parameters. So I mean, that's really strong that one intervention, the exposure is likely, you know, the cause of what you're seeing as opposed to even with randomization, there's always, you know, the risk of error. So if you really want to get to the isolate the impact of a new technique, I think it's the best controlled method.

Really just more of the same, but think about circumstances where you're trying to compare two groups, but there's a high intrinsic background variability. That's what this is. And it's even made worse because we don't even know all the factors that impact that variability quickly as relates to sustained implantation.

So it's very difficult to control and comparing one patient to another may not provide a rigorous answer unless you just have massive, massive numbers. So a pair design allows you to normalize for the things you can't measure and you can't see. And that can be all the way to in the era of two embryo transfer, that can be all the way to delivery.

Now it's really typically just through in vitro development. So my next question is really for all of the panels. So I think we all understand this power of the sibling oocyte study.

We can use fewer patients, we have increased power, and we're hearing that the confounding is limited because it's controlled for within each patient herself. But there are some limitations. If we go on to do anything past what we see in the lab, we are now sort of violating randomization because we're not transferring embryos from both arms.

So my next question for anyone on the panel is can we make clinical outcomes based on these sibling oocytes and are those valid? In other words, from these two studies we're looking at, can we look at clinical pregnancy, implantation, live birth, if both arms aren't equally represented? Kate? Yeah, I think there's actually some important differences in these two study designs that could impact the answer to that question and how they may be interpreted for the two different studies. So for example, for the PIEZO oocyte study, the study was powered to detect differences in fertilization and then they went on to look at clinical outcomes as their secondary outcome. But importantly, the embryologist was not blinded.

And so I do think that for sibling oocyte studies, one thing that's very important is, is the embryologist blinded? Because as we just heard actually in the last session here at MRSi, morphologic grading is relatively subjective. And so by the time you're getting all the way out to the point of the clinical outcomes of which embryo do you transfer, one important outcome we can look at is which one was the best, quote unquote the best. But determining the best is very much predisposed to bias by whether the embryologist knew which group that particular embryo was derived from.

So I think yes, we can look at clinical outcomes. It's not nearly as powerful and all of what we've talked about before in terms of increasing our power by a pair design, you know, is greatly minimized in that situation. But I actually have the benefit of working with Allison clinically and we're designing a sibling oocyte study right now.

And reading these two papers next to each other, because what I've read them in the past, I didn't look at them back to back. I very much want to incorporate blinding into our study design. So a wonderful comment.

And I think if you, we did a couple of sibling oocyte studies in the era of one embryo transfers and we still wanted data. And our solution for that was to randomize the patient to this patient's either going to get their best, you know, group one embryo transferred. I'm not looking at morphology.

It's just the best morphology within group one or they get randomized to a group two transfer. And so that way you don't, it helps eliminate some of the biases that could come from, you know, wanting to know more about a group or having a preference, even if it's subconscious. You know, more embryos in one of the studies was picked from one group than the other.

And that really makes it hard to interpret. So I think there's an opportunity for an additional randomization prior to transfer and that's what we've done. I'm so glad you went there, Richard.

So one thing conceptually is we randomized them up front, their sibling oocytes. Once we get to embryo transfer in the era of single embryo transfer, it's no longer valid. We're violating randomization if we just choose from one arm.

So embed a second trial and then re-randomize them there to which arm. So I think that's one great solution. So Eric and Richard, you two did something different in the era of double embryo transfer where you did DNA fingerprinting and transferred an embryo from each arm to help us drill down on day three versus day five biopsies.

Can you just tell us about that and whether you think that's an ethical solution or if that was just a brilliant design for the past that applied to that day that helped us answer an important question? Yeah, I could talk a little bit. And before single embryo transfer became, you know, essentially standard of care, before there were guidelines on all euploid being single embryo transfer, we did several studies involving two embryo transfers. Some where it was known that they were euploid, some we didn't know.

So that embryo biopsy study was actually two studies. It wasn't really comparing day three to day five. It was looking at the impact of day three biopsy and separately the impact of a trophectoderm biopsy.

And in that study, it was two untested embryos essentially transferred on day three or day five, one of which was biopsied and one wasn't. And the concept was that if the biopsy did not have an impact on the reproductive potential, that biopsied embryo should be just as likely to implant. And Richard maybe could talk more about the statistics over here, the McNamar Chi-square for that study.

But basically if neither implanted, they were equal. If they were dizygotic twins, and we could confirm that they were equal. But amongst the singletons, it should be random.

There should be as many biopsied or non-biopsied. And that's where DNA fingerprinting came in. If the singleton either in cell-free fetal DNA or newborn buccal swab DNA matched the biopsy, we knew it came from the biopsy.

If it was sibling, it was the other embryo. We didn't have DNA from that embryo. And that study showed that in a day three biopsy, there was a significant reduction in the chance of the biopsied embryo reaching live birth.

In the blastocyst biopsy, and I listened to your recent podcast and we're very critical of saying it was equivalent. We shouldn't say statistically it wasn't equivalent. They were, I would say, similar, not significantly worse.

Within about 4%. That's the confidence interval range around the trophectoderm. Day three was different.

For the fellows in the room or might be listening to this somewhere, think about a paired study where you're due a two embryo transfer is having three possible outcomes. You have twins, in which case they both had the same outcome. That's non-informative.

And you have no pregnancy at all, and that's also the same outcome, and that's not informative. And then when you get singletons, the proportion of those as Eric just described is what drives the statistical evaluation. And that's exactly what a McNamara's chi-square evaluates.

The key is if you get one, how do you know which one may be the baby? And so when you have a technique that includes SNPs, then you get so many. The ability to distinguish between sibling embryos is, you know, like 10 zeros and then the one. I mean, it's super high resolution.

So it allows you to have a lot of confidence in your result. So Josh, we've heard a lot from the REIs here so far in the embryology perspective. As the urologist in the room, are we under-utilizing sibling sperm studies? Is that a, we have dozens of eggs, tens of eggs, fives of eggs, but we have millions of sperm.

Should we be utilizing this study design more to look at sperm questions? What do you think about this? Short answer is yes. Just call me and tell me next time you guys are going to design something. But yes, I mean, you know, in the prior session we talked a lot about just the lack of randomized controlled trials in reproductive urology, and there are so many reasons for that.

There have been a lot of postmortems on some very famous trials in the reproductive urology space because of some of the technical difficulties in recruitment and conducting these trials. And so I do think this is a really great opportunity to kind of expand our repertoire on the reproductive urology side of things and maybe participate and engage in some of these trials. And I think, you know, one of the things that that may be a good segue to is how I think about, you know, my lens to these studies and the study design.

You know, reproductive urology is actually very underrepresented in these trials. So getting very specific, when we look at kind of male factor infertility as an etiology for infertility in both of these papers, it was only about, I think, 21, 22 percent of couples that were struggling with male factor. And if you think about, you know, the hypothesis, the underlying thought process around something like microfluidics, you would imagine that sperm with the poorest conventional metrics might have the poorest quality.

It might be most susceptible to improvement with some of these advanced processes, right? So if I were designing a trial, and of course nobody asked me, but, you know, thinking about the microfluidic trial, you know, in particular, they actually excluded men with the most severe male factor infertility. And I would actually hypothesize that those are the guys that are going to see the most benefit. And if you look at the supplemental data, certainly not powered for this, but if you look at the absolute numbers in the supplemental data, it actually does look like every one of these outcomes is better for the microfluidics compared to the conventional in the control group when you look at male factor infertility.

So just another plug for, hey, let's be thinking about the sperm when we design these trials, whether it's for looking at these particular interventions or down the road using this very nifty study design to look at some of the other sperm metrics. I didn't think about it until you said it, but they called that study a sibling oocyte trial, but it really was a sibling sperm and sibling oocyte trial, which maybe shows our unconscious bias. Yeah, let's get a shout out to the sperm, right? I mean.

So my last methods question is probably nuanced, and I just want your input. Is a sibling oocyte trial enough for us to say, okay, I'm going to implement this in my clinic? We have medical directors, HCLDs here. Is that enough in and of itself to say, yes, we have it at a sibling oocyte, sibling sperm study that says benefit.

Is that enough to say clinically we should do it? Maybe, I mean, Kate knows so much about this and has discussed these so well before, but I'll just comment that of the sibling oocyte studies, the findings from those studies have really not been overturned. They've been consistent, and what they have found has turned out to be true even with larger studies or longer experiences. So I think it is a powerful model.

It'll always depend on how clean your data are, how close the outcomes are. If you're not getting great discrimination, even if you can define statistical differences, the relevance would be highly in question. So I don't think there's a single answer to that, but I listen to Kate on this.

Yeah, I would agree that the devil is definitely in the details of the study design. What is the exposure? What is the outcome that you care about for your patient in that situation, and how well did the authors address it? So when we talked about it in a session the other day as well, the study that we referenced before where there was a time where we were able to do double embryo transfer, I refer to that study to this day in consults with my patients all the time in order to put their minds at ease about the impact of biopsy of blastocysts, and what I can tell them is that we know it doesn't decrease the probability of live birth more than 10 percent because that's what the study was powered to find. Now I think that that's great, and as Dr. Scott just said, as time has gone on, nobody has done a study that suggested that it was damaging to, you know, within what we would be considering clinically significant, which is less than 10 percent.

And so for that, yeah, I 100 percent think that we can tell our patients trophectoderm biopsy is safe. That said, for these studies, you know, we're looking at studies that were powered the first one for fertilization. While fertilization is something we all care about, and perhaps if they had a large enough sample, we would have gotten to a more clinically meaningful endpoint.

Similarly, earlier in this meeting, we talked about whether we should be doing ICSI for everyone. It's pretty clear that for non-male factor, ICSI results in higher fertilization rates and lower rates of complete fertilization failure, but my conclusion is we should not do it. So, you know, at the end of the day, the devil's in the details, as Richard also mentioned, and we need to look at the exposure, the outcome, and the study design and evaluate it critically.

Kate, you went to my gotcha study, because if you're going to say yes, I was going to bring up ICSI for unexplained, because the sibling oocyte trials all show benefit, but we don't all do it because the large cohort data doesn't show benefit, and we don't have good RCTs at the patient level. Yeah, I was going to say, I think it can be used from like a safety perspective, and I think labs do this when they introduce a new incubator, a new media, and that could be a whole separate discussion, like what level of evidence do you need for that, but I think, again, if you're trying to, you know, justify an add-on that has more cost and say it's better, I mean, I think this is the first step to say you can do it, and it's safe, and you get good results, but now what actually happens when you use it in clinical use, where you apply all the oocytes or all the sperm to this new technology, but that's powerful for introducing something new, making sure it's safe, making sure it works well in your system, and then if it looks promising, you know, that something is better, then prove it in a clinical randomized trial. Can I add just one thing to that? Yes, sir.

Across the board, as there are more adjunctive tests coming on, as we seek to know more and more about the embryos that are available for transfer, many of them actually change what we do in the lab, and so you can think of some of these paired studies, just like it was with the impact of biopsy, was in preparation to study PGT. They also need these types of studies to prove that the intervention itself, to be able to collect a specimen, for instance, is not harmful. Non-invasive PGT is a great example.

You know, the recommendation from one of the original groups marketing this was that you change culture media on day four and culture everyone to day six. Well, I think you have to prove that that is an acceptable paradigm before you can then say it's a non-intervention, non-invasive technique that doesn't impart some risk to the embryo, and that study hasn't been done. I think the other thing that we've touched on a couple times when you talk about using the data from these studies and applying it to your clinical practice is to remember that we're talking about relatively rare population that, given that we're requiring at least six to eight mature oocytes for each arm, so you've got a very high responder and also excluding male factor, so now you've got a high responding female with no male factor.

The applicability to your general practice is going to be very minimal in those cases, so again, just kind of keeping that in mind when we take these results. And I think the obvious reason of why the studies are designed that way, if you have a patient with two eggs or one egg, it's hard to get much out of zero, 50 percent, or 100 percent in each arm from a statistical standpoint. So, I just want to highlight and take sort of the methodologic rigor standpoint.

I think these studies are great for proving proof of concept, and as we've heard for proving things in the lab. I don't know that that always translates to the patient level, so we can think about both these studies. They have about a five percent improvement in fertilization.

Well, the average patient makes 10 M2s, so that's going to give you 0.5 extra 2PNs per patient. A 40 percent blastulation rate gets you to 0.2 blasts per patient. A 50 percent euploidy rate gets you to 0.1 euploid embryos, and if you have a 50 percent live birth rate, you're now at 0.05 extra live births, if you just do the math, which is about a 10 percent live birth per fertilized egg sort of rate.

And so, the clinical trial that you would need to show that that's clinically relevant is massive, and depending upon the cost and the amount of intervention of that technique may not be clinically useful to get that marginal gain, even though these studies show it very simply with just 100 patients. All right, so that's the stats nerdy part of it. Let's talk about what we actually think about these two systems.

Let's start with PIEZO-ICSI, and I'll start with our two HCLDs and our sperm guide. What do we think about this clinically? Do we think, so I've actually seen four different sibling oocytes trials of four different mechanisms of improving fertilization that all show about a five percent benefit over the last year. How should we think about this from what's actually happening in the lab and from that oocyte perspective? Yeah, I mean, I think we heard a lot at this meeting and going forward about, you know, automation, standardization.

So, you know, ICSI is still a micromanipulation skill that takes years for an embryologist to become fully competent with, and this may be, you know, one of several methods of automating that and maybe getting some more consistency. I think it's important to look at the control group. I think in this one had an unusually high degeneration rate in the conventional ICSI group.

So, you know, if you have, you know, again, technical challenges with your kind of standard technique, you might be able to show that something new that's more reproducible is better. But again, we don't see the 12 percent degeneration rate of eggs that undergo ICSI. So, it's hard to say what the comparison would be in a higher performing lab.

Exactly, and such a good point. And so, one of the studies had a 12, in the control group, had a 12 percent degeneration rate. That tells you that this technique may be better than doing ICSI in a way that produces a 12 percent degeneration rate.

That's probably not what any of you have in your laboratories. And so, it limits the relevance. It's excellent that they included the data, though.

I mean, it's outstanding that they presented it, but it's concerning. I think also for some of the things, you know, they're saying that ICSI should overall, the threshold for success would be 65 percent. Most places are now in the high 70s to low 80s, so I think that you have to look at the overall quality of what's going on and the pregnancy rates and implantation rates of what's going on, know how applicable that may be in your laboratory.

So, it's really important. You know, I've never done ICSI myself. You guys have.

Are you a believer that maybe this technique really does reduce stress on the the oocyte and improve? Or do you think maybe we're just rescuing some bad eggs that weren't going to fertilize but don't make good blasts at the end anyway? Or like, how do you think about this? I think that the concept of using a piezo, we, in the early days of ICSI, and this is like the dark ages, we would get the needle through the zona. We didn't make the spike as big in those days, and that was a mistake. And we would push into the oolema, and it wouldn't break through, and we would take a pencil and bop the micromanipulator, and that was our own form of PIEZO, and it would pop through.

I know that sounds ridiculous, but let me tell you, we did it all the time. And I do worry a little bit, though, about these techniques. There's more, I think, we have to see, because if you look at things that can alter activation, vibration, an acute vibratory event is one of the things you can do.

We can do an electrical pulse, you can do it with exposure to ethanol, you can do it with a mechanical insult. And so I think there's more data. I thought this was a very provocative paper.

I thought it was, you know, they said some things like, you eliminate the potential distortion of the oocyte because you can break through more readily, which kind of made sense until I saw that before they did the PIEZO, they put the needle 80 percent of the way through, and they're still not through the oolema. That's called the same amount of distortion. So I think that, I thought this was a very exciting paper, but I think we need a little more data, and we're going to have to look at other things to become comfortable.

So let's talk about microfluidics, and then I want to give us the last 10 minutes to let this wonderful audience ask questions of our expert panel. So Josh and Kate, from the microfluidics standpoint, just give me your clinical gestalt, your overview of this paper and the literature in general, because there's a lot of data on this, but I don't know that we have the answer. What do you think about it, Josh? Yeah, so I think just to follow up on the PIEZO-ICSI, I think, you know, unlike the microfluidics, I think the PIEZO-ICSI is in some ways kind of sperm agnostic, right? So that's really a laboratory technique.

I think clinically, there's not a lot to be said about the individual who's got a normal semen parameter, abnormal semen parameter. Interestingly, in that study, they actually didn't exclude severe male factor infertility, and if you look at the ranges of the semen parameters, they had some folks who actually had very severe male factor infertility, but ultimately, as I think about how I'm going to use these clinically, I'm never going to make a comment about how you should do ICSI. And so this really is, in some ways, out of the reproductive urology clinical territory, and I think for that reason, that kind of, you know, underrepresentation of male factor patients in the piezo study really doesn't bother me at all, because, you know, from a hypothesis standpoint, that really shouldn't matter.

And microfluidics, I think, is, yeah, go ahead, Kate. I just, I know we are wanting to move, but there's so many exciting things to say, PIEZO-ICSI. So, you know, when I talk about PIEZO-ICSI to our embryologists across the network, it is something that, in general, there's consensus, and obviously this is expert opinion, not a sibling oocyte study, that it probably is less damaging, and that they expect to see higher rates of fertilization in well-trained hands.

However, it is technically more difficult, and something with which most embryologists have less training, and the reason why I wanted to go back to it is because of the point that Eric was making about automation, where that may not be a concern any longer. We may soon have robots in the laboratory that can do PIEZO-ICSI, you know, quote-unquote perfectly, and then potentially paired with AI-driven sperm selection, again, once appropriately validated, can't emphasize that enough, this may actually be something, I think, that could make a clinical difference in the laboratory, and help us kind of, you know, break through little by little the asymptote that we have seemed to find ourself at in this field. Perfect, and how about the microfluidics? Let's get some insight on that.

What are the thoughts, Josh? Yeah, well, I think, to some extent, the answer to the question lies within the question, because you're right, we've had super conflicting data on this. There have been some studies that have really showed a tremendous benefit in certain circumstances. We've got others that have shown none at all, and here we have what I think we would all call maybe a modest, at best, benefit.

And, you know, I think the question ends up being, you know, what is the potential downside? We've talked a lot here about, you know, what are the, you know, potential risks of any new, you know, laboratory technique? Is there some unintended consequence that we don't know of? I don't worry too much about that here, but I think it's a reality that we don't know the answer to that. And, you know, I think in this particular case, going back to one of my other comments, when it comes to microfluidics, I think the advantage probably is going to be in men who have severe malefactor infertility. The thought that, you know, there is a sperm quality issue that is unaccounted for may be accounted for by DNA fragmentation, other advanced sperm quality metrics.

We know that there are some pretty good studies showing that sperm DNA fragmentation is enriched when you use microfluidics, but what does that mean downstream? And I think this is, you know, a really good study showing us that it's probably, it's not as great as we all hoped it would be. I do think, though, that some of, you know, the data and the supplements and, you know, you know, the signals are here for us to do a better study, focusing on severe malefactor infertility, focusing on guys with elevated DNA fragmentation. And this study does reference some of the prior papers in that space, one from Cornell, there are a few others, looking at men with abnormal DNA fragmentation.

I think that's where the money's at for future studies. In the meantime, in the absence of having that data, I look at this as probably an unharmful, possibly mildly helpful, adjunct to my other interventions. And I mentioned this in one of the prior sessions here, I never look at microfluidics as kind of a monotherapy for any of my patients.

So in the setting of elevated DNA fragmentation, whether that's RPL or recurrent IVF failure, I'm always thinking about that in addition to something such as short abstinence, right? And so I would never hang my hat on this, but I also am not out on this as, you know, as a potential helpful therapy if we study it better and we're able to demonstrate that there are some significant clinical outcomes downstream. Yeah, I think tons of great points there. And the thing about microfluidics that's nice as well is it actually decreases the work of the lab for the most part rather than increasing it.

So in the terms of the very politically correct framed up statement you made about the potential possible benefit, which is about where I sit as well here, at least we're not increasing our embryologists' workload with this add-on. You know, both of the studies starting to kind of tiptoe back into the nerdy statistical part had pluses and minuses in terms of their study design. This study was powered for the primary outcome of percent euploidy, and I think it's so important in this study who their sample was, how they chose what their inclusion and exclusion criteria were.

And so as Josh said, you know, probably the group in which we expect the greatest benefit are those who have elevated DNA fragmentation index. The one thing we know this tool can do is produce a sample on the other side of the device at the outloading portion that has a lower DNA fragmentation index. The authors say in their introduction that this has already been shown in that group, or at least they quote a study that showed that, but that study did not have anywhere near the rigor of this study.

So I do think that that's the next study that they should do. I'm very surprised given that this study was funded by ZyMōt that, you know, they were willing to take a leap on sort of an unselected population here. The other thing that I would say about this study is they did a little bit of p-hacking here in terms of looking at quite a lot of secondary outcomes and, you know, going to the conclusion that there was a statistically significantly increased fertilization rate, and I think they had a p-value of 0.03 there.

That said, you know, I think they looked at somewhere around 15 different outcomes, and I wondered whether they needed to control for multiple comparisons. So, you know, just food for thought there. Yeah, those are great points.

I think I was thinking the same thing about, you know, they could have looked at the time processing. That's a metric that's useful for their lab, where again this may be an intervention that even if it's similar may have some benefits on the workflow, and then I'm sure Dr. Scott is thinking as well with, and we know over 90 percent of aneuploidy comes from the egg, so it's going to be very difficult to use a sperm selection technique to significantly change whole chromosome aneuploidy. So, I just question that, and again, that's not, again, I don't think that's even the mechanism of if we believe high DFI contributes to recurrent loss.

It's recurrent euploid losses, not sporadic aneuploidy. Yeah, I really wondered why their primary endpoint wasn't DFI before and after or clinical loss rates, because the sperm were involved. They're involved some in aneuploidy, but not massively so.

It is, they really weren't empowered to ask that question if you considered the sperm contribution to aneuploidy, which I think would have to be done here. I mean, my other panelists can can jump in. So, that is, those are the next studies, and I think they would be fascinating.

Those are fantastic points. I feel like we're in a true journal club. I love that.

Let's open it up to the audience. We've got a few minutes. We've got a wonderful audience full of experts here.

I'd love to hear your professional comments or questions for our expert panel. I was just, sorry, I was just curious if any of you are offering microfluidics or the PIEZO technique in your labs, and if so, are you charging patients for the additional use of these sperm selection techniques or the PIEZO-ICSI procedure? We have not introduced PIEZO for microfluidics. We do, and like a nominal fee to cover the cost of the device.

If we use it like on weekend to make, you know, workflow better, then we don't charge additional. That's then our decision. We are similar.

We do offer it. We charge a nominal amount simply to cover the device, and we, you know, there's more than one, as we all know, microfluidic devices available as well with different price points. We don't do PIEZO-ICSI yet, although maybe someday, again, with automation.

I think just to go back to one of the comments that I think, you know, both Kate and Eric made, when we think about cost, there's not a lot of the things we're doing in the IVF lab that are reducing costs. Most of the things we're doing are tacking on costs, and so I do think this is not designed as a non-inferiority trial when you think about the microfluidics, but if we say, hey, there's enough data here to suggest this is really safe, and we have, let's say, a forthcoming secondary analysis where we can do some sort of operational cost effectiveness on the lab side of things, we're always looking for ways. You are all looking for ways.

I'm just an innocent bystander of being more efficient, more cost effective in the lab. You could make the case that there are significant cost savings here to do it just based on that. How much time are we saving in the andrology lab and the embryology lab by being able to use these devices, and I do think, like, that's an undersold potential upside of this.

That's outside of the clinical arena, but super important. I'd like to add that when we do those cost analyses, of course, we're going to look at what it takes for throughput in the laboratory. That's the first thing you do, but think about PGT-A.

PGT-A unequivocally lowers the cost to obtain an IVF pregnancy. It's not even close. You can read Shelby Neal's paper.

She's at Duke now, and across the board, because you avoid futile transfers, you avoid unnecessary cryo for saving embryos that are not, and the decrease in the loss rate, that actually saves quite a bit of money, and it saves more than PGT-A costs. I wonder if the decreased clinical loss rate might be impactful here, and because if you had better sperm, and you had better embryos, and potentially even a little jump in implantation rates, that that also might really make this cost effective, so when they do those studies, I would encourage them to go all the way out to delivery and look at the totality of care provided to get them that far. It may be cheaper than we think.

I didn't want to let it go without saying that the PGT-A comment is something about which reasonable people can disagree. That said, in terms of the... Do you have any data that you disagree with me about? Oops. Another day, another day.

For the microfluidic study, they did go out to look at SAB, but they were in no way powered for that and did see a small, but not enough to say statistically significant decrease in SAB, so what Eric and Richard are saying in terms of the biologic plausibility of how microfluidic sperm selection may help patients in terms of reducing losses even among euploid embryos, I totally agree that that's probably its biggest potential clinical benefit in this population. We're coming up on the end of our time. I just want to give each of the panel a chance to say a last word.

It can be anything you think that you learned from this or any clinical pearl you'd like to give from these papers. Allison, I'll just start with you as our editorial fellow. What did you learn? What did you think about these studies? Then we'll go around the table.

Well, I think as Dr. Devine alluded to earlier, we are developing a sibling oocyte trial, so it was very interesting for me to dissect these studies and apply it to the trial I'm working on, but I do think for me, I think one of the things I always take away from the paper is how can I apply this to my patients? Just remembering the patient population that we're studying and also the limitations of the statistical analysis here. These are good studies because they limit so much variability, but also our patients are not limited in variability. I think a good reminder that even relatively modest-sized programs can do trials with sibling oocyte studies.

You don't have to be part of a network with thousands and thousands of cycles when it's well controlled and you're using especially moderate to high responders. As Dr. Scott said, if we even go out to randomizing the transfer, we can even go beyond just the lab metrics, hopefully. That's a great point.

Even the smaller programs can utilize this study design to make impactful trials within your lab. Kate? Yeah, I just commend the authors on these two studies. They both provide us with, I think, clinically useful outcome data that we can counsel our patients and make clinical decisions with.

Richard? I think these are actually really, we've been kind of picking on them, but well-done studies. They're just another piece of the puzzle. They're an additional step which we need and it's foundational.

I'm not sure that it's enough by itself to justify a change in clinical intervention, but it's a piece of the pie, so that's great. Sperm gets the last word here. No, I think just to go back to the study design and thinking about engaging the reproductive urology arm, the andrology folks in both study designs in the past, which we can too, but as we move forward and we think about what the next phases of these trials could look like, how do we look at the next steps, really having the sperm experts involved in the design trying to kind of anticipate some of these roadblocks? Should we have looked at DFI? Should we have enriched for, excuse me, you know, male factor? I think it's really important for these to be multidisciplinary in their origins.

Well, thank you again to MRSi for having Fertility and Sterility Journal Club Global Live here at this meeting once again. We look forward to coming next year. As always, you can listen to F&S On Air Podcast.

This month will be recorded by Kate Devine, Eve Feinberg, Pietro Bortoletto. We'll be live at ESHRE next month and then live at Sydney the month after for the New Zealand and Australian Society meeting, so encourage you to like and subscribe to the F&S On Air Podcast wherever you get your podcasts, and thank you very much for listening today.