Would you use an algorithm to select your embryos? Enter Orchid, a company that promises parents the ability to protect their future children through genetic testing for embryos before pregnancy. The founder, Noor Siddiqui, and Ross debate the scientific, moral and ethical implications of designing a “healthy” child and what we lose in separating reproduction from sex.
Below is an edited transcript of an episode of “Interesting Times.” We recommend listening to it in its original form for the full effect. You can do so using the player above or on the NYT Audio app, Apple, Spotify, Amazon Music, YouTube, iHeartRadio or wherever you get your podcasts.
Ross Douthat: Noor Siddiqui, welcome to “Interesting Times.”
Noor Siddiqui: Thank you so much for having me.
Douthat: I want to start by talking about what it is that your company, Orchid, does and promises. And I thought I’d try and do that by presenting myself to you as a prospective client.
Now, this will require a little bit of imagination, since I’m actually a 45-year-old man who has many children. But let’s imagine for the sake of this conversation that my wife and I are about 30 years old. We are healthy and not, to our knowledge, infertile. We want to start a family, but we’ve read that genetic testing is advancing rapidly. Like most people, we have some medical issues on either side of the family tree, and we’re interested in doing right by our potential offspring.
So we come to Orchid for a consultation. What do you tell us that you offer?
Siddiqui: What Orchid can do is it gives parents the power to protect their children before pregnancy begins. What happens today in I.V.F. centers is that they’re operating essentially almost blind. This really, really critical decision about which embryo to transfer happens with extremely limited information. What happens is that the embryo that looks best under the microscope kind of wins this morphology beauty contest and is often the one that’s selected. Other times there’s a very limited genetic test that’s offered that looks at a tiny fraction of genetic diseases that could affect a future baby.
Orchid completely changes that. We’re the first company in the world that allows parents to actually sequence the entire genome of an embryo, sequencing 99 percent of the bases in an embryo’s genome, which allows parents to detect risks for some of the most serious conditions — heart defects, birth defects, pediatric cancers, developmental disorders — things that massively change the trajectory of a child’s life. And the vast majority of these diseases don’t have cures.
What’s really exciting about this possibility is that now parents have this ability to protect their children from an entire category of disease that previously we had to just hope for the best and wish that our children wouldn’t be affected by them.
Douthat: Since it’s going to be important to the larger conversation, one of your views is that this kind of testing and this kind of process isn’t just for people who are older parents or who have super high risk factors or who are struggling to conceive, right? Your argument is this is potentially for everyone who wants to protect their kids from these kinds of conditions.
So you would say to me as a healthy 30-year-old male with a healthy wife that we should consider doing this, right?
Siddiqui: I think that no one should ever be pressured into any decision, especially a medical one. So I would say it’s the same as giving women the opportunity to choose an epidural or not, or to choose a home birth versus a hospital birth. I think these are incredibly intimate and personal decisions, and I do think that it should be part of the menu of choice and everyone should consider it, but obviously, no one should be compelled or coerced to do it.
Douthat: Right, no, I wouldn’t assume that you would be locking us in the room. But you would be presenting it as a reasonable option for someone in our position to consider.
Siddiqui: Yeah, definitely. I think it’s something that every parent should consider. Yeah.
Douthat: So we’re interested. We decide to pursue your process. What happens next? We go through a cycle of I.V.F. and produce some number of embryos, right?
Siddiqui: Yeah, exactly. For people who aren’t super familiar with I.V.F., the way that it works is that you’re on medication for 10 to 14 days. What that medication does is it actually synchronizes the follicles in your ovaries, and what I.V.F. does is it captures those eggs.
If you’re familiar with egg freezing, egg freezing and I.V.F. are actually identical processes. The question is: When you tap those eggs at that retrieval that happens typically between days 12 to 14, what do you do with those eggs? Do you go ahead and freeze them? Or do you fertilize them with your partner’s sperm?
Once you fertilize them with your partner’s sperm, those embryos grow for five days. On day five, they typically have about 125 cells. If you choose to use Orchid, five of those cells get sent to Orchid’s lab.
The procedure to actually sample cells from embryos, you can think of it as a haircut. The embryo has an outer membrane called the trophectoderm — those are the cells that are actually sampled. Basically, what’s really powerful about Orchid is that instead of getting that really tiny amount of genetic information, you’re able to actually get the entire genome — 100 times the data — and then use that to make these really important decisions.
Douthat: And this is called whole genome amplification, right?
Siddiqui: It’s called whole genome sequencing, and the first part of whole genome sequencing is amplification.
Amplification means you’re copying DNA. So if you have a blood or saliva sample, there’s no need for amplification because the minimum quantity of DNA that you need is already there. You basically lyse those cells — you break open those cells — you take the DNA and the nucleus, and you have enough DNA to just throw it on a sequencer.
What we’ve invented is a new protocol for amplification, a new way to copy DNA so that you get really high uniformity of coverage and really high-quality data off that really small sample size.
Douthat: And you’re the first company to do this at all? To do it commercially?
Siddiqui: Yeah, we’re the first company to clinically validate it. That means we’re able to do it on embryos at comparable quality to blood or saliva, and that hasn’t been possible before.
Douthat: OK, so we decide to go forward, my spouse and I. We do the I.V.F. cycle. What would be a typical number of embryos that we’re testing?
Siddiqui: How old are you? And how old is your wife?
Douthat: We’re 30 in this hypothetical.
Siddiqui: For me, I got 20 eggs and I got 16 embryos, and I was, I think, 28.
Douthat: So let’s say we have 16 embryos.
Siddiqui: So what happens is that of those 16 embryos, you take five cells at day five, and they get sent to Orchid’s lab. Then we come back to you in three to four weeks with reports on each of your embryos.
Then a genetic counselor, someone who’s an expert on genetics, will meet with you and your partner, and then you guys will go through the risks of each embryo and then make a decision between you and your doctor about which one you actually want to go forward with and implant. And then that implantation will hopefully be successful and lead to a pregnancy.
Douthat: How many conditions are you doing scores for, for each embryo?
Siddiqui: It’s helpful to go through all the different analysis that happens. The first layer of testing is chromosomal analysis. Chromosomes, you can think of them as like chapters in a book. That’s where genetic testing that’s widespread today during I.V.F. stops. That’s akin to if you had a proofreader for your book and the only thing that they’re able to tell you is: Hey, your book has an extra missing chapter. Like, that’s not very good. You’d obviously want to be able to check for typos, check for whether all the sentences are capitalized, and do they have a period?
That’s the level of resolution that Orchid’s able to go. It’s able to actually read every single gene and look for typos that could lead to diseases in each of those genes.
Douthat: So that’s the first stage.
Siddiqui: Yeah. So the first stage is chromosomal analysis: Does the embryo have the correct or incorrect number of chromosomes?
Then the next stage is monogenic analysis. That’s where we’re looking for over 1,200 monogenic disorders. These are things like birth defects, heart defects, skeletal defects. These are pediatric cancers, adult-onset cancers. These are neurodevelopmental disorders. These are where over the last 20 years, geneticists have cataloged: Hey, there’s this specific gene, this specific variant, and it leads to this disorder.
Douthat: So these are monogenic, right? You’re talking about problems that are caused pretty clearly by a single gene. How many of these kinds of things did you say you’re testing for?
Siddiqui: It’s a little bit over 1,200 conditions.
Douthat: What about the next level, which is polygenic conditions? These are conditions that then are influenced by multiple genes. And you’re testing for those as well?
Siddiqui: Yep. Yeah, exactly. I think it’s just really important not to skip over this monogenic side.
Currently, you can’t test for that large of a set of monogenic diseases during pregnancy. What’s offered today is just something called NIPT, in which at 10 to 12 weeks you can get Down syndrome screening or chromosomal screening. And women are in a really difficult position, because if that test is positive, if Down syndrome is detected, the only choice that they have at that moment is either to terminate that pregnancy or to proceed.
Conversely, being able to give that information at the earliest possible stage before a pregnancy has even started avoids families having to make that really difficult decision.
Douthat: So, obviously, I have some familiarity with testing during pregnancy, and part of what comes with that is a really wide error range, where you get lots of false positives. When you’re talking about monogenic results, what is your level of confidence in these results?
Siddiqui: The testing for chromosomes and monogenic screening is, I would say, very strong. You’re talking about 99-plus percent accuracy, in that same ballpark. You’re getting the same level of performance as you would on blood or saliva. Those are very, very strong indicators.
At the same time, any embryo testing — any testing on embryos, period — is still a screening test. Until that baby is actually born, you can’t give a definitive diagnosis because you don’t have an actual human being in front of you. But in terms of the accuracy of the testing, it’s very high.
We actually also do another layer of confirmation called Sanger sequencing. So basically we use two different technologies to confirm that a variant has been detected before we actually go report it out to parents and their clinical team.
Douthat: OK. And then polygenic testing — what is that?
Siddiqui: So we talked about monogenic disease — that’s very prevalent, but there’s a one-to-one mapping between, hey, this variant leads to this disease.
For a lot of diseases that you might be familiar with, things like bipolar disorder, schizophrenia, heart disease, diabetes — they’re not driven by just one genetic typo. They’re driven by the cumulative impact of many millions of variants collectively driving risk.
What’s really interesting that’s happened over the last decade or so is that genetic data sets have gotten very, very large. So we now have both sequence data and medical records from really, really large numbers of people. What that allows us to do is to build models where we can actually quantify genetic risk or genetic susceptibility to disease.
What that means on a practical level for embryos is that, for example, I’m South Asian. We have, unfortunately, incredibly high rates of heart disease and heart attacks. We’re twice as likely to die as other ethnicities. So I don’t want my child to be at high risk for heart disease just because I’m predisposed to that.
What Orchid and genetic risk scores allow you to do is to quantify genetic susceptibility for each embryo. Our reports will tell you: What is the actual percentile of risk? Is this embryo in the 99th percentile of risk? OK, what does that actually map to? Does that mean that embryo is four times as likely or 15 times as likely to develop the disease? And then what is their absolute risk of the disease?
Douthat: But those can be quite different. An embryo could be in the 98th percentile of risk for an extremely rare condition, but the odds would be that it did not have that condition, right?
Siddiqui: Oh, yeah, of course. That’s why we report all three numbers, so people can get a really solid grasp of what the risks look like. Because it is important to understand, at a population level, how high-risk is this embryo? Are they at the 99th percentile or not?
But of course, the absolute risk is really critical to understand, too. If the base-line risk for that disease is low, even people who have a family history might be less activated or less concerned about it. So we’re trying to provide parents the fullest possible picture of how significant this risk is for their future child.
Douthat: And how many conditions are you testing for in the polygenic screening?
Siddiqui: I think it’s about 12, 12 to 15, yeah.
Douthat: So in practice, you have an embryo and you say: Embryo A has an elevated score on these three conditions and a normal score on the others; Embryo B has an elevated score on these two conditions; Embryo C has no elevated scores — that’s how the parents see the data.
Siddiqui: That’s the way it’s collapsed, but obviously, again, all of that detail, in terms of the absolute risk, relative risk and the percentile, is also in there. I guess, to walk you through it, I would say people are coming at it from a couple of angles. Some people have a specific condition that they’re affected by, that their partner’s affected by, that the parent that they’re a caregiver for has been affected by, or a child — like a first child — has been affected by, and they want to mitigate risk. They’re coming with a very specific concern. That’s one category.
Then there’s another category, which is people who just want to mitigate the maximum amount of risk. So there’s not maybe one specific concern, but they basically see this as: Hey, this is the most significant way that I’m going to be able to shape my child’s health, so I want to understand everything that’s well understood, and then just minimize the genetic risk that my child will end up with.
Douthat: I think the first baby screened by Orchid was born in 2023. Is that right?
Siddiqui: I think so, 2023, yeah.
Douthat: So how many people have used your service? And how many babies, just as a guesstimate, have been born at this point?
Siddiqui: Huh. Yeah, I don’t know. It’s honestly pretty hard to say because we’re like a blood testing company, you know what I mean? People will come to us and get their embryos tested, but they might not necessarily come back to us and say: Hey, we’re pregnant. And sometimes people ——
Douthat: How many people have had their embryos tested with you, ballpark?
Siddiqui: We don’t really talk about that publicly, but it’s in the thousands.
Douthat: It’s in the thousands?
Siddiqui: Yeah.
Douthat: But not the tens of thousands?
Siddiqui: We don’t talk about that publicly.
Douthat: And your clientele, I know, obviously, you also don’t talk about specific people publicly, so I don’t expect you to comment on the reports that Elon Musk and, I guess, one of his partners used Orchid, but generally, what you said before is that you’re dealing with a clientele that is partially people who have family histories of serious genetic disorders and partially people who are, what? Cutting-edge-minded, Silicon Valley types?
The part of the clientele that is coming to you the way that myself, as a hypothetical client, is coming to you — who are those people?
Siddiqui: Um.
Douthat: Make one generalization about those people.
Siddiqui: [Laughs.] To be totally honest, I think it’s just dishonest to try to flatten the set of people who are using the product.
Douthat: OK.
Siddiqui: I mean, really, it’s an incredibly diverse set of people. It’s people from incredibly modest means who’ve had to bury a child due to a genetic disease. It’s people who are very affluent. It’s people who are much younger, people who are much older. It’s gay couples. It’s single mothers by choice. It’s like, kind of every type of family has expressed interest in the product.
Honestly, that’s why I’m excited about building the company, building the product. Everyone wants to protect their child. Everyone wants their child to be healthy and to have the maximum amount of opportunity and ——
Douthat: That’s a good answer. So a couple of personal things: You’ve already mentioned that you’ve used this technology yourself — you said 16 embryos currently with your husband. How many of those are you planning to implant?
Siddiqui: Yeah, we honestly haven’t — we haven’t really made ——
Douthat: You haven’t ever had a conversation about it? I don’t believe you.
Siddiqui: [Chuckles.] We haven’t made any decisions yet. I’ve mentioned that I’d be really excited to be able to have four kids. You’re already beating my wildest dreams — you have five of your own. Very impressive. But we’re super excited about multiple of those embryos. We want to have two boys and two girls.
Douthat: So you would select two male and two female embryos in that case, right?
Siddiqui: In order to get four, yes. [Chuckles.]
Douthat: OK. And how old are you?
Siddiqui: I’m 31.
Douthat: When do you think you’ll have your first kid?
Siddiqui: I’ve got to chat with my husband about that. It’s not a one-person decision. [Laughs.]
Douthat: OK. Well, we can have you both back.
Second personal question: You’ve mentioned in a number of interviews that you were inspired to get into this field and focus on it because of your mother’s experience. Can you talk a little bit about that?
Siddiqui: Yeah, so growing up, my mom got a pretty devastating diagnosis. She started by losing her night vision. Then she lost her peripheral vision, and then slowly she started losing her central vision. She ended up getting diagnosed with a condition called retinitis pigmentosa. What that means is that you progressively go blind, and it was a pretty long odyssey to actually get that diagnosis. And then there was a lot of fear around: OK, is that going to affect her siblings, my aunts and uncles? Is it going to affect us, her children?
I was obviously very young when a lot of this was happening ——
Douthat: How old was she when this manifested itself?
Siddiqui: I think the first symptoms that she had admitted to at least my dad were probably in her early 30s, and then I think maybe in her mid-30s is when he pushed her enough to be like: Hey, I think this is something that you should really be looking into.
What sat with me and what I felt through that experience was just this profound unfairness, right? This idea that there’s this genetic lottery that’s unfolding, and some people win and some people lose, and through no fault of their own someone who I love bitterly isn’t going to be able to enjoy things like being able to see her grandkids — things that I think you and I take for granted, that we’re going to be able to see into old age.
Douthat: I try not to take that entirely for granted, but yes, I take your point.
Siddiqui: Well, you’re a better person than me.
Douthat: Is your mother’s condition one of the conditions that you test for right now at Orchid?
Siddiqui: Oh, yeah, absolutely. We have an entire category.
Douthat: Monogenic or polygenic?
Siddiqui: It’s a little complex, like the entire disorder, but the set of conditions that we test for are the monogenic forms of retinitis pigmentosa. We also screen for monogenic forms of blindness, deafness — a lot of congenital anomalies that are known. But the important thing to understand is for cancer, for example, as a category — we test for monogenic forms of cancer, but not all of cancer is monogenic, if that makes sense.
To me it seemed like the most important thing for me as a future parent was that I wanted to be able to minimize the chance that my child was going to be affected by what affected my mom and what affects, unfortunately, millions of people today.
Right now it’s this really horrible situation where these people who have these genetic diseases, each of them are individually so rare that pharma companies have no incentive to design a cure. The market is literally too small. That’s why I was so activated by this idea.
Douthat: Right. So embryo screening is faster, more certain, more comprehensive than waiting for a cure.
Siddiqui: And more affordable. It’s much more affordable. You’re talking about the price of consumer electronics to be able to mitigate risk for thousands of some of the most devastating diseases.
Douthat: Which reminds me, I failed to ask one client’s question, which is: What does the procedure cost for one set of embryos?
Siddiqui: It’s $2,500 per embryo. So just multiply that by the number of embryos that you and your partner ended up creating.
Douthat: Let’s talk about the science for a minute. There’s a lot of scientific controversy about the work you’re doing. There’s a lot of criticism and a lot of skepticism, not about the basic idea of genetic screening but about both the specific way you’re doing it and some of the promised results. So I want to ask you some of the questions that seem to come up in that area.
The first has to do with the process that we’ve talked about already: amplification, where you are essentially copying a full genome from a smaller sample. Something that you hear frequently is that this process itself introduces errors, that the genome that is amplified is not going to be identical to the actual genome of the embryo. And so in doing that kind of process, you are inevitably going to introduce, essentially, potential error levels and its own form of dice-rolling and Russian roulette into the test results that you get back. What’s your response to that critique?
Siddiqui: So it is true that older amplification methods suffered from those issues. The technical terms for these are allele dropout and lack of uniformity of coverage. But what does that actually mean? It means that you’re not getting a representative sample of the genome that you want.
That is one of the core technical problems that Orchid solved. The way that we know that we’ve solved it is that we take the exact same standards that are used on blood and saliva, and we’ve exceeded those standards. We’ve actually gone a step further than that, where we’ve actually sequenced babies that have been born. So we can compare the embryo sample to the cord blood and the DNA of the baby at birth, and we’ve shown that all of those results agree and that we can reliably read the entire genome from that really small sample.
Douthat: To what extent are these results publicly available, as opposed to reflecting proprietary corporate technological findings? Can I as a journalist see those results and take them to a geneticist who’s skeptical of you and have them read it? Is that out there?
Siddiqui: Yeah, of course. That would be your right. All of these validations are public. They’ve been peer-reviewed. We’ve also had state- and federal-level inspections come into the lab to also review all of these documents. That last piece around the outcome data is still in a pre-print phrase, which means that it’s going to be available publicly in a few months, before the end of the year.
Douthat: So what is proprietary?
Siddiqui: The amplification protocol, as well as the computational methods that we use in order to basically call variants and build these genetic risk score models.
Douthat: So it’s the risk score that is essentially a trade secret. You can’t tell me how you generate the risk score because a potential competitor would come along and steal that method.
Siddiqui: Well, what I’m saying is that both the chemistry and the computational side is proprietary. They’re all new methods that we’ve invented. For the chemistry side, we have a provisional patent that we filed, and for the computational side, that’s more in the trade secret category.
Douthat: OK, so let’s talk about the questions about polygenic testing for diseases that are caused or influenced by multiple genes. My understanding is that the state of our actual knowledge of how the genome works and influences things like diseases is limited in all kinds of ways, and that there are two sets of open questions about this kind of polygenic screening.
First, to what extent is it basically reliable? And second, to what extent do we just not know what other things a set of genes might be doing? You can imagine a world where you look at an embryo and you say: Ah, it has this set of genes that make it somewhat more likely to have heart disease or Parkinson’s — we don’t want to use that embryo. But then it turns out that something in that set of genes also codes for resistance to disease or something like that. I think this concept is called pleiotropy, the idea that a set of genes can have different effects in the same person.
How does your testing deal with both of those realities? One, the basic uncertainty about how genes interact with each other, and two, the idea that you might be testing for one condition and failing to capture some other set of effects that would be themselves lifelong and powerful?
Siddiqui: Yeah, let’s get into both of those concerns. I think what people are really asking when they talk about pleiotropy is this idea of genetic trade-offs. If you lower risk for schizophrenia, are you actually unknowingly increasing risk for another condition?
One thing that I love about genetics is that it’s become an incredibly quantitative science. Researchers — not just at Orchid but around the world — have actually looked into this exact question. What they do is they build something called a confusion matrix, which tests pairwise correlations between schizophrenia, or any disease of interest, and thousands of other diseases. The dominant pattern in that data is that diseases that are of a similar category correlate with each other. So basically, schizophrenia risk actually correlates with risk for bipolar disorder and other mental health conditions.
What that means is that this idea of genetic trade-offs isn’t the dominant story the data tells. The dominant story the data tells is actually that when you want to reduce risk for mental health disorders, you actually reduce risk for multiple diseases simultaneously. So heart disease has the most correlation with things like atrial fibrillation and other cardiac conditions.
It’s not the case that there’s these massive genetic trade-offs that we know of. It is the case that it happens in a small number of specific situations — those correlations have been found — but the dominant story is actually that you’re able to reduce risk for many conditions in the same category at once.
That’s a really exciting discovery because, I totally agree, it’s a fear that people have: Am I unknowingly increasing risk for another condition?
Douthat: Right, and you’re effectively saying: It could happen, but it’s not the most likely scenario.
Siddiqui: Yes. However, I would say that the other boogeyman argument of “unknown unknowns” — I just don’t find it very compelling, because if our standard for any medical intervention was that we’d have to clear the bar for unknown unknowns, we would never do anything.
When we run a clinical trial for a drug and we find that it’s effective, we don’t say: OK, but what about in 50 years if this person self-destructs or falls down dead? There’s a certain standard that we’ve accepted is a reasonable standard, and what genetic risk scores do is they actually exceed the standard in all of medicine. You’re talking about ——
Douthat: But that gets ——
Siddiqui: Hold on just one second. The thing is that genetic risk scores exceed the standard for any drug that’s currently on the market. You have hundreds of thousands of people that these scores have been validated in. You’ve validated these scores not just in a single country but across multiple countries, multiple populations. So I think it’s very difficult to argue from a quantitative perspective that we cannot measure the genetic predisposition to disease.
We have seen over and over again, in many different populations across time, that we can quantify the genetic susceptibility to disease. We can see individuals who are four times, three times, 15 times, 30 times the likelihood, depending on which genetic risk score that you’re looking at. I think that that’s a much stronger evidence set than we have for blockbuster drugs.
Douthat: But that gets to my next question. You mentioned the idea that whether someone drops down dead at age 50 does not necessarily have to be central to the calculation of any given intervention. How do you know that your interventions in some of these areas are working, given that some of the conditions you’re testing for don’t manifest themselves for years or decades?
So yes, you have a set of tests where, if a set of children are born through Orchid’s work and they all turn out to have severe abnormalities, you could say: OK, this isn’t working. But when it comes to, let’s say, schizophrenia — setting aside all the difficult questions about what actually causes schizophrenia — it tends to manifest itself often in late teenage years, in people’s 20s. From our point of view, an Orchid baby born today who you’re trying to offer a polygenic risk score for on schizophrenia might not manifest schizophrenia until the year 2046.
Isn’t it going to be decades before you will actually have data to be able to say definitively: Our risk assessments for these longer-term conditions are working?
Siddiqui: No. So what’s amazing is that we actually already have this data today. The way that we know that it’s working is that we already have data from people who are alive today who have schizophrenia. We have data from their families. And what you can do is run the model and see how often you are able to identify that risk in individuals.
You can actually go a step further and look at data within families. What that means is that you look at data sets where you have multiple siblings — one who is affected by the disease, one who isn’t — and then you can actually see: How often is that risk score correct? How often are you able to reliably see that this individual developed the disease, and they also had a very high risk score?
And what that data shows you again and again — and again, this data isn’t just Orchid alone; this is the entire genetics community who’s published these results — is that what these genetic scores can do, how powerful they are, is that they can identify individuals who are anywhere from three to six to 30 times, depending on which disease it is that you’re looking at.
And that’s the exact same framework that we use, again, for clinical trials for drugs. You look at cases and you look at controls, people who are treated with a drug and people who weren’t.
How often is the outcome that you’re measuring mitigated? If you’re trying to mitigate heart attacks, how different is the rate of heart attacks in people who took the drug versus who didn’t? Similarly, in the context of genetic risk for people who are quantified at high genetic risk versus people who are at low genetic risk, how much more often do the people at high genetic risk get the disease?
And what you see over and over again across populations is that happens. It would be ridiculous to tell people: Hey, we’re going to stop offering any drug because we have to wait 50 years to see if it continues to work. If it works in the population that you studied, you offer it because it’s clear that it works based on a reasonable standard.
Douthat: But what you are promising is that you have, again, a proprietary algorithm that is not publicly available that assesses these risk levels. And tell me if I’m wrong, what you’re saying to me is that you are validating this algorithm retrospectively by saying: Look, we’ve based it on tests of people and families that have schizophrenia already. We know when it manifests, and based on that, we are able to project forward and say: This should work for the next set of populations down the line.
I completely accept that that’s possible. But again, the claim of a proprietary algorithm is that you’re doing something special that not everyone else could do. And you are making a prospective bet on the effectiveness of that algorithm. You can’t prove it to me by saying: Look, people who didn’t go through this testing, we promise you we’ve matched the algorithm onto them. In the end, it will be a while before the algorithm itself is vindicated, right?
Siddiqui: No, that’s not how any medical test is evaluated. Any medical test is evaluated on existing data, and you proceed assuming that existing data is valid. For example, a drug company doesn’t go and tell you: Hey, this is exactly the molecule and this is the proprietary — I mean, of course, they patent it. They show you the results. And we publish the results, too. For every single risk score that is offered, we’ll tell you exactly what the performance is.
Douthat: Forgive me if I’m confused, but what I’m saying is when a drug company tests a drug, they’re testing it on people. They’re treating a cancer, and they give the drug to people with cancer. Some of the people with cancer get better and some of them don’t. And they can say: We gave these people this test. These people over the course of our process manifested these results.
With an algorithm that’s being applied to embryos, you don’t get those results until the person actually develops or doesn’t develop schizophrenia. That’s all I’m saying.
If you’re comparing the way a drug is tested, it’s not just looking back at people in the past who have had cancer. You are actually giving it to people and seeing what happens. And you are running a test on embryos and making a set of predictions about where the embryos go.
I completely buy that you’re trying to make those predictions based on existing data and that it might be possible to make those predictions. All I’m saying is you won’t have the equivalent of clinical trial data for some of these conditions for a long time. That’s all I’m saying. Is that wrong?
Siddiqui: I think it’s a bit misleading. The difference is that we’re not giving embryos drugs. What are we evaluating about embryos? We’re evaluating their DNA. So the DNA of existing people is a perfectly legitimate way to evaluate whether or not a risk score works.
And if you think of ——
Douthat: Right, you’re not giving the embryos drugs. You’re just discarding the ones that you — I mean, in a way, you’re doing something more extreme to them, right? You’re discarding them.
Siddiqui: No, we’re not discarding any embryo. Orchid has absolutely zero to do with discard. We provide genetic reports and parents make their own decisions about which embryo they want to implant and whether they want to discard it. Orchid actually advises against discarding any embryo for any reason. We have absolutely nothing to do with discard.
Douthat: That’s a good bridge to some moral questions I have. But before we cross that bridge, I just want to ask you about I.V.F. itself.
Right now, I.V.F. for a lot of women is painful. It’s difficult, it’s expensive, and it’s something that people only really seek out in many cases when they’re facing serious fertility challenges.
You’re imagining a future where I.V.F. becomes, if not the normal, at least a normal way to have kids. So does I.V.F. itself need to change in some way for that future to be possible?
Siddiqui: No, I actually don’t think so. From my personal experience, I had zero side effects. I took zero days off work. It was a nonevent, to be totally honest. I think that some people have really heavy periods and it ruins their whole day or week.
I.V.F. itself was — it’s pretty simple. You’re taking medication for 10 days. The results were amazing. And if you think about what people already do for trivial reasons — people get Botox, people get plastic surgery, people do much more invasive things that are more expensive for really trivial gain — this is talking about: Is your child going to suffer for a lifetime from a disease that you could have prevented? Are you going to spend two weeks to make sure that doesn’t happen? I think that a lot of parents would choose that.
This question has actually already been asked nationally to Americans, and 70 percent of Americans are in support of using these types of genetic risk scores to actually understand risk before pregnancy. And 30 percent of Americans would consider I.V.F. in order to mitigate risk. And if you think about the costs — they are just artificially inflated in the U.S. In Europe, you can get I.V.F. plus the medication for less than $5,000.
Douthat: I think one should be skeptical of applying the answer to a poll question to the way people actually make choices about these things in real life. And I think one should be skeptical about applying your experience — and I’m glad that it was as easy as it was for you — to the experiences of the wider range of women.
I would say, just from personal knowledge, there are a lot of different experiences with I.V.F. I would also just note you haven’t actually had a child, right? So you’re describing part of the I.V.F. process as having been incredibly simple for you, but there’s another part, which is trying to actually get pregnant through it.
Siddiqui: Sure, but data shows that if you have three chromosomally normal embryos, you have a 95 percent chance of a successful pregnancy. That’s from data on thousands and thousands of cycles. And yes, I don’t want to diminish that people have really rough and difficult I.V.F. experiences, but I think there’s a difference between the practicalities of what needs to be done — you need to take shots for 10 days and do a 10-minute procedure to extract your eggs — and then battling with infertility. Battling infertility is equivalent to battling cancer. I don’t want to diminish that. That’s a very difficult thing that people have to go through.
Douthat: Back to the client, the healthy 30-year-old who isn’t infertile: You’re saying that a certain amount of modest pain and challenge is worth it for the benefits that you describe.
Siddiqui: For me personally, and for people who choose this, it’s not — again, no one should be compelled to do something that they personally find is the wrong decision for their family. But if you look at the history of medical innovation in pregnancy, you see this over and over.
When epidural was first introduced, there was this massive moral panic. People were told that women have to endure pain during pregnancy, that we shouldn’t remove it, we shouldn’t overmedicalize it. Now it’s completely standard. I mean, 70 percent of women choose an epidural when they deliver their baby.
If you think about hospital birth versus home birth, it used to be that everyone gave birth at home. Now 98 percent of Americans choose to give birth in a hospital. Why did they do that? When you’re hemorrhaging, you have a lifesaving intervention they’re able to do in the hospital that you don’t have access to at home.
For the same reasons, people are going to vote with their feet where they think the better outcomes are. And I think that there’s this huge knowledge gap where people think that genetic disease maybe won’t affect them or isn’t as prevalent as it actually is. And if people really understood that developmental delay, intellectual disability, autism, pediatric cancer, birth defects — all these really horrible things — that actually, we now understand the molecular basis of them, I think people are going to vote with their feet and say: Hey, yeah, two weeks of injections — it’s annoying and it’s a hurdle that I would rather not cross, but the payoff is worth it.
Douthat: But people don’t just vote with their feet in terms of how medical processes work. The United States has a very, very high rate of cesarean sections, for instance, which are not necessarily medically indicated, but have become a source of convenience and pressure from doctors and so on. That has, I think, in certain ways warped the experience of childbearing in the U.S. It becomes easy to see a scenario where you get certain kinds of medical pressures and expectations that push people toward doing the kind of interventions you’re describing because, again, I agree, no one is going to be locked in a room and told they have to do this. But you can end up with certain kinds of medical systems creating pressures for things that aren’t necessarily just people making free choices and voting with their feet. Both things can happen. People make choices, but the system imposes certain expectations.
I think it’s very likely, in a future where the Orchid process becomes a kind of norm, that you would then get a whole set of medical pressures. Like, what are you, a bad parent? You’re not going to test your kids? You could imagine a kind of tipping point or a cascade where you go from “this is a choice that a certain number of parents do” to “this is the expectation that the medical system imposes on expectant parents.”
Which is why I think it’s important to think about some of the moral questions and cultural questions associated with this kind of change.
Siddiqui: I just want to agree with you and push back on that. I think that maybe it’s a universal opinion that there’s too much finger-wagging at women. I think that women are attacked if they choose to breastfeed; I think they’re attacked if they choose not to breastfeed. I think that they’re attacked if they want to have a lot of kids; I think they’re attacked if they don’t want to have kids. And I think that all of that societal pressure, whether it’s cultural or whether it’s from the medical establishment, needs to stop, honestly.
I think that this should really be about parental choice, parental freedom and parental autonomy. So I totally agree with you. I think that for any intervention or for any procedure that’s touching something so intimate and so personal, it should be about just making an informed decision where people are just told honestly: This is what is possible; these are the risks and benefits. And then they make a decision that’s right for their family.
So I totally agree with you there, that there shouldn’t be pressure from the medical establishment one way or the other.
Douthat: Let’s talk about some of the moral and societal questions that I think this tech raises. Let’s start with a basic one: Orchid technology, relative to having babies by having sex, involves creating substantial numbers of embryos. Many of those embryos will not be used. They’ll be kept on ice. They’ll eventually be discarded.
This is something Americans are comfortable with under the existing I.V.F. system, but they’re comfortable with it in a context where most of the people who are doing this are struggling with infertility and are in a crisis environment. What you’re talking about is doing this at a much, much larger scale. So I just have to ask: Do you think the embryos that are created in the Orchid process have any kind of moral status whatsoever?
Siddiqui: I think that embryos are extremely precious. They’re the most miraculous and magical cells that we’ve ever discovered. They somehow differentiate into the trillions of cells that make you and me. Unfortunately, I think a lot of people don’t understand biology.
So what happens the old-fashioned way, when you and your partner have sex at home, a lot of embryos are discarded through that process. Nature is extremely brutal. A lot of people don’t know that even people in their 20s have about a 20 percent chance of getting pregnant every month.
So yes, an egg can fertilize. Great. Now you have an embryo. But that doesn’t mean an embryo is going to implant. There’s been multiple really large-scale studies on this — lots of embryos are discarded at home. It’s just that I.V.F. makes that process visible that is currently invisible to people that’s happening at home.
Douthat: But that’s not true. The term “discard” implies agency. If you and your spouse ——
Siddiqui: I don’t think it implies agency.
Douthat: OK, fine. The obvious difference between embryos that fail to implant when a husband and a wife have sex and embryos that are discarded in a laboratory is that in the first case, the embryo dies with no human being deciding that it’s going to die. And in the second case, the embryo dies because the laboratory decided it should die. In the same way ——
Siddiqui: That ——
Douthat: Human beings — all human beings — die over a long enough time horizon. Everyone’s survival rate is zero. But if you had a system that was set up that required discarding hundreds of thousands of adult human beings, we wouldn’t say everybody dies; Mother Nature is much more cruel.
In order for your argument to work, you just have to say: The embryo doesn’t have any moral status that we are obliged to respect, and therefore it’s OK to discard it. All this talk about Mother Nature, I think, is a distraction from that issue.
Siddiqui: I think that your framing is dishonest, and I want to tell you why it’s dishonest. You can take embryos that have been created through I.V.F. and transfer them back into the mom if you really want to make it not a human choice whether or not those embryos become babies. It is exactly equivalent, and many people choose to transfer those embryos into Mom — it’s called compassionate transfer. They don’t take the I.V.F. meds, and there’s a very low chance that that’s going to become a pregnancy, just like when you and your partner at home do it the old-fashioned way. There’s plenty of times of the month that it’s not going to lead to a pregnancy.
Douthat: I actually completely agree with you. I think it is quite different to transfer embryos into a woman’s body when they have a really low chance of survival than it is to discard or permanently freeze and then forget about existing embryos.
Which do you think is most likely to happen at scale in a world where Orchid technology takes off: the compassionate method you describe, which does have some chance of leading to pregnancy? Or millions upon millions of embryos just being discarded? Clearly it’s the latter, right?
Siddiqui: It’s overly simplistic to say that everyone is going to make the same decision. Maybe a lot of people really want to do compassionate transfer. Maybe that becomes the default way that embryos are handled. It really depends on people’s personal decisions.
I think the bottom line is that we shouldn’t be shoving our moral beliefs on other people. If you believe that you want your embryos to be frozen indefinitely because you believe in this future where sci-fi is going to be able to provide gene therapy on those embryos, more power to you. If you feel very strongly that every embryo is a human life and you want to do compassionate transfer, then you should have full right to do that.
Douthat: OK, but you’re ——
Siddiqui: I don’t think we should be making assumptions about what choices people will make, because when you stand in that position and when people are actually in that decision in that moment, they’ll make a potential decision that would surprise you.
Douthat: But you are promoting a technology that has profoundly potentially transformative effects on society. Clearly you’re arguing for it because it has beneficial effects, and you are essentially saying you don’t know what’s going to happen to all the extra embryos. Who can possibly say? It’s just up to people.
But we know in practice what is likely to happen.
Siddiqui: I don’t think that’s true.
Douthat: I feel like I’d like to know what you think about the moral status of the embryo. Isn’t that something I should be able to know? Just to go back to the example you talked about, with your mother and her illness: In the world you’re describing, your mother would exist as an embryo — not your mother as an adult human being, your mother as an embryo — and someone running an Orchid-style program would look at that embryo and say: We’re not going to implant that embryo.
Does that bother you at all? Would you say: Oh, I’d like to have my mother have a chance at life through some kind of low-probability compassionate injection. Are you comfortable with that embryo being frozen or discarded? What’s your actual view?
Siddiqui: Look, that question has, again, extremely dishonest framing and dichotomous reasoning. Let me just walk you through my grandma’s life situation.
Do you support girls getting the chance to finish high school? Do you support girls getting the chance to choose who they marry and when they marry? Well, my grandma wasn’t ——
Douthat: Someone’s doing some dishonest framing here, but I don’t think it’s me.
Siddiqui: No, I don’t think it’s dishonest framing. That is the actual situation that my grandmother was in. She was married at 16 and had my mom at 17.
Now we’ve progressed as a society, and we give girls the choice to finish high school. We give women the choice to choose when they marry and who they marry, and all of those decisions have profound consequences on who gets born. Different people are getting born because girls are finishing high school. Different people are getting born every time a woman rejects a man on a first date.
Douthat: I’m not asking about the set of choices that bring embryos into being. I’m asking about the embryos themselves. You are running a business and you are saying: It would be good to live in a society where the embryo that became my mother was assessed and ruled out of bounds.
I’m just asking you what the actual status of that embryo is. Does it matter? If I took the trays of embryos that contain you and your husband’s embryonic children and I threw them in the river, what kind of crime have I committed? Have I committed a property crime? Should I pay a fine? What have I done?
Siddiqui: So are you saying that you want to criminalize I.V.F. and that you should ban I.V.F.? Is that your stance?
Douthat: I have convictions on abortion and I.V.F. that are generally out of step with the mainstream of America, and I completely own that. I would say no, I don’t want to ban I.V.F., but I think there should be limits on the number of embryos created, absolutely. And I would favor, at least at a level of moral suasion, encouraging the kind of compassionate — what’s the term you use? Compassionate ——
Siddiqui: Compassionate transfers.
Douthat: Compassionate transfer — I would support that. So I’ve answered your question.
Siddiqui: Sure.
Douthat: Now I’d like you to tell me: Does the embryo have any moral status? That’s all I’m asking.
Siddiqui: Sure. I think that the question of an embryo that is going to get adult-onset blindness, what do I think about that embryo? My mom doesn’t want to be blind. She doesn’t want me to be blind. She doesn’t want her grandkids to be blind. So I think that it is a positive moral choice, it is the responsible decision as a parent, to detect that risk at the earliest possible stage and to transfer the embryo that has the best probability of a healthy life.
I don’t think that there’s any moral question there. I think almost the opposite. I think that creating stigma or creating some sort of taboo around the idea that parents would want to proactively get that information is a dangerous idea to propagate.
For me personally, I do think every single embryo is precious. I think it’s an absolutely amazing thing that we’re able to make this process work outside of the body. It would be amazing if we had cures for all of these genetic diseases so we didn’t have to make these difficult decisions, but unfortunately, the medicine is just not very good. We do not have cures for these diseases. My mom has no options for being able to reverse the vision loss.
Douthat: All right. I’ve badgered you on this point enough. Two last questions. The first is about the kind of sci-fi dystopian future that a lot of people see hanging around this issue. It relates to the fact that I think most people’s expectation is that if you can figure out what polygenic scores mean for diseases, you’re going to be able to figure out what polygenic scores mean for athleticism, for intelligence, to say nothing of a whole host of superficial traits.
In which case, if you have a set of genetic technologies that cost a certain amount of money ——
Siddiqui: $2,500 ——
Douthat: Per embryo, right?
Siddiqui: Yeah.
Douthat: $2,500 per embryo. If this technology works as well as you say it does, then you are essentially advancing toward a future where there will be a caste system in terms of how rich people versus poor people are genetically sculpting their offspring. Is your view that you are ushering in that kind of future?
Siddiqui: No, I don’t think so. And I think that society fundamentally rejects that idea because it is fundamentally so disgusting. We’re moving toward a world where I.V.F. itself is something that is going to hopefully be covered for everyone. It’s a really sad history, I think, for the last 40 years that rich people get to have babies, and poor people who can’t afford I.V.F. don’t get to. That’s a fundamental human right that I think was violated, and I think it’s nice that we’re finally seeing steps toward that being corrected, where everyone will be able to have access to I.V.F.
In addition to that, hopefully we’ll be able to get access to screening technologies like Orchid. Because it spurs such moral outrage, this idea that something that would create such an advantage, such vastly different outcomes — I don’t think that’s something that Americans will choose to only leave to the few that can afford it. I think it’s something that in the very near future hopefully we’ll be able to mobilize enough excitement around so that it’s something that’s going to be covered for everyone.
Douthat: So essentially, you need a kind of socialism to avoid the genetic hierarchy, eugenic dystopia ——
Siddiqui: I think it’s just insurance coverage. I mean, it’s insurance coverage for a very nominal amount. We’re currently covering, in the last year of your life or the last two weeks of your life, extremely expensive life support machines and medical procedures that are extending your life by a few days.
I think that even today, with the money that we have, if we spent it more appropriately, if we spent it more in line with what people’s actual preferences are, we could already afford this for everyone in America today. It’s just a question of will. That’s why I’m so excited to talk to you ——
Douthat: Will and budget deficits. But yes.
Last question. You are excited about a world in which lots and lots more babies than is the case right now are born from laboratory fertilization. And I’m just curious if you think, allowing that this might be desirable in certain cases, if a world where this became the norm would be losing something that is very fundamental to human beings and human families and human relationships, and that’s the relationship between sex and procreation, between you and your husband having sex — apologies — and the future generations that come into being?
I’m going to take the podcaster’s privilege — I apologize for this — I’m going to read you a piece of a poem. It’s by a poet named Galway Kinnell. And the poem is called “After Making Love We Hear Footsteps.” And the idea is contained in the title that the husband and wife make love and it wakes up their child and the child comes and gets in bed with them. And Kinnell writes:
In the half darkness we look at each other
and smile
and touch arms across this little, startlingly muscled body —
this one whom habit of memory propels to the ground of his making,
sleeper only the mortal sounds can sing awake,
this blessing love gives again into our arms.
Do you worry about removing or diminishing from human experience that aspect of being a husband and a wife, in a relationship, with a child?
Siddiqui: What do you mean?
Douthat: I mean, in a future where Orchid technology becomes the norm, the feeling that poet is expressing, where a man and a woman make love, and by making love they bring a new life into the world that they haven’t sculpted or engineered in any way, that is given to them out of the self-giving from each other, that forms this profound connection between sex, the way you love your partner and the family that you brought into being.
And again, allowing that we’re in a world where people have genetic disorders and there are all kinds of reasons that people would consider other ways of having children, you’re imagining a future where that just goes away. And I’m wondering if you think anything would actually be lost if that goes away? If 90 percent of babies are born through I.V.F., and having sex and having a baby out of that becomes this weird thing that the Amish do, aren’t you pushing some really intimate and important aspect of human experience out of human experience?
Siddiqui: Yeah, I think people will obviously continue to have sex. I mean, it’s a profound source of connection. I think it’s just that people will — I mean, it’s actually funny. This quote that I’ve said — “Sex is for fun, Orchid and embryo screening is for babies” — it’s actually kind of ——
Douthat: Yes, I didn’t want to quote that to you because I thought it was so ridiculous.
Siddiqui: Oh, I actually think it’s already true. I mean, already people are having sex much more often than they’re having babies. Sex is already for fun and not for babies 99 percent of the time. So it’s actually not so strange of a concept. So when it ——
Douthat: But when you get a baby, most people get it from having sex. And yes, there’s plenty of people who have sex without having babies, but most people who get a baby — it is linked inextricably to having sex with your spouse.
Siddiqui: Yeah, today ——
Douthat: And you are saying it’s time to sever that for the sake, I concede, of potential medical benefits. I’m just saying, I think, pretty clearly, something that poets write about would go away.
Siddiqui: Yeah, I think that sex is a beautiful thing, and I think that if you have enormous genetic privilege and for you to roll the dice and to get an outcome that isn’t going to lead to diseases is in the cards for you, then of course, go ahead and roll the dice.
It’s just that I think that the vast majority of parents in the future are not going to want to roll the dice with their child’s health. They’re going to see it as taking the maximum amount of care, the maximum amount of love, in the same way that they plan nursery, plan their home, plan their preschool — all of these decisions are actually extremely insignificant in terms of the difference between whether your child is going to live with pediatric cancer, with a heart defect that we can’t surgically fix, or born without a skull and never going to be able to make it to their first birthday.
I think when people think about it really concretely in terms of: What are they giving up? What are the risks that could potentially affect this child? I think that then it becomes about stewardship. It becomes about: How do I make a responsible choice for my family? How do I make sure that my child doesn’t have to suffer in the same way that I do? In the same way that my sibling does? In the same way that my parent, who I’m a caregiver for, does?
I think sex is obviously a very beautiful thing. It’s a very profound part of the human experience, but I think that it’s denigrating and dismissive to I.V.F. parents and to I.V.F. babies to say that somehow science babies are inferior to babies that are made the old-fashioned way.
Every human life is equally valid, and I think no parent who chooses to take the maximum amount of love and care and information going into that decision should be stigmatized in any way. I think it’s their personal choice, and I think freedom and choice is what makes America a great place to live and to be.
Douthat: Noor Siddiqui, thank you so much for joining me.
Siddiqui: Thank you so much for having me. This was an awesome discussion.
Thoughts? Email us at [email protected].
This episode of “Interesting Times” was produced by Katherine Sullivan, Sophia Alvarez Boyd, Andrea Betanzos and Victoria Chamberlin. Research by Raina Raskin. It was edited by Alison Bruzek and Jordana Hochman. Engineering by Pat McCusker and Sarah Benedict. Mixing by Pat McCusker. Cinematography by Marina King and Derek Knowles. Video editing by Arpita Aneja. Original music by Isaac Jones, Sonia Herrero, Pat McCusker and Aman Sahota. Fact-checking by Michelle Harris and Mary Marge Locker. Audience strategy by Shannon Busta. Video directed by Jonah M. Kessel. The director of Opinion Audio is Annie-Rose Strasser.
The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: [email protected].
Follow the New York Times Opinion section on Facebook, Instagram, TikTok, Bluesky, WhatsApp and Threads.
Ross Douthat has been an Opinion columnist for The Times since 2009. He is also the host of the Opinion podcast “Interesting Times.” He is the author, most recently, of “Believe: Why Everyone Should Be Religious.” @DouthatNYT • Facebook
The post The Next Parenting Trend Starts Before Conception appeared first on New York Times.
