The President's Council on Bioethics click here to skip navigation


Thursday, October 17, 2002

Session 1: Choosing the Sex of Children: Current Technology and Practice

Arthur F. Haney, M.D., Department of Reproductive Endocrinology and Infertility, Duke Medical Center

CHAIRMAN KASS: Our interest in the topic of sex selection is in part a continuation of questions that we touched on in the cloning report, namely opportunities to select in advance some of the genetic traits of children.

It also represents at least in part a use of biomedical technologies not for therapy, but for the satisfaction of client or patient desires.

This is also an interesting case in which the aggregated effects of choices that might be innocent in themselves might produce results such that even the people who use the technique might be unhappy with the result as a result of its being used by everybody.

And it is also an interesting case for us because there are international implications of biomedical technology where techniques developed for one purpose and one use and one country, say, in the United States, will be used for other purposes and in different ways abroad with consequences that may, in fact, come back to have an influence here at home.

We're very pleased to have with us this morning Dr. Arthur "Cap" Haney, who is the Roy T. Parker PROF. of Obstetrics and Gynecology and Director of the Division of Reproductive Endocrinology and Infertility in the Department of Obstetrics and Gynecology at the Duke University Medical Center.

Dr. Haney is a past President of the American Society for Reproductive Medicine, a researcher, and a leading authority in this area, and I'm very delighted that you could be with us, and especially after all of the travels that were required to make it possible.

We're in your debt, and we look forward to your presentation.

Thank you.

DR. HANEY:  Can everyone hear me?


DR. HANEY:  Okay.  I appreciate the opportunity to be here.  I'm actually a stand-in for Sandy Carson [President ASRM] who couldn't be here.  She's attending our meeting, and hopefully I can give you something similar to what she would present.

I think I reviewed your briefing booklet, and it has virtually everything I'm going to talk about in it, and so I'll try to do this relatively quickly and then respond to any questions that you have.

So I choose — and this is all personal — I choose to separate two terms:  sex selection from sex determination, and you'll see as we go through this what I mean by that.

Now, by way of indications currently one would define two, a medical and a nonmedical.  There are probably 350 or plus diseases known to be linked to an X chromosome, either autosomal dominant or recessive or X-linked dominant or recessive.  But an X chromosome as a potential carrier of disease.

The Y chromosome to my knowledge — and I'm not a geneticist,  so I apologize if there are some others that are around — the AFG mutations creating severe oligospermia have been well – documented now.  With many of the X-linked recessive diseases, the families who had these children or they're in the families would like to avoid having another child born, and hence, having a female with two X chromosomes and screening — by definition the father doesn't have it — would lead to that outcome.

The Y-linked diseases, we've actually been able to, using ICSI, intracytoplasmic sperm injection, have these very severely oligospermic men father children, but they will automatically pass on to any male offspring, which is going to be 50 percent, the same mutation that created their oligospermia.

And there's two responses to that.  One is I'd like to avoid that and have daughters, and the other response is that if we were successful in 2002, that meant by the time my son is 27 and wants to have children, he'll be that much more successful in 2029.

But in any event, that would be a medical indication for sex selection.

Now, the nonmedical ones, I choose to break into two categories:  primary gender selection, that is to say, the first conception, and the second one would be family or gender balancing, which tries to equilibrate to the social desires of the family, the opposite gender from preceding children.

Now, I don't need to go through the glossary in great detail, but we're all in our business acronym – related. 

IUI means intrauterine insemination with washed sperm, getting rid of the seminal plasma, which contains lots of prostaglandins and being able to put the sperm themselves in the uterine cavity without reaction.

COH is stimulation with — it's supposed to be — I'm a little off on the right-hand side over here —  gonadotropins, but controlled ovarian hyperstimulation, that is to say, inducing multiple ovulatory events in a single cycle.

IVF, pretty traditional, in vitro fertilization and embryo transfer.

ICSI, or intracytoplasmic sperm injection used to inject single sperm into oocytes to create embryos.

Blastomere biopsy, typically done at day three post – fertilization, which is removal of blastomere, typically an eight – cell embryo for some sort of DNA analysis.

Fluorescence – activated flow cytometry, or FACS, which is prominent in this arena here that we are discussing today.

Discontinuous density gradient centrifugation, and this is essentially layering the sperm or the seminal plasma on a gradient and centrifuging the sperm through it trying to separate the weight, subtle weight difference between X and Y bearing sperm to separate them.

DNA analysis, I'm not a geneticist, but there's tremendous numbers of these opportunities, which there will be more.  The ones that are most frequent for us are typical PCR or whole chromosome amplification or FISH, fluorescent in situ hybridization, and those are simply techniques to identify unique DNA, whether it's an X chromosome or a specific gene mutation.

Amniocentesis, removing the cells around a fetus which have the fluid in it and the cells from the fetus are present there.  You then culture them and do genetic analysis on those, and that's typically done by 15 to 18 weeks of gestation.

Chorionic villus sampling is a similar technique, but actually biopsying the trophoblast much earlier in gestation, much more rapidly growing cells.  They are proliferating trophoblast cells.  So in 48 hours you can have information or even shorter time.

And then traditional karyotype, just looking at the actual chromosomes.

Now, this is not a new phenomenon, the attempt to control the gender of offspring.  There have been for basically eternity people attempting this in one way or the other.

There are clearly very unsuccessful methods.  Coital timing, trying to be close to ovulation, there's a whole literature on that.

Changing the environment in the vagina where ejaculation would occur.

Electrophoresis, looking for trying to distinguish the X and Y sperm based on electric charge.

Transferring embryos with the most blastomeres because it's been the observation that the male embryos, genetically male embryos, proliferate.  The blastomeres proliferate a slightly bit faster than the female ones do.

And then density gradients, which I sort of listed as unsuccessful, but it's still practiced and many people in the United States use it, and you have a variety of materials:  albumen,  Percoll,  Ficoll, Sephadex, et cetera, that you could use to do the separation.

Now, possible methods — and I stress the word at the moment "possible" because in the absence of clear randomized clinical trials demonstrating efficacy, I really cannot tell you there is a method that effectively works, but at least these are a step above what was on the preceding slide.

And if you try to do this before fertilization, that focuses on selected sperm, and this is a relatively new phenomenon.  The technology was pioneered by someone in the Department of Agriculture named Johnson, and that's been applied in animal husbandry in a variety of species and simply adapted in the mid-'90s to try to do this to human sperm, and we'll talk a little bit more about what it means.

But when you preselect sperm, you have several options to use them.  You can use them in intrauterine insemination, simply in a natural spontaneous cycle timed reasonably proximate to the ovulatory event.

You can stimulate the patient with gonadotropins to increase the efficiency so that the likelihood of conception goes up, but there is also some hazard of multiple gestation.

In vitro fertilization, simply taking the selected sperm that you have and placing them in the dish with the collected oocyte.

And then trying to preselect the sperm and actually inject it into the oocyte directly with ICSI, and these all basically are the same philosophy with different efficiencies.

And then preimplantation would be not using selected sperm, but simply doing IVF, having embryos developed, and then doing blastomere biopsy and identifying the genes of the embryos to be transferred.

Now, sex determination is certainly not new, and that's a very old technology, and when ultrasound became more sophisticated and the resolution improved sufficiently, then one can identify the gender of a fetus simply by the anatomy of its genitals in utero, and that's probably about 15 weeks to 16 weeks of gestation that can be done.

And then chorionic villus sampling, as I mentioned, is biopsying the trophoblast, and that will be typically done between 11 and 13 weeks of gestation.

And then amniocentesis, which collects the fetal cells, and that would be done closer to 16 to 18 weeks of gestation.

Now, the human sex chromosomes are different than many other animals in that the difference between the amount of DNA is relatively small in the total genome.  So it's 2.8 percent between the X and the Y sperm, and that does leave the possibility at least theoretically of trying to separate it by a centrifugation.

As I alluded, that has not been demonstrated  to be effective, but I'll continue to discuss it a bit because it's being currently used.

And then the X and Y sperm can be differentiated to a degree by the amount of binding of fluorescent dye which allows their separation in a FACS, or fluorescent activated cell sorter.  A difference of 2.8 percent would then bind a greater amount of dye in a Y chromosome.

Now, the typical albumen or the Ericsson method, which has been used for many, many years, or at least propagated for many, many years, is albumen density gradient centrifugation, and it's discontinuous, and this is the various concentrations of human albumen, et cetera, have been more empirically defined, but the method has never been demonstrated to be effective.

And despite claims to the contrary, when objective observers have used it, they cannot see any difference in the offspring resulting.

Now, fluorescence activated flow cytometry is the current one, and it's licensed.  I'm not sure exactly who owns the patent, but it's licensed to Genetics and IVF Institute and under the trade name MicroSort, and essentially sperm are mildly sonicated, stained with a Hoechst vital dye, which reversibly and as best one knows does not alter the DNA, and then this dye fluoresces at 359 nanometer of ultraviolet light, and one can then be using the difference in the amount of dye bound to the two gender specific sperm.  They can be separated by fluorescence activated sorting and trying to simply enrich the fractions of X and Y bearing sperm for subsequent use.

So that's basically the principal involved, and this is just an illustration to show you the stained sperm coming through and then deviated left and right with X and Y.

And there's actually in practicality — they are mostly done not simultaneously, but sorted for X or sorted for Y.  So don't sort them all together for both.

And this is what the proof as has been proposed for the efficiency of that process, and that is that when you extract — this isn't done in the actual treatment procedure, but to demonstrate that they are accurately doing that, they then use FISH, fluorescent in situ hybridization, and look for probes for either X or Y, and this is in an XSort, and you'll see that there's a — it's not so clear here, but these are pink, and that one's green, showing you that there's a substantial shift toward X bearing sperm, and they estimate, based on their FISH numbers with some data that it is 80 to 90 percent effective in separating the two sperm haplotypes.

The FISH analysis with the Y sort is similar, and you see they have the sperm.  An X is marked in red, and the other is green.  This is a little less effective.  Somewhere between 60 and 70 percent shifted to Y bearing sperm with this technology.

And then a case report was forthcoming for the first initial case of a birth that occurred after this procedure in 1996 and consistently the reports have been, except for one, related to female selection.  And this was, I believe, — I believe this is an X-linked hydrocephalus problem in the family.  I'm not sure exactly what it is, but it is a medical problem that this child — they opted for a female offspring.

Now, there are several utilizations or there are several ways of utilizing the sperm that you get, and the number and quality of the sperm selected in this process determines how they can be used.

Now, you have to make the assumption that if you use them, that the selection process did not harm sperm function, how it actually works, and we can't really test sperm function other than the fact that fertilization and embryo development occurs.

So you can look at the number of sperm and accrued semenalysis or a swim-up sample and look for a variety of surrogate markers, capacitation, et cetera, in the sample to see if it had an impact, but function is still difficult.

So if you get what appears to be after your whatever the procedure of sorting that you get normal qualitative and quantitative parameters for this washed sample, then you have the potential for putting it in the uterus, and if you do that in a natural cycle, it turns out to be about an eight percent cycle fecundity, that is, term delivery after one treated cycle.

And the human fecundity at max is probably 20 percent, and people argue between 12 and 18.  That is to say one in every five to six cycles turns into pregnancy.  That's clearly age related, mostly maternal age.

If the selected sample is not in the normal range, but there's more than a million hyper modal sperm, and by that I mean when they're layered in media and allowed to swim out of the pellet with centrifugation.  Those are the most actively modal sperm and the highest quality fraction.

If that is greater than a million, but you're not in the normal range, then you have  the option of doing an IUI with stimulation.  Now, that increases your cycle fecundity slightly, and the reason that's probably true has to do with the estrogen stimulation of the female with higher numbers of oocytes and follicles, and the estrogen in the female genital track facilitates sperm transport into the distal tube where fertilization occurs.

You run the risk of multiple gestation with that.

In vitro fertilization significantly increases your cycle fecundity, but at much significant cost, and we'll talk a little bit about cost at the end, but depending on the circumstances and the quality of sperm, age of the mother, et cetera, you would sort of currently go between 15 and 35, maybe up to a 40 in women below 30 for a per cycle delivery rate.

The conception rate, that is to say, missed period and clinical pregnancy rate, would be slightly higher, but you would see some spontaneous abortions.

And then if the sample is noticeably lower in quality  and the motility in addition may well be abnormal, then instead of traditional IVF where the sperm are simply placed in with the retrieved oocyte, one typically microinjects the sperm or the egg with the sperm, ICSI, and that offers per embryo, which you generally can get the same number of embryos you would if you put healthy sperm in with the standard egg retrieval.  You get the same success rate.

So ICSI, even though it's just a little more manipulation, turns out to be virtually identical to IVF for oligospermic either samples or males.

Now, intrauterine insemination is simply, just to illustrate what we're doing, after you separate the liquid from the sperm and get rid of all of the prostaglandins and the seminal plasma which normally stay in the vagina and never enter the uterus, only the modal sperm actually enter the uterus.  If you can separate the two, you will lose a few sperm in the process, but you then place a small catheter into the uterine cavity and deliver the sperm there.

And the attrition that normally occurs in the vagina to the uterus is on the order of 100-fold loss.  One percent actually get there, maybe two percent.  So tremendous attrition.

So placing them higher, even if you lose a few, will increase the numbers of sperm in the distal tube where fertilization occurs.

So that's the strategy between intrauterine insemination.  You simply have to wash them so seminal plasma is not transferred.

Controlled ovarian hyperstimulation is a more complicated scheme where one injects LH and FSH, the human gonadotropins that are normally made in the pituitary.  Humans are monotocous, that is, we release one oocyte even though a crop of follicles matures every cycle.  Polytocous animals don't have the selection that goes from a crop to a single ovulation that we do.

So the difference between a monotocous species like us and a polytocous species is not the number of follicles you move forward each reproductive cycle, but the selection process to release one.

That's clearly involved with gonadotropins, and when you simply overdose the patient, if you will, with gonadotropins, you can get a larger proportion of that crop to mature and hence you can stimulate more than the single ovulation in a given cycle.

All those follicles appear functionally equal in genetic quality, et cetera.  It's just in monotocous species the metabolic demands of the mother are such that you should have one offspring most of the time.

But that's the way it's done and simply has daily injections in the follicular phase in the first half of the cycle until ovulation. 

And this is what it looks like and how it's controlled.  One stimulates the patient, and then you do standard transvaginal ultrasound exams, and you can very accurately with the technology today measure follicular diameter.

And the stimulation is maturing the follicles and typically human ovulation occurs about 22 millimeters.  We can trigger with HCG ovulation about 18 millimeters, and from a lot of other information with IVF, we get healthy oocytes and mature pre-ovulatory oocytes when you do that.

So this stimulation is then by injection.  We teach husbands how to do it.  They do the shots.  We do the monitoring, and you can measure with the estrogen because each of these follicles makes estrogen as well.

So a combination between ultrasound and estrogen gives you the follicle number and the timing for triggering release of those follicles.

Now, IVF with ICSI using selected sperm, you do the same stimulation with the injectable gonadotropin.  You do the same ultrasound, but at the appropriate — we block the LH surge, blocking the pituitary so the patient cannot mount the terminal 36 hours of that maturation of the oocyte and release of the egg.  We block that so that we're in control of it.

We then do a transvaginal ultrasound guided oocyte retrieval.  It sounds kind of gruesome, but it's very efficient, and it's not very uncomfortable.  We just use contrasedation and retrieve the oocyte, and it's age – related how many you'll get per patient.

And then you have the oocytes in vitro, which you then inject with whichever of the two sperm that you had selected.

The embryos are then transferred via the cervix into the uterine cavity and implantation and all subsequent pregnancy events.

This is what it looks like.  This actually isn't a needle, but this is very accurate.  It's the same picture you saw before, and we simply take our ultrasound probe and put a guide on it and slide the needle through, and you can literally puncture each of these follicles in succession and with minimal suction aspirate the contents, and about 90 percent of the time an oocyte will come with the follicle, the follicular fluid that you've aspirated.

So that's basically the process.  You stimulate with gonadotropins and then do an egg retrieval rather than allow ovulation to occur by itself.

Now, ICSI then to inject the sperm, this is done with micromanipulators in vitro.  There's a cumulus mass.  I should have shown you that, but a cumulus mass is normally present, the zona pellucida.  We strip the cumulus mass, and then with very low pressure fix the oocyte up against a pipette.  This is done with micromanipulators.  So it's not by hand, and then a drawn glass pipette has become very small. 

You can crimp a sperm tail and get a single sperm in the pipette, puncture the zona pellucida and the plasma membrane of the egg, inject the sperm, and then withdraw the pipette, and both the plasma membrane and the egg and the zona pellucida seem to tolerate this reasonably, and fertilization then occurs at the same rate as the normal sperm would penetrate in vitro and have fertilization occur.

So that's used both for oligospermic men, and it was simply adapted here to be able to take sorted sperm and do the same thing.

Now, host fertilization pre-implantation selection is a different phenomenon.  The preceding slide showed you that you injected selected sperm of the gender that you were looking for.  With this process, in vitro fertilization is performed with just the routine semen sample available in the same series of events, except there's no ICSI.  It's standard fertilization in vitro.

And then at about three days, blastomere is removed for determination by one of these analytic methods, FISH, PCR, whatever, to determine the presence of an X or Y, and then one has a series of embryos which are kept in each of their own wells, and you've identified their genetics.

And this is exactly the same technique you would use for pre-implantation genetic diagnosis, except there you do PCR first, with a specific probe for gene defect you're looking for.  Here they're looking for a gender differentiation.

There's a biopsy. You can pull out of a blastomere and then do some sort of DNA analysis.  This is actually on leukocytes, but it looks the same.  FISH is identified with colored probes.  You can do whole chromosomes.  You can do whatever the most efficient analytic technique you have and identify the gender.

Now, here's the clinical data that has been presented for fluorescent activated cell sorting to date, and this is the initial paper in Human Reproduction in 1998, and all of it except for  this one little bit here for the YSort, which I'll tell you about, was last week at ASRM, was on X, selecting for Xes.

And so they did 208 cycles, and they did a variety of things.  Two hundred and eight cycles of IUI, intrauterine insemination, where they simply got samples and then placed them in the uterine cavity of the wife at the appropriate time.

They had a ten percent cycle fecundity with that.  They then did either IVF or IVF with ICSI, and they did 36 cycles in 27 couples and had an improved success rate, though it's not as high as one typically sees with just straightforward ICSI for oligospermia.

And of those, they ended up with 29 pregnancies.  They had seven losses, one ectopic, nine deliveries, and at the time of the report, 12 ongoing clinical pregnancies, and they have never gone back and validated in the literature what those turned out to be.

But they then claimed that they had 15 of 18, or 88 percent, of fetuses with known gender, and some of these presumably were ultrasonically identified, were of the gender of choice, female, for the selection they had.

Now, that's what existed up until Tuesday, this past Tuesday, as best I can tell in all of the literature, and if you think of all the activity that has gone into this up until Tuesday, it's based on this, and there isn't a single male paper out there demonstrating an equivalent YSort to show that you could actually increase the likelihood of male delivery.

And a comment was made without data in an abstract at ASRM that they thought that the babies born were 78 percent when they did YSorts and 92 percent when they did XSorts, and that didn't have numbers associated with it.  It was just a sentence in the abstract which was actually done for a different reason.  It was there to show that there was no abnormality in the offspring that were generated.

And I tried to deduce based on some issues they had for abortion, spontaneous abortion with that.  They were assessing, and you'll see in a minute, the number of losses that might have occurred and was that different than normal? 

And I think I could deduce 304 total pregnancies in the group, and out of 1,900 sorting procedures.  So that's as best trying to extract from them what's published as to what's actually going to validate that this works.

Now, their current usage, and by that I mean request usage, this is an abstract, two abstracts the group presented, 245 ICSI cycles.  They had roughly two thirds were looking for Xes to select for females and one third for males, and they did note that the goal is to be able to send samples frozen through the mail, get them sorted and send them back, and then do ICSI with them.

And they did note that when they did that, they had a reduced fertilization and cleavage rate, which isn't terribly surprising with the trauma of cryopreservation.  When it works, it works, but the efficiency is going to be declined.

And the other abstract that I mentioned that had 1,900 separation procedures, they separated them without the numbers in each category as family balancing and medical indications and found the same spontaneous loss rate and then noted a 2.5 percent major anomaly rate, which is favorably compared to the three to four percent in the general population.

And importantly, if you look at those sorts of anomalies, there is no pattern.  They're all random and different.

But, again, the fleshing it out to the actual numbers wasn't in the abstract.  It's a little hard to do.

Now, the current availability, if you will, there is no comprehensive place.  ASRM or any other doesn't compile any programs  that offer sex selection, and basically this is an Internet advertised offering.

And I could find six sites, although I'm sure there are more, but there are many locations.  Some of these programs have locations in different states and advertise them in different countries, and the technologies that they advertise  — and sometimes it's very difficult even in the Internet site to figure out what they're doing for sex selection.  So you can't tell if they're doing the Ericsson method of density gradient centrifugation or actually doing the fluorescent activated cell sorting or even PGD.

And some of the sites separate them and say you can do PGD where it's extremely accurate, where you've actually pulled out the blastomere and know the genes to the extent that PCR or FISH would do.

Others — and the couple that would choose that can pick whichever level of security they want for increasing costs to do them.

So they're all over the place, but it's basically an Internet business that's being advertised, and this is the sort of ad that one sees.  This is one of the franchised MicroSort centers, and complete with a full debate about whether it's appropriate or not.

And then if you're really interested, click this button and go further.

These are roughly what the costs are, and they will vary a lot.  New York City is very different than Durham, for example, in IVF costs.  IUI is, with the preparation and insemination, about four to $600; COH, about 2,500, including the drugs and the ultrasound monitoring.  IVF in our institution is ten.  Other places, in more expensive markets, 15, 14, $15,000.

ICSI will be an additional fee with the IVF.  So this fee is the entire thing.  It's a couple thousand dollars more than you traditionally do with IVF. 

Most of the others are not done often enough in enough places to get a good idea what the costs are, but these are estimates for a blastomere biopsy equivalent to ICSI, $1,500 for a manipulation, the same basic sort of technology.

PCR I've seen in several places around $3,000; FISH, a little bit less.  The MicroSort varies, and depending on the franchise place for it, between 2,500 and 5,000.

Ultrasound is relatively inexpensive.  When you look at the three determination as opposed to selection technologies, they're relatively cheap.  Ultrasound exam is very simple, $300, and amino with the karyotype is probably between 1,500 and 2,000, and then a chorionic villus sampling, slightly higher.

And that would be a rough idea, but there is some variation from institution to institution and locale to locale.

Now, the questions that immediately come to mind with fluorescent activating cell sorting, which is the most heavily advertised version, is the relatively small number of reported clinical outcomes, and this is particularly true since the success of the technology is not validated by anyone other than the people who own it, if you will, or who are franchised for it.  So it's not independently done.

It has not been a technology that has been validated by anyone else.  So very small numbers, and I think you have seen what is in the literature.  I could have missed something somewhere, but I scanned everybody's name, whoever was on the MEDLINE and any of those papers and tried to find their name on anything, and that's the only thing I came up with.

So there could be an abstract somewhere else that isn't in MedLine, but other than that, I think that's it.

This is the dye utilized to Hoechst 33342 dye, bisbenzimide, and that's the binding to the sperm that's reversibly binding.  But the question is:  what is the true impact?  Is there any subtle impact in altering the DNA?

And similarly, we're using the wavelength of ultraviolet light to cause it to fluoresce.  So you have the marker to separate the light.  I think there certainly have always been some concerns about the exposure of sperm to that and DNA to that.

And then the relatively small recovery of sperm after you do this process makes the efficiency of doing it with insemination much lower.  So you then, to make it an efficient process after cell sorting, is you see as time has gone on the reports that occur, always focus on doing ICSI and IVF, and certainly those carry their own concerns as well.

Now, future technologies — and this is more free association, and believe me, as you well know, smart people will do things I haven't possibly dreamed of, but future technologies for sex selection would be selective elimination of an  X or Y bearing sperm on the basis of something on the cell surface biochemically or immunologically, and that's been talked about  a lot, but has never come to fruition.

But you can think of it very similar to an assay using complete mediated cell lysis where you lysed sperm of the sort you wanted to remove, and then selection of sperm by some noninjurious DNA analysis rather than simply the density of binding of the fluorescent, something that specifically bound the sperm.

Now, the limitation to date, as you can appreciate, is the sperm is a very condensed DNA package.  It's basically a DNA packet with a tail, and it's very hard to get any probe into this condensed DNA where you could determine anything, but I won't say that isn't possible at some point.

Preimplantation you can potentially look for a gender specific gene expression within an intact embryo analogous to the Fisher dye staining.  So you could potentially without — if you could do it in a noninjurious way, without doing a biopsy, you can potentially look for any genetic trait within an embryo if you can get to the point of not being able to injure an embryo.

You can look in the media and see if there's a differential uptake of one precursor or another.  I don't think that's terribly likely to be profitable just given the undifferentiated state of the early embryos we have, but I can't exclude that.

And then you can do a FISH analysis similar to something like that on cells that remain after you've hatched, an embryo is hatched.  And we've seen something like that happen.  We've transferred just when you get to blastocyst cultures, some of these are beginning to hatch, and you're going to see some cells that are just going to fall away from the embryo and you potentially have those to use as well.

So there's a lot of options, and there are going to be more that come along.

Now, sex determination, that is, having a clinical pregnancy and determining what the gender is, just so that you know if you haven't seen these things before, this is an ultrasound showing a tear shaped uterus with a gestational sac and a fetus in a yoke sac, and the amniotic fluid.

And so one then at a much later stage of gestation can aspirate the fluid from the amniotic or aspirate the amniotic fluid in fetal cells that are in the amniotic fluid, are then available for cell culture.  They use fetal fibroblast and slow to grow.  So it take a long time to culture them.  It's a laborious and difficult process.

The chorionic villus. sampling where you're getting trophoblasts, they're proliferating very actively.  So it's extremely rapid, and you basically place a needle and remove a bit of trophoblast.

And then the fetus itself, you can — I'm sorry.  The red isn't really clear — but you can begin to look at the characteristics of the fetus, such as its yoke sac and other things, by ultra sound and begin to look at gender differentiation.

So the future for sex determination as opposed to sex selection is going to be a variety of things:  a collection of exfoliated trophoblasts which may be present in the vagina.  The membranes of the fetus, there may well be very small numbers of cells shed into the endocervical canal and virtually equivalent to a Pap smear, you could extract those, determine that they were fetal, not maternal, and then have a genome to look at.

Detection of factors in the maternal circulation responsible for gender differentiation in the fetus, and these are primarily related to the duct systems of the two sexes, the Mullerian duct or Musophen [sp?] ducts and Wolffian ducts, and there's some very, very specific hormones involved in female differentiation and male differentiation, and potentially small amounts of those will cross the placenta and be detectable in blood.

There's a phenomenon called 4D ultrasound.  It's 3D with a time element for the fourth, and they're getting extremely sophisticated with high resolution, ten, 12, 11 megahertz to be able to see much more details in the fetus, and that's for a lot of prenatal diagnosis.

And then fetal red blood cells.  Once the fetus begins to make them, they are nucleated so they show up.  And they do show up in very small numbers in the maternal circulation, and you can try to filter them out, and people have for years been trying to find ways of doing that.  It would save you doing an amniocentesis if you could efficiently retrieve them because they have the nucleus, and you could then do a DNA analysis on the fetus by fetal nucleated cells.

CHAIRMAN KASS:  Sorry.  Could I ask?  Could you give the rough age of gestation when these things might be possible?

DR. HANEY:  You're going to stress me there.

CHAIRMAN KASS:  Well, roughly.  What's the earliest?

DR. HANEY:  I don't honestly know.  I would hesitate to say.

Sandy [Carson] and Joe Leigh [Simpson] can probably tell you better than I can, but I'm sure it's no earlier than 14 weeks.  I think it's minimum then, but I honestly don't know the exact week.

And then ultrasound guided amnios transvaginally, we aspirate many, many things we never dreamed possible before down to four and five millimeters.  So easily one may well see transvaginal much earlier amniocentesis for fetal material as you saw in the picture.

And just two slides, and I don't want to steal the thunder of the subsequent speaker, but to show you the impact of the sex determination and all of these gender differences, I would tell you are probably related to sex determination, not sex selection, and ultrasound showed up somewhere in the early to mid-'80s that was then capable, sophisticated and with high enough resolution to actually begin to look at genital differentiation in utero, and you began to see that's when the largest change in disparity of male to female ratio by birth occurred.

But this would be both.  Whatever gender selection might have been going on, but I'm going to bet it's all sex selection, but I'll bet it's all sex determination, and this would be for people in this category because they're all parities.  It would be people who selected primarily to get the fetus as opposed to people who had three children of one sex and wanted another.

And then if you look at the impact of gender determination for balancing, that is, in subsequent pregnancies, you can see it's very dramatic; that the further they go in the number of pregnancies, the greater the desire to balance the offspring.

And I think I'll stop at that.

CHAIRMAN KASS:  Dr. Haney, thank you very much.

The floor is open for discussion.  Mike Gazzaniga.

DR. GAZZANIGA::  I'm just curious to know how many cases, if we call it, an event where there's a desire for sex selection or sex determination; how many in the United States per year are seeking this sort of service?

DR. HANEY:  I don't think there's any way whatsoever to determine that.  There's not a record kept anywhere, and the best you have, I think, is an estimate of the total number that are reported in the abstract you saw there, but there's absolutely no way of knowing.

DR. GAZZANIGA::  So say within your own center.  What percent of births do you think would be governed or guided by this technology?

DR. HANEY:  We don't do it at all.  We wouldn't do it.

DR. GAZZANIGA::  Oh, you don't do it?

DR. HANEY:  No, no.


DR. MAY:  You used the word "franchise."  Are you talking about the ownership of technique and then franchising locations?  Is that what you mean by franchising?

DR. HANEY:  Maybe that's a loose — I don't know all of the financial arrangements of their program, but they do have the label on MicroSort and various programs around the United States, and they have programs hither and yon.  So that's just a loose way of me describing.

They must be related to Genetics and IVF Institute in some fashion.


DR. HURLBUT:  I want to ask you about a couple of the scientific sides of this.  When they use the dye that interpolates into the DNA, is there a way of washing it out before the gametes are mixed with the oocytes?

DR. HANEY:  As best you can read the technology, it's not actually washed out, other than the amount that's washed out in the processing that would occur subsequent to the sorting before you put it in, and it's thought to be noninjurious, and it will begin to — if you wait an interval of time, you get less and less fluorescence.  So the dye is constantly being disassociated with the DNA.

DR. HURLBUT:  What I'm thinking of in asking that is we're coming to appreciate more and more how transcription is affected by large scale operations on the centrosome and around the histones and so forth so that something that interfered, even if it seems to be innocently intercalating, might actually be affecting something.

DR. HANEY:  I think that's the concern.  That's I think what prompted the abstract with the 300 offspring and looking for anomalies, but they're all young and you don't know what's going to be there over time.

DR. HURLBUT:  Other questions on this line.  I know that at least I think it's established that there are often events within a normal embryo where aneuploid cells are produced, for example, or even cells without nucleus in a given eight, ten cell embryo.  There may be a couple of cells that are abnormal.

They normally gravitate into the trophoblast apparently and don't actually make their way into the embryo.  So this isn't a matter of sex selection primarily, but you could have instances in post – implantation or pre-implantation diagnosis where you are actually getting a misimpression from looking at one cell over another, right?

DR. HANEY:  There's no question that when you do pre-implantation genetic diagnosis for anything, you're going to have a limit to the technology because of heterologous, if you will, or heterozygosity of this, the particular agent, and particularly if you have some error in one cell doesn't contain that.

They don't use polar bodies that much for that reason.  Polar bodies could be used, too, for the maternal mutations, and they are less reliable than blastomeres, and you're talking about blastomeres that aren't all the same.  But these are not chimeras.   So it's some other technical abnormality that would have to be there.

But I think that's known inherent with the technique.

DR. HURLBUT:  And finally, the recent evidence seemed to suggest that there is already polarity even in the early embryo, and that there is asymmetric cell division with regard to cytoplasm, which may contain certain determinant factors in the ultimate outcome of the embryo.

This is a rather abstract question, but does it worry you at all that even if you produce a normal, apparently normal pregnancy out of this process that you're actually altering the outcome of the individual life?

DR. HANEY:  By what?

DR. HURLBUT:  By taking —

DR. HANEY:  Blastomere biopsies?

DR. HURLBUT:  — a blastomere out of the developing embryo.

DR. HANEY:  At an eight – cell stage, I mean, the blastocyst is five to six days, and you have a blastocoll cavity and innercell mass, and then the trophectoderm on the outside.  And that's clearly — I don't know if polarity is the right word, but it's clearly differentiated into those compartments.

At the eight cell stage and based on animal models, one would not anticipate that you would alter that blastocyst development by any methodology I'm aware of.  I can never say it couldn't be there, and the more subtle you look, the more reassured you are, but as best I know, when you biopsy that early when there are undifferentiated cells, before that and at an eight cell stage you don't see polarity.  They're just eight cells.

So I think you have to get another day or two out to be able to see that, but I cannot tell you there's some, but not biochemical polarity occurring, if you will, in that process.  And all you can ultimately do is look at the offspring that are born, and hopefully in an animal model under experimental conditions, and determine that as you move forward.

DR. HURLBUT:  There was an article in Nature magazine about four months ago.  Its titled "Your Destiny from Day One," and it tracked the asymmetric cell divisions and certain cytoplasmic factors that were disproportionately assigned between the cells of the embryo, showing that there is a predicted cell fate even at that early stage with disproportions of cytoplasmic factors.

So theoretically at least you  might be getting a different outcome than you would.  The only reason I raise this is because in the thinking about this issue, is the general assumption in the community that you're taking one or two blastomeres out at the six to ten cell stage is not affecting the outcome in any way, or is it accepted that you're affecting the outcome in some way, but not adversely in an abstract sense?

DR. HANEY:  I think that the general, if I had to characterize it, it would be that you're risking that the embryo will not survive, but you're not risking that you're going to have an anomalous embryo.

DR. HURLBUT:  What percent increase of failure to survive do you think you're affecting the embryo with by doing this?

DR. HANEY:  If you looked at IVF success, and I won't say that this is good enough data, but I would guess that you're at least cutting the success rate if you looked at standard IVF without a biopsy at let's just say 35 percent and you do a biopsy, it's probably 20.

DR. HURLBUT:  Thank you.

CHAIRMAN KASS:  Gil and then Michael Sandel.

PROF. MEILAENDER:  I understood you to say you don't do this at your clinic; is that right?

DR. HANEY:  My institution, that's correct.


DR. HANEY:  I wasn't going to talk about ethics and that sort of thing.

CHAIRMAN KASS:  You're free to.  You can take the Fifth if you'd like.

DR. HANEY:  No, we would —


DR. HANEY:  We don't philosophically agree with gender selection.

PROF. MEILAENDER:  I mean obviously you don't have to talk about anything you don't want to, but I'd be interested if there's a short version of the philosophy and if you're willing to say a word or two about why you don't agree with it.

DR. HANEY:  I'm fully supportive of pre-implantation genetic diagnosis for medical indications, for medical diseases.  We just don't believe that gender — influencing gender birth by medical manipulation at my institution personally for my division, and all three of us who work there, we're gynecologists.  We think women are good people, too.

PROF. MEILAENDER:  So you engage in client selection?

DR. HANEY:  I'm sorry?

PROF. MEILAENDER:  You engage in client selection.

DR. HANEY:  Define "client selection" for me.

PROF. MEILAENDER:  Well, the purposes for which one wants —

DR. HANEY:  Okay.  Medical indication.

PROF. MEILAENDER:  — medical indication.

DR. HANEY:  Correct.  In that sense, yes.

CHAIRMAN KASS:  Technically speaking, by the way, it is — I think you might agree with me in the suggestion that what you're doing when you're doing pre-implantation and genetic diagnosis that involves the gender of the offspring, as in the cases of the X-linked diseases, if there were a way of finding the presence or absence of that disease marker, you wouldn't be doing sex selection at all.  You would be looking for the marker.

So this is sort of incidentally sex selection as a way of making sure that you do not produce the afflicted child.  I mean, it belongs really with pre-implantation genetic diagnosis for diseases, and it happens that the quick way to screen for the possibility of the disease is to screen for X-Y.

DR. HANEY:  You're exactly right.  If you looked at the — you have two Xes.  One is affected; one is not.  And if you could not only look for the X, but you could look for the specific mutation on the X, then you'd be doing exactly the same thing as any other autosome.

And I think it's just an efficiency, quick and dirty, simple.


DR. HANEY:  It's much easier to screen for the X chromosome than it is to be looking for the specific gene mutation.

CHAIRMAN KASS:  Right.  So that technically speaking, I don't think — this is a kind of accidental sex selection as it were.  The intention really is disease prevention.

DR. HANEY:  Correct.

CHAIRMAN KASS:  Could I, Michael, before?  Just to clarify, to see if I can sum up, and I think this is what Mike Gazzaniga's first question was getting at, just sum up where we are technically speaking here.

There is certainly no cheap way of doing this.  There's no do-it-yourself way of selecting in advance.  The best figures right now from MicroSort are if someone were interested, for example, in producing a male child, they could get 70 percent instead of 50, roughly speaking.

So that at the present time it doesn't look like that there is anything that is likely to be used on a wide scale in the way of selecting the sex of children. 

Would that be —

DR. HANEY:  I'm even hesitant to say it's effective until I see randomized clinical trials showing me data.

CHAIRMAN KASS:  Okay.  Then second, on the pre-implantation genetic diagnosis, there the diagnosis is pretty accurate.  There's some questions about the safety of —

DR. HANEY:  An estimate would probably be 90 percent.  You'd be 90 percent accurate with a prenatal — pre-implantation genetic diagnosis using PCR or something like that.

CHAIRMAN KASS:  Only 90 percent?

DR. HANEY:  Un-huh.

CHAIRMAN KASS:  Now, if this technology were used for screening for diseases on an increasingly large scale, in other words, if the future holds much more pre-implantation genetic diagnosis, wouldn't it be — will it be just as easy to simultaneously do the screening for X and Y?  Get the information whether you want it or not?

DR. HANEY:  Not if you're doing specific gene probes.  So if you're looking for Huntington's, you're screening for Huntington's, and you're not doing anything to look at the —


DR. HANEY:  — gender determining chromosome.

CHAIRMAN KASS:  So there's no necessary — if someone were to say, "Look.  If we have a lot more PGD coming, then it will become a lot easier for people who are using PGD also to engage in nonmedical sex selection."

DR. HANEY:  I think you can argue that the better you get a PGD, the less gender you care about.  that's what you were alluding to before, and you're going to use probes —


DR. HANEY:  — that are specific for the mutation, and you'd have to do something additional. Now, if the lab is working, sure.  I mean, and you had another probe for something on an X or a Y.  If your lab is functioning, that's true, but the more specific you get, the actually less concerned you are about the —

CHAIRMAN KASS:  Well, you're looking for the disease, but I'm thinking now in terms of the client.  The client says, "Look.  I'm going to have PGD anyhow.  By the way, I would like a girl," or, "By the way, I would like a boy.  Can you do it for me, Doc?"

DR. HANEY:  I think it's exactly the same argument that you had if you did it just without a disease.

CHAIRMAN KASS:  Technically, I mean, it's —

DR. HANEY:  It's just a matter of having a functional system.


DR. HANEY:  And if you're good at it and you have a lab doing it well and you're looking for a lot of different mutations, I mean, no center is probably going to do everything.  You're going to have some centers that do cystic fibrosis more and Duchenne's muscular dystrophy and all the various ones, and there's going to be some centers that like to do a lot more of it, and they're just good at it, and they have a system that's very efficient.

And so if you said, "Fine.  We had one of those other probes," yes, it will be easier, but it's actually going to be less important from the geneticist's point of view.

CHAIRMAN KASS:  Yeah.  Let me try one last time.  Sorry.  I don't think I'm — what I'm trying to figure out is whether this is a problem we should worry about in the United States or not.

DR. HANEY:  Okay.

CHAIRMAN KASS:  And that was in a way Mike Gazzaniga's question.

DR. HANEY:  One of the most interesting parts about the numbers is you have no idea of where the country are that they came from.

CHAIRMAN KASS:  I'm sorry?

DR. HANEY:  You have no idea where the countries are that they came from.  The data they presented because it isn't even listed in the abstracts as happening in the United States.  So I have absolutely no idea what it is.  It's a very small number.

CHAIRMAN KASS:  Well, it would seem to me that it would not be — I mean, people who want to do sonography and abortion can do that.  I mean, people who want to try to find some way of selecting in advance, either this MicroSorting technique is going to have to be perfected or you're going to find surface specific antigens that will differentiate X and Y carrying sperm and so on, stuff that we don't yet have, or those particular people who offer — there are apparently some 30 or 40 or 50 clinics in the United States that already are offering sex selection.  If more and more people are using PGD for other purposes and they don't have your scruples about doing this for nonmedical reasons, the question is:  could this get to be a sizable phenomenon simply by piggybacking on an increased rate of PGD?

That was the point of the question, and Ó-

DR. HANEY:  There's no question more PGD you do, the more things you can screen for.

CHAIRMAN KASS:  Yeah.  I'm sorry for holding you back, Michael.

PROF. SANDEL:  I have a general question, but this exchange leads me first to a specific one for Dr. Haney.

Did I understand that the thing that you haven't seen, the kind of trials that would persuade you that it works, that's the pre-fertilization sex selection?

DR. HANEY:  I mean, if you're at the FDA, you do a randomized trial.  You're looking for an outcome, and you wouldn't accept something that didn't have that.


DR. HANEY:  And I don't care if you're a surgeon or — it's sometimes harder to do, but you need to look at a more substantial database to be able to do it.

And the disconnect is when you look at FISH related semen or you were talking about 70 percent of the sort being male.  That's not going to translate to 70 percent males born.  It's going to be less than that.

So that it's a fairly inefficient system.  So before I would accept that it's really going to work, if I was simply looking at it, I would want more rigorous testing, and certainly by a variety of different participants, not just the franchisee.

PROF. SANDEL:  So that's the sorting.

DR. HANEY:  That's the sorting business, right.

PROF. SANDEL:  But the thing that works is the pre-implantation.

DR. HANEY:  The PGD works much more — it would be much more highly accurate because you're taking the actual blastomere from the genetics of the embryo you have created and screening for it, and there's very simple whole gene amplification and looking for other epitopes that you could identify, satellites, alpha satellites, and whatnot that you can clearly identify.

So far more expensive, far less efficient, but much more accurate.

PROF. SANDEL:  Well, thank you.

This has been a fascinating account of really what's out there, especially for those of us who are not knowledgeable about this field.  It's of enormous interest.

The general question I had really was for Leon and for the group, and it's a question of what we do with this fascinating overview of the technology.  We could kind of probe Dr. Haney to see, well, just what's the risk of harm and so on and at what stage is this done and what are the technologies that involve destruction of embryos and which ones not and how prevalent.

But is there an opportunity — and maybe this isn't the session for us to take up this question — to address the underlying ethical question itself if there were a way of doing this without harming, without harm and without killing embryos?  Would it be objectionable and on what grounds?

Are we supposed to shoehorn that into this discussion or is this discussion just acquainting us with the technologies so that we can then be in a position to decide whether later we want to take up that ethical question or not?

CHAIRMAN KASS:  Well, I mean, my sense was that we invited Dr. Haney, as we've been inviting other researchers in the field, to get us up to speed as to what's possible and even to — and I'm very grateful to Dr. Haney also for not shying away from suggesting what might be possible, thinking ahead to various at least conceivable techniques so that we have a way of thinking about this — but that we amongst ourselves are free now also to discuss the implications of this and the ethical questions.

The American Society for Reproductive Medicine has — and I think these were distributed with the briefing books — has policy statements both on the use of pre-implantation genetic diagnosis for nonmedical sex selection and also on the sperm sorting, and it seems to me perfectly reasonable for us to discuss amongst ourselves and Dr. Haney insofar as he wishes, I think, to enter into this discussion.

So before we went there, are there any other technical or use questions before we went into that area, just so that we don't leave out?

Janet, did you want to?

DR. ROWLEY:  So as I understand it from particularly one of the slides that you had of oocytes or developing blastomeres in a Petri dish, that particularly if you have super ovulation you may have four or five oocytes that you can do for fertilization.

DR. HANEY:  Typically what's done is all mature oocytes are — I shouldn't say "all" — but almost all.  If you also do ICSI, you practically will do eight or ten or 12.  If you had 15 embryos it gets a little much to do them all, but you do basically all you can.

And then if you're just in standard IVF, you're going to put sperm, hypermotile sperm, with all of them because the fertilization rate probably won't exceed 50, 60 to 65 percent, and then the two pronuclear egg, the pronuclear egg with two pronuclei won't get to a cleaving blastocyst.

So if you started with ten oocytes, you'll probably have six or seven that are fertilized and probably no more than four or five that are cleaving embryos in the process.  So it's a matter of trying to get as many in the beginning so that you can get down to reasonable numbers at the end.

But if you end up with healthy, which occasionally you do, healthy embryos that you're not going to transfer, you freeze them.

DR. ROWLEY:  Okay.  That was actually my question.  So then you have these frozen embryos available either for the couple if they choose to have a second pregnancy or for whatever purposes.

DR. HANEY:  Correct.  In every IVF center in the United States there's large numbers of frozen embryos.

DR. ROWLEY:  Right, because this is one of the issues that we dealt with earlier on, is that at some point some of these developing embryos may actually be discarded, and then if that's the case, what are the acceptable uses of those developing embryos?

And what's the practice at Duke?  Just continuing to accumulate these, or do you have a time after which you discard them?

DR. HANEY:  Every institution has their own  pragmatic and philosophic way of dealing with this.  So you have our attorneys have worked out an agreement.  The patients sign it.  They pay for and store extra embryos, and they're their embryos, and many of them will come back.  If they don't conceive, they'll be back to get their embryos thawed out, much less expensively transferred.

If they conceive, I would say 90-plus percent will be back a year or two later for another transfer.  If you can get both pregnancies out of a single embryo or a single oocyte retrieval, it's much more efficient.

There are people who will get a set of twins and deliver their babies and say, "That was our family size expectation and I have three extra embryos that are still in the incubator or still in the freezer," and then we have legal issues that they have to go through, what to do with those.

I don't know what the actual numbers are, but most of them go through a legal adoption process and put their embryos up for adoption for couples that have no sperm and no oocytes, cancer patients, premature renal failure patients, Turner's Syndrome, people who have no oocytes at all, and they'll adopt embryos.

In our state, we have no embryo adoption law, but we go through an adoption process with an attorney that basically the couple that's going to receive the embryos, before they get them, goes through an adoption process, and the couple that's giving them up goes through a very legal process as if, virtually identical to what they would do if they had a living child doing the same thing.

And then some couples will say, "I want the embryos destroyed."

We haven't destroyed any embryos.  So even the couples that say, "We don't want them," we haven't yet destroyed them.  Our lawyer keeps telling us sooner or later we have to destroy them if the couple says, you know — we haven't had people tell us they want them destroyed.  They have just left them in the freezer, and we're not going to destroy them.

But our lawyer tells us sooner or later couples are going to tell us to destroy them, and we have to.

CHAIRMAN KASS:  Thank you.


PROF. DRESSER:  These are probably overlapping facts and ethics questions.  I was wondering do you know whether the FDA has tried to regulate this MicroSort process.  It sounded to me as though you personally are concerned about  the objectivity of the data.

And is there concern in the organization about pushing toward more rigorous testing and about inflated claims that might be made and advertising similar to the, quote, success rate of IVF in the past?

DR. HANEY:  Yeah, I think I wouldn't exactly characterize it as objectivity of the information.  I just think it has to be done in a rigorous, scientific design.  And it's not a scientific design that's currently in the literature.  That's number one.

Number two, the FDA is going to exert, I'm sure, regulatory authority over most gamete tissues in the lab comparable to blood banking, and the ASRM has spent some time trying to help educate them.

For example, they wanted us to do the kind of testing we do for semen samples to oocytes, for oocyte donation, and for semen you can freeze a sperm.  You can double check the donor six months later for all of the diseases.  You can take an aliquot of the sample and test it if you choose.  You have a lot of opportunity to do that.

With an oocyte you don't have that opportunity.  You either use it or it's gone.  So they wanted us to freeze oocytes as their preliminary to do donor oocytes, and that just isn't possible.  You might as well ban the procedure because it's not going to be functional.

So they needed some education, and they got it.  I think they're going to be reasonable about what they do and be rational and do it the same way.  They just need a little scientific updating because they're used to blood banking and other kinds of tissue things.

And there are some other issues that they just needed education on, and I think it can be ultimately anything that bothers programs.  We all now have certified CAP, College of American Pathology, or whatever certified programs, and I think that's just going to be one more layer.

PROF. DRESSER:  So you think they probably will start looking at safety and efficacy of this MicroSort procedure?

DR. HANEY:  I suspect they will.

PROF. DRESSER:  The other question I had was I was a member of the ASRM Ethics Committee during a lot of the years when these two statements came out, and I was wondering whether your program's position was at all influenced by those statements.

Because one of the problems is when we think about trying to formulate professional standards or, you know, ethical recommendations and so forth, if they don't have any legal effect or there are no professional consequences to not following them, you  know, in terms of the organization perhaps suspending a member or something, how much influence do they really have?

So I wondered whether you all paid any attention to them or you've just reached your own program's position based on —

DR. HANEY:  I would simply say it's a work in progress.  It's becoming much more influential than it was in the past.  Programs now in SART [Society for Assisted Reproductive Technology], they're going to very seriously look at the average number of embryos you transfer and things like that, that there were ethical comments, and in practice those are translated into practice guidelines.

And you're going to have very specific criteria that you should work under for maternal age related influences  in the number of embryos you transfer, et cetera.

I mean all of that, I think is becoming much more codified, but it started with absolutely nothing, and it has taken some time to get there.

And in the absence of being able to do NIH research on any of this because of no funding, you know, it's difficult.


CHAIRMAN KASS:  Michael, did you want to raise a particular question on the ethics of this?

PROF. SANDEL:  Well, I don't know whether — did you want —

CHAIRMAN KASS:  Please.  I think we should get started on it in any case.  So please.

PROF. SANDEL:  Well, rather than advance it, I think there is a widespread sense, though I don't know how widely shared, that there is something ethically troubling even apart from the question of harm and even apart from those technologies that would involve killing embryos in nonmedical sex selection, but it's one thing to have that reaction and it's another to try to articulate the actual reasons and to assess those reasons to see if the initial reaction is correct.

And it seems to me that independent of the harm consideration there are at least two different kinds of reasons that might be operating to explain this and to ground the unease, and this is really just by way of inviting reactions.  It's not a worked out view certainly on my part.

But one kind of ethical worry has to do with the way in which this practice, were it to become widespread, might change the norms that inform the practices of procreation, childbearing and child rearing by changing the relation between the parents and the children in some of the ways that we worried about when we were talking about the designer baby objection to cloning, reproductive cloning.

So that would be one set of objections that we might investigate, and then another set of objections might have to do with a different worry, which is the disposition or the character of the desire to control, to choose the sex of one's offspring.

Independent of the effect on social practices and the effect on childbearing and child rearing, is there something troubling in the stance?  Maybe the short label is the hubris objection, something objectionable in the stance of the person who has the desire and acts on the desire to control the sex, to choose the sex of his or her offspring.

These seem to me different kinds of objections, though they may overlap, and there may be others, and to further question how weighty the two are and how we would make them up, this really would be just by way of inviting discussion on them.

CHAIRMAN KASS:  Someone want to join on this?  Sir, Mike.

DR. GAZZANIGA::  It's sort of a prior question to that.


DR. GAZZANIGA::  When hearing about sex selection and seeing the fact that it's sort of being done ad hoc and without any massive government program or monitoring, it raises the question, of course, of how American medicine works.

And how American medicine works, my understanding is that it's sort of monitored locally, monitored by professional groups, monitored by the sociology of a specialist field, and that there isn't a grand monitor somewhere in Washington making sure that all the procedures that go on in a hospital have been given some stamp of approval or being carefully followed up as to their efficacy and so forth.

And if I'm wrong, I would like to —

CHAIRMAN KASS:  That's absolutely right.

DR. GAZZANIGA::  — I'd like to have comment on that so that we all understand that as one thinks about this issue, because it is a red button issue, and if there was any recommendation to somehow monitor this from a government point of view, it would be one of the first because that's not how we do medicine, and I think that's just worth a comment.

CHAIRMAN KASS:  Well, I think the point ties in with the topic we'll be talking about tomorrow with the help of our British visitors, where they have official bodies that in some cases simply advise and in other cases, in fact, regulate.

The question of regulatory activity was part of both sides in the cloning report and something that this Council wants to take up in a serious way.  So the fact that it's unprecedented, while true for the time being, might be an invitation to think through whether we really want that precedent to remain.

But would someone go back and pick up a response and then Robby and then Bill?

DR. GÓMEZ-LOBO:  Yeah.  I'd like to continue along the lines opened by Michael.  I thought it was interesting.

And just to contribute to the conceptualization of the problem, I honestly think about this as questions.  It seems that there is a goal which has to be questioned first, the goal of choosing the sex of the child.  I think that there are various problems there.

And then comes the question of the means, and if I understood correctly, Dr. Haney, there are really two methods:  selection and determination, right?

Now, selection seems to involve the discarding and destroying of sperm.  Now, of course, from a moral point of view that may not be a problem.

On the other hand, determination which can be both pre-implantation and post implantation seems to entail the discarding and destroying of embryos, of human life, human embryos.  And of course, that is very troubling for anyone who tries to think about it.

Now, a last remark on this.  I was very impressed by those charts about three countries, China, South Korea, and Singapore.  Now, the charts about South Korea were really very, very impressive.  They get to, what is it, 130, 140 males per female?

It would seem to me that's a massive discrimination and destruction of females either via abortion or even infanticide.  Is that a possibility?

So that would seem to me to be the extreme to which the acceptance of the goal can take.  Again, these are questions.

CHAIRMAN KASS:  Someone else.   There was Robby and then Bill, yeah.

PROF. GEORGE:  Michael's comment earlier obliquely raised a different question for me, Michael Gazzaniga's comment a minute ago, and it brought me back to thinking about Dr. Haney's comment that his clinic doesn't do this and doesn't do it on ethical grounds.

When Michael was referring to the way in which we practice medicine or the way in which medicine is monitored in this country, it left me with a question.  When it comes to sex selection, Dr. Haney, in thinking about whether your group would do it, is part of your thinking governed by the question whether this is medicine?

 What's your own thinking on it?  Let me just ask.  Is sex selection medicine?  Whatever else it is, I mean, is it medicine?

DR. HANEY:  I think that's the question everybody is going to ask themselves.  We would probably at my institution for nonmedical reasons say no.

PROF. GEORGE:  And how do you decide the question of what constitutes medicine and what doesn't, just sort of in a rough and ready way?  I mean, in a borderline case, how do you think about that?

DR. HANEY:  Define a borderline case.

PROF. GEORGE:  Well, someone wants to insure that they don't have a mentally retarded child.  Is that medicine?

DR. HANEY:  I would think that is.


DR. HANEY:  I would view that, if you knew the process that created the mental deficiency, Tay-Sach's disease, something, not having a child born with Tay-Sach's disease I think is a medical decision, just as it is for Down's Syndrome or other things.

PROF. GEORGE:  How about a borderline case like color blindness?

DR. HANEY:  I'm color blind.


DR. HANEY:  And I don't think that's — it's probably not fair to ask me.


DR. HANEY:  I'm only mildly color blind.


DR. HURLBUT:  When you try to differentiate this issue of sex selection from what you said was medical, and I assume you mean therapeutic in the sense of healing something or preventing something that has a detrimental implication, buried in your statistics was something that struck me as potentially troubling.

I know this might be pulling too much out of a very limited sample, but I think it was with MicroSort.  One of your statistics showed a decreased rate of congenital abnormalities.  You said half, roughly a third to half of the rate.  It was 2.6.

DR. HANEY:  I don't think that's decreased.  In other words, they found in their sample 2.5 percent of offspring had a major anomaly.  The accepted rate in Washington, North Carolina, wherever, is about three percent.  So three to four at the most, depending a little bit on the definitions and how thorough you are.

Those are not different numbers.  I have no illusions.  That's not a decline.  That's just within the range one would anticipate for a general population.

DR. HURLBUT:  Oh, okay.  I thought you had pointed out that it was somewhat lower than the otherwise noted rate.

DR. HANEY:  In this particular sample it was 2.5 compared to —


DR. HANEY:  — what most public surveys would be a little higher.

DR. HURLBUT:  Not making too much of that particular sample, I want to ask you a theoretical question then.

Suppose it turned out that this procedure, either sex selection by gamete sorting or pre-implantation genetic diagnosis for something neutral or even IVF just done without some sort of sex selection, actually had a therapeutic effect in the sense that it produced better outcomes than the natural way of doing procreation.  Would that then change your view if the procedure that produced this better outcome actually produced a healthier subset?

Do you see what I'm getting at?  I mean suppose —

DR. HANEY:  Are you asking me if Caesarean section is less traumatic than vaginal delivery would I tell —

DR. HURLBUT:  No, no.

DR. HANEY:  — everybody to get sectioned?

DR. HURLBUT:  Oh, okay.

DR. HANEY:  Is that what you're asking sort of?

DR. HURLBUT:  Maybe.  That's not — that carried a little different atmospherics, but what I'm kind of getting at here is are we heading to the realm where as we understand the biochemistry and then the early development, media into which the blastocyst is sculptured; are we getting to the point in your opinion where we might be able to produce a better than natural outcome?

DR. HANEY:  I would be highly suspicious that's not true.  I can't fathom that's true, but I'd have to face that if it actually had some data to bind it.

DR. HURLBUT:  Well, isn't it true that the implantation of blastocysts at blastocyst transfer actually have a higher success rate than would be statistically implied by normal sexual intercourse?

DR. HANEY:  Oh, I don't think that's true

DR. HURLBUT:  You think 30 percent of successful blastocyst formations go on to —

DR. HANEY:  No, I don't think it's anywhere that high.

DR. HURLBUT:  Isn't that the rough success rate with blastocyst transfer?

DR. HANEY:  Well, all I can tell you is our experience in 1998.  We thought this was blastocyst — in other words, the biggest problem with IVF is your inability to pick out healthy embryos.  We don't understand implantation well.  Human embryos implant very nicely in the fallopian tube.  They don't need endometrium to implant.

If they're healthy and at the right point in time, aggressive — I think humans are the only species that gets ectopics.  So our embryos evolutionary have learned the ability to implant themselves at the appropriate time.  So they don't really need endometrium to do it in.

Then we get to IVF where we're putting embryos a little dissynchronous into the uterus, and we don't know how to pick the good embryos that are going to potentially be developed into humans, and so we compensate by adding more embryos for transfer.

In 1998, the notion of — we were able to then, by media changes and some understanding better of in vitro culture technology, were able to keep embryos on the same growth curve that they would occur in vivo in vitro.  So the opportunity then to grow them further out and do blastocyst transfers and allow developmental selection to pick the best embryos, that idea finally became possible to test, and that started in 1998 or '97-'98, and there was a great enthusiasm for growing embryos out to the point you let the healthy ones continue to grow and transfer those, and you will cut down the multiple gestation rate by being able to maintain a relatively high pregnancy rate with your multiple gestations.

And my reading of the general approach today is that that didn't work out; that it has not been as successful, and there are very few programs exclusively doing blastocyst transfer today and very selected patients who are getting it.

So if anything, I would say it's not, as you allude, a better implantation rate than in vivo.

DR. HURLBUT:  So what you're saying is nature is better and looks to you like will be better in the long term.  It's better not to intervene in nature is —

DR. HANEY:  Well, we're definitely treating disease.

DR. HURLBUT:  Pardon me?

DR. HANEY:  We're treating disease, infertility.


DR. HANEY:  Or we're treating patients who can't get pregnant or we're treating patients who in this case if you talk about PGD, of people who have genetic diseases.

But we're not enhancing normality or improving upon it.

CHAIRMAN KASS:  We're going to come up on the break.  I have Gil and then Michael.

PROF. MEILAENDER:  I want to try to just think a little more about the issues that Michael Sandel raised, and if I use Alfonso's distinction between the goal and the means, I want to try to think about the kind of means that would be least problematic presumably, namely, the pre-fertilization, the ones that you don't think are very successful right now.

But if we just keep it to that, then we're not raising the kinds of issues that pre-implantation genetic diagnosis would involve.  Though I have to say, by the way, that on another occasion I'd like to pursue with you the question of exactly in what sense you're treating disease when you do that.  I mean, I'm not actually persuaded by that.

There's a difference between eliminating diseased embryos and treating disease, but let's let that go.

I can understand how you might be worried about problems that would happen with the sex ratio as a result of doing this, and I can understand simply saying we don't want to use medical resources that are scarce in this way.

But what I'm interested in is those ASRM policy statements that we read that Rebecca has claimed some responsibility for here.

No, the interesting thing is that the way the issue is couched is that you shouldn't do this in order to choose nonessential characteristics of human beings, and I'd like to see us figure out what that means in a way.

In certain contexts I would never describe being male or female as being nonessential.  In fact, I'd describe it as much closer to what's essential to a human being than lots of other things.

So I'm not sure what role that language is playing actually there.  I can understand some other things it might mean, but the longer I think about it, the less clear I find myself actually, and so I just put it forward.

If the kind of dis – ease that countless people feel with respect to sex selection is articulated or is to be articulated, is that the right way to articulate it? 

I don't know.


DR. MCHUGH:  I, like Gil, would like to follow up on what Michael was saying, that we should be laying out some of the ideas that are a concern to us, and the one that I think is wrapped up in the issue that Gil mentioned is sex ratio, but probably needs another expression.  Probably an expression felt by PROF. Haney, that is, that you know, when we produce our children, we're not just producing it for our generation.  We are producing it for a community, a community ultimately that goes generation after generation.

And although one can appreciate perhaps that you would like one or the other at this time simply for somebody to be able to use your baseball gloves or somebody to be able to do something else, there is a higher purpose that we all serve from our families to the community at large.

And I feel that one of the things that concerns me about this is that we alter something which is very natural and which we sense only as we move from one generation to the next.

We saw that very interesting data that you showed us from South Korea, and you used interestingly the nonchallenging term "family balancing."  It seemed remarkable that the family balancing all tended towards producing males.  So it seemed to me to the motivation was male specific, and maybe we should talk about this as a growing male hunger in some cultures because, after all, you could have three or four children and they could all be boys, and one would have thought if it was balancing, there would be a balancing out of those.

So I just wanted to make the point that the issues that we're talking about have more to do with just the pleasure of one child or one sex now with all that that relates to the designing of the relationship and that we deal with a community that ultimately has a long term ahead of us.

CHAIRMAN KASS:  Yeah, the practices in other countries is the topic of the next talk, and we'll talk about that then. 

I want to intervene briefly in my own name, too, if I might.  It seems to me that however uncertain are the technologies at the moment, it's also clear that there is an increasing pressure whether created by the people with the licenses to have the techniques used or whether there is a growing demand in the culture for the use of these things.  There's a lot more interest in sex selection now than there was even five or six years ago.

And particularly with PGD there are clinics that are offering this service for non-medical reasons, even though the society has discouraged its practice.  I mean, there are groups that are using this, and we understand from our friends in Britain that they have now to reconsider this question there as well.

So whether we like it or not, it seems to me it's a question that's going to be coming on the table.

And here it does seem to me that a couple of things that have been mentioned are worth our attention.  One is the question of what the arguments are both for and against.  And I guess I have to say that I find the arguments in the ASRM documents puzzling, where, on the one hand, one begins by worrying about contributing to sex bias as if sex doesn't matter at all; on the other hand, arguing that it's probably okay for sex balancing, which suggests that sex matters a great deal, providing that you treat it absolutely equally, that you don't give additional preferences.

And I'm not sure, Paul, whether one could say — I mean, one worries really about what it means not just to pray for a child of a certain sex, which doesn't necessarily produce the result because the Good Lord doesn't necessarily give you what you want, but there's a difference between that and actually having exercised the control over it and have the parents be responsible to the child for the choice made, regardless of what the choice is.

That's a new step, but I'm not sure that we're going to be able to say in this rather libertarian climate of ours that you can't use it for these and these reasons.

And in the Jewish tradition, orthodox tradition, there was a great demand for the son because, among other reasons, there's an obligation that falls on the male child to say the Kaddish for the dead, an obligation that falls on a male child.  Female children can do it.

And people would describe a male child as "now I have my Kaddish."  Now I have the child who is going to actually say the prayer when I die.

I'm not sure that one is going to be able to say to members of those subgroups that's an illegitimate reason for making use of this technology should it be available.

So, I mean, I think that there are a variety of reasons why even if this is practiced on a small scale and when it's practiced on a small scale, by the way, it becomes a question for people to decide do you want to take advantage of it when it now becomes offered.

I think that it's much more complicated than the ethical argumentation that we've seen in those documents indicated, and I think the pressures on this are going to increase, and it might mean that if this is to be left for professional self-regulation, to avert to Rebecca's point earlier, that maybe the society has to think about ways in not simply saying this should be encouraged or this should be discouraged, but these are the guidelines we expect people to follow for these and these kinds of reasons, or perhaps it is for us to recommend other kinds of bodies for at least promulgating such recommendations, subject, of course, to review and reconsideration as time goes by.

DR. MCHUGH:  Yes, Leon.  I was not when making my point saying that this should be go to legal issues.  I just think that we're at the point of raising consciousness about what our problems are or what our difficulties represent when sex selection is brought forth.

We want to just sense is there anything behind our feelings other than the "ugh" factor that we've talked about before.


DR. MCHUGH:  And I'm just saying that one of the things that lies behind it is a sense that we all have, believe it or not, even though we never feel it at the time the first child comes along, that we have a generational and community responsibility that comes from bringing up our children well, making sure that they are members of this community, as well as being suitable to go out and find new mates and add to our world.

CHAIRMAN KASS:  let me ask.  Since the general discussion can come back and we've run way over here, let's take a ten minute break.  We'll have the second discussion, and the general questions will come up after Nick Eberstadt's presentation.

(Whereupon, the foregoing matter went off the record at 10:44 a.m. and went back on the record at 10:57 a.m.)

  - The President's Council on Bioethics -  
Home Site Map Disclaimers Privacy Notice Accessibility NBAC HHS