FRIDAY, JUNE 23, 2006

Session 5: Newborn Screening for Genetic Disorders

Duane Alexander, M.D., Director, National Institute of Child Health and Human Development, National Institutes of Health

CHAIRMAN PELLEGRINO:  This morning we turn to a new topic, Newborn Screening for Genetic Disorders.  Our first speaker will be Dr. Duane Alexander, who is Director of the National Institute of Child Health and Human Development, National Institutes of Health.

Dr. Alexander, we usually do not go into a lengthy curriculum vitae, simply because it's so long and so impressive that we want to leave time for you to speak rather than spend it all on your biography.  So if you would be kind enough, we're delighted to have you address us.

DR. ALEXANDER:  Thank you very much.  That's the best kind of introduction, is a very brief one.

Thank you for the opportunity to speak with you today on a topic that's very dear to my heart, something that I've been involved in for a long time, something that I am not involved in as a geneticist or a professional genetics screener, but as a pediatrician who has had some experience in the ethics arena, but has been dealing throughout his career with the field of developmental disabilities and mental retardation and ways to alleviate those particular problems and disorders, by prevention when possible and by treatment to the fullest extent that we can.

I'm going to be talking with you about newborn screening, and I'm going to begin with a highly visible disclaimer.  These are usually in small print at an end, but the fact is that much of the factual information that I'm presenting is clear and referenced and well defined, but I will also be presenting to you some opinions that are quite clearly my own, not necessarily those of the Department or of the NIH or of the NICHD.

Some of these have been vetted and cleared for publication by the Department, an article by Dr. Van Dyck and myself that's I believe in your folders, but that has a disclaimer on it as well as representing our views and not necessarily those of the Department.  And it's very clear that that be understood.

Now, newborn screening has been around really as long as medicine has for particular defects or problems with the infant.  We look at the newborn infant, we count numbers of fingers and toes, we listen to the heart for heart murmurs or successions of heart disease, or look at the skin color.  We look at the genitalia for abnormalities and other parts of the anatomy as well. 

We even look at the baby with the idea of genetic disorders in mind, for Down's Syndrome, we look for ephocanthal folds, low-set ears, transverse Palmer crease .. this is something that's called a simian crease .. and we were diagnosing Down's Syndrome, a genetic disorder, long before we knew it was a genetic disorder, and far longer than we knew that it was caused by an extra chromosome 21.

So the idea of newborn screening itself is not new.  What is new is using laboratory studies to help with that process of screening the newborn infant for genetic disorders.  And it really begins with the story of phenylketonuria.  It's important that you understand that background, because it not only was the first but it's the model on which we learn about newborn screening for other disorders. 

And some of the problems we've had with it, some of the unexpected things that have occurred, and how we've dealt with them are all important for you to have as background for understanding what's going on with newborn screening today and presents to us some of the ethical issues and concerns.  So I'm going to spend a fair amount of time about .. talking about the PKU story. 

It really began in 1934 with a description of the condition for the first time by Foling, who observed several patients with phenylpyruvic acid in the urine in association with their severe mental retardation, and he called this condition phenylketonuria. 

It was clearly due to a deficiency of an enzyme that converts phenylalanine to tyrosine, so that phenylalanine accumulates in the blood, crosses into the brain, and the high levels of phenylalanine are toxic to the developing brain and caused the mental retardation and some of the other adverse symptoms of the condition.

No therapy was possible.  People started to question whether it might be possible to reduce the amount of phenylalanine in the diet and, therefore, reduce the phenylalanine levels in the blood.  There was no good way to do that at the time.  Phenylalanine is one of the essential amino acids.  It's part of protein, and it's essential because the body can't make it itself.  It has to be supplied to some degree in the diet.

But in persons with this condition, just the amount that they get in a diet is too much for their systems to handle, because they can't get rid of the phenylalanine by converting it to tyrosine, and so it accumulates and causes the damage.  It was very difficult with the knowledge from the '30s to the '50s to make a diet sufficiently reduced in phenylalanine to really do studies, but some people did this. 

They tried the treatment of people who already had full-blown PKU with all of the symptoms, and some people reported some suggested improvement, but for the most part there was no evidence of any benefit by this kind of late treatment of people who had already been damaged.

The only people who could benefit were the siblings of people who were born to some .. to parents who had already had a child with PKU and who could be identified in the newborn period and be treated with a low phenylalanine diet.  This was done, and there was enough evidence of benefit that people thought that they ought to try to do this in a better way.

This required two things .. first, a better source of dietary protein that was free of phenylalanine; and, second, a way to identify not just siblings of affected individuals who are very few contributing to the problem, but people who had no family history of the disorder and were born with hyperphenylalanine and were people who could potentially benefit from treatment if we had a low phenylalanine diet.

In 1958, the infant formula industry developed a product called Lofenalac, which was markedly reduced in phenylalanine.  It didn't taste very good, but it worked.  And people who were identified early .. sibs of affected individuals .. did show clear benefit from this reduced phenylalanine diet.

Then, the remarkable breakthrough occurred by a microbiologist, not a pediatrician, named Bob Guthrie.  Dr. Guthrie had a mentally retarded son, and he had a niece who had phenylketonuria, and he attacked this as a problem of trying to develop an early diagnostic test that would work in infants.  And what he came up with was something that was quite ingenious, that a microbiologist, not somebody else, might think of.

He identified a species of bacterium, a very common organism, called bacillus cedillas that was deficient in the ability to produce phenylalanine and required phenylalanine to be provided in order for it to grow.  Without phenylalanine, it didn't grow.  What he did then was plate agar plates about the size of a cookie sheet with cultures of bacillus cedillas, and held those in reserve and then would put onto that cookie sheet of agar the .. a sample of blood from newborn with or without phenylketonuria.

Without elevated phenylalanine in the blood the organism didn't grow, didn't have the substrate that it needed for it to grow.  But if there was elevated phenylalanine in that blood spot, the organism would grow and you get a growth ring around the spot of blood.  The blood was collected on little pieces of filter paper, just little dots about half an inch in diameter, and used in a project to test and see whether you could screen newborn infants this way.

And Guthrie's test worked.  You could clearly identify if you .. if the baby had been on milk formula for one or two days, those who had elevated levels of phenylalanine.  Guthrie announced this test in 1960, and there was a rush to apply it across the whole arena.  Even though it was very clear that phenylketonuria was not a very common cause of mental retardation, it was represented in institutions for individuals with MR (mental retardation), but it was a relatively small contributor.

When these individuals were picked up as newborns, and the dietary treatment with Lofenalac was instituted within the first couple weeks of life, these infants grew normally, they developed normally, and appeared to be spared the ravages of phenylketonuria.

There was not a study done immediately about this, but just a few anecdotal cases were sufficient to spark an interest in the whole mental retardation community and really helped to start a revolution in mental retardation research.  The picture of mental retardation research before 1960 when Guthrie's test became available was not a pretty one. 

It was mostly conducted in institutions for individuals with mental retardation.  It was not a topic that was either taught or researched very much in medical schools.  It was frowned on as a third rate field.  The people who were doing the research were pretty much limited to people who worked in state institutions for individuals with mental retardation, and you never saw a person with mental retardation on a medical school campus or in a hospital.

This was the sad state of affairs, and one of the reasons was the belief that this was a futile area for study.  There was nothing to be done for these individuals.  We could not improve their condition, we had no means of treatment, we had no means of intervention.

The PKU test changed that.  All of a sudden there was a means of detection and prevention of these terrible disabilities, and the idea was if there's PKU, there may be other PKUs out there.  Can we find them?  If we can find them and identify them,  particularly in newborns, we may be able to develop treatments akin to what we are able to do for PKU that will alleviate other disorders that cause mental retardation, and the search began.

People who previously had shunned the field of mental retardation entered it.  This was spurred in part by a stimulation of the Joseph P. Kennedy, Jr. Foundation, which the Kennedy family started, and they funded scholars in mental retardation, they funded centers in mental retardation, and this shifted and emphasis as well from places .. from these people working in institutions to working in academic centers .. again, the idea being, can we find and treat other PKU-like conditions?

The timing couldn't have been better, because President Kennedy .. John Kennedy was elected in 1960, the same time as the PKU test became available.  And he, as part of his transition team in Health Education and Welfare, included Dr. Robert Cooke, who had .. who was Chair of Pediatrics at Johns Hopkins and had two retarded .. mentally retarded children of his four children, and had a very profound interest in mental retardation.

He proposed establishing a new institute at the NIH to focus on mental retardation and development in general, to be called the National Institute of Child Health and Human Development, to provide a focus for research on mental retardation.  President Kennedy also appointed a President's Committee on Mental Retardation that still exists today.  

It has just changed its name to President's Committee on Intellectual Disabilities in keeping with the trend of the time, but that still exists.  And that committee made a number of recommendations, including efforts .. making efforts to try to reduce the prevalence of mental retardation by 50 percent by the year 2000.  This was the extent of the optimism that was triggered by the ability to screen for and treat and prevent PKU.

Legislation was also passed as part of President Kennedy's program to construct 12 mental retardation research centers at universities and academic health centers, not at the institutions, to bring this research into the mainstream of biomedical research in the country. 

They also established .. again, part of President Kennedy's legislative program .. university-affiliated facilities for mental retardation in a number of states, and now each state has at least one of these, for research on .. not medical treatment but behavioral treatment, educational treatment, social treatment, for individuals with mental retardation, and for training personnel to work in this field.

Also a part of this was a campaign begun by the Advertising Council spurred by the President's Committee on Mental Retardation, by the National Association for Retarded Children, to make PKU screening universal.  Their slogan was:  PKU tests should be a must for all babies everywhere.  This was done without real evidence of efficacy other than basically anecdotal case reports.  There had not been a combined study really looking at this in a scientific way.

So the Children's Bureau joined with NICHD in funding a study of efficacy.  This was based on a recommendation from a report from the American Academy of Pediatrics in 1965 that said a collaborative study to evaluate management of this disease would be valuable. 

They also pointed out that without any citation that there was evidence of some infants actually being harmed by unnecessary treatment for phenylketonuria when they really didn't have it, because they had some variant of the disorder that was relatively benign, and other infants who had suffered malnutrition because the restriction of phenylalanine from the diet was so severe.  You have to have some.  You can't eliminate it totally.  And some people in the zeal for this treatment eliminated it too much, more than was appropriate, and some babies did suffer malnutrition as a consequence.

So in 1967, this collaborative study was launched of children with PKU, and the data from this study showed clearly that early treatment led to normal growth of these kids, their IQ levels were comparable to those of their unaffected siblings, and that the diet should be maintained at least throughout childhood.

The best predictor of efficacy was the age at which the dietary treatment began, and second was the degree of dietary control that was maintained. 

So with this information, then, more and more states began to pick up and screen.  Massachusetts, New York, Louisiana, and Rhode Island led the way, and many other states joined.  By 1975, 43 states had mandated PKU screening.

We had learned that its prevalence was 1 in 14,000 births.  We had discovered variants, particularly a non-PKU hyperphenylalanine anemia, that did not require treatment.  We had also started to learn about what we knew was going to happen, the problems of false positives and false negatives.  At that time, in the early days, about 95 percent of the babies who screened positive were false positives.  Only 1 in 20 who tested positive actually had the condition, so there was a fairly high ratio.

False negatives, up to 10 percent in the early days were missed.  Most of these misses were people who were not appropriately .. who were not screened in the hospital.  Today our figures are a little better.  The prevalence is 1 in 14,000.  We identify in 14,000 about 10 kids who test positive by screening.  Only one of those has PKU, so we've lowered it from only 1 in 20 being positive to 1 in 10 being positive.

We still have a rare, rare false negative.  These are almost non-existent.  And if there are kids who were missed, they were not generally screened in the hospital.  It's not the lab test that fails; it's the fact that they were born at home or for some other reason didn't get screened in the hospital.

So we've made major efforts to improve these, and that's about where we are at the present time.  You can't totally eliminate your false positives, because of the way these levels are distributed in a population, and what you have to do is draw a line, your cutoff, of babies who are called back for a retesting for a confirmatory test, so that you can not miss kids who should be tested.  And that's a very tricky thing.  It has to be done differently for each disorder we screen for.  But, again, here's the lesson, the model from PKU.

Okay.  Other problems with PKU.  One was dietary management.  We didn't know at the beginning how low we had to keep phenylalanine in order to allow for normal growth but to avoid the adverse consequences of phenylalanine being too high.  Over the years we learned that the level should be about six milligrams per deciliter, and that has been the target for control.

We also didn't know how long to continue the diet.  The studies early on suggested that it probably should be continued through childhood.  Initially, people said age seven and then let up.  Now people believe I think it certainly should be continued throughout childhood. 

Adolescence is tough.  There's an awful lot of phenylalanine in a hamburger, and these kids have to stay on some dietary restriction that's hard to do and have a good bit of their protein intake coming still from Lofenalac in order to avoid still some consequences in their behavior, their intellectual functioning, their general sense of well being, as teenagers, and it looks like this carries over into adulthood as well.

Then, there's the problem of maternal PKU.  This really did take people pretty much by surprise.  What we did was create a population that had never existed before .. that is, a group of women who instead of being severely retarded and basically out of the reproductive pool, were now functioning normally, marrying, and having children.  They had high phenylalanine levels during their pregnancy, and it turned out that the phenylalanine crosses the placenta and is toxic to the development of the brain and other body parts in the fetus.

Even though the fetus does not have phenylketonuria itself, and it is an obligate carrier, but it does not have PKU, but it is adversely affected by the high levels of phenylalanine it is exposed to during the pregnancy, because of the mother's high levels.  And as a consequence, those babies are born microcephalic, many of them have heart disease, they stay small, and they are mentally retarded.

Generally, their mental retardation is not as severe as people with untreated PKU, but it is severe enough to be significantly problematic.  What the data showed, then, was that if each of these women had two children .. and all of them are affected, there was no escape .. we would have as much mental retardation due to PKU in a generation as we had before we started the screening program.  Something had to be done.

Rather than give up screening, we've looked for a way to try to deal with the problem of the high phenylalanine levels in pregnancy.  And the obvious approach was to try to reinstitute the diet during pregnancy, ideally before pregnancy, keep the mother's phenylalanine levels as low as possible, six was the target, and see if this would not reduce the likelihood of hyperphenylalanine anemia in the mom, crossing in the placenta into the baby and affecting its developing brain.

This was tried.  NICHD actually organized a collaborative study that started in 1984, went on for about 15 years, and that study demonstrated very clearly the benefits of this dietary approach.  It also showed how difficult it was.  Many of the women were off the diet before they got pregnant and had instituted it again during early pregnancy.  It was very difficult to do.  Some women just could not do it.

The ones who did the best and had the best outcomes, basically normal babies, were the ones who had instituted the diet before they got pregnant, kept their dietary control around six during the pregnancy, and the babies basically were minimally affected, although that was not 100 percent, and essentially were normal.

The moms who were not able to achieve this still had the problems with babies who were damaged by the phenylalanine exposure.  So NICHD held a consensus conference on this topic after the study was done trying to bring attention of the scientific community and the OB/GYN and pediatric communities to this as a problem .. the fact that there was a solution, it was a hard one, but there was a way to avoid this as an issue.

And today I think there is general knowledge in the OB/GYN and pediatric communities about this as a problem.  What we don't always know is that the women are phenylketonuric, because they appear normal in so many ways, unless the moms tell and unless we remember to ask.

The other thing that happened was the addition of new disorders to the screening.  With the success in PKU and the development of means to detect other conditions, it was pressed for these to be added to the conditions we screened for.  The most successful of these was congenital hypothyroidism.  This has a prevalence about four times as high as that of PKU, about 1,000 babies born in the United States each year with congenital hypothyroidism, compared to about 250 with PKU.

So if the problem with congenital hypothyroidism is that, again, these babies don't look any different than other kids at birth.  They are very hard to pick up, because the mother's thyroid has protected them during pregnancy.  But once they lose that maternal thyroid hormone, they start to develop symptoms. 

Thyroid hormone is necessary for normal growth and brain development.  Without it, the kid's growth slows, their appearance changes, and their intellectual development suffers irreparably.  If we don't pick these kids up by three to four months of age and institute thyroid hormone replacement therapy, they are marked for life. 

They do not improve their mental function.  They may change their physical appearance, but that brain development that is so critical in those early months has not taken place in the absence of thyroid hormone stimulation, and they permanently are mentally retarded.

Dr. Del Fisher, a pediatric endocrinologist at the University of California-San Francisco, in the 1970s said, "I think we can develop a test to screen for congenital hypothyroidism like we screen for PKU."  It's different, it's not genetic.  It's like a birth defect that is sporadic.  The thyroid gland basically doesn't grow.  There is a very small portion of this that is genetic that's based on an enzyme deficiency in making thyroid hormone, but that's a very small subset of the group.  Most of these kids do not have thyroid glands.

If we replace that absent thyroid gland with thyroid hormone, which is very cheap, very easy to do, these kids grow normally, their brain development is normal.  So he set out to develop a screening test using the same filter paper blood spots that we used to screen for PKU and was able to develop a micro-assay for thyroid hormone and thyroid-stimulating hormone, was able to pick up virtually every kid with congenital hypothyroidism.

The thyroid hormone is very low.  The thyroid-stimulating hormone, TSH, is very high, because it's responding to the lack of thyroid hormone and overworking, trying to make something make a hormone that isn't there.  So with the combination of these two assays, basically you've demonstrated you could pick up all the kids.  Dr. Dussault in Quebec demonstrated in a province-wide study that this screening system worked and states rushed to add it to their screening armamentarium.  And very quickly it became mandatory, along with PKU.

But in 1975, a National Research Council report on newborn screening, particularly for PKU, was issued and it was cautionary in several ways.  First, it said, "Do not rush to add too many new tests to screening, until you have systems in place for medical education to know how to deal with these, to counsel parents, to provide information so that misinformation is not there, and kids don't get treated inappropriately."

They also issued what sort of was practiced in the community but became dogma, which was a recommendation that you should not screen for anything that you don't have an effective treatment for.  And that dogma has persisted until just .. until recent years.

I'm going to skip now over the intervening years until today and look at where we are.  Here's where we stand today with newborn screening.  This is data as of June 1st.  You can see that there are nine states that screen for more than 50 disorders.  What has made this growth possible is the development of what we call tandem mass spectroscopy, where we are able to use a technique that looks at proteins, enzymes in the blood, as well as accumulated abnormal products, and detect those in a way that multiple things can be detected at once.  So you basically with tandem mass spec can detect as many as 50 disorders, and the number of those is growing. 

So there are nine states that are screening for 50 or .. for more than 50 disorders.  There are .. and you can see the other numbers right here.  There are only three states that screen for less than 10, so that this system of screening has grown markedly, and a lot of that growth has occurred in the last two or three years with the advent of tandem mass spectroscopy. 

With the report from the American College of Medical Genetics recommending the expanded use, and the adoption and endorsement of that report by the Secretary's Advisory Committee, this has markedly expanded.  So you can see the current picture of newborn screening.

How fast is this?  How quick can we get answers?  I can give you one example.  The State of Iowa screens for 53 disorders.  They pick out or they get samples sent to the state laboratory every day by 5:00 p.m.  All those samples are at the laboratory by midnight.  They work 24 hours a day, seven days a week.  They run those analyses.  The results are back to the hospital by 5:00 p.m. the next day, 24-hour turnaround.  That's very, very, very fast. 

So we are able to pick up these kids by really a day after the heel stick is done.  Not every state is that fast, but the rest are not far behind that kind of a record.

Now, we've learned, as I said, a lot of lessons along the way from PKU screening, from congenital hypothyroidism screening, from galactosemia screening, from screening for many of these other disorders.  And basically, we've learned from this process enough that I believe we can proceed with caution to avoid harm.  But we must proceed to take advantage of the preventive potential of this technique of newborn screening.

And if we apply the lessons we have learned as we have broken this new ground, we didn't do it perfect the first time around.  We learned along the way, but we have learned a lot and there is carryover applicability to virtually each of the other conditions that we would screen for.  And if we apply these lessons we learned as we broke this new ground, we can implement expanded screening responsibly.

Now, how do we proceed?  There are a number of guidelines here that I'm just going to run through very quickly, and then I'm going to spend some time on each of them in order.  First, I think we need to make the programs comparable across the states.  We need to improve the technology that we use in screening, emphasizing DNA-based approaches.

We need to aggressively pursue development and testing of therapies for disorders that currently we could screen for, but we don't have effective treatments for them.  We need to evaluate the tests to minimize the false positives and the negatives, and have a confirmatory test system in place that can operate very quickly to evaluate kids who screen positive and have a good system of parental counseling, especially for those with the false positives, so they don't continue to think "something is wrong with my kid."

We also need for the treatable conditions to have a science-based service delivery and a followup system in place, so that these children do get appropriate treatment, and expand our screening to include disorders that don't yet have proven preventive therapy available, and also provide a registry of these patients with parental approval, so that they will be available for future research participation if the parents agree to do that, because only in this way can we develop, test, and evaluate effective treatments for the conditions we don't have treatments for now.

Let's take a look at each of these now in order.  First, making newborn screening programs comparable across the states.  These programs vary from state to state.  Remember the map.  These are state programs; they're not Federal Government programs.  The Federal Government has provided assistance to the states through the Maternal and Child Health Bureau grants to develop and implement these screening programs.

But the government has not dictated what you screen for.  The states choose that.  They choose it in a variety of ways.  Sometimes the legislature itself dictates which conditions will be screened for.  Sometimes they set up a commission.  Sometimes they entrust the Health Department to do it.  But each state winds up with essentially a different group of conditions for which they screen, and, again, remember the map.

They decide this in a variety of ways.  It can be either based on legislation, on a payment system, what they'll pay for, what the costs of the individual tests may be.  Some are more expensive than others, and essentially you're paying for each test separately, except for tandem mass spec, and a variation in prevalence of the conditions by state.

North Dakota or Montana may not have very many people who would be even at risk for sickle cell disease, and they may not want to spend whatever additional it costs to screen for that condition.

So this is really what got me involved, I will confess to you, in newborn screening, was this problem of variability across the states.  And this is .. I got involved with this, I will tell you, by a letter that was referred to me by Hillary Clinton when she was First Lady. 

She had gotten a letter from a constituent.  She passed it on to me to respond to.  I'm going to read to you a piece of this letter, because it had such a profound effect on me, making me determined to try to do something about this if I ever got the chance, that I think you ought to .. because it sets the case forward so well.  And I'm going to just take a minute to read this.

The letter was from the mother of an infant with a very rare condition called glutaric acidosis.  The cause of that condition was first described in 1985 by Steve Goodman at the NICHD-supported University of Colorado, Mental Retardation Research Center.  It's a genetic disorder that causes an enzyme defect.

The affected children appear normal at birth with no suggestion of abnormality until enough brain injury occurs at around one year of age to cause poor muscle tone and movement problems.  Treatment by dietary restriction of lysine and tryptophan is at least partially successful if instituted before the brain injury occurs, again like PKU, but is useless once symptoms appear.

This mother's poignant story illustrates the problem of a policy of screening for a disease that is based on cost and rarity.  Here's her letter.  "There's a two-year old boy with golden curls, bright blue eyes, and an incredible will to walk and run, talk and play.  His story, however, is one of constant struggle and heartache.  He was born prematurely.  Although small for his age, he did all of the things that normal babies do. 

"On Christmas Eve, one month before his first birthday, the real tragedy began.  With the entire family at home ready to celebrate all the joys of the season, he was sick.  Held close in his father's arms, his limbs were limp.  His golden curls dangled as his head fell back.

"No one could possibly imagine the diagnosis.  In the local hospital he was subjected to a variety of tests.  The doctors were baffled; all tests were negative.  But his condition remained the same, even worse.  His motor skills had regressed to those of an infant.  After months of prodding and poking, he was diagnosed with a rare genetic disease .. glutaric acidosis Type I, a genetic disorder that could have been detected at birth with a $17 routine test.

"Instead, it is a disease that results in brain deterioration.  There are approximately only 70 cases known in the United States, less than double that figure worldwide.  Each story is equally heartbreaking.  The statistics are grim, and research minimal, due to the small number of children affected."

That's the end of the letter.  This letter goes on to talk about a foundation that parents set up to support research and education and provide support for other parents.  Resolving this screening dilemma is a major problem we face when we have a preventable disease that we can screen for, but the rarity and the expense mitigate against doing so.  So that's the situation we face. 

Equity across states, I believe, is an ethical imperative.  Mandating it is difficult.  Another approach would be to try to develop a test that tests for everything at once, with one cost, no difference between how many tests you test for.  If you test for five, it's the same cost as if you test for 50 or 100 or 200.  And that test could then be so attractive that every state would want to adopt it.

For a long time such an idea was only a dream, but with new genetic technology it now becomes possible to think about doing that and try to develop a single, unified test system.  Tandem mass spec comes close, but it is limited in the number of disorders it can screen for.  It is not a gene-based test.  It's not a DNA-based test.  It's a product .. a gene product-based test.  And so we're still trying to find this single DNA-based test. 

So the current system basically is a different test for everything you want to screen for.  Congenital hypothyroidism is different from the bacterial assay of PKU, which is different from the gene assay for something else, or a biochemical assay, or an endocrine assay, or whatever.

Tandem mass spec has changed this a lot, but it still doesn't go as far as we need it to go.  And so we're looking at potential DNA-based systems.  If we could have this, we could screen for basically anything we have the gene for.  We could screen for all of the genetic metabolic diseases that have mental retardation or neural degeneration associated with them. 

We could screen for all the immunodeficiencies, all the hemoglobinopathies like sickle cell disease or thalycemia, all the coagulopathies like hemophilia, all the muscular dystrophies, Dushen's, spinal muscular atrophy, for cystic fibrosis, for hereditary deafness syndromes, and others as well.  The numbers go into the hundreds.

And each time we discover a new gene or a new abnormality of a gene the number of conditions would go up.  The only exception that couldn't be done this way is congenital hypothyroidism, which as I said is not a genetic disorder.  We would still have to screen for that separately.

So NICHD almost a year ago put out a solicitation for contract proposals to develop and test new approaches to newborn screening focusing on DNA-based systems.  Now, we have received applications for those.  Those are in the review process and will be negotiated and hopefully funded early in the next fiscal year.

We are optimistic that these systems will work.  Now, there's two basic systems.  Excuse me for a minute.  I want to pick up one of these things.  This is a microchip, DNA microassay, microarray.  This one is for the mouse.  This chip has 45,000 genes on it, basically a whole genome of a mouse.  That's how fast this technology has come along.

Basically, there's a separate little dot, a well for each gene, and basically what we can do is digest the DNA, break up the DNA of the genome of a mouse or of an infant, newborn, and expose it to .. with a special technique to this chip, and any abnormalities that are detected will be .. can be detected by a color change of the little dot on this chip that's read by a computer.  I'm going to pass this around so you can see it.

Potentially, we could screen every newborn with this kind of a system.  If we put on a chip, just not 30,000 genes of the whole human genome, but the genes that we are interested in screening for, for conditions that we're interested in screening for, either the normal gene to make sure that what we see in the infant is normal or the abnormalities, the genetic .. specific gene abnormalities of the conditions. 

That gets a little complicated.  Some of these conditions only have one or two known gene abnormalities.  Some have several hundred, like cystic fibrosis.  So the technology for doing this still needs to be developed, but we're determined to proceed with trying to do this.

There's a second technology that has come along and is in early testing phases.  It's called Luminex microbeads, and it replaces this gene chip with a microbead system.  Each of the beads has annealed onto it oligonucleotides, small fragments of a gene of particular interest.  Here we can make a bead for a given gene of interest, and, again, expose the chopped up DNA of the infant to the beads in this case rather than to the microarray chip, and, again, with lasers and computers identify which .. whether there are any abnormalities of interest.

We don't, again, have to put the whole genome on it.  We put on genes of interest, ones that are associated with the abnormalities that we want to screen for.

This is another technology that's coming along.  It has some potential advantages over the microarray chip.  It's simpler to use, it's easier to change, it might even be cheaper.  So these are things that are coming along and that we are investing in, trying to develop an enhanced capability to screen, and to have a test that is so attractive, so simple, and not too expensive, so that every state will want to use this in their screening program, and no longer will there be this state-to-state variability, so that what you get screened for depends on the state in which you're born.

Imagine being the physician who had to tell a mom that .. who has an injured infant that, "If you had been born in the neighboring state, your child would have been screened for this disorder.  It would have been detected, we could have initiated treatment, and your child would be normal."  That's what we've got to get away from.  That's why we've got to push the development of this technology.

Okay.  So the next thing, if we're going to develop capabilities for screening like this, we also need to develop improved approaches to treatment. And what we need to do here is remember that part of the justification for screening for these disorders is to identify a population that is presymptomatic, so that they don't have damage before we have an opportunity to try therapies.

Once they've developed these .. the brain deterioration or lost the brain development, it's too late for any known potential therapy.  And so they have to be picked up as newborns.  Only newborn screening can do that.

So what screening has the potential to do is to make available populations of patients with these rare disorders .. there's not many of them .. for study of new treatments as we get ideas for these approaches and be able to use to .. approach the families of these children before they get symptomatic ideally, and roll them .. with their permission .. in trials of new therapy developments.

In companionship with our solicitation for new screening technology development by contracts, NICHD also put out a program announcement to the scientific community asking them to submit their ideas for new therapeutic approaches to conditions that we could screen for but don't have effective treatments for at the present time.

We have already received a number of proposals.  Some of these look very good and are all set to get funded.  This is an ongoing solicitation that runs over three years, and I hope that we will get many projects, many ideas from investigators for therapeutic interventions.

We also by doing this will have an opportunity to enter patients in a registry with the consent of parents, and I'm going to spend a little more detail on that in a minute.

We also need to evaluate tests to minimize false positives and negatives as they are entered into the screening system.  Clearly, as I indicated before, there's a balance in where you draw your cutoffs, particularly so that you don't have too many false positives, but you don't miss patients at the same time.  And this is tricky.

There's also some variability from lab to lab, and there's needs to try and make practices in labs as comparable as possible.  Again, with these new techniques this lab-to-lab variability should basically be eliminated.

There's concerns of the costs and the parental anxiety with these .. particularly by the false positives.  Each false positive requires a subsequent visit and a workup and evaluation.  They cost money.  They also worry parents.  You get a phone call, "Your child screened positive.  We need to do a followup.  It may not be anything, but we need to check."  And immediately the anxiety meter goes off the scale.  So we need to minimize those false positives to the extent we can to avoid both the cost and the anxiety. 

The Secretary's Advisory Committee on Inheritable Disorders and Genetic Diseases in Newborns and Children is evaluating each of these proposed new tests before recommending its addition to a treatment regimen, taking into account this issue of false positives and negatives.

There's also an ongoing monitoring and standardization needed, that the system of regional collaboratives .. there are seven .. each state is part of one of seven regional collaboratives network, and it's headed by a national coordinating center.  And this gives us the opportunity for monitoring and standardization across the state sites.

We have to have in place a confirmatory test system with parental counseling.  Again, the problem with the false positives, we have to have .. these are just part of a screening.  Remember, this is screening.  This is not diagnosis. 

The screening must be followed by diagnosis, often a second screen.  If the second screen is negative, then we just can kind of relax.  If the second screen is also positive, then we go on to a confirmatory test and make sure whether .. that the person in fact has the disorder and what the nature of it is and can institute therapy.

Again, we have the system of the regional collaboratives and the national coordinating center in place to make sure that patients who screen positive get referred for this confirmatory testing, and this has to be done very quickly in order to initiate therapy when it's needed, and the results have to get confirmed by experts, a system of referrals has to be in place, so that we get the best information for these, and this information has to be passed on to the families.

The treatment initiation has to be done.  There has to be a system in place, so that these kids don't fall through the cracks, so that there is an expert system of care for those who test positive and have a disorder that we can treat.  And, again, the regional collaboratives follow .. provide this, and we need to learn as we go, so we must have a system in place with these regional collaboratives as they begin treatment, particularly for disorders where we don't know what standard treatment should be, or we're testing new ones, to follow these, to gather data, so that we can evaluate these new treatment approaches.

Finally, we need to expand screening to diagnose .. include disorders without definitive treatment.  Now, this goes against the dogma, and this is probably the biggest challenge that we have.  Why do we want to do this?  Again, as I mentioned before, even though the standard belief has been don't screen for something you can't treat, we now have confronting us not just the definitive preventive or therapeutic treatment, but other ancillary treatments for these kids as well.

There are some things that you can do for these children that don't necessarily mean a curative treatment, but they improve their quality of life and their ability of parents to care for them, and their ability to get programs and services in the community.  So there is benefit to the child from having this diagnosis, even if there is not definitive therapy.

In addition to that, the only way we are going to develop therapies for these conditions is to be able to diagnose them before symptoms develop.  And the only way to do that is through newborn screening.  The requirements here I believe should be .. include consent.  Most of the state programs for newborn screening for conditions now do not have a consent.  A few states do, but they are part of standard treatment, and their parents do generally not sign a form to consent for these.

If you're going to screen for conditions that don't have a treatment, some parents might like not to know about that, and they should .. this is being done in part as research, and parents should have the opportunity to be informed that this is being done and have the opportunity to opt out if they wish to of this part of the screening, not for the mandatory parts where we have clear benefit to the child.  But if there's not evidence of clear benefit, this, again, is .. this is my belief, they should be asked for their consent to this screening.

There should also be counseling for the parents, and there should be supportive treatment available for those who have conditions that are diagnosed that we don't have effective treatment for.  I also believe that there should be an opportunity offered to the parents to have their child listed in a registry that would be maintained either by the regional collaboratives or by the Centers for Disease Control or some other possibility.

This, again, would be offered to the parents, an opportunity to list their child on this registry by disease and disorder with contact information for the parents, agreeing to be contacted by investigators if someone comes up with a potential trial for a therapy that .. for these disorders.

What drives this is that many of these disorders are so rare nobody can get enough of a population at their own academic health center to test.  And it requires not just a collaboration of 10 or 20 academic health centers, but the whole national network of screening to get a sufficient number of children presymptomatically to do the studies once they're ready for human testing of these therapeutic interventions.

So this would be kept confidential, and parents would just agree to be contacted, not necessarily to participate, but to weigh the study and decide on each study's basis whether they want their child to participate in this or not.  And the outcomes of these studies would also be maintained by the registry, so that we gain knowledge of what the effects of these interventions are.

Okay.  That brings us to my final slide, and this is what I call the legislative and ethical imperative.  And this is perhaps the key to the whole success of this operation of newborn screening.  Basically, it relates to a belief that nobody's DNA information should be used to discriminate against them in employment or insurance. 

And unless we assure by national legislation that such discrimination won't happen I think parents are going to be reluctant to have their newborns screened in a public program, particularly for conditions if we don't have a treatment for them.  What they will worry about is that this will disqualify their children for insurance, it will disqualify them later as adults for employment if they have a condition that, you know, may not develop until later on, and this proceeding with caution .. as we've talked about from the beginning, "proceed with caution" includes providing this protection as well as all of the medical and the laboratory procedures identified.

I think that if we have learned enough from the experience that we've had starting with PKU, and involving numerous other disorders as well, that if we apply what we've learned and we can have the necessary funding support and continue to learn as we go, that newborn screening can take its full place among the most significant and effective public health measures of all time.

Thank you for your attention.  This has been a long talk.  I'll be glad to answer any questions.

(Applause.)

CHAIRMAN PELLEGRINO:  Thank you very much, Dr. Alexander, for a very enlightening overview of a very complex problem.  We appreciate it very, very much.

I'm going to ask now a Council Member, Dr. Floyd Bloom, to open the discussion.  Would you, Dr. Bloom?

DR. BLOOM:   I enjoyed reading your paper and hearing your talk to spell out the details of your paper, but it brings up the fact that, while I don't think that anyone could argue with the concept that if we could prevent mental retardation due to a post-natal screening procedure that we'd want to do so, obvious as that is. 

The question is .. you've named the points .. the cutoffs, the need for evolving technology to improve the breadth and sensitivity and accuracy of the tests.  And when are we ready to implement such tests?  As you say, the Luminex technology looks as though it will displace DNA chip technology.

You raise a very radical concept of screening for things even if we can't treat them, because unless we can identify who has them we'll never be able to develop treatments for them.  I think those are all very laudable, but I foresee great difficulties in getting states, particularly poor states whose health budgets are already compromised, to invest in such testing, particularly with such rare disorders.

Maybe just to get your discussion aspects going, I'll take you to one aspect of your paper with Dr. Van Dyck in which you said perhaps the biggest set of discussions will involve deciding what not to include in the set of tests, and there you list Huntington's Disease as something not to include.  But it seems to me that fulfills all of the criteria by which you listed the tests that you want to include, even though we can't treat them. 

And Nancy Wexler has many poignant stories of people who grew up thinking they had the gene and lived their lives as though they were going to die and then found out they didn't, whose lives were then wasted, and people who grew up not knowing creating additional children with the genes and thereby perpetuating the problem. 

And one could contrast that with the prenatal testing for Gaucher's Disease where the community involved and vulnerable to the disease decided to have all their children tested, so that that recessive disease could be identified and to avoid matings among people in whom the recessive pairings would perpetuate that disease.  And that has been largely eliminated.

So my question is:  why did you put Huntington's Disease on the question of not to be tested for, and why do you not also focus intensively on prenatal testing as well as neonatal screening?

DR. ALEXANDER:  Okay.  Good and difficult questions.  Huntington's was thrown out as one example for discussion for consideration, and probably the one where there has been much .. most discussion and controversy about screening.  I think where people have come own on that is allowing individuals to decide when they have capabilities for decisionmaking, whether they want to be screened for that or not.

The consequences of that disease are so devastating, but they fail to appear until relatively late in life, after child-bearing is over, that parents knowing about that particular one where there really is nothing that we have to offer at the present time may well treat that child differently.  If that child is told what they have, and they may well not be, knowing .. just knowing that information may be extremely difficult in terms of the life choices it presents, and so forth.

I did not put that down as an absolute, just as an example of something that might be considered, maybe at the top of the list, as a prototype for conditions that you might exclude from this. 

So I think you're quite right in raising the question, and responding exactly as I hoped people would, and thinking about, are there conditions that we might not want to screen for, that we can't treat for, can't treat, and just use Huntington's as probably the most familiar example of something that might be in that category?

As far as, you know, the state decisions, and so forth, part of the drive here is to develop a technology that can screen for with basically one test instead of multiple tests, at one cost, many conditions, so that you don't have to look at the individual cost of adding/not adding each test.  And that's what we would hope the microarray or the beads might provide.

As for the issue of screening for conditions that we can't treat, again, there are benefits to and arguments for this above and beyond the issue of making available a population to test presymptomatically, to try interventions.  There are benefits to the child and the family, the benefits of knowing what a diagnosis is, the benefits of avoiding what's often called the diagnostic odyssey, when the child does start to develop some symptoms, as you well know .. the search that the parents undergo to try to come up with a diagnosis.

Many of these are such rare diseases that physicians have not seen or even heard of them, and that odyssey may go on for two or three years before they get to a point where the diagnosis is finally made.  During those three years, that child has been poked, prodded, radiated, everything you can imagine.  And having that diagnosis made as a newborn would avoid that diagnostic odyssey and what is done to the child in the course of such a diagnostic odyssey.

It also enables the parents to make decisions .. decisions about future child-bearing, decisions about care and treatment for that child, decisions as practical as if you have a child with muscular dystrophy, should I buy a one-story house or a two-story house?  Very practical things like that.

And you hear stories about families whose kids were not diagnosed with muscular dystrophy until they were three or four, and in the meantime they bought a two- or three-story house that they would not have done if they knew they were going to have a child with this kind of disability getting around.  They would have bought a one-story house.

So decisionmaking on practical things like that are added to benefits I believe of being able to give parents a diagnosis before the child shows any symptoms, and even for conditions that you don't have treatment for.  I have not done a large sample, but I have talked to a lot of parents of kids with a variety of conditions and asked them, many having completed the diagnostic odyssey, "Would you rather have had this diagnosis at one or two weeks of age, or gone through what you have gone?"  And to a person, there hasn't even been one who hesitated with an answer, "I would rather have known about it right from the beginning."

Now, that's not a big sample, but it's a diverse sample, and it's without even any hesitancy or any divergence of opinion.  So I think that parents, offered the choice, are likely to opt for having this screening available as to their newborn infant, even if we don't have a specific preventive, curative treatment to offer.

CHAIRMAN PELLEGRINO:  Dr. Rowley?

DR. ROWLEY:  Well, I think it's very common.  It has to do with the Guthrie blots.  And just as an aside, it's interesting that some states, and particularly countries in Europe, save the blots for quite a long time and they were used quite imaginatively by Mel Greaves in London to show that children who developed leukemia, sometimes even as adolescents, but initially he used them for children who developed leukemia in the first few years of life to show that they had leukemic cells circulating at the time of their birth, namely at the time that Guthrie's file was taken.  And for those of us in the arcane field of leukemia, and the etiology of leukemia, that came as a surprise. 

So that's .. unfortunately, in Illinois, because of financial reasons, they store blots only for about three months, so they're not useful.  Some states do store them longer.

Let me now get to my questions, and there are two.  It was my impression as a member of the American Society for Medical Genetics that at least the Senate had considered the question of making genetic information not available to insurance, or at least disallowing insurance discrimination based on genetic information, and that would go to one of the points or concerns that you raised.  And I could be incorrect, because this was just coming for a vote rather than having been for a vote.

And I know that isometrics chips at least at the University of Chicago cost a great deal of money, and so I'm interested about cost.  And can you give more insight as to why, when the Federal Government is willing to intervene on many different arenas, why the tests aren't either federally mandated or uniform across states?

Thank you.  And I should say I enjoyed your talk very, very much.

DR. ALEXANDER:  Thank you.  Excellent questions.  First, let me say something about the leukemia issue.  Clearly, the studies have shown this.  The problem is that they also find leukemic cells in individuals who never developed leukemia, and so they are not 100 percent predictive.  So you're not quite sure what to do with them.  And until we try to figure that out, I don't think we're ready to, you know, start screening and treating individuals with leukemia. 

We still have to figure out what the natural history of that phenomenon is.  But it's a very interesting observation, and it's something that could in fact lead to earlier treatment, if we can separate out those kids who are destined to go on and develop full-blown leukemia from those who for some reason lose these cells.

The Senate vote .. a year ago the Senate voted 99 to nothing to support this legislation to ban discrimination based on genetic discrimination in insurance or employability.  And it was just totally overwhelming .. 99 to nothing.  Legislation is now before the House, and the House has not taken action on it.  The reasons for that are not totally clear.

You would presume that there is objection to it from insurers, perhaps from employers, but, again, I don't have the information to really comment on that, other than to say the effort now must be with the House.  And we really need to push that agenda, if we are, I believe, going to be able to take full advantage of what the potential is for newborn screening into public health measure.

You mentioned the cost of the chips.  This chip, if you try to .. if you buy them individually it's about $400.  If you buy them in bulk it's about $200.  Now, there's not four million of these sold and run a year.  What we are hoping and anticipating in our discussions with the companies that make these and the beads is that by bulk purchase and bulk use that cost would come down dramatically.  And what we have been talking about is trying to get these under $100.

Many of the states now that screen for this many disorders spend $100 a kid or more.  And if you could substitute this for many of these other tests, you would not be money behind.  You might come out ahead.  And, in fact, you could screen for so many more disorders with these that you might come out way ahead.

So what we're hoping is that if you have a purchase of 4.1 million of these a year that that cost will come down significantly.  Now, that's just the cost of the chip.  That's not the cost of running the assays and the personnel and the system, and so forth.

DR. ROWLEY:  Or the oligonucleotides that are used.

DR. ALEXANDER:  Yes.  So, but I think with the large numbers of these that we can get .. we should be able to get the cost down.  That's the goal.

Finally, about a federal mandate, one of the things that the Federal Government has generally tried to stay away from is directing health care policies in states.  And one of the things that they definitely have stayed away from is dictating what conditions states screen for.

States pretty much jealously guard their prerogatives in the medical arena, and the Federal Government has, to a large extent, not always, stayed away from dictating these kinds of policy.  So I don't see much likelihood of that changing.  That's why I think the carrot, rather than the stick approach, is the one that's likely to be the most successful.  And that carrot is a chip or a bead.

CHAIRMAN PELLEGRINO:  Dr. Meilaender?

DR. MEILAENDER:  This is just a small question that follows up, really, on something Janet just asked about.  But with respect to the worry that parents would not want to have children tested for non-treatable .. currently non-treatable disorders because of the possibility of not being able to get health insurance or something like that, and your answer to that is the need for legislation to prohibit that.

I'm just wondering .. and I don't know the answer to this exactly, but with .. and we don't know what the future direction of health insurance will be in this country, but with the proliferation of consumer-driven plans and cafeteria policies, and so forth, aren't there plenty of ingenious ways to get a select population that don't involve anything that would be called discrimination?  I mean, I'm just wondering if .. it's not so easy to get a solution to that problem, it seems to me, and I just wondered what you'd say about that.

DR. ALEXANDER:  Well, it was in the Senate.  You're right.  This is a changing field, a changing situation, and there are opportunities for things like you're talking about.  But they're not universal; people have to know about them, have to search for them, need to know what they have to look for something like this.

It won't take too many cases, if we start down this path, of families being denied insurance for their child who tested positive for something we can't treat to get publicized and scare people away very quickly.  That's what worries me.

And I think it is imminently solvable.  We have it half-solved, and we need to go the rest of the way.  That's my personal view.

CHAIRMAN PELLEGRINO:  Dr. Kass?

DR. KASS:  Thank you very much, and thank you for the very illuminating and clear presentation.

The Council has been given, and no doubt you have seen, this commentary from the May 5th .. May issue of Pediatrics by Dr. Botkin and colleagues, "Newborn Screening Technology:  Proceed with Caution."  And ..

DR. ALEXANDER:  You may have recognized some of those words in my talk.

DR. KASS:  No.  I recognized "proceed with caution," but they seem to be much less ready than you are to proceed along the lines you've suggested, and the particular offer for recommendations that ought to take place before, in fact, one institutes a plan for which you are so enthusiastic.  And I wonder what comments you might offer in response to this critique.

I mean, I could raise some of the issues, but let's .. Council has all read this, and I would just .. these are all people in the field.  I mean, these are not just outlying ethicists who ..

(Laughter.)

DR. ALEXANDER:  This field has been one of widely divergent opinions in some of these issues from the beginning.  PKU screening was characterized by many widely differing opinions about whether it should be done at all, whether it should be mandatory, etcetera.

Sam Besseman made a life work of critiquing PKU screening, and making many claims, some of which were true, many of which had no foundation in truth at all, about PKU screening.  It has persisted in spite of that.

I think that the concerns and the issues that are raised by Botkin and his colleagues are concerns that many people have thought about but come down differently in terms of our readiness for moving ahead.  I think that numbers of problems that have occurred and infants that have been affected by adverse consequences in the past are not documented, and probably are vastly exaggerated. 

If you really try to go to the literature, the courts, or anywhere, finding these is .. the numbers are very, very small, and we have learned from those frank mistakes how to try to avoid them, and things that we should avoid, not just with PKU but with other disorders as we move into those.

My contention is that we know where the minefields are, we know where the problems are, we know enough that we can proceed with caution if we put into place the things that I listed.  And those are pretty much not very different from Botkin's.  Where we differ I think is in the availability of this system to carry the stuff out.

And we need .. we have the beginnings of it.  We have the foundations for it.  Every state screens .. everybody.  Every state has a list of conditions they screen for.  Every state has a system of standards.  Every state is a member of one of the regional collaboratives .. the seven.  There is a national coordinating center for those regional collaboratives. 

We hope to build on that foundation to construct the things that I have listed here, so that we have in place that whole system, starting with obtaining the blood spots, doing the analyses, confirming the analyses, doing the counseling with the parents, initiating the treatment, following the treatment, that can make this go right.  That has not always been present in the past, and that has accounted for some of our failures and our problems.

I think that we are .. we now have the foundation to build on, and what we have to make is the commitment to do that building, to proceed with caution.  I don't disagree with that at all.  I fully agree.  I think we must proceed with caution.  We'll get in trouble if we don't, and I think we're only going to get one shot at this kind of expanded screening program, and we have to do it right.

And I think that the field knows what they have to do, there's a commitment to do it, and wise people who are in positions to try to make sure that that happens, and critics out there who are going to point out the problems as they develop, so that we can fix them.

CHAIRMAN PELLEGRINO:  Yes, go ahead, Leon.

DR. KASS:  May I just follow up?  Let me land on just one of the questions that I at least would have before one moved to any kind of mandatory screening.  And I grant you all the states have them, so this is not something new.  But the yield is very small.  The number of false positives .. the ratio of false positives to actual cases is very high.  I mean, if you looked at the total of the data that's pooled, 95 percent of the initial screens are false positives over all of the different diseases.

As you can say, you know, one should reassure those parents whose subsequent tests reveal that their child doesn't have a disease, but we all know that simply telling somebody once doesn't sort of settle the anxieties.  And the question is:  are there built-in programs before you make this thing mandatory and not requiring parental consent that one do the careful research to find out what the consequences are simply of having identified somebody as a positive incorrectly on the first screen?

Now, maybe we don't do enough of that already, but we are on the threshold, it seems to me, of screening not just for the 29 things here but with the new DNA arrays for potentially thousands of things, many of which will not have treatment.  So there's a kind of complication, and the question is whether one has built in enough research to make sure that one is not in those areas doing harm as opposed to the harm done by giving someone a treatment which is itself harmful. 

I mean, that wasn't well put, Duane, but perhaps you could address that.

DR. ALEXANDER:  Yes, that's a very good point, an extremely important consideration.  The issue of false positives has haunted this field from the very beginning, and major efforts have been directed to get that number down as low as possible, with PKU screenings now down to about 1 in 10, a little less.  But you still have nine families that you are going to subject to the anxiety at least of a repeat test.

The field is very aware of this.  One of the major topics of discussion in the Secretary's Advisory Committee is the issue of false positives, of laboratory standards, and of trying to draw the boundaries as precisely as possible.  It's not easy, and you're always going to have some number, some proportion of false positives.  You're going to have more false positives than true positives, and you just try to keep that number in balance as low as possible.

What we do need to learn .. and I think you're absolutely right here .. is better ways as we counsel families, as we deal with families, of conveying the information about false positives, what they mean, and then, when we go back and do the confirmatory tests and it's normal, your child has nothing.  This was a fluke, this was a temporary phenomenon, we don't have an explanation for it, but your child does not have this disorder.  And you can go home and not worry about it, treat them as normally.  There's no risk of this condition.

We need to be .. to learn how to do that better and more effectively than we're doing in the past, because the studies that have been done have in fact shown that even some years later parents still have some concern.  "Well, they said he was okay, but I'm not sure he really is." 

And that's a valid concern, and we do need to study that better and get better at our counseling techniques of conveying that information and that assurance that if you have been retested and the confirmatory test is negative, and even the screening test is now negative, you can stop worrying.

CHAIRMAN PELLEGRINO:  Professor Schneider?

PROFESSOR SCHNEIDER:   I just wanted to say a word about the confidentiality issue, which showed up in this conversation and which showed up in the paper we're going to be discussing later.  I'm aware that part of the confidentiality problem is not a problem about actual leaks of information or misuse of information.  It's apprehension about misuse of information.

I think for that reason it's especially important that bodies like this one and people who are supposed to be knowledgeable about the kinds of problems people are going to encounter speak accurately about what the actual behavior of insurance companies is. 

And there is a small literature that suggests that insurance companies in fact, in reality, do not discriminate on this basis, and that they have no good economic reasons for discriminating on this basis, partly because an awful lot of genetic risks never materialize, and it just .. and the numbers are so small that it isn't worth the insurance companies trying to take these things into account, partly of course a lot of .. an awful lot of American insurance is insurance acquired through employers and the companies essentially don't have an opportunity to discriminate on that basis.

In any event, since there is this empirical information suggesting that at least now insurance companies are not discriminating and have reasons not to discriminate, partly of course market reasons not wanting to be branded as companies that discriminate on this basis, I think it would be a good thing if we were able to tell people that even if Congress has not acted that no problems seem currently to be materializing. 

I'm also apprehensive about the idea of asking Congress to legislate to get rid of a problem that does not exist, because when Congress and the Department of Health and Human Services have tried to act to protect confidentiality, as through HIPAA, they have often done so in ways that have been damaging and clumsy and extremely burdensome without being remotely beneficial.  So I urge us to keep this empirical information in mind.  It may not be accurate, but it's the best information I've been able to locate.

CHAIRMAN PELLEGRINO:  I have ..

DR. ROWLEY:  On this point ..

CHAIRMAN PELLEGRINO:   I'm sorry.  On this point?

DR. ROWLEY:  On this point, concern for discrimination, while it's true that these newborn screening disorders have been on aspect of it, the other is the concern of patients who have genetic changes predisposing to cancer.  And this is particularly true of women with BRC-1 mutations, and there are many women who have a strong family history of breast cancer in young parents or other young female members of their family who have refused to have BRC-1 mutation analysis because of this.

Now, you're saying, as I gather the gist of your remarks, that they are either misinformed, that there is no discrimination, or they are concerned about a problem that doesn't exist, because that's another whole group that has refused diagnostic DNA analysis for fear of discrimination.

PROFESSOR SCHNEIDER:   Yes, that's quite right, and that was one of the reasons that the research that I'm describing was done, and it's quite clear that genetic counselors, for example, themselves often believe that any such testing ought to be done outside of your insurance program, so the insurance company won't find out.

But the fact that it is widely believed that insurance companies are going to discriminate on this basis doesn't make it true, and what alarms me is the extent to which it is apparently universally believed that this is a real problem currently going on at a serious level when there doesn't seem to be actually any information about it.  So I'd like to be able to calm people's apprehensions.

CHAIRMAN PELLEGRINO:  We're getting close to the end of the time for this session, so can I ask Dr. Carson and Dr. Hurlbut, who have asked to speak, to pose their questions, and Dr. Alexander to hold until they've both presented, and then you respond to both.  Thank you.

DR. McHUGH:  Am I not in the queue?  Can I join the three of them?

CHAIRMAN PELLEGRINO:  Yes, by all means, Paul.  I didn't see you.  Thank you.

DR. CARSON:  Okay.  Very interesting and well presented dialogue.  This is obviously a rapidly-moving train as we are able to diagnose more and more things.  And as you're working with various others to try to create, you know, legislation that will protect children in the long run, is there any interest on your behalf of looking at prenatal genetic diagnosis?

Because it seems to me like if we go through a great deal of trouble dealing with just post-natal, and then we .. while the prenatal diagnosis is moving along very rapidly, particularly with more and more older women involved in in vitro fertilization, and the possibilities of multiple types of screenings there, shouldn't there be some effort to try to couple these things, so that we don't have to go through the whole thing all over again?

CHAIRMAN PELLEGRINO:  Could you hold?  Dr. Hurlbut, and then Dr. McHugh.

DR. HURLBUT:   Well, my question/comment relates to that as well.  Obviously, it's a very exciting time, because we're getting at the molecular foundations of disease, and yet at the same time there is a troubling reductionistic foundation for .. a reductionistic kind of paradigm that is operating here, and that raises some questions I want to probe with you a little bit.

You spoke of the false positives and negatives.  Obviously, in some cases you're speaking not just of a missed detection of a presence of an indicator, but a lack of correlation between a real indicator and a phenotypic outcome.  In other words, you can have the presence of a gene, but balanced with other genes it will never express a disease.  Right?  That's a fair statement at this point?

DR. ALEXANDER:  In some rare instances, that could be the case.

DR. HURLBUT:   I mean, if we start with a one-to-one correlation between genotype and phenotype, that's one possibility, but there are many other possibilities .. variable penetrants and various even lack of manifestations of a phenotype.

DR. ALEXANDER:  Right.

DR. HURLBUT:   I mean, identical twins, for example, are largely thought to be on the popular level identical, when in fact their concordance of gene expression, at least at one study done at Stanford, was only 18 percent relative to fraternal twins, 18 percent higher. 

So the point is that what we're really looking at is not one-to-one correlations for many things.  Some traits will be one to one, but many will be just statistical probabilities.  And, therefore, you finally have to come down to what the gene tests really mean, and that means you have to do even further tests and maybe further tests of the further tests.  Isn't this fair to say?

And that finally you are going to enter into that strange zone of, what's the phenotype really?  The secondary question of that is:  wouldn't this logically translate out .. and here I'm not objecting, I'm just raising these concerns and questions.  Wouldn't this logically play out ultimately to much more complicated phenotypes, not just obvious genetic diseases, but finally even behavioral phenotypes?  And wouldn't this then carry us into the realm of racial implications and backwards into prenatal diagnoses that would also relate to potential treatments in the womb? 

And I'm just raising this because I think there are lots of issues here that we haven't yet introduced.  And a troubling implication I think we all ought to face is the idea that they're talking now about the $1,000 genome, where every gene is tested.  That would obviously, is you were doing a purely scientific approach to it, logically be correlated with phenotypes all the way through the history of the individual's life, not just at birth.  Maybe mandatory testing at two, five, and 10 years, and 22 years, and so forth. 

And a huge data bank of the relationship between genotype and phenotype, and then backwards into prenatal diagnosis, maybe even pre-implantation diagnosis.  I mean, you can see this opening up into a hugely positive but hugely dangerous thing.

And let me just close with one comment here.  There was a public .. it wasn't .. it was a public forum at a major hospital in the United States recently in which somebody stood up and said to the physicians, "Well, what are you doing in this university about Down's Syndrome?"  And a physician told me that the physician who answered this question stood up and said, "The cure for Down's Syndrome is abortion."

And that .. regardless of how you feel about the status of the fetus at the stage in which these tests are done, it struck me as a strange transformation of the historical attitude in medicine that we are a healing profession.  And I find there is troubling dimensions to all of this, and I also find very positive things in what you said, certainly the poignancy of the letter you read.  It's powerful.

Could you just give us some general comments on how we need to frame this to make sure that the positives come out and not the negatives?

CHAIRMAN PELLEGRINO:  Dr. McHugh?

DR. McHUGH:  Dr. Alexander, I want to thank you very much for that splendid presentation, and the demonstration of how lucky in America we are to have public servants like yourself.  I particularly appreciate this, of course, as a neuropsychiatrist who has looked over and has lived in this field for over 60 years and have turned to the PKU achievements and the achievements that have followed after it as one of the few advances that have really occurred in the field, and to see it described and developed the way you have and show how carefully our people are .. doctors and public servants are working to make all the messages of that achievement clear and the advances go further from it.

It was a great drink of water in the desert, so ..

(Laughter.)

But I wanted also to ask you, and I'm sure you are thinking in these terms, but they come up for someone like myself in the neuropsychiatric domain.  And that is that as biology and particularly neurobiology advances, the issues illuminated by mental retardation and the issues of mental retardation will blur with the .. will blur the seemingly and obviously biologically inappropriate sharp division between the cognitive disorders that turn up in childhood, and the cognitive disorders that turn up in adulthood.

And those are the matters that are little touched upon by what Floyd Bloom said to you about Huntington's Disease, another condition that I am very .. have been very interested in, and the two .. the two issues that come up, and I wonder whether your group is thinking about them .. in deciding about what to study for and what to recognize early on, the first one is that other matters besides the cognitive problems of patients with these genetic disorders are accessible to treatment, after all, if they are early identified.

In the MR group, of course, it's the autistic features and other kinds of things that go along with the cognitive problems, some of which can be treated with medications today.  And particularly in Huntington's Disease there are two aspects of that that are appropriate to identify. 

One of them is the great vulnerability of these patients to show classical bipolar disorder before the onset of either the motor or other cognitive disorder, and in that time the suicide rate amongst Huntington's patients, whether they know or not about the presence of their genetic thing, is extremely high, not only suicide but homicide.  You probably know that.

But it does make it clear that in the discussion of the biology of these matters, from MR until now, people like yourself can champion a spread of these identified .. I understand all that we .. at Hopkins we know a lot about the problems of counseling and the issues of identification, particularly in Huntington's Disease.  So I know it's a fraught area, but on the other hand there are things to be said other than just that we can cure the disorder now to make it identified.

And the other thing is that, you know, one of the great messages of PKU and congenital hypothyroidism is that the preventive treatment post-natally is the action.  You work because the disorder isn't symptomatic at this point, and you can prevent it by treating now.  Well, it .. if I'm correct that there will be a blurring of the boundaries, then conditions like Huntington's Disease that don't show up until you're in your late thirties or early forties .. by the way, there are some Huntington's patients, as you know, that turn up at age 12, 13, or 14, very difficult problems.

But if Huntington's Disease or Alzheimer's Disease or things of that sort take from conception until 40 or 50 years later to produce their effect, presumably there are things that are going on that we may be able to interfere with and interrupt, again, given the model of PKU and congenital hypothyroidism.  And, once again, I wondered whether those issues are considered in terms of the scientific advances, the state programs, and, to some extent, very much the ethical principles that we have as a caring society.

CHAIRMAN PELLEGRINO:  Thank you very much, Dr. McHugh.

Dr. Alexander, we have presented you with three very complex questions, so we're going to extend the time so you have a chance to give justice to those questions and to yourself.  I hope all of you will permit this.  Thank you.

DR. ALEXANDER:  Well, they are terrific questions.  It's hard to do justice to them.  Let me just try to respond without too .. not too extensively.  The question of using comparable technologies for prenatal diagnosis is obviously one that people have thought of and talked about. 

Basically, what you need is cells, cells of the fetus and those are obtainable by either amniocentesis or chorionic villi sampling.  That's basically what we use now. 

We're not developing it for that purpose.  We're dealing .. we're working on this for newborn screening.  Whether people will take it and apply it in prenatal diagnosis, it's likely that that will be done.  But I have no way of predicting that for sure, and the applications that we're working on right now are focused very precisely on prenatal .. on newborn screening, and all of the technologies that are associated with that, where you're screening essentially everybody in the population.

So that's a short answer to a question.

With regard to the issues of gene variability and behavioral phenotypes, and so forth, clearly this is a problem and a concern.  The classic example here is cystic fibrosis.  There is hundreds of gene variants in cystic fibrosis.  There's an extremely variable phenotype of people who are affected severely in infancy and early childhood to people who show no symptoms at all even until late adulthood.  And it would be nice if this would correlate with the particular gene variant, but it doesn't seem to precisely.

So one of the issues that we have to deal with in this .. and, believe me, we have our eyes open, it's not like we're doing this for the first time, like we were with PKU.  We know the issue of gene variability, and the .. how the .. not only the genes vary, but the phenotype varies as well.  And to take this into account as we make our diagnoses.

This is a problem much more for some conditions than it is for others.  For others there is very little gene variability, and the clinical course and the phenotype show very little variability.  But for others it's enormous, and, again, we've learned a lot about that.  We know for many of these conditions what that variability is. 

We know the different gene variants that are associated with the disorders, and at least we are aware of the issue and the problem and the concern, and it is humbling, believe me, to people who are working in this, trying to deal with it, and to know that there is no 100 percent predictability for most of this, because not only is there variation in the gene, there is variation, as you say, in what other genes may be present and what environmental stimuli may be, etcetera.

So it's a very good question, a very important point to raise.  I should have included it.  I didn't.  I appreciate your raising the issue.

Dr. McHugh, the issues that you raise about more than just the newborn in terms of the expression of these issues, clearly there are some of these conditions that don't have expression until later on.  We know most about the ones that express early, and we are just in the process of learning about some of these others that have perhaps some suggestive signs but not the expected phenotype earlier in life.  Your Huntington's example is a perfect one.

So this is in the category of things we still have to learn about, and why we have to proceed with caution as we go about this screening process, and implement it in a way that we learn about the conditions as we go.  We don't assume that we know everything, and that we don't start screening for some of these things where there is so much variability that we can't predict until we know more.

So people are making decisions about what should be in the test regimen, and part of that is based on how much predictability there is about phenotype from genotype, and when these things are likely to occur.

Another issue with the Huntington's is that the whole picture of Huntington's may change within the .. before the disease might be expressed.  Hopefully, it's going to.  I've got my fingers crossed and my hopes up.  And so I hope that .. another reason for maybe not screening for that condition that isn't manifest in most people until later is that the picture may change. 

And why worry people unnecessarily for 30 years if we are going to have a much better treatment prognosis outlook for them?  So that, again, is part of the thinking.

DR. McHUGH:  Just to follow up on that, given that you have been saying that we need to have these registries and things to go forward to know these treatments, and that's about the kids that show their disorders in childhood ..

DR. ALEXANDER:  That's right.

DR. McHUGH:  .. how much more is it necessary to do the same thing for things which are slower in onset but still have the causal factors present at conception?

DR. ALEXANDER:  That's a very good point.

CHAIRMAN PELLEGRINO:  Again, on behalf of the Council members, let me thank you for a superb presentation.  We are going to have our break until 10:30, and then pick up the next session. 

Again, thank you very much, Dr. Alexander.

DR. ALEXANDER:  Thank you.

(Applause.)

(Whereupon, the proceedings in the foregoing matter went off the record at 10:14 a.m. and went back on the record at 10:33 a.m.)