The New York Times reports couple using in vitro fertilization (IVF) and pre-implantation genetic diagnosis (PGD or PIGD) are screening for genes that affect cancer risk in order to select against embryos that have a higher risk of cancer.
Prospective parents have been using the procedure, known as preimplantation genetic diagnosis, or P.G.D., for more than a decade to screen for genes certain to cause childhood diseases that are severe and largely untreatable.
Now a growing number of couples like the Kingsburys are crossing a new threshold for parental intervention in the genetic makeup of their offspring: They are using P.G.D. to detect a predisposition to cancers that may or may not develop later in life, and are often treatable if they do.
For most parents who have used preimplantation diagnosis, the burden of playing God has been trumped by the near certainty that diseases like cystic fibrosis and sickle cell anemia will afflict the children who carry the genetic mutation that causes them. The procedure has also been used to avoid passing on Huntington’s disease, a severe neurological disease that typically does not surface until middle age but spares no one who carries the mutation that causes it.
I suspect as meaning of more genetic variations get identified that selection to lower cancer risk will eventually lose out to selection for other qualities. This will happen in part as a result of the development of better treatments for cancer. Also, as the significance of more genetic variations become known people will weigh cancer risk against advantages and disadvantages of other genetic variations found in each embryo.
Australian fertility clinics that do IVF (in vitro fertilization) also are now offering tests for genetic variations that increase cancer risks.
The tests pick up mutations in the BRCA1 and BRCA2 genes, Dr Leeanda Wilton, head of the Genetic and Molecular Research Lab at Melbourne IVF, told a recent international conference in Brisbane.
...
This technique can also detect mutations in the BRCA genes, which confer an estimated 65-85% risk of the carrier developing breast cancer by the age of 70.
Suppose a gene therapy to breasts to correct the BRCA gene only in breast tissue is developed 10 or 20 years after an IVF baby is born. Well, selecting against BRCA variants for breast cancer risk reduction will turn out to be unnecessary.
I suspect the people who are selecting against BRCA1 and BRCA2 in their offspring are making a poor trade-off without even being aware they are making a trade-off. If Greg Cochran and Henry Harpending are correct the genetic variations that occur at higher rates in Ashkenazi Jews which cause health problems (including BRCA1 and BRCA2) are there due to selection for higher intelligence during the Middle Ages. Then selecting out BRCA1 and BRCA2 mutations using PGID and IVF will result in dumber offspring. Henry Harpending explains that BRCA1 probably accelerates early central nervous system (CNS) development at the expense of higher breast cancer risk later on:
Re mechanism: The argument (well known to breeders where there is no argument) goes like this: In a drastic new environment there is big fitness payoff to IQ. In this new environment there is a payoff to "turning down" BRCA1 to free up early CNS development but at the cost of higher cancer rates later in life. Eventually, especially in a big population, a BRCA1 variant with the optimum activity will show up. Meanwhile carriers of one normal and one broken BRCA1 gene have a big fitness advantage because they have, say, 90% of normal suppression of early CNS development. So the broken BRCA1 allele is favored by selection even though homozygotes for it die. After a long time it would be replaced by the optimum allele but it takes a long time for that optimum allele to show up.
But if biomedical advances will produce cures for cancer within 20 years (and I'd be very very surprised if they didn't) then for the BRCA variations the trade-off for getting lower cancer risk at the expense of lower intelligence is a bad choice to make today. Better to boost the risk of a disease that will become easy curable before the disease is likely to develop and give your daughter higher IQ. The higher IQ will be a tremendous asset throughout life. The cancer problem will be fixable. If you believe as I do that the rate of biotechnological advance is accelerating so rapidly that cancer will become curable before babies born today reach their 30s then the risk of increased breast cancer is a price worth paying in order to give your daughter a smarter brain.
I predict most of us will live to see the day that donor egg banks offer eggs with the BRCA1 and BRCA2 mutations as selling points.
By Randall Parker at 2006 September 03 12:25 PM Bioethics Reproduction | TrackBackI disagree, unless you believe that the risky variants necessarily accompany the beneficial ones. The real premium will be for donor genetic material which has beneficial variants but lacks somewhat-associated risky ones.
Desirable rarities usually trade at nice premiums in a market.
wcw,
If you have the set of all known genetic variations to choose among there are probably other ways to boost early CNS development. But that is not the case at the present time. I'm talking about now. Right now if someone selects against BRCA1 and BRCA2 carrying embryos they are selecting against having smarter kids.
Desirable rarities: Yes, I've previously written posts arguing that once we have really cheap DNA sequencing and know what many more genetic variations do then egg and sperm donors with the most sought after genetic variations will be able to sell them for large premiums.
But right now we do not have accurate DNA tests to determine which eggs, sperm, and embryos are best.
Why do we always assume intellect tech will stay constant and unchanging? Even right now, lifelong low-dose (really low dose) lithium probably gives an accretion of 5 IQ points. BMP-7+2 seem to regen nerves in the CNS. Phosphatidyl Serine restores plasticity. All this means that one can drive intelligence up artificially and it will happen bigtime eventually. We could equally postulate in the next 20 years we develop tech that adds 20-50 IQ points to the avg human -- are we going to then say this ivf human is disadvantaged being a 145 IQ vs 152 IQ while dodgeing cancer? I don't think so.
As usual I'm the contrarian here: It is utter hubris to speculate on the overall outcome of genetic-environment sorting against long timeframes - especially in the light of the tech uncertainty. But that's OK -- that's why we are here. However, it is just plain wrong to state that the BRCAx selection works against IQ. It's the difference between association and causation. E.g. the the selective pressure process that led to the Ashkenazi Jew IQ resulted in BRCAx selection as it is today. This does NOT imply that the BRCAx mutation is responsible for the IQ. There are probably hundreds of genes involved and one should not make sweeping generalizations based on a singleton gene. And the two quoted authors should have known better!
Additionally, it is utter arrogance to assume that the tradeoff of lower IQ vs cancer, even if it were valid, made by the humans MOST at risk, is somehow not beneficial. As I tried to instruct my students back in the day, one of Dr. Bob's Commandments: Thou shalt not substitute thy judgement for thy patient's wishes.
Also, IQ, IS over-rated. Between stimulus and response sits consciousness. It is more important (today) to process well than to process fast. IQ improperly selects for the fast and the accurate while shafting those who are slower but even more accurate. Guess which group has more millionaires? The race does not always goto the swift.
Think twice folks, think twice.
I doubt BRCAx alleles could account for even 1 % of total IQ variation. Testing for those alleles has been going on for some time. Breastfeeding and prenatal fish oil consumption have far stronger affects.
Here’s an interesting article,
http://www.nature.com/mp/journal...l/ 4001327a.html
Prion allele M129V: "Genotype accounted for 2.7% of the total variability in Full Scale IQ"
Here’s an editorial piece on this topic,
http://dx.doi.org/10.1016/ j.inte...ell.2006.01.001
The quest for quantitative trait loci associated with intelligence
Robert Plomin, Joanna K.J. Kennedy and Ian W. Craig
Abstract
Progress towards identifying quantitative trait loci (QTLs) for complex traits like intelligence and common disorders like mental retardation has been slower than expected. An important factor is that most QTL effects may be much smaller than expected—not just 1% effect sizes but perhaps effects as small as .1%. If so, this would mean that studies have been seriously underpowered to detect and to replicate QTL effects. We have used microarrays to genotype DNA pooled for groups of low versus high intelligence in order to screen very large numbers of single nucleotide polymorphisms (SNPs) on very large samples in the quest for QTLs of very small effect size: We find no effect sizes greater than .5%. Microarrays with 500,000 SNPs are now available that facilitate genomewide scans which will make it possible to identify nearly all SNP associations that account for 1% of the variance of intelligence—if there are any QTL effect sizes as large as 1%.
I wonder why the genome wide studies didn’t identify the M129V locus. Also such studies only find simple, additive affects. More complex studies with larger sample populations are needed to identify gene-gene or gene-environment interactions.
Rsilvetz: “Why do we always assume intellect tech will stay constant and unchanging? Even right now, lifelong low-dose (really low dose) lithium probably gives an accretion of 5 IQ points. BMP-7+2 seem to regen nerves in the CNS. Phosphatidyl Serine restores plasticity. All this means that one can drive intelligence up artificially and it will happen bigtime eventually. We could equally postulate in the next 20 years we develop tech that adds 20-50 IQ points to the avg human…”
I agree. There will be nutritional, drug, training, and cybernetic enhancements that significantly boost IQ and other mental traits in the next two decades. Medical treatments for birth defects, injury, and disease will drive early development. As the technology improves, personal enhancement will take over.
Fly,
Your URLs are broken. Looks like you copied them as text rather than as URLs.
Robert Silvetz,
With all due respect, you do not begin to approach Greg Cochran's level of knowledge and mathematical ability with population genetics. No, it is not as easy as you think it is to poke holes in his theory.
As Greg et al pointed out, the Jewish genetic diseases occur at far too high of frequency to be there by chance. Far too many of the genes that cause problems are concentrated in a small number of pathways to be there by chance or founder effect. There had to be positive selection to produce the relatively high frequencies for so many alleles that cause diseases. Most of them do affect neuronal growth or neuronal function. Selection for higher IQ is a highly plausible theory to explain this result.
Read their paper. I've linked to it.
Sorry about the broken URL's.
Prion allele M129V:
http://www.nature.com/mp/journal/v8/n11/full/4001327a.html
Editorial Review abstract
Hi Randall,
Apologies if I pushed the envelope. The issue isn't Cochrane's mathematical ability (or mine tho I am a triple degree holder). It's the burden of proof. The assertion is made that BRCAx has central impact. Not only is it unlikely to be true based on 9-th grade biology (intelligence is not eye-color and not subject to a several-gene reduction), there are exactly three (3!) published articles on BRCAx in the CNS. NONE of which support the contention that turning it up (or down) makes one whit of a difference, so BRCAx function in CNS is a matter of extreme speculation.
If you prefer, the quoted authors have a case right upto the point they claim that the selection process that led to the intelligence also selected for disease.
Plus, the authors in question forget their own systems knowledge: Break a pathway and disease is almost certain. Since there is a strong correlation between pathways and genes, break a gene - get a disease. But for a systems behavior, which is an overlapping of outputs, there is no such correlation for obvious reasons. Thus, the converse assertion, that the diseased genes are responsible for intelligence, is of extraordinary difficulty to prove. The authors do not meet that burden of proof. In essence, they have data for the selection process, not the causation of the intelligence.
Dr. Silvetz,
The pathways aren't broken, they are slightly altered. If I recall, they do cite a paper on increased dendritic growth with the BRCA1 mutation.
Are you arguing that a small genetic change cannot cause a situational improvement? Sickle cell does, for heteros in some environments. Cochran (who certainly does not need me as a defender) et al made a case that if there are different allele frequencies in several genes in the same or related pathways, this is pretty unlikely, and probably has a cause.
For example, Austrailian Aborigines have a stronger inflammatory response than white Austrailians, and differences in a couple of genes in the pathways. Both the differences up inflammation. It is probably an adaptation. Certainly, no.
I do think (but it's ask a bumblebee) that they made the weakest case for the BRCA1 mutation. But IQ and educational acheivement are closely related, and tons of women have been tested. We might be able to do pair-wise comparison of siblings who do and don't carry the BRCA1 difference.
If carriers are smarter than noncarrier siblings, would you change your view on the matter? If they aren't any different, I'd want to test men too, and if it having the mutation does not correlate with IQ, I'd say it wasn't selected to up IQ.