Parents or Gods?

The birth of Louise Brown in 1978, and with her that of human in vitro fertilization (IVF), was a landmark in medical science. Surgically harvesting eggs from a woman's ovaries, fertilizing them outside her body, and transferring the resulting embryos into her uterus enabled effective treatment of female infertility caused by irreparably damaged fallopian tubes. Since then, rapid innovation has led to new applications for IVF and other assisted reproductive technologies.

Many infertile couples now turn to such advanced technologies when other "low tech" options fail, and they are the treatment of choice not only for tubal damage, but also for significant forms of male infertility. For example, intracytoplasmic sperm injection is a technique in which a single viable sperm is injected into an egg, allowing fertilization to occur even in cases where few healthy sperm are available. Freezing unimplanted embryos is now standard procedure; freezing unfertilized eggs is under development.

Perhaps inevitably, our access to human eggs and embryos now enables us to extend prenatal genetic diagnosis to the pre-implantation embryo. Conventional prenatal diagnosis entails removing fetal cells, either from the amniotic fluid (amniocentesis) or from the placenta (chorionic villus sampling, CVS). Both procedures are routinely offered to pregnant women 35 years and over to diagnose chromosomal abnormalities such as Down's syndrome, or to screen for cystic fibrosis, sickle cell disease, or Tay Sachs disease.

However, amniocentesis and CVS both entail a 0.5-1% risk of pregnancy loss, and if an abnormality is diagnosed the only clinical option is to terminate the pregnancy. Preimplantation genetic diagnosis (PGD) is fundamentally different from its prenatal counterpart. Because the diagnosis is made prior to placement of the embryo in the woman's uterus, the exclusion of specific abnormalities is possible without terminating an established, ongoing pregnancy.

Clearly, couples carrying a transmissible genetic defect and otherwise requiring IVF should be offered the option of PGD, and if necessary, referral to a center that can perform this specialized procedure. But genetically affected couples who do not suffer from infertility may also be candidates for IVF with PGD. They could well prefer this option to natural conception and the prospect of therapeutic abortion following a conventional prenatal genetic diagnosis.

But now consider a more morally complicated scenario: a couple has a naturally conceived child affected with a life-threatening genetic disease for which the only possible cure is a bone marrow transplant from a matched donor. Understandably, they may want PGD to prevent similar defects in their future children. But they also want those embryos to be tissue typed in search of a sibling who could serve as a bone marrow donor to their first child.

In this case, an embryo is being selected not only to avoid congenital disease and thus directly benefit the resulting child, but also to produce a child whose very existence could provide a cure for an older sibling, and with virtually no risk to the donor child. Ethicists have grappled with such scenarios, and many would find this clinical option to be ethically sound. But what if the older sibling requires a donated kidney? The more the donor child is placed in physical jeopardy the thornier the issue becomes.

Gender selection is another hotly debated potential application of PGD. Gender selection may be clinically justified in order to prevent transmission of a sex-linked disease, such as hemophilia. Under such circumstances, it should be no more controversial than using PGD to screen for sickle cell disease. But is PGD ethically sound if an infertile couple merely prefers a boy or a girl? If so, then would it not be acceptable for "family balancing," when a fertile couple with three boys strongly desires a girl, or if a couple wants to choose the sex of their first-born child?

Obviously, analogous questions can be extended to the selection of many other traits that are not essential to the health of the resulting offspring. To all of these questions, there are no simple answers.

The essential goal of medicine is diagnosis and alleviation of disease, and infertility is a disease. The extension of prenatal diagnosis to an embryo in a laboratory dish will, in turn, reduce the incidence of certain types of genetically transmitted diseases. But new medical technologies also offer applications that fall outside of these conventional precepts. As always, the rapid advancement of science and technology should be accompanied and tempered by careful and thoughtful consideration of the appropriate uses of newly realized capabilities.