Introduction
Newscom
Connor Sternberg had his first surgery before he breathed his first breath. Doctors worked to close a small hole at the base of his spine through which his spinal cord protruded, a severe form of a birth defect known as spina bifida. Although spina bifida is relatively common—affecting about one out of every 1,500 babies—this was no run-of-the-mill operation: Connor was still in his mother's womb.
Surgeons have been operating on fetuses for more than two decades. Because of the danger to both mother and child, doctors reserve the procedure for just a handful of the most severe defects. As fetal diagnosis and treatment progress, however, physicians might be able to combine less invasive surgical techniques with stem cell and gene therapies to treat a wide range of diseases in utero, with considerably less risk. According to Alan Flake, a pediatric surgeon at the Children's Hospital of Philadelphia, operations such as Connor's—where doctors must slice through the mother's abdomen and womb to reach the fetus—will be nonexistent in 20 years. "I think these techniques will be considered somewhat Neanderthal," he says. "To me, the real future of fetal treatment is applying stem cells or gene therapy to the fetus."
Foreign donation
Stem cell therapies hold particular promise, according to some researchers. By providing therapy in utero, doctors have a better chance of treating a disease before debilitating symptoms appear. In addition, the fetus has a poorly developed immune system, so it is more likely to accept foreign cells. Lastly, because of the fetus's small size, relatively high doses of cells can be delivered, making the therapy more likely to succeed.
Flake is investigating the potential of blood-forming stem cells to tame sickle cell anemia, a genetic disease that doctors can already diagnose in utero.
Sickle cell anemia leaves an individual with crescent-shaped blood cells that don't live as long as normal blood cells and tend to clog small vessels. The disease isn't easy to treat: bone marrow transplants can allow the body to produce normal cells, but matched donors are rare. And without a matched donor, the procedure has a mortality rate of about 15%, Flake says. "So it's not something that people are willing to have applied to their children."
Delivery of stem cells in utero might make the procedure less risky. If doctors could deliver blood-forming stem cells from the child's mother or father—a half match—in utero, the fetus's weakened immune system might allow the cells to take hold. The fetus would become, in effect, a chimera, composed of two genetically distinct cell lines. "It would make the fetus tolerant to the donor," Flake says, so that after birth, a bone marrow transplant from the same donor would be less risky.
Flake and his colleagues have already proven that injecting blood-forming stem cells in utero can work in large animal models, including sheep and dogs (Science 233, 776–778; 1986). "I don't think it's unreasonable at all to expect that, within a few years, we'll be ready to proceed with clinical trials," Flake says. And, if the procedure works for sickle cell disease, Flake says, it should work for other genetic blood disorders or as a way to prepare babies for organ transplants. The technique will also work for immune disorders in which the donor cells have an advantage over the normal cells. In fact, 14 years ago Flake used the strategy to cure a boy with X-linked severe combined immunodeficiency disorder—known as bubble boy disease (N. Engl. J. Med. 335, 1806–1810; 1996).
Dario Fauza, an associate in surgery at the Children's Hospital Boston, says Flake's expectations are "well founded." But some obstacles have yet to be overcome, he cautions. The large animal models are not yet consistent and predictable, he says. "We also need to learn more about feto-maternal traffic of these transplanted cells in higher species."
Gene therapy may be even farther from the clinic than stem cell therapy. Researchers have had some success in rodent trials (Curr. Opin. Mol. Ther. 9, 432–438; 2007). "There are quite a few proof-of-principle experiments in fetal gene therapy that show that it works," says Anna David, a clinician at the University College London's Institute for Women's Health in Britain. But no studies have yet been done in humans. "I think the jury is still out on which is the right disease to go for," David says.
And gene therapy involves substantial risks. For example, the viruses used to deliver the desired gene could elicit a dangerous immune response or the inserted gene could interfere with fetal development. Consequently, David says fetal gene therapy will probably be confined to severe, life-threatening diseases. So she and her colleagues are focusing on fetal growth restriction, a condition in which the fetus, for unknown reasons, stops growing. David says the current treatments—bed rest, breathing extra oxygen, drinking lots of fluids—don't really work. "You're basically left saying your baby is doing to die in utero, or it's going to be delivered early, and it will probably die afterwards," she says.
The earliest approach
Children's Hospital Boston
Future fix: An engineered fetal trachea
For now, doctors must make do with what is currently available: fetal surgery. But not everyone agrees about which defects should be treated in utero and which are best left until after birth.
Take, for example, congenital diaphragmatic hernia (CDH). Babies with CDH have a hole in their diaphragm, allowing organs normally situated in the abdomen to migrate into the chest. These organs take up space meant for the lungs. CDH is rarely fatal in utero, but babies with a severe form of the condition often die after birth as a result of their underdeveloped lungs.
In 1989, fetal surgeons at the University of California, San Francisco (UCSF) began treating CDH in utero. To repair the defect, they would cut open a mother's abdomen and womb, reposition the baby's organs and patch the hole. A special surgical instrument minimized bleeding, allowing them to preserve the amniotic sac, though they had to refill the sac with fluid after the surgery. But the strategy had limited success: giving the lungs more space did help them grow, but in cases where the liver had migrated into the chest, the umbilical cord often became kinked during the delicate procedure, cutting off blood flow and killing the fetus (J. Pediatr. Surg. 32, 1637–1642; 1997). In addition, any time surgeons open the womb, they increase the risk of premature delivery.
In 1999, the UCSF surgeons tried a new tack. Instead of patching the hole in the diaphragm, they used a small balloon to block the inside of the baby's windpipe. (The fetus gets its oxygen through the umbilical cord until birth). The fix was based on studies that showed that lungs grow faster when the windpipe is blocked. This time, the surgeons also tried a less invasive procedure. They still opened the mother's abdomen, but they left the uterus intact, using a small endoscopic instrument to place the balloon, which requires only a pencil-sized hole.
The procedure seemed to work, but when they compared this type of prenatal CDH treatment with standard postnatal treatment in a randomized trial, they found no difference in survival (N. Engl. J. Med. 349, 1916–1924; 2003). Most fetal treatment centers stopped performing the procedure. They couldn't ethically continue to put the mother's health at risk given that fetal surgery didn't seem to give their babies a leg up over surgery after birth.
When the trial results were published, "we considered very carefully whether we should continue," says Jan Deprest, a gynecologist at the Gasthuisberg Hospital of the Katholieke Universiteit Leuven in Belgium. But the procedure that Deprest and his colleagues in Europe were using was less invasive than one that had been tested in the US. They didn't open the mother's abdomen, and they used an even smaller instrument to insert the balloon. And the Europeans made another improvement: instead of leaving the balloon in until birth, Deprest and his colleagues removed it after six to eight weeks to allow for more normal lung development. Because they were only operating on the group with severe CDH, which normally has an 80% mortality rate, "we considered it ethically and scientifically sound and clinically relevant to continue our study," Deprest says. So far, he and his colleagues have performed more than 150 surgeries at three different centers across Europe. And they now have a clinical trial underway comparing fetal surgery with postnatal treatment in babies with a less severe form of the defect.
"Jan Deprest and I have sort played ping-pong," Hanmin Lee, director of UCSF's Fetal Treatment Center, says about the two institutes' involvement in treating CDH. This year, Lee and his colleagues started treating CDH in utero again. But now they use the exact same procedure as the Europeans. The surgical instrument and balloons were recently approved by the US Food and Drug Administration. Lee says his team will do their own trial to test the procedure.
In utero surgery for spina bifida, Connor's defect, is still experimental as well. The surgery is only offered as part of a randomized trial that began in 2003 to compare prenatal treatment with standard postnatal treatment. Spina bifida isn't typically fatal; the hole in the spine can be patched after birth. But scientists have discovered that much of the damage caused by spina bifida happens in the womb. By repairing the defect early, doctors hope to minimize the severity of the side effects, which can include paralysis and collection of fluid in the brain.
In the future, there may also be a way to combine surgical techniques with stem cell therapy to get at the defect even earlier, which might improve a child's outcome.
Fauza has developed a method to extract fetal stem cells from the amniotic fluid. Eventually, surgeons might be able to use minimally invasive techniques to place a synthetic scaffold over the defect. "You might seed that [scaffold] with stem cells that would grow into skin or would grow into a tissue that would basically cover and seal the defect," Fauza says. "It's not too difficult to envision that as something that would be applicable within a decade."
If such a combination approach works, it would be good news for kids with spina bifida, like Connor, and their moms. Patching the defect earlier might reduce the severity of the side effects. Connor is doing well, but his father says not all of the babies who had surgery are as healthy. At the same time, performing the surgery without opening the mother's womb decreases the risk of premature delivery. Moreover, moms recover from minimally invasive surgery much quicker than they do from open fetal surgery. And that is an enormous incentive. "Any time you talk about [in uetero] intervention, the main concern should be the mother," Fauza says. "She is an innocent bystander."
