Blood Matters

by Masha Gessen

  The building I visited—the original barnlike construction plus an addition raised in 2000—looked like a Hollywood rendering of the Amish aesthetic: soaring ceilings, exposed wooden beams, shining hardwood floors, spacious offices for the two pediatricians, sparkling-clean exam rooms, and a general sense of friendly calm so rarely found anywhere, especially at a medical institution—particularly one where parents brought their very sick, their inexplicably crippled, their hopelessly unable to communicate, impossible-to-understand children. The secret to the calm, aside from an apparently boundless capacity for acceptance characteristic of both the Amish and Old Order Mennonite cultures, was a belief Holmes Morton seemed to project from the moment he set foot in Lancaster County: He knew what was wrong with the sick children here, and if he did not know, he could figure it out, and then he could cook up something that would help. Cook he did: Morton spent the first twenty minutes of our acquaintance telling me about an experimental pig-brain-bouillon treatment he and his junior partner at the clinic, Kevin Strauss, had devised for babies born with a disorder that apparently robbed the body of necessary compounds called gangliosides, causing children to become severely microcephalic. The process sounded downright bizarre—Strauss had gone to the butcher’s to purchase fifteen pig brains, which he boiled down to forty cubic centimeters of “brain soup.” And the logic was suspiciously simple—pig brains contain a lot of gangliosides—but the approach was probably quite similar to those that had led Morton and his colleagues to devise ways to manage diseases like glutaric aciduria and maple syrup disease.

  The Clinic for Special Children came a few years too late to help Susie’s children. If this tardiness caused her particular regret, she had come to peace with it a long time before I met her. From the way she told her story, in fact, it was clear that Morton’s appearance in Lancaster County had marked a new era in her life. He named her children’s disease. He helped keep other children from becoming similarly disabled—although some 20 to 30 percent still suffered strokes, despite all the precautions. The words would certainly be too grand for Susie’s restrained speech, but perhaps Morton had managed to make her feel that her children’s suffering and her family’s grief had not been in vain. Susie told me of Morton’s work in great detail: She followed it in part because her sister, Rebecca, had worked at the clinic nearly from the beginning.

  All of Susie’s own children were grown now. Steve was able to find work helping a family in whose basement apartment he was now living. Livanna was renting an apartment with two other disabled young women, helped by a home-care aide who came once a day. She visited her parents once every two weeks, arriving on a bus for the disabled. She lived on Social Security—a level of integration into the larger world that pushed the margins of what the Amish found acceptable, but church elders had apparently chosen to tolerate this, perhaps out of consideration for all the hardships the family had had to endure. Alvin, too, had found work out in the larger community, through a workshop for disabled adults, where he had the task of counting screws and packing them into boxes. He had a paycheck—between two and ten dollars for a day of work—which made him very proud.

  Susie was showing me around the house, explaining the battery-powered lift used to get Alvin on and off the toilet and in and out of the bath, when a van for the disabled announced its arrival outside with a series of squeals and beeps. Alvin rolled in on his rechargeable wheelchair. He was very thin, dressed all in black, and he sat tilting noticeably to one side, his knees drawn up and his hands hovering close to his long face. He had never learned to walk or talk, so he communicated using jerky gestures and a lot of lopsided smiles. Susie introduced us. Alvin said he had had a good day at work.

  ***

  “So the big joke now is that Hugo got his oil and got to be in the movie and Morton went to Lancaster County.” Seventeen years later, Morton still resented the way his mentor had undercut him. Hugo Moser had received his grant to do research into Lorenzo’s Oil, so named for a little boy (now a grown man) whose parents educated themselves in chemistry and medicine and devised a compound that seemed to help soften the symptoms of his adrenoleukodystrophy. Moser’s initial research had failed to prove that Lorenzo’s Oil worked—a fact Morton noted with obvious satisfaction, even though more recent studies of the compound had been much more hopeful. But Hugo Moser got to be in the movie.

  It is one of those films with a plot so heart-wrenching as to make perception of its aesthetic qualities impossible. Nick Nolte, armed with a grotesque Italian accent, and Susan Sarandon play the parents of an adorable little boy who suddenly begins to have temper tantrums on the one hand and coordination problems on the other. He begins to lose his abilities to walk and talk before the family finds out what is wrong: adrenoleukodystrophy, a metabolic disorder caused by a mutation in the X chromosome. It is one of several mutations—like the one that causes a kind of hemophilia—that only women pass on to their children, of whom only the males are affected. This is another genetic condition that seems to have a conspicuous built-in cruelty. Since women have two copies of the X chromosome, they always inherit one healthy copy of the gene from the father, and this allows them to remain healthy. It is the boys who stand to inherit the one bad copy from the mother, with nothing to counterbalance it, and get sick.

  Boys with adrenoleukodystrophy generally develop symptoms before the age of ten, then suffer from progressive neurological deterioration and die within a couple of years. Lorenzo Odone, whose story inspired the movie, lost his speech and all of his mobility but did not die. Years after he became unresponsive and should have been dead, he began to show signs of consciousness and learned to communicate using his eyelids and one of his fingers to sign “yes” and “no.” His parents believed this happened because of Lorenzo’s Oil, and this may or may not be true: Studies seem to show that the treatment is possibly effective only as a means of prevention and becomes useless once symptoms set in. Quite clearly, though, Lorenzo survived because his parents refused to let him die, which means they cared for him so closely and so attentively that they certainly did things that could never be measured and published in a medical journal to be reproduced in a patient population and checked in a proper trial. Their best medicine was not pure love: It was a thorough understanding of the disease earned through years of sleepless nights spent at libraries (Augusto Odone, the father, ultimately published several medical papers in major journals), combined with unblinking observations of the way the disease worked in their particular boy. In other words, it was the essence of medicine as it will be practiced in the future.

  In the course of researching this book, I interviewed dozens of medical doctors, geneticists, and biochemists who defined their research in terms of “personalized medicine.” Some were looking for genetic markers that would allow doctors to identify patients who are likely to vomit when coming out of anesthesia—a common problem that is a major source of complications following surgery. Some were already testing patients with colon and other kinds of cancer for genetic markers that would predict their response to one or another sort of chemotherapy. The idea is that eventually even community-based physicians—those outside academic research centers—would routinely be using genetic tests to tell them which medicine to administer, and how, to a particular patient.

  By the time I conducted these interviews, Holmes Morton had for seventeen years been practicing medicine one child, one life, and one illness at a time. He had conducted clinical trials that involved one, two, and three children. Some of these were trials of his own invention, like the brain-soup experiment. Others involved the use of experimental treatments on children who for one reason or another could not enroll in one of the larger academic trials. When I spoke to him, for example, he had begun administering valproic acid to a child with type 1 spinal muscular atrophy, a condition that destroys cells in the spinal cord, robbing a child of the ability to use his muscles, and usually leads to death in early childhood. The parents of this particular baby, a h
orse-and-buggy Mennonite family from Ohio, had located a large clinical trial of valproic acid in California, but the trial turned out to be randomized and placebo-controlled, which meant that the child stood a 30 percent chance of being assigned to a placebo group. “To make a long story short,” Morton said, “the family asked me if I would prescribe valproic acid in the hopes that it would help this child. Well, I prescribe valproic acid all the time for less interesting reasons, so I said, ‘Sure, why wouldn’t I?’ And we monitor liver tests, and we monitor blood counts to make sure the valproic acid is not causing any toxicity. And what’s the downside of a trial like that? Not much, you know.”

  It was just the opposite of what was still medical gospel in the United States at the turn of the twenty-first century: that proper medical knowledge is gained only through evidence gathered from large-scale double-blind placebo-controlled randomized trials. Few people had yet noticed that the medicine of the future could probably at no stage be either blinded or randomized. But in southeastern Pennsylvania, a small group of slightly odd people was already practicing personalized genetics-based medicine.

  ***

  Genetics-based medicine begins with a genetic diagnosis. In the old days—the 1990s—determining the genetic cause of a disease generally required a large group of patients and DNA samples from members of their families. Following the logic that genes generally travel in clumps—meaning that if a person inherited a mutation from one parent, she or he probably also inherited a sizable group of genes in the vicinity of the mutant—researchers looked for areas of the genome where all the samples seemed to be similar. This usually allowed them to focus first on the right chromosome, then on the right neighborhood in the chromosome, and gradually zero in on the gene. The search for a mutation took months or, frequently, years. Often, before a test for the mutation could be developed, a test for a genetic marker appeared. This happened at the “neighborhood” stage of the search and essentially amounted to looking for signs that the person had inherited the suspect region from the affected parent; the test invariably required a DNA sample from the affected person, too, to ensure that both the person being tested and the affected parent carried the marker. Marker tests could also yield inconclusive results.

  Technology that appeared in the new millennium allowed scientists to look at more of the genome much faster. The Clinic for Special Children received a donation of equipment from Affymetrix, a company that made machines that could rough out the map of a person’s genome in the space of about three days. The technology was based on an array of genetic markers: Erik Puffenberger, when we met, was using chips that contained ten thousand markers, and he told me breathlessly that Affymetrix was about to come out with a five-hundred-thousand-marker chip. By the time I was writing this chapter less than six months later, Affymetrix had announced the coming of a million-marker chip, and, less than a year after that, when I was editing it, the 1.8-million marker chip had already arrived.

  Where it used to take months to scrutinize three or four hundred places in the genome for conspicuous matches, the new machines took three days to look in ten thousand spots. The number of markers meant that, instead of using a large patient population, Puffenberger could now take just five, three, or even two people with the same disorder. Alternatively, he could test a single affected person and compare his genome to that of his unaffected siblings, looking for areas of difference that could be significant. Puffenberger had done this in the case of a family that had had three children born with an extremely severe seizure disorder—the babies seemed to start seizing in utero and after birth continue to have thousands of seizures a day. Two of the children had already died, so Puffenberger compared samples from the single affected child to those of the child’s six healthy siblings, and located the mutation.

  The clinic had had the equipment for all of two or three months by the time I visited. Puffenberger had already used it to map four or five disorders and had grown accustomed to comparing the new era of genetic mapping to the dark old days when one had to labor for years—as though many decades had passed. I arrived on the second day of Puffenberger’s three-day mapping cycle. He was arranging tiny vials of samples on a palm-sized tray.

  “These two have an unusual disorder,” he said, looking at small drops of liquid in a vial. “Mental retardation, and I believe they have seizures. We think this might be a new disorder. They are Amish from out in Indiana, and we think this might be a new disorder.

  “Then these are four patients, Mennonite patients from Canada, who have a muscle myopathy.

  “These are two patients here at the clinic who have an unknown mitochondrial disorder. They are brothers.

  “And here we have a child that died, and we are trying to figure out if they had a chromosomal abnormality. And the next child then had a seizure disorder that we know is not the one we recently mapped, so we are adding him to our big panel of genotype seizure patients.”

  Puffenberger used a tiny pipette to put a fluorescent “tag” on each sample and left them in the fantastical Affymetrix machine to “hybridize.” He then led me into another room, a very large one, where the level of background noise from all the lab equipment approached that of a roaring engine. There on a small desk Puffenberger kept his PowerBook, where a collection of Excel files held the fruits of his labors.

  Puffenberger opened one of his files with apparent pride. It contained the data on three siblings with a severe dystonia disorder, and two other children, from two different families, who had turned out to have the same disease. We were looking at tens of thousands of Excel cells with the letters AB, BA, AA, or BB in them. The first order of business, once the ten thousand markers had been mapped, was to look for AAs and BBs. “These patients all have the same disorder,” explained Puffenberger, “so we are interested in finding a region where this patient, this patient, and this patient are all homozygous for the same allele.” Every so often an area of repeated pairs of letters showed up, but some of these could be easily dismissed—either because they were only very small pieces of DNA or because control samples (people with a different disorder or healthy people from the same population) were also homozygous in that area.

  Once Puffenberger identified a large area of homozygosity, he worked on the assumption that the defective gene was found there. The next step was going to the online database of known genes maintained by the National Center for Biotechnology Information, a resource of the National Institutes of Health, and looking at genes that are known to reside in the suspect area. The NCBI data, culled from the Human Genome Project, included a list of genes whose functions had been studied or theorized, as well as a list of genes that were not known to do anything in particular—either because they did not or because no one yet knew what they did. For this disorder, Puffenberger decided to look first at two genes, one of which was known to cause seizures while the other was known to cause a movement disorder when it was knocked out in a mouse. He would sequence both genes. If he did not find a mutation, he would go back to the database to identify other candidate genes and sequence them.

  Puffenberger made the entire procedure as exciting as one could make a very long process of chasing numbers and letters. It did seem that the quest for the cause of disease had lost much of its detective-novel luster and turned largely mechanical. The impression was slightly deceptive. The straightforward number-crunching route worked only some of the time. Other times, Puffenberger had been stumped. Indeed, one condition had stumped the entire clinic for years. It was a kind of retinitis pigmentosa, a disorder that causes progressive loss of vision. In the Amish, it was accompanied by a loss of reflexes and position sense: Affected people have an odd gait and trouble gauging their movements. Caroline Morton told me of one young man who would crush a paper cup if he were given one to drink from, because he could not quite tell what his fingers were doing.

  Dr. Morton and I passed another affected person, a middle-aged woman, on the winding road to the clinic. “She shouldn’t
be out here,” he said. “She has retinitis pigmentosa, like we were talking about. She has bad night vision and also almost no reflexes. She also has deafness. At some point—I’m not sure exactly how it happened—but Caroline hired her to clean our house. And my daughter Sarah walked in, and she was there with the vacuum cleaner, dragging it around. And she couldn’t hear well enough to know that it wasn’t turned on, and she couldn’t see well enough to know that it wasn’t cleaning.” Something told me this was not the only time the Mortons tried to supplement their good work at the clinic in ways that were not entirely practical.

  It was back in 1999 that a group of scientists, including Dr. Morton, identified a likely location of the gene that caused posterior column ataxia with retinitis pigmentosa, as the condition was formally described. They theorized that a single mutant gene would cause damage both to the spinal cord and to the retina. And now Puffenberger had spent years looking for the gene. “We sequenced every gene that is expressed in the retina, and they are all normal,” he complained. “Then we went and sequenced all the other genes that are expressed in the brain, and they are all normal, and now we are sequencing the ones that really make no functional sense whatsoever, but it’s got to be one of these! So it’s not always easy and obvious.”