Hidden Valley Road: Inside the Mind of an American Family (30 page)

Jim Galvin had been in and out of the emergency room at Penrose Hospital in Colorado Springs for weeks, complaining of headaches and tingling in his extremities. The staff sent him home again and again, writing off what he was saying as signs of his usual paranoia.

Toward the end, Jim believed he had a hole in his chest. “Don’t you see I’ve been shot?” he said.

Jim died alone in his apartment in Colorado Springs on March 2, 2001, at the age of fifty-three. The doctors recorded the death as heart failure, related to his use of neuroleptic drugs. His family took this to mean that he died of a condition called neuroleptic malignant syndrome—a rare, life-threatening disorder most often caused by the drugs meant to help.
Researchers predisposed against the reflexive use of medication to treat the mentally ill have attributed tens of thousands of deaths to this syndrome. Some of the symptoms, like agitation and delirium, can easily be mistaken for psychosis, which explains why the syndrome is often only identified after the patient dies. Still other symptoms, like cramps and tremors, are often the same as the side effects of the drugs. Jim’s symptoms were so pronounced that at the time of his death, he had been prescribed procyclidine, a drug usually used to curb the effects of Parkinson’s disease.

For Margaret and Lindsay, the lesson of Jim’s death was clear. The cure was as bad as the disease. The sisters looked around and saw four more brothers—Donald, Joe, Matt, and Peter—and wondered who might be next.

Mimi, meanwhile, continued to explain Jim’s mental illness in terms of his life circumstances—a bad marriage, and maybe even the trauma of being made into an abuser by a nefarious priest. She was not ready to cast any of her children aside completely, not even Jim, despite what her daughters had revealed about him. “It was the strain of the marriage, I think, as much as anything,” Mimi would say, “and probably his own guilty conscience. But he was so well liked by
all
the children.”

Jim’s ex-wife, Kathy, and their son, Jimmy, did not attend the funeral. They were living in California now, rebuilding their lives, trying to forget the man who had tormented them both.

THERE WAS ONE
question about their childhoods that Margaret and Lindsay had left unasked until now. With Jim gone, they saw an opening. Their parents had known that Jim was mentally unstable—even hospitalized—early on in his marriage. Why had they allowed both girls to spend nights in Jim’s home, weekend after weekend, alone with him?

One day in 2003, tape recorder in hand, Margaret put the question to Mimi, point-blank. “Why did you let me go to his house that whole time?”

Mimi answered at once. “Because he had a recovery,” she said. “He had a recovery. He went back to work. His wife took care of him, and everything seemed fine. And he had subsequent breakdowns—he was seeing an outside doctor—and he would recover and be fine for six months.”

Margaret’s voice broke as she responded, her voice faint as a child’s.

“No one ever told me he was sick.”

“Oh my God,” Mimi said—less shocked than exasperated about having to rehash this again.

“I never knew,” Margaret said.

“See, they didn’t
know
in those days, Margaret,” Mimi said, speaking quickly now. “You know, they snap right out of it, it seemed. And he did. He came back, he would go back to work and hold it together. But he always overdid. He’d not just do the one job, he’d try to do two. And he was working like eighteen hours a day, things like that, and so he’d have a collapse. And he was drinking. Right. He was drinking.”

When cancer finally took Don Galvin—on January 7, 2003, at the age of seventy-eight—he had wasted down to less than a hundred pounds. He received a funeral with full military honors at the Academy Chapel, the architectural showpiece of the Academy that he had helped to launch into the world. One of the Academy’s roster of performing falcons was in attendance, perched on the fist of an Air Force cadet for the entire length of the service.

Michael played classical guitar as people entered. The boys’ old piano teacher played “Be Not Afraid” as the opening hymn. Mark, the former chess prodigy, read aloud from the description of a wise man in Chapter 39 of Ecclesiastes that fit well enough with the way Don wished to be seen in the world:
He researches into the wisdom of all the Ancients, he occupies his time with the prophecies / He preserves the discourses of famous men, he is at home with the niceties of parables….

Richard read from the book of John. John, the music teacher, in from Idaho, conducted the prayer of the faithful. Donald—nearly sixty now—read from the Beatitudes about the four virtues: prudence, justice, fortitude, and temperance. Michael and Lindsay each read poems. Margaret delivered a eulogy. “His memory failed him late in life,” she said, “but that didn’t mean his life was ordinary. It was extraordinary.”

The service closed with “On Eagle’s Wings.” And shortly before 11 a.m., the Air Force conducted a flyover in Don’s honor, planes tipping their wings as they soared over the grounds of the Air Force Academy, the place where he had been happiest.

Don’s body was sent to the University of Colorado, where Robert Freedman’s team examined his brain. They were surprised to find that Don’s brain did not have any of the physical attributes that were associated with mental impairment or illness.

Mimi had little to say about that. She knew what she knew.

The last time Joe, the mild-mannered seventh son, living alone in his Section 8 apartment, talked to his mother, he told her that his feet were numb and that he couldn’t walk. It was snowy, and Mimi couldn’t drive in bad weather. She said she would see him in the morning. By then, it was too late.

Joseph Galvin died alone at home on December 7, 2009, at the age of fifty-three. The county coroner’s report marked the cause of death as heart failure, caused by clozapine intoxication. The powerful atypical neuroleptic had proven helpful to Joe in some ways, but the drug’s physical side effects seemed to slowly wear his body down. The echoes of Jim’s death were unmistakable. Here was another likely instance of neuroleptic malignant syndrome.

When Jim had died, neither Lindsay nor Margaret had been tempted to mourn him. Joe was different. As children, both sisters had fantasized about not having brothers. The truth was, seeing the brothers they loved like Joe develop schizophrenia felt a little like watching them falling off the face of the earth. And so when Joe really did die, it was hard to grapple with what seemed like both a loss and the echo of that other loss, experienced years earlier, when mental illness took him away.

The family gathered to scatter Joe’s ashes. Peter’s face was ruddy, his clothes shabby and smelling of cigarettes, but he was still boyish, his bright blue eyes twinkling, his hair still jet black. Donald told the group that when he died, he wanted to be eaten by an elephant. Michael wanted his ashes scattered, but he wasn’t sure where quite yet. Richard had a place in mind: Boreas Pass in the Rocky Mountains. Margaret said she’d like to be in Maroon Creek in Aspen. Lindsay picked the back bowls, a skier’s paradise at Vail.

They all reminisced about their best times with Joe. Donald mentioned watching him play hockey. Mark remembered Joe racing his GTO against a Datsun 240Z and winning. Peter recalled living with him briefly in Chicago, when Joe was still throwing bags for United. Margaret talked about him teaching her how to drive a stick shift, off of Arapahoe Road.

Mimi went further back in time than anyone else—reflecting, perhaps, not just on the boy Joe had been, but about the time of her life when all the boys were still young, and when happiness still meant the promise of something wonderful to come. When he was a baby, she said, Joe was so beautiful while he slept. Like an angel.

2009

Cambridge, Massachusetts

In 2009, Stefan McDonough was entering his seventh year at Amgen, a neurobiologist lured out of academia by the prospect of developing real-life treatments and cures for one of the world’s largest biotechnology companies. After a few years researching new pain-management drugs, McDonough’s portfolio in the neuroscience department had broadened to diseases of the brain, including schizophrenia. Amgen was looking for a gene that could be targeted, something that needed rejiggering to help people with schizophrenia; if McDonough could find such a gene, Amgen would set to work on developing a drug to attack it.

From his office in Cambridge, McDonough threw himself into the work. He was so enthusiastic about the genomics revolution’s potential that he arranged to audit an undergraduate genetics course at Harvard, sitting after work in an old wooden one-armed desk chair, week after week, dreaming of finding the gene that would prove to be schizophrenia’s smoking gun. But very quickly, McDonough grew frustrated. Despite all the fanfare, it was clear every genetic location that had been associated with schizophrenia since the completion of the Human Genome Project—and there were more than a hundred of them now—had an effect so tiny that the idea of making a drug targeted at any one of them seemed ridiculous. That was when McDonough started looking around for another way—a shortcut to narrow the search. Wouldn’t it be easier, he thought, to find a smaller haystack to rifle through? Instead of searching the genetic code of many thousands of unrelated people, why not study a limited group of people who had seemed to inherit the disease because of a genetic irregularity they all shared?

Why, he thought, wasn’t anyone researching families?

McDonough was hardly unaware of the drawbacks. He knew that one family’s genetic mutation—or, as the field now calls it, “disease-causing gene variant”—could be unique to that one family, and pointless to spend resources on. And yet he also knew that one family’s abnormality might reveal something fundamental about the illness that everyone had been missing. He needed to find someone who felt the same way—an expert on schizophrenia and families who could teach him more. He found a professor at Harvard, an easy stop on his commute home from work in Cambridge, who was kind enough to talk about imaging the brains of schizophrenia patients. Families weren’t her thing.

But she did know Lynn DeLisi.

“I’VE GOT MY
name on more papers than I need,” DeLisi said. “I just want to find these genes and help solve this disease.”

She was working not far from McDonough, in Brockton, where she had just joined the staff at the VA Boston Healthcare System’s psychiatric facility. That very year, 2009, she had moved from New York to Massachusetts, where she also was teaching classes at Harvard Medical School. Since her split with Pfizer in 2000, DeLisi had been estranged from her own research; no company seemed interested in picking up where the sale of Parke-Davis had left her, until now.

As she listened to McDonough talk about what he wanted to do, it was hard to say what she felt more intensely—surprise that a pharmaceutical company was interested in her work after all this time, or impatience to get started again. For McDonough, DeLisi checked off every box: a world-class researcher who had broken ground in this field; a devoted clinician who cherished one-on-one interaction with patients; a determined geneticist who yearned to find a cure. Best of all, she had been collecting pedigrees of families with schizophrenia since before McDonough graduated high school. And she was nice—something McDonough appreciated, given how territorial and guarded some academic researchers can be around pharma people.

She invited McDonough to join her on rounds at the VA hospital’s inpatient psych ward. This would be the biotech researcher’s first face-to-face contact with people suffering from the condition he wanted to cure. He watched as DeLisi, soft-spoken but direct and firm, visited with one patient who seemed perfectly calm, his delusions controlled, only to learn later that the man had committed an unspeakable crime. Other patients at the VA seemed completely soothed by the medication, but matter-of-factly noted that yes, they were hearing voices. “They’re telling me to kill people,” one said.

McDonough began to see how many of the patients were cognitively present, but without the emotions that typically make a person seem truly there. Only when he finally saw one patient having a violent breakdown, barricading his room and hissing with rage at the attendants around him, did he understand the plight of everyone there. “They have been warehoused where nobody can really deal with them,” he said.

Here was the real reason, he thought, why big pharma could afford to be fickle about finding new drugs for schizophrenia—why decades come and go without anyone even finding new drug targets. These patients, he realized, can’t advocate for themselves.

DeLisi made a deal with Amgen to work with McDonough on a new schizophrenia study. There was a maze of bureaucracy to contend with. First came a question of whether DeLisi was, in fact, the owner of her family samples, since the work of collecting most of them had taken place while she was with an institution, SUNY, that she since had left. Next, Amgen needed documentation showing that every donor in DeLisi’s collection had consented that their biological data could be used for research. Hundreds of emails later, she and McDonough finally retrieved a selection of samples that DeLisi had stored at the Coriell Institute—DNA from some three hundred families, faithfully preserved in culture.

When McDonough got his first look at DeLisi’s old research, he was astonished. The complete sequencing of a genome had been impossible in the 1990s, and yet the level of analysis she had done on these samples was ahead of its time. Now, these samples had woken, Rip Van Winkle–style, in the age of computer-assisted genetic analysis. The analysis would be easier than ever now—and more precise, more nuanced, more detailed.

FOR THIS NEW
study, they wanted only the most blatant, egregious multiplex family cases they could find. At least three people in each family had to have schizophrenia, and at least three others in the family must not. They settled on nine families—four whom DeLisi had contacted while she was at the VA, and five from DeLisi’s old collection. The Galvins were from the latter group, the largest group by far of siblings of any family in the sample.

McDonough and DeLisi’s goal was to see if any of these families carried a rare genetic mutation or irregularity that was shared by the sick family members. This was what made the size of the families in their analysis so important: Any gene variant found in a person with schizophrenia, DeLisi and McDonough knew, might also be present in an affected parent or sibling simply by chance—and not because it is a cause of their shared disease. Parents share half of their genes with each child, after all, and a variant present in one sibling has a 50 percent chance of being present in a brother or sister. But as the number of family members with schizophrenia grows, it becomes increasingly meaningful if each and every one of them has a specific gene variant. The likelihood that the mutation is harmless or unrelated to their schizophrenia dwindles and, as it tracks faithfully with the disease in more and more family members, the mutation looks increasingly likely to be the cause. The assumption they were making was that any rare mutation they found would offer a fresh way of understanding the illness. “Even though that particular mutation may be unique to that family,” DeLisi said, “it’s possible that the abnormality in that gene is part of an overall biochemical pathway that may be abnormal in schizophrenia.”

Sure enough, with the Galvins, DeLisi and McDonough found something tantalizing: a mutation shared by every Galvin brother for whom DeLisi had collected samples back in the eighties, in a gene called SHANK2. The mutation they found was connected to an important process in the brain—a process that seems vitally related to schizophrenia. SHANK2 is a communications assistant for brain cells. The SHANK2 gene encodes the proteins that help brain synapses to transmit signals and the neurons to react quickly. The Galvins’ mutation significantly alters the protein that SHANK2 produces. “The mutation was found right at one of the known functional guts of SHANK2,” McDonough said, “right at one of the spots that’s known to be critical for SHANK2 function.” In this way, the SHANK2 mutation was pointing the way to something new, potentially, about the illness—a glitchy molecular process that might be shared by more people than just this one family. Schizophrenia might take shape in that process. “It’s certainly not a proof, to a scientific standard, that this mutation caused the disease,” McDonough said. “What it really tells you about is a mechanism of schizophrenia.”

Similarly rare variants have shaken up the research into other diseases. Parkinson’s researchers, for example, found a genetic mutation affecting the
α
-synuclein protein in one family in Italy that pointed the way toward developing new drugs. The greatest example may be the development of statins, the drugs that decrease cholesterol levels for thousands of people at risk for developing heart disease. Scientists knew for years that high cholesterol contributed to heart disease, but no headway was made in lowering cholesterol until two researchers at the University of Texas Southwestern Medical Center in Dallas found some families with very early onset cardiovascular disease who had unusual mutations that impaired the body’s ability to remove LDL cholesterol from the blood. This mutation was not present in most people with heart disease. But that didn’t matter—the study of those mutations exposed how cholesterol levels can be lowered, not just in those families but in almost everyone. The drug that was developed to correct that particular LDL issue turned out to revolutionize the treatment of heart disease.

This may be the real miracle offered by the Human Genome Project: not the chance to find a smoking-gun gene that may or may not exist, but the ability to see how schizophrenia takes shape in the brain. SHANK2 is just one example of this; the way Robert Freedman’s CHRNA7 gene shed a light on information-processing issues is another. And at the same time that DeLisi and McDonough were doing their work,
a team from the Broad Institute in Cambridge, the Harvard-MIT collaboration that had taken charge of the GWAS efforts for schizophrenia, published its own highly publicized study identifying a mutation in a gene called C4A—more common than the mutation in SHANK2, but still far too rare to target with a drug—that seemed to play a role in the overpruning of brain synapses. Their research suggested that people with schizophrenia might end up cutting some synapses as adolescents that they would need later in life—another angle on the process of schizophrenia. While it is not clear if the Galvins have that C4A mutation, they played a small role in that study, too, being among the earliest families to donate their DNA to the pool of data analyzed by the Broad Institute’s team.

DeLisi and McDonough’s study was published in
Molecular Psychiatry
at the end of 2016. It was not possible to say for sure that this particular SHANK2 variant in this specific gene caused the Galvins’ schizophrenia. But that conclusion was consistent with what DeLisi and McDonough saw. Thirty years after she first met the family in their living room in Woodmen Valley, DeLisi had arrived at what looked like an answer to the question that beset the Galvins: Why?

THAT ANSWER CAME
with some surprises. The first involved the connection of the genome’s three different SHANK genes—SHANK1, SHANK2, and SHANK3—not just to schizophrenia but to other mental illnesses. Before this study,
others had conducted separate studies of each of the SHANK genes’ relationship with autism and other brain disorders. Now, taken together, all the research indicated that at least some varieties of mental illness exist on a spectrum: Some people with certain SHANK mutations may have autism, while others are bipolar and still others have schizophrenia.

The concept of a spectrum of illness seemed highly relevant to the Galvin family. Peter, for example, wandered between diagnoses, from schizophrenia to bipolar disorder. Donald also was diagnosed with mania and prescribed lithium early on, before the doctors moved on to the usual assortment of neuroleptics. Joe’s collection of symptoms was different from Jim’s, and Jim’s was different from Matthew’s—and surely there was no one else like Brian. Yet seven of the brothers—the seven who provided DeLisi with samples, including at least a few nondiagnosed brothers—all had this same mutation, in a gene that also figured prominently in other mental illnesses.

“Lynn was right,” McDonough said. Studying families with multiple occurrences of mental illness was, in the end, the study of a shared genetic issue—one that, depending on each person, manifests itself in a different way. “These are multiplex families, and it sure looks like the same genetic determinants can give rise to subtly different diseases.”

It’s possible that discoveries like the Galvin family’s mutation could point the way toward a completely new conception of mental illness. That could come sooner rather than later; in some corridors, it’s already happening. In 2010, the psychiatrist Thomas Insel, then director of NIMH, called for the research community to redefine schizophrenia as “
a collection of neurodevelopmental disorders,” not one single disease. The end of schizophrenia as a monolithic diagnosis could mean the beginning of the end of the stigma surrounding the condition. What if schizophrenia wasn’t a disease at all, but a symptom?

“The metaphor I use is that years ago, clinicians used to look at ‘fever’ as one disease,” said John McGrath, an epidemiologist with Australia’s Queensland Centre for Mental Health Research and one of the world’s authorities on quantifying populations of mentally ill people. “Then they split it into different types of fevers. And then they realized it’s just a nonspecific reaction to various illnesses. Psychosis is just what the brain does when it’s not working very well.”