Angie Rojas's cells are about to be returned to her. Taken from her bone marrow five days earlier, genetically altered and nourished in the lab, the cells nestle in the tip of a syringe, about 45 million of them, a pale nib barely visible in the liquid. When the doctor nods and the nurse starts the "push," the cells trickle through an IV lock into the teenager's bloodstream. It is September 1, 2001, and gene therapy is mounting another try.
Patients with a form of severe combined immune deficiency disease (ADA-SCID) have a defect in a gene that is crucial to immune function. The diagram above outlines a recent gene therapy trial that could correct the condition: (1) extract defective cells from the bone marrow, (2) insert a virus bearing a healthy gene into the cells, and (3) inject the altered cells into the patient.
José Rojas gets up from his seat at the window, eager to keep his daughter's face in view. He cranes his head around the monitor at the foot of the bed. Lucy, her mother, stays seated, placid as usual. She has been with Angie in hospital rooms more times than she can count. This time, she hopes, the doctors will arrest the severe immune disorder that has stalked her daughter's life. She knows the treatment is experimental. Pressing close to Lucy is a young cousin, Denise, whom Angie has invited to the hospital. The girl looks around wide-eyed, not sure how to react.
The air outside the window at Childrens Hospital Los Angeles is bright and hot. Palm trees dot the steep hillside. From where the relatives sit, the famous "Hollywood" sign can be seen, large as life. Although the room is full of witnesses besides the family to this unfolding medical drama, there is no buzz or chatter, no lights or cameras. A sense of anticipation is mixed with anticlimax, as if this were a rerun of a movie that wasn't that great to begin with.
What a difference a decade makes. In September 1990 the first human gene therapy trial took place at the National Institutes of Health in Bethesda, Maryland. The young patient in that experiment, Ashanthi DeSilva, Ashi for short, suffered from the same condition as Angelica Rojas. Their rare genetic disease is called ADA-SCID—severe combined immune deficiency (SCID) resulting from the failure of white blood cells to produce a critical enzyme, adenosine deaminase (ADA).
Then as now the ultimate aim of the experiment was to insert a healthy form of the ADA gene into the patient's cells to restore her immune function. Officially, the aim of the trials was to see if transferring the gene was safe. The Food and Drug Administration approved both procedures as Phase I experiments. In introductory trials like these, an actual therapeutic benefit is considered a bonus.
Other coincidences: Ashi and Angie, both daughters of immigrants, are nearly the same age—Angie, who turned 16 in December, is nine months older. Donald Kohn, the doctor-scientist in charge of the trial, had trained with the National Institutes of Health team that conducted Ashi's trial. But where the publicity for gene therapy in 1990 was intense—the NIH researchers held a big press conference before Ashi's trial, yet journalists still tried to sneak into the hospital to see her—Kohn is cautious about attention. "It's good for me and for Childrens Hospital in some ways," he says. "But publicity probably isn't a good idea for a gene therapy researcher in 2001."
Kohn is alluding to the death, in 1999, of Jesse Gelsinger, a gene therapy patient in a quite different experiment in Philadelphia. Gelsinger had an extreme reaction to an engineered virus bearing therapeutic genes into his liver. The incident has prompted the FDA to tighten rules on researchers throughout the field and raised general concerns about safety. "Profound damage to public confidence in the discipline of gene therapy has been done," declared a past president of the American Society of Gene Therapy last summer. The president also took note of the alleged financial conflicts of the lead investigator in Philadelphia, whose biotech company stood to profit if the trial worked, and of the subsequent revelations at other medical institutions of "massive underreporting of patient adverse events." The "adverse events" included a half-dozen cases in which patients died of their underlying conditions, evidently unrelated to gene therapy. More often it happened that fevers and other worrisome complications in the test subjects were never brought to the government's attention.
Even before Gelsinger's death, results of direct gene therapy had been bafflingly bad. Since the groundbreaking experiment on Ashi, more than 450 gene therapy trials have been launched in the United States. Although the trials have enrolled 4,000 subjects, not one trial has demonstrated that a disease can be cured or controlled by altering a person's genes.
Still, the idea of gene therapy remains compelling—to fix a disease at its root in the DNA. But the technical problems have been twofold: First, how to transfer enough of the gene into a patient. Most often viruses are harnessed for the task, because viruses are able to insert their genes into cells and force them to make foreign proteins. With few exceptions the viruses used for gene therapy are altered so they cannot reproduce. Their sole function is to sneak into the cell and deliver a packet of DNA, the desired human gene.
The second challenge for gene therapy is to get the gene, once inserted, to make enough of the desired protein. Natural viruses carry molecular switches that regulate their genes, but engineered viruses need help. A DNA sequence called a promoter is inserted into the virus; it turns the gene on, producing a protein that is supposed to do the work of correcting the disease. In ADA-SCID the protein needed is the ADA enzyme.
In 1990, when Ashi got the ADA gene, scientists did not understand the hurdles they faced nor did they imagine the string of disappointments to come. The initial experiment seemed to work, at least partially, and the results were heralded. The "success" of Ashi's gene therapy spurred researchers around the nation to launch their own human tests. It was true that roughly a quarter of the T cells (a kind of white blood cell) circulating in the girl's body did produce the ADA enzyme, an indication that the gene was active. But the added gene wasn't making enough of the enzyme to protect her from infections. So Ashi, like Angie Rojas during the 1990s, has continued to require very costly medication. As the cells trickle today into Angie's arm, the jury is still out on the case of her predecessor, as well as on gene therapy in general.
Don Kohn, who has been sitting at the bedside, has a sudden thought and goes over to Lucy and José. He gives the parents the most promising news about gene therapy it is possible to give. Since the spring of 2000, researchers in France have reported increasing success in their treatment of X-SCID, an immune deficiency similar to ADA-SCID but more menacing because there is no medication for it. First there were two children, then five; now there are seven, including a teenager, whose immune systems have been restored by gene transfers and are still going strong. A permanent cure? It is still too soon to say. The best part is that some of the lab techniques employed by the French team were developed by Kohn himself.
At the very outset, medical genetics focused on children and enzyme deficiencies. The first disease that was linked to a single genetic trait was a condition called alkaptonuria, in 1902. Sir Archibald Garrod, an Englishman, studied families in which some of the children produced red or black urine and earwax of the same strange color. They also had symptoms of arthritis. Their siblings had none of these traits.
Donald Kohn relies on grants to support his team's gene therapy trial at Childrens Hospital Los Angeles. "I'm poor but honest—and maybe stupid," he says.Photograph by Gillian Laub
Although he didn't know what genes were, Garrod suspected that the children with the disorder had inherited a double copy of a "factor"—one factor received from each parent. Garrod thought that the factor produced an enzyme, and if both factors were defective, the lack of the enzyme led to the disease. Applying the laws of Mendel to his findings, Garrod sketched family lineages for this "inborn error of metabolism."
Alkaptonuria is now listed as a recessive gene disorder, one of hundreds categorized since 1902. Genes, as Garrod surmised, are inherited in pairs, and a person with a recessive disorder has had the misfortune to acquire a bad copy of the gene from each parent. Having a single bad copy of a gene isn't generally a problem, because the disorders often involve enzymes, and the person's normal copy of the gene can usually make enough of a particular enzyme to keep the metabolic pathway in good order. If the enzyme is too little or absent, however, the ill effects show up not long after birth, when the child has to rely on his or her own metabolic system instead of the mother's. According to a recent literature survey by Johns Hopkins researchers, "An extraordinarily high fraction of diseases with onset in the first year of life are caused by defects in genes encoding enzymes."
The metabolic breakdown that causes ADA-SCID was clarified in the 1970s. The ADA enzyme clears the T cells of certain waste products. If the T cells cannot do this because of an ADA deficiency, they die. And when T cells die, the immune system fails and a SCID disorder erupts.
In the 1980s the gene at the heart of the disorder was identified. Moreover, the ADA gene was cloned so that it could be used in experiments. Researchers inserted the ADA gene into an engineered mouse virus and then coaxed the virus to infect a culture of cells, thus compensating for the genetic flaw. The stage was set to bring Angie Rojas and Don Kohn together.
Kohn started his medical career as a pediatric immunologist. At the time he left the NIH for Los Angeles, in the late 1980s, Lucy Rojas was trying to keep her infant daughter alive. The normal ailments that Lucy had experienced with her two older children were so much worse when Angie came down with them—asthma attacks, fevers spiking to 105, ear infections bad enough to require surgery. The mother was told, "If you don't watch this child 24 hours a day, she'll pass away."
Kids with undiagnosed SCID usually do die before the age of 2. If diagnosed properly, they spend their days sealed in a plastic compartment, like David, the heart-wrenching "boy in the bubble," who finally succumbed to X-SCID in 1984. Although Angie's immune system was full of holes, she appeared to have generated just enough ADA enzyme to fight off commonplace germs and viruses. As one doctor commented, "She was lucky she didn't walk in front of a bad bug."
When the specialists at Childrens Hospital first saw her, in 1991, Angie had pneumonia. A bone marrow transplant, which would use cells withdrawn from her brother, was considered. Kohn and the other doctors of the transplant unit had performed this procedure as a last resort for children with leukemia and other grave diseases of the blood. The survival rate for SCID patients after a bone marrow transplant was just two children in three, however, and those patients did not have pneumonia.
Fortunately, Angie had another option: drug therapy. Her disease was diagnosed just as a drug form of ADA, derived from cows, was made available. Called PEG-ADA, the medication was injected two times a week. Her mother was shown how to give the shots to Angie at home. The treatment cost several hundred thousand dollars a year. For the past decade, while bringing Angie to the hospital for regular monitoring, the Rojases have managed to have their health insurers and a state agency pick up the staggering expense. However, the aid will run out long before Angie stops needing the medication. Not surprisingly, SCID families and doctors are eager for an alternative to high-risk transplants and high-cost medicines.
In Ashi DeSilva's trial 11 years ago, T cells had been filtered from her blood, exposed to the gene-bearing virus, and then put back into her system. A better target, researchers knew, would be stem cells. Harbored in the body's bone marrow, they are not the same as stem cells in an embryo, but they are flexible in the same way. They give rise not only to the T cells and other components of the immune system, they also program cells to recognize a host of different infectious agents. Stem cells lacking the ADA gene are not impaired, but every T cell they produce has a fatal flaw. Fix the stem cells, the idea is, and all the subsequent T cells should thrive.
It is June, three months before the procedure. The patient and family are being briefed. "We'll try to put the gene into the bone marrow," Kohn says to Lucy and José, "and then we'll give Angie's cells back to her through a vein in the hand." He turns to Angie. "It's like getting a bone marrow transplant from yourself, but hopefully the marrow has been fixed, and it can make all the different cells that you need to fight everything."
The examining room is packed. In addition to Kohn and the Rojas family, there are two members of the research team, a hospital public affairs representative, a translator, and a social worker. The translator puts Kohn's words into Spanish to be certain that Lucy and José understand him, and the social worker, taking notes, monitors the ethical aspects of the meeting.
This is a consent conference where risks and benefits are discussed, after which the parents must decide whether or not to enroll Angie in the clinical trial. Wearing black sneakers, black pants, and a blue sweatshirt, Angie sits cross-legged on the examining table. Her family and members of the hospital team are in chairs in a semicircle around her. Used to being the center of medical attention, the girl is quick to smile and quick to speak her mind. She is compact and broad-shouldered and plays saxophone in her high school band.
Kohn tells the family that "this is a research study, not treatment. It's fine if Angie doesn't want to be in the study. She'll get the same care. We'd like you to feel you could make this decision without any pressure."
Angie and her mom are already in favor of proceeding, having met with Kohn previously. It is José who has to be satisfied. Declining to sit down, he asks, "Is this an experiment, then?"
Kohn says yes. Chewing gum and pacing his words to allow translation, he reads from the research protocol. He starts by describing ADA-SCID, "a genetic disease that Angie inherited. One gene doesn't work in Angie. . . . "
A hand shoots up in the corner. "I have a question," Angie says, grinning, as if she's back in school. "How did I get the gene? Do you know from which parent?"
The doctor asks her to hold the question until later. He methodically summarizes the drug and transplant treatments for her disease. Then he takes up the precedents for gene therapy. Ashi was the forerunner in the SCID field, but Kohn himself conducted a trial in 1993, when excitement about the new technique was at its height. Three infants born with ADA-SCID were infused with the missing gene. Because the patients were too small to provide bone marrow, Kohn had used stem cells from the cord blood of the placentas instead. When put back into the children, a small portion of the altered stem cells made T cells with the ADA gene, but the gene did not make the protein. Still short of ADA, the patients had to remain on medication.
"The T cells with the gene have increased," Kohn says, "but the T cells don't have the gene 'on.' We hit a double, not a home run." So in the mid-1990s Kohn went back to the drawing board to improve the rate of gene transfer. And more work had to be done on the laboratory culture of stem cells. So Kohn and his research team developed a more powerful viral vector and also improved the growth factors—the supplements that nurture stem cells while the virus moves the gene into them. The new growth factors were used by the French scientists in their promising treatment of X-SCID.
"Will this newer way work better than what we did before?" says Kohn. "That's what we're trying to find out."
In discussing the risks of the procedure with the family, he mentions the death of Jesse Gelsinger. He doesn't hurry past it, but he doesn't dwell on it either. "It was a different virus, and they were trying to put it directly into his body. He had a strong reaction and he died. It was unexpected. But only he [of all the subjects in the Philadelphia trial] had that problem."
José and Angie stir uneasily, but Lucy doesn't blink.
Kohn repeats, "It was unexpected. It taught us bad things can happen, though." He points out that the anesthesia Angie must undergo when her bone marrow is extracted is more risky, statistically. He goes to the exam table and touches Angie for the first time, on the hip. "We take the bone marrow from the top-back part of the pelvis." The girl's sneakers crinkle on the white tissue paper. "We do it on both sides."
José asks, "Is it possible she won't need shots after this?"
"It's possible. If the gene is working, we'd try to take her off the shots."
Angie's hand shoots up again. "Question. I have always wanted to know: Which one of the family had the disease?"
Kohn explains the nature of a recessive condition. "Probably each of your parents has one defective gene. You will pass on to your children the ADA gene that doesn't work, but if the father has a good gene, your own children won't have problems."
"How rare is what I have?"
"We don't know. One in a million? Ten kids each year come forward."
"Wow, lucky me."
The doctor smiles. 'Yes, it's pretty rare. You're unique."
"Would I be the first [in the trial]?"
Kohn indicates yes. The consent is a done deal.
The SCID disorders make a good target for gene therapy experimentation for several reasons. One, just a single gene needs correcting. Two, just one type of cell needs the gene, the bone marrow stem cell. Three, the manipulation of the stem cells can be done in vitro, outside the body, providing greater control than does the in vivo technique. In the latter the vector has to be more or less squirted toward an organ or a tumor, with no assurance that the virus will evade the body's defenses and deliver its genetic cargo to the right cells.
Angie Rojas, left, sits with her sister Suzy, who's 18, in their bedroom in Los Angeles. Angie's condition affected her two healthy siblings, she says, "because my mom was always at the hospital taking care of me."Photograph by Gillian Laub
Finally, researchers have learned that the regulation of the inserted gene does not have to be exact. To combat ADA-SCID or X-SCID, the gene can make a lot of protein or a little. Either way, the benefit occurs. Other diseases demand a more precise regulation of gene activity.
In the heyday of clinical trials, the early 1990s, gene therapists took on many single-gene disorders in addition to SCID. The most prominent was cystic fibrosis, a lung condition that kills half its victims by age 30. The gene for cystic fibrosis—or rather the gene whose flaw spawns the disorder—was discovered in 1989, raising hopes that researchers could heal the damaged airways of patients. After much publicity and a number of attempts, the trials were deemed failures. Mucus in the lungs of the patients impeded the vectors from penetrating the cells of the respiratory tract, and the genes that did manage to get in were neutralized by an immune response. Even if the new gene had taken hold, the treatments would not have lasted, because the cells lining the airways are constantly replaced. The ideal targets would be the parent or precursor cells of lung tissue, but the identity of those cells is unknown.
The story of cystic fibrosis is typical. Yet despite the difficulty in treating single-gene disorders, researchers have gravitated toward even more complex conditions. According to a review in late 2000 by Bret Ball and W. French Anderson at the University of Southern California, the great majority of gene therapy efforts since 1990 have been directed at cancer, although cancer offers no single target in the genome. In distant second place was AIDS, with 33 trials launched or proposed. The classic, single-gene diseases have accounted for only 50 trials, of which 20 were for cystic fibrosis.
Thus the conditions most likely to yield to the new technology received the least attention. The reason is not hard to figure. For rare, inherited disorders like SCID, wrote Ball and Anderson, "there is little potential for return on investments in expensive research and clinical trials." Translation: Drug companies have no motive to invest. By contrast, the "more common diseases," such as cancer, heart disease, and AIDS, "make attractive targets for pharmaceutical companies." Moreover, experimenters had a ready supply of sick people, for whom the risks of enrolling in a gene study seemed well worth the while.
The NIH issued a report in 1995 criticizing researchers because they had neglected the basic science, especially the animal models, which might have worked out the kinks in the procedure. Don Kohn recalls the period: "There was such zeal for doing clinical gene therapy. At medical centers the social rewards were higher for doing trials, especially on cancers like melanoma, where cases were desperate, than for mouse research. The thinking was, you didn't know if it would work, but you didn't think it would hurt, considering the disease."
It took Jesse Gelsinger's death to shake the field and tighten standards. Now, in the wake of the French team's breakthrough with X-SCID, attention is once again focused on basic diseases resulting from enzyme deficiencies. Anderson, who was co-leader of the trial on Ashi DeSilva, firmly states, as he has all along, "If you can't treat SCID, you can't treat anything else."
Angie Rojas lies on her back in the hospital bed, fighting to keep alert. While the stem cells are slowly pushed into her arm, she is growing drowsy from an infusion of Benadryl, an antihistamine administered before the test to guard against inflammation at the injection site. Kohn doesn't fear an acute, life-threatening reaction. He believes that 15 years' experience with both animal and human SCID subjects, though not guaranteeing a success, has minimized the danger to Angie.
Privately, Kohn thinks there's just a one in four possibility that healthy T cells will arise in the coming months and rescue her immune system. He is uncertain because he doesn't know how many of the stem cells were altered by the vector or if they will take hold in her bone marrow.
The nurse making the push, Debra Tumbarello, leans over the patient. "How're you feeling, Ange?"
Blearily, she says, "Fine. I'm not dying." Her heart rate has crept up to 94 beats per minute, probably from anxiety.
In five minutes it's done, and the room starts to empty. Kohn recalls that a colleague in his 1993 SCID trial urged that the syringe be saved for the Smithsonian Institution because it was a national treasure. This one won't be.
Lucy says, relieved, "I expected a big container [of cells]. That's nice it was so little." She is wearing a crucifix around her neck.
Shaking José's hand good-bye, the doctor says, "We won't know for sure for another two years."
"Ten years of coming here," Angie's father muses. "How many times?" he asks his wife.
"If Angie was going to have a reaction," says Tumbarello, "she would have had one already."
The girl's heart rate has settled to 80. She sleeps peacefully, new genes swirling in her blood.
For general information, including a wonderful primer on the basic science behind gene therapy, see the University of Pennsylvania's Institute for Human Gene Therapy page: www.uphs.upenn.edu/ihgt.
For more about gene therapy research at the University of Southern California, see cwis.usc.edu/schools/medicine/academic_departments/biochem_molbiol/igm.html.
For updates on political issues and links to ongoing clinical trials, see the American Society of Gene Therapy's site: www.asgt.org.
For more about SCID, including various forms and treatments, see www.scid.net/scidpid.html.