Introduction

The patients were at the end of the line – unable to climb a flight of steps, constantly short of breath, and prepared for a last, fatal heart attack. Using a long device vaguely resembling a fishing rod, a medical team led by Texas Heart Institute cardiologist Emerson Perin navigated through a small incision in the groin to reach the patients' hearts, where he made a series of injections.

The content of the injections was bone marrow mononuclear cells – a type of adult stem cell that gives rise to different blood types – freshly pulled from the back of each patient's hip. This would be one of the first attempts at cell therapy for heart disease. The theory, grounded in an important study performed at the NIH, was that some of the bone marrow would develop into cardiac cells and regenerate damaged heart tissue. Because the cells were derived from each patient, immune rejection was not anticipated to be a problem. Still, the procedure was a bold one: what if the cell mixture developed into bone? What if the cells caused abnormal rhythms in the heart? What if they clumped together and triggered an embolism?

Brazilian regulators took a favorable view of the protocol, and Perin, who was trained in Brazil and had working relationships wiThcardiologists there, took his study to Rio de Janeiro in 2001. The team, and their regulators, it turned out, guessed right: among the fourteen patients, none developed worrisome complications. Though the study was not rigorously controlled, volunteers receiving cells seemed to do better than patients who did not.

Introduction

Gene transfer researchers and others with historical inclinations like to describe an experiment performed by virologist Stanfield Rogers in the early 1970s as the first attempted gene transfer in human subjects.The Oak Ridge National Laboratories scientist had observed that Shope papilloma virus infections, which normally cause warts, also depress blood serum levels of the amino acid arginine in rabbits. Studies also showed that workers handling the virus also had lower serum arginine. Rogers then postulated that the virus contains a gene that codes for the enzyme arginase, which breaks down arginine

Around the same time the German physician H. G. Terheggen in Cologne encountered a series of patients with various neurological impairments due to a deficiency in the enzyme arginase. On learning of the report, Rogers contacted Terheggen and proposed administering the virus to patients with the enzyme deficiency. The study, notwithstanding its bold vision, proved unsuccessful in either improving disease symptoms or in producing biological insights. In fact, the virus used in the experiment “degenerat[ed]… in storage,” and much later studies revealed that in fact the virus did not encode arginase after all.

There is, of course, a sense in which the conceit behind contemporary gene transfer can find precedent in this episode. But Rogers's experiment predated recombinant DNA technologies, which emerged in the mid 1970s and enabled manipulation of genetic sequences. It also predated the development of a biotechnology industry, or the knowledge economy, or the emergence of the “triple helix” configuration of universities, the private sector, and the government.

One clear bag of fluid

In September 2001, one-year-old Rhys Evans became the poster child for gene transfer. The Welsh boy had been diagnosed with X-linked Severe Combined Immune Deficiency (X-SCID) after months of infections, oxygen tents, intensive care, and finally, an isolation room. His parents – one a teacher, the other a pipe fitter – were offered two options by his caregivers at the Great Ormond Street Hospital in London: bone-marrow transplantation plus a brutal regime of chemotherapy (to allow the new tissue to “take”), or an experimental procedure – gene transfer – that had shown success in two infants at Paris's Necker Hospital. Fearing the potentially lethal consequences of chemotherapy, the parents elected for gene transfer, and in July Evans received what his mother recorded in her diary as “one clear bag of fluid.”.Within weeks, immunological parameters began normalizing, and as Rhys approached his second birthday he was discharged from the hospital. Today, Evans leads a normal life – aside from occasional public appearances for charity, and lots of medical monitoring.

The latter has proven especially important in Evans's case because, across the English Channel, several of his X-SCID peers who underwent the same procedure were not as fortunate. Under the leadership of immunologists Alain Fischer and Marina Cavazzana-Calvo, several children had been cured in a protocol initiated in 1998. But by the early summer of 2002, the French team detected a never-before observed lymphoproliferative disorder in one of their patients. By December, a second patient began showing identical indications. Two years later, a third patient developed the disorder.

The science, ethics, and policy of gene transfer are rapidly evolving. As a companion to this book, I maintain a running commentary on developments in ethics, gene transfer and cell transplantation, and translational clinical trials at my blog Lost in Translation (http://lostintranslationethics.blogspot.com/)

The blog is designed to report on relevant news items, identify emerging ethical issues in translational research, apply frameworks developed in the book, and extend them; readers are invited to use the forum to discuss and debate the ethics of translational clinical research.

Introduction

Patients with undiagnosed glioblastoma multiforme – an advanced form of brain cancer – typically arrive at the doctor's office complaining of headaches and weakness; a good many might also have experienced a seizure. Within a year, approximately half will be dead.

Progress against glioblastoma has been halting, and even modest improvements in treatment, like a median extension of survival of two and a half months, are greeted as breakthroughs. And so it was that, in 1992, a 51-year-old woman, whose husband was a prominent lawyer and chair of the board of the San Diego Cancer Center, was diagnosed with glioblastoma. Conventional therapies were unsuccessful in controlling her disease, and the woman sought enrollment in a gene-transfer study being planned at the San Diego Cancer Center. The problem was that the protocol had not yet been reviewed by the RAC, which met every three months; the patient was expected to live no more than two.

A former client of the woman's husband knew Iowa's Senator Harkin, and offered to contact him to see whether political pressure might be brought to bear. Shortly thereafter, Harkin forwarded a letter to NIH head Bernadine Healy requesting that she waive normal RAC review, though the developer of the intervention, Ivor Royston, had previously stated that he was “not optimistic that this therapy will work for his patient.”

RAC members bristled at Healy's involvement; they furthermore considered the supporting preclinical data insufficient to justify the protocol.

Introduction

French and British researchers have treated X-linked severe combined immune deficiency syndrome (X-SCID), otherwise known as “bubble boy” disease. Italian and UK researchers have also treated a related disease, ADA-SCID. Though only three patients were enrolled in the study, Swiss and German researchers have treated yet another severe immune disorder, chronic granulomatous disease. And some commentators believe American researchers are on the cusp of a durable treatment for hemophilia B. In these instances, it would appear that gene transfer – briefly, the administration of genetic materials to human beings – has finally earned the title of gene therapy.

But the field's development has been, and continues to be, a long, strange trip. As I write, the most visible name associated with gene transfer, W. French Anderson, is serving a fourteen-year prison sentence on child molestation charges. Another leading figure, James Wilson, has nearly finished a five-year, FDA-imposed ban on leading clinical studies. Other sanctions in the field's thirty-year history include one of the earliest ever violations of rules for human research issued by the US Department of Health and Human Services (Martin Cline, for initiating a study without proper IRB review), and a widely publicized rebuke of two other leading figures (Ronald Crystal and Jeffrey Isner) for not reporting trial deaths to the NIH. In 1995, a high-level panel at the NIH faulted the field for rushing into clinical trials. 5 In 2000, two prominent researchers editorialized in the pages of Science magazine “gene therapy has many of the worst examples of clinical research that exist.”

Introduction

In 2000, Lancet published results of a phase 1 study testing augmerosen in volunteers receiving the standard chemotherapy dacarbazine for melanoma. Augmerosen belongs to a class of genetic therapies – called antisense agents – that work by blocking expression of specific genes. In this case, the target was BCL-2, which prevents cells that have sustained DNA damage from undergoing “cell suicide.” The idea of the study, then, was to use augmerosen to release the genetic “brake” on cell suicide so that tumor cells would die after chemotherapy.

The study results were favorable: augmerosen combined with the chemotherapy drug dacarbazine proved safe even at the highest dose tested, with no volunteers developing life-threatening and/or unanticipated toxicities. In addition, six of the fourteen volunteers showed tumor shrinkage, and for two others, tumors stopped growing. In one figure, numerous thick, pigmented growths crowd a volunteer's forearm before the start of treatment; after, the tumors are reduced to a series of small moles. Whereas patients treated for advanced melanoma might typically live another five months, these study volunteers survived a median of twelve.

The study was valuable in the way phase 1 studies are typically valuable. According to one authoritative source, “the primary purposes of classic phase I studies are to investigate toxicity of organ systems involved, establish an optimalbiological dose, estimate pharmacokinetics, and assess tolerability and feasibility of the treatment. Secondary purposes are to assess evidence for efficacy, investigate the relation between pharmacokinetics and pharmacodynamics of the drug, and targeting.

Introduction

In June of 2001, a team of gene-transfer researchers led by Matthew During and Michael Kaplitt presented the RAC with a clever new strategy against Parkinson's disease. The second most common neurodegenerative disorder in North America, the cardinal symptoms of Parkinson's disease – tremor, rigidity, and inability to initiate movement – are caused in part by the excessive firing of a structure deep inside the brain called the subthalamic nucleus. The researchers proposed to genetically modify this structure with a gene encoding the inhibitory neurotransmitter, glutamic acid decarboxylase (GAD). With the subthalamic nucleus churning out GAD, the investigators postulated that nearby brain structures would quiet down, and Parkinsonian symptoms would abate.

The protocol required that a surgeon insert a needle through the volunteer's brain and inject small quantities of vector. But you didn't need to be a brain surgeon to recognize the peril. This would be the first administration of AAV vectors to the brains of non-terminal patients, and the investigators could not rule out the possibility that the vector might trigger an autoimmune reaction. Nor, according to members of the RAC, had the investigators decisively established the efficacy of their intervention in non-human primates. Though the RAC never formally advised against initiating the study, controversy trailed the team out of Bethesda. When the study was initiated in 2003, several leading Parkinson's disease researchers denounced it as “a crazy experiment” and “terra incognita.”. Another accused the lead researcher of “raising hopes in people with minimal evidence of benefits.”

Introduction

Monday morning, September 13, 1999, a team of physicians led by Steven Raper and James Wilson at the University of Pennsylvania threaded a thin tube through an incision in the groin of human subject OTC.019 to administer thirty milliliters of a modified adenovirus to his liver. The subject, like the seventeen who had preceded him, suffered from a rare hereditary disorder, ornithine transcarbamylase deficiency (OTCD), in which patients accumulate toxic levels of ammonia. Half of all infants born with severe forms of the disorder die within seventy-two hours of birth. However, participants in this study suffered from a mild form – one that could be controlled with medication and a restricted protein diet.

Like the previous subjects, OTC.019 developed the usual symptoms of viral infection: achiness, fevers, and a headache. Eighteen hours after the infusion, however, the response of OTC.019 headed into unfamiliar territory. He began showing signs of jaundice, a troubling sign for someone with a liver disorder. He became disoriented and agitated. Tests indicated dangerous levels of ammonia in the volunteer's blood, and soon thereafter he slipped into a coma.

Thirty or so hours after receiving the vector, OTC.019 began hyperventilating. Elevated oxygen threatened to increase the rate at which his body broke down the proteins released by his deteriorating blood cells; this, in turn, would drive his ammonia levels higher, portending brain damage. To control his breathing and reduce his blood oxygen, physicians placed OTC.019 on a ventilator. But OTC.019 continued to hyperventilate.

Introduction

In 2001, a report in Nature Genetics raised the possibility that a rare, hereditary form of blindness, Leber's congenital amaurosis (LCA), might soon have a cure. A team of researchers led by Jean Bennett of the University of Pennsylvania's Scheie Eye Institute had successfully restored vision to three dogs with LCA. According to news stories, the report “electrified” families with the disease; said one mother of an LCA child, “we are bursting at the seams.” Word spread like “wildfire,” according to Bennett, who received hundreds of inquiries from expectant parents.

The next year, the Alliance for Eye and Vision Research and the Foundation Fighting Blindness dispatched a team of researchers to meet with lawmakers during hearings on NIH appropriations. Traveling with them was Lancelot, one of the Briard mix dogs whose blindness had been partially corrected. One member of the research team, Cornell's Gustavo Aguirre, directed lawmakers to note the dog's posture: Lancelot always stood to one side – he favored his right eye because his left had served as the uncorrected experimental control. “If any of the investigators in the study had not received funding from the National Eye Institute/NIH,” Aguirre stated, “this amazing breakthrough would never have come to pass. Without increased funding, much of this promise will languish on the laboratory bench.”

The hearings followed one year on the heels of the first leukemia diagnosis in the Paris X-SCID study, and two years after Jesse Gelsinger's death. In addition to embodying the promise of vision research, Lancelot's eye also projected a favorable image for the embattled field.

Introduction

Adenosine deaminase-deficient severe combined immune deficiency (ADA-SCID) is a parent's nightmare. The disease is caused by a deficiency in an enzyme responsible for breaking down toxic metabolites that would otherwise kill lymphocytes. Untreated, infants develop recurrent infections and fail to put on weight; they rarely live past two.

But to scientists trying to establish proof of principle for gene transfer, ADA-SCID is a dream: only small amounts of gene correction are needed to restore immune function, tissues are easy to procure for genetic modification, and theoretically, treated cells should have a survival advantage over those that are not corrected. The attractiveness of ADA-SCID is so high, and its incidence so low (approximately one case per 100 000 births) that researcher Stuart Orkin once commented “more [gene transfer researchers would soon be] working on ADA deficiency than there are patients who have it.”

But there's a catch. Since the late 1980s, a relatively safe and effective enzyme replacement therapy – PEG-ADA – has been available for ADA-SCID patients. Denying PEG-ADA to children in gene-transfer trials would be unethical, because it would expose them to the risk of relapse. Yet concurrent treatment with PEG-ADA would have scientific costs: it would confound interpretation of the subjects' responses (if volunteers improved, how would investigators be able to tell whether this was owing to the gene transfer or the enzyme replacement?)and it would sustain uncorrected cells (thus dampening the potential therapeutic effects of gene transfer).

Throughout the 1990s, several teams attempted ADA-SCID gene transfer. Each, however, used concurrent enzyme replacement therapy and none produced clear therapeutic successes.