Trends in Genetics, June 2000, 16/6, Doris Teichler Zallen
US gene therapy in crisis
The recent death of a young man, Jesse Gelsinger, in a gene therapy experiment has sent shock waves throughout the US research community. This tragic event has raised new questions about the prospects for human gene therapy. And it has also reopened old controversies about how best to guide the development of this field.
Human gene therapy experiments are being actively pursued in the US and around the world. Results, however, have been meagre*. Introducing genes into target tissue and getting them to work has turned out to be harder than anticipated (an assessment of the status of the field can be found in a report by S. Orkin and A. Molutsky at: http://www.nih.gov/news/panelrep.html). Despite the lack of efficacy, gene therapy researchers had come to believe that, at least, the procedures were safe'. However, the recent death of a research subject, Jesse Gelsinger, as a direct result of a gene transfer has challenged that view. Brought to wide attention by newspaper headlines such as 'Death Leads to Concerns for Future of Gene Therapy' and 'Some had said research moved too quickly’, the field of gene therapy is now under intense scrutiny from scientists, the government and the public.
Forging the infrastructure
The present policy framework for gene therapy research in the US can be traced to
events that occurred in 1973. In that year, the new-found ability of scientists to
recombine DNA between species raised fears that these new techniques might result in the
creation of microbes with dangerous properties. The scientific community took decisive -
and unprecedented -action: researchers declared a voluntary moratorium. An international
group of scientists conferred to set up procedures that would guide the continuation of
recombinant-DNA research. Human gene therapy was not high on anyone's agenda at that time.
When the voluntary moratorium was lifted in 1976, a new oversight body had been created.
Within the US National Institutes of Health (NIH) there was now a national-level body: the
Recombinant DNA Advisory Committee (or the 'RAC'). The RAC was established to formulate
standards for recombinant DNA research and to oversee its progress. Local biosafety
committees were set up in institutions where recombinant DNA experiments were underway to
work in conjunction with the RAC.
And there was a new partner in decision making: the general public. Alerted by extensive
media coverage, various citizen groups began to be formed to add their voices to the
discussions that were taking place'. In 1978, the RAC expanded to include 'public members'
- typically lawyers, ethicists, political scientists and consumer advocates - as a means
of incorporating a wider range of views into its deliberations.
Recombinant DNA research then proceeded in a fashion so dramatic that what many call a
biological 'revolution' was set in motion. As experience was gained, much of the oversight
responsibility was delegated to local biosafety committees or, in the case of agricultural
organisms, was transferred to the Department of Agriculture and the Environmental
Protection Agency.
Anticipating human genetic manipulation
In 1980, the possibility of applying genetic engineering procedures to humans
catapulted to the center of attention. The US Supreme Court decision allowing patent
protection for a living organism, a genetically engineered 'oil-eating' bacterium,
provided powerful financial inducements for private companies to expand their research
efforts. Also that year, a California researcher made a premature attempt to alter human
genetic makeup in order to treat the genetic disorder thalassemia. (This was the so-called
Cline affair, in which an investigator proceeded to do studies overseas in the absence of
approval from his home institution.)
Worried that scientists were embarking on 'a new era of fundamental danger triggered by
the rapid growth of genetic engineering', three leaders of the Protestant, Jewish and
Catholic religions requested that then President Jimmy Carter act swiftly to 'provide a
way for representatives of a broad spectrum of our society to consider these matters and
advise the government on its necessary role' 17 . The President's Commission, a body
impaneled (from 1980-1983) to study ethical issues in medicine and research, responded in
a report Splicing Life. That report strongly endorsed the acceptability of
research intended to treat, and perhaps cure, genetic iseases through genetic manipulation
of body tissues (somatic cell gene therapy). It was decidedly more cautious about genetic
manipulations that could alter gametes and be transferred to future generations (germline
gene therapy) or that might be used to 'improve' otherwise healthy-individuals
(enhancement gene therapy).
The Commission took seriously what it termed the 'Frankenstein factor': the public's
deep-rooted concerns that human genetic manipulation could lead to harmful outcomes for
individual research subjects or pose a threat to widely held societal values. It
-recommended that human genetic engineering activities should proceed only under the
watchful scrutiny of a body with both scientist and public representatives. The process of
deliberation was to be open. Secrecy was to be avoided. To underscore this point, the
report cited a particularly prophetic comment made by George E. Brown, Jr, a member of
Congress with expertise in science policy: 'If ... industry follows the path that
appears easier initially, the cloistered avoidance of other forces of society, it will pay
a penalty years hence should some event force a public inquiry."
Developing a multi-level oversight mechanism for gene therapy
The RAC responded to the Splicing Life report by establishing a Working
Group on Human Gene Therapy (later renamed the Human Gene Therapy Subcommittee). The
group's recommendations built on the preexisting policy framework.
The RAC was to be central in the oversight process. Gene-therapy protocols were to be
considered on a case-by-case basis, just as the RAC had done for almost a decade for other
forms of recombinant DNA research. Protocols would be reviewed first by local-level
biosafety committees and by institutional review boards. Already required by federal law,
institutional review boards would render judgements on whether human subjects could be
used in the research. The protocols would then proceed for national-level review by the
Human Gene Therapy Subcommittee and finally by the RAC itself. The presence of public
members on both the Subcommittee and the RAC would ensure that broader ethical and social
issues, not just scientific feasibility, would be considered. All meetings would be held
in public to permit input from interested individuals or members of citizen and consumer
groups. The RAC would make its recommendations to the Director of NIH, whose official
approval was required before the research could proceed. Thus, US policy regarding human
gene manipulation would emerge, over time, from the decisions rendered by the RAC
operating in full view of the general public.
In reality, RAC jurisdiction only extended to research conducted at institutions receiving
public funding for recombinant DNA research through the NIH. However, the expectation was
that private companies engaged in human gene therapy research would bring their protocols
before the RAC voluntarily, just as they had done for other types of recombinant DNA
research. Companies had seen it as an advantage to gain RAC endorsement of their research.
Along a parallel pathway, the Food and' Drug Administration (FDA), would also have a
central role in the approval of gene therapy experiments. The FDA has statutory authority
over any drug (including, in this case, any system for DNA delivery) intended for human
use. In contrast to the RAC, however, the FDA focuses on safety and efficacy, not ethics.
To protect the proprietary interests of the sponsors of research, the FDA is compelled by
law to carry out its work in private and keep all its data confidential.
The standards to be used by the Human Gene Therapy Subcommittee and the RAC in making
decisions had to be developed. The issues were complex and the uncertainties profound.
What were the real risks - and how safe was safe enough? Could patient privacy be
protected? How would the selection process be managed?
To help address this ethical dimension, the Subcommittee added a new element. It produced
a document, known as the 'Points to Consider in the Design of Human Gene Therapy
Protocols', that was incorporated into the NIH Guidelines". The Points to Consider
contain seven areas of questions for investigators to answer that would help determine
whether investigators had met the three ethical criteria undergirding all human subjects
research - beneficence, respect for persons and justice. (Beneficence refers to an overall
balance of benefit over harm. Respect for persons requires that adequate and accurate
information be given to each prospective subject and that the decision to participate is
uncoerced. justice is seen as the fairness of selection criteria for enrollment in the
study".) By 1985, the RAC was ready to render judgements in a controversial area of
research.
Three years later, the very first experiment to alter human genes wended its way through
the approval process. This experiment was not an attempt at treatment. It was a
study to insert a neomycin-resistance gene into lymphocytes of individuals with advanced
melanoma in order to mark the cells and determine their distribution and duration in the
body. The first attempt to use genetic manipulation to correct an inherited genetic
disorder was approved in 1990 (Ref. 12). This was a study to transform lymphocytes from
children afflicted with a form of severe combined immunodeficiency disease (SCID-ADA) by
introducing the gene for adenosine deaminase. Since then, the pace of research has picked
up remarkably. According to current estimates, about 350 experiments involving gene
transfer are now underway in the US. Similar research in progress in the UK and other
countries brings the worldwide total to over 400.
The protocols that have been approved employ different types of vectors and seek to alter
human cells, either in vivo or ex vivo, in a variety of ways (Table 1). Early on, the
expectations had been that most protocols would be concerned with single-gene disorders.
In reality, the emphasis has been on experiments designed to introduce new genetic
information into cells as a way to treat major health problems such as cancer and AIDS.
This emphasis reflects the increasing involvement of private industry as collaborators in
research efforts. Such collaboration is nearly inevitable because most university
scientists lack the resources to bring their research along from the lab bench through
clinical trials while meeting stringent FDA standards. However, industry's quest for a
profitable return on its investment leads it to prefer gene-therapy applications aimed at
widely prevalent disorders.
Reducing RAC oversight
Almost as soon as it started, calls came to modify the oversight process. These
efforts first aimed to eliminate the Human Gene Therapy Subcommittee so that protocols
would proceed directly from the local-level committees to the RAC. In November 1991, the
Subcommittee responded to the growing pressure by voting itself out of existence.
Then, a crescendo of criticism about the RAC itself began to be heard. Often criticism
took the form of complaints from biotech companies about possible delays resulting from
the RAC's quarterly meeting schedule. It was also likely that they were concerned about
the sometimes harsh critiques of scientific merit or of the proposed informed consent
process expressed at the public meetings. These critiques were often in sharp contrast to
the enthusiastic reports of disease-curing possibilities simultaneously provided by
biotech companies to the media.
The NIH also had its complaints. The media hype surrounding any protocol approval was
sufficient to drive up the net worth of a company and make it a prime candidate for a
lucrative buyout. The NIH found itself used to promote the financial self-interest of
private companies. It was an awkward and untenable position for the nation's premier
medical-research agency.
In 1995, NIH Director Harold Varmus, perhaps disillusioned by this link between NIH and
industry or by the low yield of much of the research, decided to examine the role of the
RAC itself. He convened an Ad Hoc Review Committee, chaired by scientist Inder
Verma. The Committee's Executive Summary can be found at:
http://www4.od.nih.gov/oba/adhoc-re.htm.
The Verma committee ultimately recommended that the RAC continue its work, calling it a
'credible forum for airing a wide range of public concerns'. It endorsed the streamlining
procedures, already in place, that ceased the RAC practice of case-by-case review of all
protocols. The RAC was to consider only protocols with novel features (e.g. those using
new vectors or targeting new diseases) or which moved the field into problematic areas
(such as in utero interventions or germline modifications). Additionally, the Verma
committee urged that the RAC 'continue to be provided with the data needed for monitoring
clinical gene transfer protocols'.
It was under this system of consolidated review that the ornithine transcarbamylase
deficiency (OTC) protocol, into which Jesse Gelsinger was enrolled, was approved by the
RAC in December 1995. It had come before the RAC because it possessed a number of novel
features: the disease targeted, the adenovirus vector used, the delivery directly into the
liver, and the use of relatively healthy individuals as research subjects. RAC reviewers
raised a number of concerns and made recommendations to alter the mode of delivery and to
improve the consent form.
In 1996, the RAC faced an unexpected challenge to its existence, one that came from the
NIH itself. Contrary to the recommendation of his Ad Hoc Committee, NIH Director
Varmus floated a plan to disband the RAC and replace it with a six- to ten-member
committee operating in house, but out of public view. This new committee would look at the
broader policy issues related to gene therapy but neither it nor the NIH Director would
retain any role in the evaluation of protocols. Through occasional conferences and from
the database, the NIH and the public would be informed of areas of promise as well as of
problems.
The Biotechnology Industry Organization (BIO) and individual biotech firms offered strong
support to the plan to dismantle the RAC. However, a surge of opposition to the proposal,
especially from consumer groups, forced a compromise. In the end, the RAC was retained but
it was reduced in size to 15 and, while it could discuss them, it could no longer vote to
approve or block any protocols. Nor would the NIH Director play any part in the approval
process. The only decision-making authority resided with the FDA. It is this system of
oversight that governs US gene therapy research to the present day.
The current crisis
What the death of Jesse Gelsinger has shown, beyond the still-to-be-determined details
of the reasons for the overwhelming response by his immune system to the adenoviral
vector, is that the present oversight system is seriously flawed. At a hearing held on 2
February 2000 by the Subcommittee on Public Health, US Senate Health and Education
Committee, it became clear that the marginalization of the RAC has produced a number of
unfortunate outcomes.
Gaps in communication with the FDA left the RAC with diminished ability to know what was
really going on or what actions, if any, had been taken on its recommendations. For
instance, the RAC was never informed that the FDA had authorized, a change in the
mode of administration of the adenoviral vector in the OTC protocol.
A pattern developed of failure to report serious adverse events to the NIH. The
researchers responsible for the OTC protocol did inform the NIH, and thereby the public,
as soon as the death occurred. However, many investigators and sponsors have either failed
to report serious adverse events to the RAC or have requested that such reports be
considered 'proprietary' and thus shielded from public view. According to an NIH audit of
adenoviral protocols following Mr Gelsinger's death, only 5% (39/691) of observed serious
adverse events, including deaths, were ever reported to the NIH (Varmus, H., letter to
Waxman, H., US House of Representatives, 21 December 1999 and Ref. 13). Investigators
maintained that the unreported events were disease-related, unconnected to the gene
transfer. All of them had been reported to the FDA.
The quality of the consent process has deteriorated. Once the consent process and
associated consent documents were no longer subject to public review, some investigators
reverted to the use of consent documents that the RAC had previously found to be
inadequate. (One example of this can be found in RAC minutes at:
http://www4.od.nih.gov/oba/dec96.htm; Item IX, Data Management.) Even in the OTC clinical
trial, investigators could not account for the deletion of information on animal deaths
from the consent documents.
There is a lot at stake. The NIH and the RAC itself have formed separate working groups to
determine what appropriate procedures and standards should be. Two areas are of prime
concern.
Ensuring public participation
The lesson is clear from the growing public resistance to genetically modified foods:
failure to guide gene therapy properly can lead to wide public disillusionment and even
opposition. Decisions made behind closed doors create a public distrust that ultimately
can obstruct entire lines of work. Unless there is public confidence in the way those in
authority make decisions regarding gene therapy research, there is the real risk that this
area, despite its promise, can be similarly derailed.
The argument has been often raised that gene therapy research should not be subjected to a
stiffer standard of oversight than other areas of frontier research. This view overlooks
the fact that genetics has an unfortunate history. In the US as elsewhere, people have
been punished for their genes under the banner of eugenics movements. Moreover, possible
in utero, germline and enhancement applications raise ethical and social dilemmas and
require that whatever decisions are made be open to public view.
The RAC is still widely respected. It would appear essential that RAC - with its
well-established links to the public - be returned to a central role in the oversight
process with its ability to veto protocols restored. The Internet can be a valuable tool
in communicating the status of gene therapy research to public audiences.
Reporting procedures for adverse events
Labeling adverse events as proprietary information might make sense from a business
standpoint, but it carries with it unacceptable consequences. Quite simply, it puts at
risk those generous human beings who are putting their bodies on the line on behalf of
science. Knowing what has happened to others in the same or similar research is vital
information that prospective subjects are entitled to have. To deny prospective subjects
this information makes a mockery of achieving informed consent.
Furthermore, without full and accurate information on the status of clinical trials, it is
impossible for the RAC or any other similar public group to make sound decisions or to
identify potential dangers. Part of the purpose of any clinical trial is to look for
patterns of unexpected outcomes and to explain them. What seems at first to be simply a
disease-related event can, in conjunction with other similar events, provide enough data
to reveal that certain treatments or practices may be harmful. It is hard to justify
keeping such matters a secret when human lives are at stake.
Some researchers have offered the view that a publicly accessible database could
compromise confidentiality. The current NIH Guidelines have anticipated this concern.
Investigators are required to have appropriate procedures in place to protect the privacy
of research subjects. Additionally, prospective subjects must be alerted to the fact that
they may be sought out by the media. Serious adverse events are to be communicated with
any identifying data removed.
These efforts seem to have worked. When individuals in gene therapy studies have become
known, it has been by their own choice. Ashanti DeSilva, a subject in the first SCID-ADA
experiment, has agreed to be featured in numerous articles. One day after the RAC
discussed gene therapy for limb-girdle muscular dystrophy in September 1999, a photograph
of the first subject treated was a front page story". If investigators are not loathe
to share their work when it can bring credit to them and their sponsors, it would appear
hard to justify any refusal to share data when possible harms have occurred. The FDA and
NIH are now working to clarify the reporting requirement so that adverse events are
transmitted to both agencies. The safeguards already in place should be sufficient to
provide the desired privacy protection.
This might well be a defining moment for human gene therapy research not only in the US
but around the world. A credible partnership between the public and scientists is needed,
not only to deal with the current challenges of the field but to allow it to face those of
the future when gene therapy experiments move into uncharted and contentious areas. Other
countries may wish to consider the US experience as they evaluate or design their own
systems of gene therapy oversight.