The tgAAC94 Study Consent Form: Raising More Questions than it Answered

We previously posted, most recently here, about the unfortunate case of a young woman who died after participation in a Phase I safety study of a gene-therapy agent for inflammatory arthritis, tgAAC94, made by Targeted Genetics.

An astute comment on our last post directed me to the actual consent form for this study, available here, courtesy of Citizens for Responsible Care and Research. Review of the form, coupled with what is already available from the media about this case, (see discussion in previous post and links backward), raised even more questions about this benighted human experiment. These included questions about what was in the form, and what the form left out.

Did the Consent Form Exaggerate the Safety of the Treatment?

The consent form included various statements that seemingly were meant to persuade study subjects that the treatment was safe, even though the treatment's safety was unknown, and the ostensible purpose of the study was to test its safety.

First, it included statements in the "Study Agent" section that seemed to suggest that the adeno-associated virus (AAV) which would transfer the gene was innocuous:

AAV infects many people in everyday life, but does not cause disease in humans. Although tgAAC94 was modified from AAV, tgAAC94 cannot grow in your body because all the AAV genes, including those that it needs to grow, have been removed.

Then, the "Risks, Hazards, and Discomforts" section seems to contain multiple arguments why the treatment may be safe, which took more space than discussion of its possible hazards.

It noted that the AAV vector (without the gene for etanercept) could produce an "immune response," but this causes "no side effects" in previous studies.

Targeted Genetics Corporation has used AAV to introduce genes into 200 people. These included about 140 subjects with a genetic disorder called cystic fibrosis (CF) who received does of a similar AAV vector into the nose, maxillary sinus, and lung, and about 65 healthy volunteers who received an injection of a similar vector into the muscle in an HIV vaccine study. Some subjects administered the highest doses developed an immune response. This immune response consisted of elevated proteins that interact with AAV in the blood. No side effects related to the development of this immune response have been noted so far.

Later, it tried to minimize the relevance of reports that high doses of AAV could cause cancer in animals.

The U.S. Food and Drug Administration (FDA), which oversees clinical studies of investigational drugs in the United States, was made aware of an animal study where [sic] a number of newborn diseased mice injected with very high doses of an AAV vector developed liver tumors. This was a research study using an AAV vector that was not designed or produced for human use. The vector used for the mouse study did not contain the TNFR-Fc gene and it was not designed to treat inflammatory arthritis. Tumors have not been observed in other similar studies of different types of mice injected with higher doses of AAV vectors and watched for over a year. There have been no reports of tumors in the limited number of human subjects who have received an AAV vector. Tumors have not been reported in any other animal studies of AAV vectors, but the number of studies that have been done so far is small. It seems unlikely that the tumors are linked to the AAV vector, but we do not know for sure.

It reported that injection of tgAAC94 appeared safe in animal studies:

A single dose of tgAAC94 has been injected into the joints of both rats and monkeys without raising any safety concerns. A single dose of a different AAV vector, containing the rat version of TNFR-Fc, has also been injected into the joints of rats with arthritis without causing any problems.

It noted that tgAAC94 had been given to small numbers of people without ill effects.

A single injection of tgAAC94 has been given to the joint [sic] of approximately 10 humans in another study of tgAAC94 without raising any safety concerns.

It noted that repeat injections of tgAAC94 given to animals only produced mild side-effects.

Repeat injections of tgAAC94 into joint have been given to rats once a month for three months. After the second injection of tgAAC94, approximately 20% of the rats developed mild swelling to the joint [sic] that resolved after a few days.

Only at the very end were there these two sentences.

There may be other not-yet-identified side effects that could occur during the time you participate in the study or years after receiving the study agent. Unknown side effects could be mild, serious, or life-treateneing, and could result in pain, discomfort, disability, or, in rare circumstances, death.

Thus, much of the ostensible discussion of possible adverse effects of the treatment consisted of arguments that the treatment was safe, again, made before any safety data had been collected. This seems particularly inappropriate for a study designed to assess the safety of treatment.

Did the Consent Form Imply that Patients Would Benefit from Study Participation?

The study was ostensibly only meant to test the treatment's safety, not its benefits. Whether the treatment has any benefits for humans has never been assessed.

However, the "Study Agent" Section included,

By injecting tgAAC94 directly into an affected joint in your body (called the target joint), we hope it will help the body make a protein that stops the inflammatory process and reduces the progressive joint destruction and resulting disabilities associated with inflammatory arthritis.

Furthermore, in the "Risks, Hazards, and Discomforts" section are statements that, rather than relating to the agent's possible risks, seem to draw parallels that argue for its effectiveness.

The gene that is transferred to the body codes for a protein that is the same as the approved medication called etanercept or Enbrel. Etanercept has been given by an injection under the skin to many people with inflammatory arthritis with remarkable improvement in their symptoms.

It also included a statement about use of a gene transfer therapy for rats,

The arthritis in these rats seemed to improve.

Thus, the consent form clearly contained arguments that the treatment could be beneficial, even though its benefits had never been assessed. Furthermore, some of these arguments were interspersed through a section of the consent form that was supposed to discuss possible adverse effects of the treatment, possibly distracting the reader.

Could A Research Subject Understand the Form?

The form contained numerous scientific terms and used technical jargon. It is not obvious that participants who were not familiar with biology or medicine could understand most of it.

The form also raised questions about what it did not include.

Who Actually Designed and Implemented the Study?

The form did identify the study's sponsor, i.e., the organization that paid for the study to be done, as Targeted Genetics Corporation.

The version of the form available on the CIRC web-site listed the study investigator as Robert G Trapp MD, with an address in Springfield, Illinois, and gave a 24 hour phone number for him. It identified the study site as the Arthritis Center, at the same address. Dr Trapp appeared to be one of the physicians enrolling patients, and the Arthritis Center appeared to be the name of his practice location. The form did not identify Dr Trapp as the Principal Investigator, i.e., the person in charge of the whole study. Apparently, his role would be that of "research doctor," a term used frequently later in the form. It is possible that different versions of the form used at other sites identified other "investigators."

However, the form was silent about who was in charge of the study, who designed the study, or who would implement the study. In general, clinical research studies have a Principal Investigator, or perhaps two Co-Principal Investigators, who take responsibility for the design, scientific conduct, analysis, and dissemination of the study. Furthermore, clinical research studies are generally carried out by study teams, or could be based at a medical school, research institute, teaching hospital, or contract research organization (CRO).

The form, in fact, did not identify anyone as the Principal Investigator, or anyone other than Dr Trapp as being involved with the study. It did not explain who had designed the study. It did not mention who would carry out the study, for example, by collecting data, ensuring that patients had appropriate follow-up, entering data, analyzing data, writing up results, etc.

The form did identify the Institutional Review Board (IRB) which reviewed the study and the consent form itself as the Western Institutional Review Board (WIRB), stating "WIRB is a group of people who perform independent review of research." This statement seems disingenuous, since WIRB is also a profit-making company (see Emanuel EJ, Lemmens T, Elliott C. Should society allow research ethics board to be run as for-profit enterprises. PLoS Med. 2006 July; 3(7): e309. DOI:10.1371/journal.pmed.0030309).

Thus, although the form identified the "investigator" who would be the study subject's first point of contact with the study, it failed to identify all the other people who were doubtless involved in designing, and carrying out the study, and failed to identify anybody who had ultimate authority for the design and implementation of the study.

Summary

It seems the more we learn about this case, the less we realize we know. There have been questions all along about the appropriateness of the consent form and how it was presented to potential study subjects: whether it de-emphasized possible risks, hinted at benefits that may not exist, and was written and administered in a way that was not conducive to somber consideration by potential research subjects.

Now, it seems that the identities of the people who actually designed, controlled, and implemented this study are mysterious as well. I do not understand how experimental research can be done on humans without telling the research subjects who is actually responsible for the experimentation to be done on them. Furthermore, these identities remain mysterious even after a well-publicized NIH hearing on the fatal outcome of this study. I also do not understand how an official government hearing on this project did not involve, nor reveal, who these people were.

Perhaps some investigative reporting will solve some of these mysteries.

Meanwhile, this troubling and now too mysterious case should cause patients, physicians, and policy makers alike to re-consider whether we should continue to allow experimental testing of drugs or devices on humans to be done by those with vested interests in the success of the drugs or devices being tested.