Ethical Issues in HIV Vaccine Trials

 

Chapter 11

Real risks

 

A preliminary disquisition on The Other


In considering the risks detailed in the following sections, those who hold the Antithesis position would wish readers and researchers to keep in mind the wise little proverb: "The burdens that appear easiest to bear are those that are borne by others." This is particularly important to keep in mind because virtually all volunteers in phase III HIV vaccine trials will be persons from groups which may easily be seen by research sponsors as The Other. The doctors at Buchenwald and Auschwitz, for example, had little trouble distancing themselves from the pain that they inflicted on subjects in their medical experiments because the subjects were all Other: they were Jews or Poles or homosexuals or Gypsies or Slavs or mental patients, and especially important, they were all prisoners, they all had tatoos, and their heads had all been shaved. They did not look like "Us," and they did not act like "Us". It was easy, though of course not morally justified, for doctors to perceive the prisoner-subjects as The Other.

I do not wish at all to hint, not even indirectly, that today's research sponsors of HIV vaccine trials are somehow like the Nazi doctors. Today's vaccine researchers are highly compassionate and are not racially prejudiced as were the doctors of the Third Reich who worked in Auschwitz and Buchenwald and Birkenau. No one who has seen the effects of the pandemic in Uganda or Rwanda or any of the other hardest hit nations could fail to be greatly moved by the deep tragedy of it, and by the powerful sufferings of whole families and whole regions. What I do wish to underscore, though, is how easy it is for ordinary human beings to feel differently toward The Other than we feel toward someone whom we perceive as Us. Whenever we perceive someone as Other we become less likely to understand them, and even less likely to make an effort to understand them. We are less likely to take their sufferings and their burdens as seriously as we take the sufferings and burdens of people whom we perceive as Us. I am inclined to think that this tendency in us weak and ignorant human beings is a rather prevalent tendency, but it is certainly not a good one. It stems from a weakness and a narrowness of perception, and we have a strong moral obligation to strive to overcome it. We must strive to appreciate the sufferings and burdens of others as much as we appreciate the sufferings and burdens in our own lives. Research sponsors particularly should take very seriously the insight of Mrs Eva Mozes-Kor.

Eva is still alive. She and her twin sister Miriam were human subjects in Dr Josef Mengele's experiments on twins, at Birkenau.

 

To look back at my childhood is to remember my experiences as a human guinea pig in the Birkenau laboratory of Dr Josef Mengele. To recount such painful memories is to relive the horrors of human experimentation, where people were used as merely objects or means to a scientific end.

 

Mrs Mozes-Kor proceeds to describe the horror of these experiments in which she and her sister were subjects, some of which involved studies in genetics and some of which concerned germ warfare, and how they were conducted. Interested readers can find and read (in the book cited below) her moving six page account of what it was like to be one of those twins, but for our purposes here I want to focus on her conclusions.

 

I hope that what was done to me will never again happen to another human being. This is the reason I have told my painful story. Those who do research must be compelled to obey international law. Scientists should continue to do research. But if a human being is ever used in the experiments, the scientists must make a moral commitment never to violate a person's human rights and human dignity. The scientist must respect the wishes of the subjects. Every time scientists are involved in human experimentation, they should try to put themselves in the place of the subject and see how they would feel. The scientists of the world must remember that the research is being done for the sake of mankind and not for the sake of science; scientists must never detach themselves from the humans they serve. I hope with all my heart that our sad stories will in some special way impel the international community to devise laws and rules to govern human experimentation.

 

We will examine later in this book a document, sponsored and published by the Council for International Organizations of Medical Sciences (CIOMS) and the World Health Organization, titled International Ethical Guidelines for Biomedical Research Involving Human Subjects, published in its final form in late 1993, after long years of consultation and research. We will be exploring this document in detail, but for now it would be well to consider Eva's key words: "Every time scientists are involved in human experimentation, they should try to put themselves in the place of the subject and see how they would feel."

I again wish to emphasize that the human experiments performed by the Nazi doctors in the concentration camps are not here being compared to today's HIV vaccine trials. These are two entirely different categories of human experiments. The former were done with utter disregard for human beings and human suffering, and the latter are being planned by compassionate researchers with the greatest concern for the well being of humanity in general, and the research subjects in particular. The only similarity is that both involve the participation of human beings as research subjects, and whenever human beings participate as volunteers in research protocols, there is some degree of risk that they will be seen more as a means than as ends, more as an Other than as an Us. Eva's words are reminders to us to overcome this tendency to see other human beings as Other, particularly when those others look and act differently than we do. Eva asks us to "put ourselves in the place of," particularly when it comes to medical research involving human subjects.

The unspoken conclusion to Eva's words is a variation on the Golden Rule (to do unto others as you would have others do unto you). The variation, sometimes called the Silver Rule, is phrased thus: Do not do unto others what you would not wish others to do unto you. This is Eva's main request to researchers: Put yourself in place of the Other. Anticipate the Other's concerns. Do not do to them what you would not wish done to you. If ever any principle should be kept foremost in the minds of those who undertake human subjects research, this is that principle.

 

Now let us consider, from the perspective of the Antithesis position, some of the potential hazards, large and small, more probable and less probable, that may have to be borne by those who would choose to volunteer in these vaccine trials.

11.1 No future protocols

The opportunity to participate in a future vaccine research protocol is, in all likelihood, effectively sacrificed when one volunteers for these present experiments. Should a more promising candidate vaccine be developed in the future - and future candidate vaccines would almost certainly be in some ways superior - today's volunteers would be excluded from participating in that research. Reasons for exclusion might vary, but would probably be built around the notion that these subjects had been "contaminated" by another significant variable, namely, exposure to a previous vaccine. Volunteers in today's experiments should know that they will probably not be able to participate in future vaccine research. This should be considered a "harm" for them, but (in my opinion) perhaps not the most important one. That this "harm" would actually happen to subjects in the present studies is a virtual certainty.

11.2 Immediate systemic reactions

A usually mild, but still real, systemic reaction sometimes immediately follows upon a vaccination procedure. Sometimes the reaction is as mild as a headache, local pain at the site of the injection, and a fever, but sometimes the reaction can be more severe, including symptoms as serious as convulsions. Subjects will want to be aware of these possibilities.

11.3 Potential immune tolerance

The possibility exists, though it may be a relatively small one, that a subject who receives this present candidate vaccine could become immune to the effects of any future HIV vaccines. This circumstance, if it were to come about, would be called "immune tolerance" and would have the effect of rendering a person

immunologically unresponsive to the original antigen [HIV], i.e., viral protein(s) capable of inducing immune responses. Immune tolerance may prevent the immune system from mounting any response upon "seeing" this antigen again.

No one presently knows how likely or unlikely this occurrence might be, nor will they know until phase III trials are well under way, but it is within the realm of possibility.

If this were to occur it would have two serious consequences: 1) It would mean that this person's immune system would no longer mount any response at all against HIV. It would mean that if this person were to become infected with HIV, the immune system would not fight the virus at all, so progression to disease would likely be more rapid than normal. 2) Immune tolerance would also mean that if an effective vaccine against HIV were to be developed and proven successful at some time in the future, this volunteer would not be able to use it because they would not respond to it at all. It would do them no good. They would probably not even be offered a future successful vaccine, especially if it is a vaccine of the whole attenuated virus type, for fear that it would simply infect the person and cause them to develop AIDS.

This risk, though perhaps small, may be a significant one in the minds of some potential volunteers.

11.4 Enhanced infectivity

A more important hazard for volunteers is the uncertain possibility of "antibody enhanced infectivity".

It has been shown by a number of investigators that some HIV antibodies actually help HIV enter host cells, primarily monocyte cells. This phenomenon is called antibody enhanced infectivity. Antibodies that enhance HIV infectivity have been identified in the serum of HIV-infected patients and in HIV infected and immunized animals.

 

This potential hazard would mean that subjects inoculated with the candidate vaccine could actually be at greater risk for disease than non-vaccinated persons. It would mean that persons who have developed antibodies to HIV (which is what most vaccines stimulate the body to do), and who then subsequently became exposed to HIV through one of the usual routes of infection (which is what phase III subjects will be doing), would then be

1) more likely to become infected (that is, more easily infected) with the virus,
2) more likely to progress to disease,
3) more likely to progress to disease faster, and/or
4) perhaps more likely to develop a more serious form of disease than non-vaccinated persons.

Antibody-enhanced infectivity in candidate viral vaccines is not, unfortunately, either unknown or unprecedented. According to studies cited below, the phenomenon has been shown to exist in at least three or four candidate vaccines for various viral diseases. This hazard became enough of a danger that it blocked research into a vaccine against respiratory syncytial virus (RSV, the major cause of infantile pneumonia) thirty years ago. When an RSV vaccine was undergoing testing in infants in the 1960s "researchers found to their surprise that a strong immune response against the virus can, for unknown reasons, actually enhance the disease". Research was stopped because the hazard was considered too great a risk for the subjects in the trials. This candidate RSV vaccine had already been through phase I and II trials and had been "shown to be nontoxic and antigenic in adult volunteers and nontoxic in infants and children before the larger efficacy study was initiated. The apparent sensitizing effect of the vaccine was entirely unexpected."

The problem of antibody enhanced infectivity has also "been shown to be important in other viral diseases such as dengue fever and rabies". It was a problem in some earlier measles vaccines as well.

I believe this risk, if it is indeed as real as the literature suggests, is a significant one. Subjects in phase III trials may have the experience of the subjects in phase I and II trials to give them some estimate of the likelihood of this occurrence, but perhaps not (as we have just seen with the RSV experimental vaccine). The number of subjects in phase I and II trials may simply be too small to demonstrate the risk, or the length of the trials too short (only one or two years) to establish just how serious or likely this risk actually is. Or the subjects in phase I and II trials may simply be at too low a risk to demonstrate the enhancement that would occur only in the presence of encounter with wild virus. In any case, subjects in phase III trials will probably have no more information on what level of risk to expect for antibody-enhanced infectivity than we have right now. (The problem of enhanced infectivity with the measles vaccine mentioned above was not discovered to be a significant problem until long after phase I and II trials had been completed, and the vaccine had already been administered to "hundreds of thousands of recipients". ) In sum, the risk for antibody-enhanced infectivity in HIV vaccines may be fairly high or not very high. We simply do not yet know. And unfortunately we cannot know until we test the candidate vaccines in human populations.

I imagine that some volunteers will see this risk as important enough to decline participation in the trials. When it is made clear to them that "antibody-dependent enhancement of HIV is [truly] a genuine concern," they may conclude that this risk alone offsets the potential (perhaps small potential) for protection against HIV infection offered by the vaccine. All potential volunteers should be made to fully understand this potential hazard.

11.5 Discrimination

11.5.1 The problem

Social discrimination due to the volunteers' new HIV antibody status should be expected. Probably for the rest of their lives these volunteers will be HIV antibody positive (HIVAb+) as a result of being given an HIV vaccine. They may find that when they wish to donate blood or other organs or tissues, or enlist in military service, or apply for health or life insurance, or travel internationally, or even be treated by certain health care or dental practitioners, that they are subjected to unfair discrimination. They may be treated with hostility by neighbors, friends, or work associates. They may have difficulties finding a marriage partner.

 

The potential for harm to one's personal reputation also comes with trial participation. The peers of potential participants may view cooperating with government researchers as evidence of gullibility or foolhardy complicity with authority. Participants should know that such social discrimination could seriously affect their everyday lives and may not be compensable.

 

Moreover, the degree of social discrimination faced by volunteers in phase III trials is liable to be significantly higher than the risk of discrimination faced by those in phase I and II trials, since most of the subjects in phase I and II trials will probably be from the middle classes of a society and will also be from groups at lower risk of infection.

Already, even before trials begin, the social discrimination suffered by persons who are HIV+ is severe. In Thailand, for example, it is reported that one of the biggest problems facing HIV+ persons is the social discrimination leveled against them. "Disclosure of a positive status," says one account, "usually leads to the loss of employment, isolation from the community, and problems within the family". It may lead to numerous other problems as well. And we can be sure that the fear of HIV+ persons is not unique to Thailand.

Researchers will probably provide each volunteer with official documentation certifying that their HIVAb+ status is due to their participation in a vaccine trial, and will probably also provide a toll-free phone line for doubters to call and confirm. Researchers will probably also make pre-arrangements with insurance companies and other agencies so they will accept such documents, unless there are extenuating circumstances. But there may be good reason to doubt that these documents would ever be taken seriously. In the first place, documents can be forged and phone numbers can be bogus, so an effective method of insuring authenticity would need to be devised. But more importantly, the persons to whom you would present such documents might not accept them because they may be secretly thinking, "Of course I see that you were in a vaccine experiment a few years ago, and that caused you to become HIVAb+, even though you were not infected with the virus. But perhaps you have also become infected with the real virus in the interval since that vaccine experiment, and I have no simple and quick way of testing for that. Also, I wonder if the researchers might have chosen you to participate in their vaccine trials because you are a person who is at higher than normal risk for HIV infection. So I think I'll just play it safe and not hire you, not let you join the army, not let you donate blood, not sell you our insurance, not etc etc."

This hypothetical person's thinking is not illogical, since the volunteer could very well have become really infected with the virus since participation in the vaccine experiment. In fact, it would be a safe bet that the volunteer probably has engaged in some risky behaviors. As we will see below, most volunteers would not have been chosen as subjects in the vaccine trials if the researchers did not have good reason to believe that they would in some manner expose themselves to the virus now and then. Phase III vaccine trials must, after all, be done with subjects who are at some significant risk of exposure to the virus, otherwise the research would be pointless. Thus, our hypothetical secret thinker is correct in reasoning that this candidate is probably a person who is, or who was at some fairly high risk of becoming HIV infected. In any event, whether we consider this thinking to be correct or not, it is very likely that this sort of thinking will occur, namely, thinking which leads to discrimination against volunteers.

It is indeed reasonable to assume that such real discrimination will occur at some times in that volunteer's future life. In fact any interested reader can peruse some first person accounts of real discrimination against HIV+ persons who live in developing nations. Richardson and Bolle have collected a wide variety of such personal interviews in their book titled Wise Before Their Time. Some persons have even been socially stigmatized by their mere association with an "AIDS" trial, and the potential for such discrimination is unlikely to diminish.

In order to help minimize such social discrimination, research sponsors should provide volunteers with all the protections they can offer, including certificates, toll free phone numbers for confirmation, and free ELISA and Western Blot tests for HIV antibodies for as many years as necessary. Unfortunately, according to one study, in up to 40 per cent of vaccine trial volunteers, even the more specific Western Blot test will be unable to distinguish between a positive test that is due to receiving a vaccine, and a positive test that is due to being actually infected with HIV. This means that, even with the tests presently used to determine whether a subject is actually HIV+, or just appears to be HIV+ because of having received a vaccine, up to 40 per cent of volunteers could be mis-identified as actually infected with HIV, using both of today's standard HIV tests, the ELISA and the Western Blot. More complex candidate vaccines are already being proposed for new studies this year, vaccines in which it will be even more difficult to distinguish apparent HIV infection from actual HIV infection.

There are viral detection methods, such as antigen capture assay, viral culture and the PCR test, however, which test for the presence of the actual virus, rather than just for the presence of antibodies to the virus. These methods are able to distinguish between a vaccine-induced positive response, and an infection-induced positive response. Unfortunately, these methods are more expensive and require some degree of laboratory sophistication and may, therefore, not be as readily available in developing nations. If possible, however, some sort of free viral detection method should also be made available to volunteers for as long as necessary. Free intervention services and/or legal services in the event of discrimination should also be provided to all subjects.

These are some of the precautions that trial sponsors will probably provide to their volunteers. But volunteers should also be made aware of the very real likelihood that in their future lives they will probably suffer some forms of discrimination, and that some of these acts of discrimination (for example, in hiring or firing, in housing, in blood and organ donations, in application for military service, in application for some kinds of jobs, in application for insurance, for travel visas, and so on) will not be resolved in the volunteer's favor. Persons seeking travel visas to Russia, for example, may at some point be required to be tested for HIV. Any applicants who test positive would simply not be granted entry visas.

Volunteers who live in developing nations face additional struggles. In many cultures a right to privacy and to confidentiality about personal medical information has not been as strongly established as in some industrial nations. (Although not typical, in some regions of China, for example, there is sometimes public posting in medical clinics of "highly personal information about individuals' health status, inoculations, women's menstrual cycles, and other data". )

In addition, it will probably be commonly known in many communities who the people are who are participating in these vaccine trials; or at least it may be easily discovered who they are. Some community members might easily make uncomplimentary assumptions about persons who volunteer for HIV trials: people might assume that these volunteers had been "dirtied" by the vaccine, or that they had been infected by it, or that they had been chosen as a volunteer because they were at high risk for infection, and so on. And additionally for those subjects who do become infected with the virus during the course of the trials, it may be (in some close-knit communities) very easily discovered who got infected and who did not. Those infected persons can expect potentially serious discrimination.

Nor is HIV serostatus the only information that is potentially sensitive for subjects. Drug use practices, types and frequency of sexual practices, and identity and/or number of sexual partners are all equally sensitive information, and if volunteers are not confident that such information will be kept private, they may be reluctant to participate in a trial which will require them to disclose such information to researchers.

11.5.2 Confidentiality: relativism vs essentialism

Furthermore, it may be that confidentiality and privacy are not values at all in some communities. In those communities, then, researchers and public health officials would be faced with a serious meta-ethical issue: should they impose strict confidentiality standards on the community in order to protect the welfare of the volunteers, even though confidentiality and privacy do not have a place in the values of that community? Or should sponsors simply accept the prevalent standards of the community and not act to protect the privacy of individual volunteers? Should a value such as "the right to privacy," which might be perceived in that community as a "western world" value, be simply imposed on the community in order to protect individual subjects, or should the privacy of the individuals be ignored in order to respect the ethos of the community? How should this question be resolved?

This question is sometimes referred to as the thorny question of "medical-ethical imperialism". If you answer the question in one way, and say "Yes, the value of protecting individual volunteers is so important that it must be provided for even though the prevailing standards of that community do not endorse the value of the individual's right to privacy," then you risk being labeled a medical-ethical imperialist. If on the other hand you say "No, the value of respecting the local community standards is so important that we will not impose the value of privacy and confidentiality, even though in many parts of the world (and in the WHO/CIOMS Guidelines) privacy is seen as a significant human right," then you run the risk of simply running rough-shod over the personal safety of the individual volunteers in that community.

At the root of this issue is an even more basic philosophical question about the universalizability of ethical norms. I call it the conflict between ethical relativism and ethical essentialism. The position of ethical relativism holds that no values are universal and hence that no values hold true absolutely, always and everywhere. Ethical relativism sees that different communities do empirically have different ethical norms, and it concludes from this observation that therefore these communities also ethically have the moral right to have whatever norms they choose.

Ethical essentialism, on the other hand, holds that not all values or actions are equally valid or worthy. Ethical essentialism holds that some acts are wrong, even if there is a whole community that endorses them. Ethical essentialism would hold, for example, that torture is wrong, that slavery is wrong, and that genocide is wrong, even though whole communities might endorse those practices and live by them. Ethical essentialism holds that some acts are essentially, in themselves, in their essence, wrong, and should not be endorsed by anyone or any community.

This issue is most complex and multifaceted. Some philosophers argue that the tension between these two positions is the defining ethical issue in any modern pluralistic democracy; others have even claimed that this tension is the key philosophic issue that defines what modernity means. The issue is indeed a thorny one, and much too complicated to deal with inside the scope of this small book. And yet, as is the case with many philosophical questions, the way the issue gets decided will have tremendous practical consequences.

In the case we are discussing here, for example, the practical consequence will be either that regulations protecting the confidentiality and privacy of volunteers will be enacted and enforced, or they will not be. What position does the WHO/CIOMS International Ethical Guidelines take on the question?

The Council for International Organizations of Medical Sciences, in collaboration with the World Health Organization, which drafted the International Ethical Guidelines for Biomedical Research Involving Human Subjects, was sensitive to the issue of medical-ethical imperialism and took steps to insure that they would respect the values of the world's wide variety of cultures. The committee which prepared the document - over more than a decade of research and consultation - was composed of nearly 150 participants from thirty-five diverse countries. Participants came from both developed and developing countries, "including representatives of ministries of health and medical and other health-related disciplines, health policy-makers, ethicists, philosophers and lawyers". Representatives attended from all over the world, including some from sub-Saharan Africa, from Asia and from South America (including one representative each from Thailand and Brazil, two of the countries in which WHO plans to initiate phase III HIV vaccine trials; Uganda was not represented, nor was Rwanda nor Tanzania, but Kenya was.) After the last conference ended in February of 1992, the Guidelines were then revised using all the research, discussions and position papers that had been presented by participants. It was then sent out to the 150 participants for final comments. Dr Bankowski, chair of the committee, explains the process:

The draft guidelines were revised to reflect the consensus of the conference, but with due regard to minority points of view. The revised draft was then sent for comment to the conference participants, to international associations, and to medical research councils and other interested bodies and institutions in both developed and developing countries. The final text duly reflects the comments received. It has been endorsed by the WHO Global Advisory Committee on Health Research and the Executive Committee of CIOMS, which have recommended its publication and wide distribution.

I mention all this to make it clear that the WHO/CIOMS Committee which drafted these Guidelines and published them in 1993 did not act hastily. Committee members made every attempt to face the complex issue of medical-ethical imperialism and to deal with its difficulties as fairly as they could. In the end, however, they had to come down on one side or the other of the question of privacy and confidentiality for research volunteers.

The document they finally published, International Ethical Guidelines for Biomedical Research Involving Human Subjects, comes out clearly in favor of supporting and even requiring the protection of confidentiality. It insists that all research subjects have the right to confidentiality concerning their serostatus, their medical records, and even confidentiality as to whether they are participants in the study or not. This message is stated very clearly in the CIOMS Guidelines.

11.5.3 Weak protections for confidentiality

However, although this document clearly supports the policy of protecting confidentiality, it also recognizes that there are very real de facto limitations on that policy, and that these limitations could severely weaken any significant implementation of the policy. Guideline 12 of that document, which addresses the issue of safeguarding confidentiality, states:

The investigator must establish secure safeguards of the confidentiality of research data. [However,] subjects should be told of the limits to the investigators' ability to safeguard confidentiality and of the anticipated consequences of breaches of confidentiality.

This general principle is indeed important, but it should also be made clear to prospective subjects that the "limits to the investigator's ability to safeguard confidentiality" may be very grave, and they should not be minimized. Some jurisdictions, for example, will probably require mandatory reporting of HIVAb+ persons. Trial sponsors, of course, will seek exemptions from such reporting requirements, but if they are not able to acquire such exemptions, prospective volunteers will need to be aware that their confidentiality would probably be seriously compromised if they chose to join the trial.

Even in jurisdictions which do not require mandatory HIVAb+ reporting, protections for confidentiality are often not very strong. One serious instance, to mention only one, of having ineffective safeguards for the protection of subjects' confidentiality, concerns the usual manner of protecting confidential medical records. The usual method of protecting the privacy of these personal documents is to require that the individual subject sign a consent form before releasing any confidential medical information. This method, though valuable in many circumstances, does have serious limitations: 1) One limitation is that the policy makes it clear to everyone that such records do already exist, and it specifies where they exist and who is the custodian of the records. Clever people with strong motivation will perhaps be able to obtain some of the information they want without getting anyone's formal consent. 2) Another limitation on this method of protecting privacy is that great social and personal pressure can be brought to bear on an individual volunteer to sign a release form. 3) A third serious limitation on this method of protection is that refusal by a volunteer to give consent for release of information may well be considered damning in itself, much as invocation of Fifth Amendment rights by someone testifying in a trial in the US makes people suspicious that the testifier probably has something to hide. This suspicion alone could lead to acts that would discriminate against the volunteer.

This is all to say that protecting volunteers from discrimination by protecting the privacy of their medical records might be only partially effective.

Another method of protecting against discrimination is to enact laws that proscribe discrimination. Such laws, as everyone knows, can sometimes be extremely difficult to enact and even more difficult to enforce. Witness, for example, the difficulty of enacting and enforcing civil rights legislation in the southern states in the US during the early 1960s, and the violently heated confrontations that have accompanied attempts at gay rights legislation in the US in recent years. Laws protecting civil liberties are sometimes difficult to enact even in the best circumstances of relative civil order, but they can be especially difficult to enact in situations of unstable social order, or if the proposed laws do not really represent the community's actual standards and beliefs. If laws protecting individual rights are enacted in such circumstances, they will probably be exceedingly difficult to enforce. Nevertheless, it is probably better to have such laws on the books than to not have them at all.

A third method of dealing with discrimination is to in some way compensate its victims. (Cf chapter 13 below, Compensating Volunteers for Injury.) This of course does not prevent discrimination, nor can any compensation ever replace the individual's actual loss. Furthermore, it might be questionable whether it is even realistic to expect that compensation will actually be given in most cases.

 

To sum up this section: a person's participation in HIV vaccine trials carries "significant... risk of social discrimination or harm". Volunteers should be "apprised in advance of those situations in which trial sponsors might be required to disclose confidential information," and should be fully informed about the extent and implications of this risk.

11.6 Whole virus vaccines

Vaccines come in three basic forms: killed virus vaccines, live, attentuated virus vaccines, and subunit vaccines. Each type of vaccine uses a slightly different method of stimulating an immune response. The first two types, if used against HIV, would carry significantly more risk than subunit vaccines.

11.6.1 Inactivated virus vaccines

A killed or inactivated virus vaccine uses the whole virus, minus the genetic material in its core, to stimulate a successful immune response. (Pseudovirion vaccines are genetically engineered virus-like particles which closely resemble a live three dimensional virus, but also have no genetic material in their core.) Inactivated virus vaccines have been used successfully against hepatitis B, rabies, influenza, and polio (Salk vaccine). These vaccines do stimulate an immune response, but the risk is that the process used to inactivate the viruses may have been imperfect and not all viruses got killed. If this happened while using a killed virus vaccine for HIV, then some persons in the vaccine trials would be accidentally injected with an active virus, and hence may become infected with HIV and progress to AIDS. That, of course, would be an important risk.

Such a tragedy did actually occur at one point with the Salk polio vaccine, an inactivated virus vaccine, shortly after the conclusion of the efficacy trials for that vaccine, in 1954.

 

Of the 400,000 people inoculated with certain preparations, 79 contracted polio. Another 125 individuals became infected through contact with those vaccinated. Three-quarters of these cases involved paralysis and 11 cases were fatal.

 

This tragic event, widely known as the Cutter incident, gave warning to all vaccine researchers of the possible consequences that could result if not all viruses had been completely inactivated.

11.6.2 Live, attenuated virus vaccines

A live, attenuated virus vaccine uses active viruses, but viruses which have been in some way weakened. Many of the vaccines currently in use against viral diseases are of the live, attenuated form, including the Sabin oral polio vaccine, and the vaccines for measles, yellow fever and mumps . The advantage of using attenuated virus vaccines is that the virus continues to exist in the vaccinee's body, continues to replicate, and therefore continues to stimulate an immune response as long as the person lives.

Attenuated viruses continue to reproduce, thereby acting as a constant source of antigenic stimulus to the immune system. Thus attenuated vaccines appear to provide lifelong immunity without requiring periodic boosters.

Thus, the advantage of attenuated virus vaccines is that they could theoretically continue to protect persons against disease for the remainder of their lives. For this reason alone an attenuated virus vaccine for HIV would be extremely desirable.

The risks, however, of using an attenuated virus vaccine for HIV are very high. One of the risks of vaccinating people with a weakened form of HIV is that, since HIV replicates and mutates so rapidly, it would simply eventually mutate back into a highly virulent form. It would, after all, have the rest of the person's life to so mutate. Then all we would have done is infect volunteers with real HIV. (With this form of vaccination, the attenuated virus particles could also be transmissible to others via blood and body fluids, which may be beneficial, if the vaccine is safe, but may be deadly if it is not.)

Such mutations that revert to virulence are by no means unknown. The Sabin oral polio vaccine, for example, is an attenuated virus vaccine and it does happen, even with that extremely safe vaccine, that in very rare cases (one in every two or three million vaccinations) the attenuated virus will mutate back to its virulent form within the body of the vaccinated person and cause a case of active paralytic poliomyelitis. Also, in present SIV vaccine studies, a recent study in the UK using an attenuated SIV vaccine in rhesus macaque monkeys, resulted in the attenuated virus actually reverting, over a rather short time, back to a virulent form and causing one monkey to come down with simian AIDS.

With HIV, which is so highly and rapidly mutable, this risk is serious; so serious, in fact, that I know of only one candidate HIV vaccine in development now that is of the attenuated form, and it has not reached even phase I trials. Many doubt that it ever will. Researchers recognize that one of HIV's most successful attributes is its ability to replicate rapidly and mutate often. Whether it would mutate toward greater virulence or toward lesser virulence is anybody's guess. Evidence so far, however, seems to overwhelmingly indicate that, in an environment with relatively mild selective pressures, HIV generally mutates toward increased virulence.

Another problem with attenuated virus vaccines is that they pose additional risks to persons who are in any way immunocompromised. Such persons, because of their suppressed immune systems, may be overwhelmed even by a weakened form of the pathogen against which they are being vaccinated. In some lesser developed communities, of course, malnutrition may contribute to some cases of immunosuppression. Undetected HIV disease may also put persons at risk for immunosuppression. Any persons or groups who may be immunosuppressed for any of these reasons would probably need to be tested for immunocompetence before administering a live virus vaccine to them. This would require screening those populations for various markers of immunosuppression before administering a live virus vaccine that might put them at serious risk. Such large scale screening would add one more enormous logistical challenge to an already complex set of logistical difficulties, when it came time to actually deploy a successful HIV vaccine.

11.6.3 Subunit vaccines

Subunit vaccines are not made from whole live viruses, nor from killed viruses, but from certain key fragments of the virus, often surface proteins. In these first generation HIV-1 vaccines, the subunit that has been most often used is the glycoprotein gp120 (or its precursor, gp160), which is the portion of the viral surface that actually binds to the CD4 molecule on the surface of human cells. Sometimes a viral core protein, such as p17, may be used instead. Often, chemical or biological adjuvants are injected (or ingested) along with the vaccine in order to help the immune system detect the subunit, "thereby producing a more potent immune response".

Subunit vaccines can be developed in three different forms. In one form, a) the subunits themselves are simply grown in vitro (for example, in ovarian cells of Chinese hamsters), collected, and then injected directly into the vaccinee. In another form, b) a peptide of the subunit - that is, a subunit of the subunit - is developed and injected into the vaccinee. And in a third form, c) the gene that encodes for that particular subunit is taken from the DNA stage of HIV replication and spliced into the DNA of another virus, a vector virus such as vaccinia or canary pox virus. Then the newly engineered vector is injected into the vaccinee, and when that vector virus replicates in his or her cells, it will express the selected viral subunit for which the gene encodes. The presence of that subunit will then (it is hoped) stimulate an immune response against any wild virus that displays the same subunit. These are called recombinant vector subunit vaccines, and they do actually use a live virus, though it is not the virus which is responsible for the infection that is being vaccinated against.

The advantage of the a) and b) forms of subunit vaccines (the subunit and the peptides) is that they are not whole viruses, so have absolutely no capacity to infect the vaccinee or anyone else the vaccinee may contact. Their primary disadvantage, however, is that they seem to produce an immune response that is specific to only those subtypes and strains of HIV that have the exact same form of the subunit as is in the vaccine (that is, homologous strains), and the response seems to be relatively short-lived. These would be significant disadvantages. The advantage of the c) form using a vector virus would be that the immune response would probably last much longer, since the vector would continue to replicate inside the vaccinee, would continue to express the desired HIV surface subunit, and hence would continue to stimulate an immune response in the vaccinee as long as it continued to replicate. This would be a significant advantage.

The disadvantage of the c) form, however, is that it requires the introduction of a live, albeit usually harmless, virus into persons who may be immunocompromised already and not in good enough condition to fight off yet another virus. There have purportedly been deaths from fulminant vaccinia disease caused by the vaccinia vector virus in some recombinant vaccines for other diseases. Furthermore, besides this risk to the vaccinated individual, there is the additional risk that the process of immunization itself will probably produce purulent sores that could put other nearby immunosuppressed third parties at risk of infection with a disease that they are not capable of fighting off. Deaths from fulminant vaccinia disease, while rare, did sometimes even occur during the years that vaccinia virus was being used as a vaccine against smallpox. As recently as 1968, for example, two decades after the US had seen its last case of smallpox, "15 million persons were vaccinated that year [against smallpox], and because of vaccine complications, 240 required hospitalization, 9 died, and 4 were permanently disabled". Vaccinia virus, then, while a generally safe vector virus, is still potentially a disease-causing agent, and its use in immunocompromised communities may be problematic.

These latter risks, however, seem to be more of a problem for those recombinant vaccines in which vaccinia virus is the vector used for carrying in the subunit gene, and vaccinia is indeed the vector being used in some of the recombinant subunit vaccines presently in phase I and II testing. Vaccinia, after all, is the agent originally responsible for the disease of cowpox. Cowpox is normally quite a mild disease in humans, but we do not yet know the full effects cowpox might have if it were to be introduced into persons who are already immunocompromised. Because of this potential risk, other viruses are also being explored as potential vectors. Avipox virus (canary pox) vector vaccines are just beginning phase I testing. This virus has the peculiar trait of replicating only one time in human host cells, so the infection is actually aborted before it can develop into anything that would cause disease. As Seattle researchers Julie McElrath and Larry Corey point out, the canary pox virus

 

can infect avian fibroblasts but induces [only] one round of replication in the human host. Because it causes an abortive infection, the risk of transmission to immunocompromised patients is minimized.



Other vectors that are presently being investigated for potential HIV vaccine use include adenovirus, rhinovirus, BCG (presently used as a tuberculosis vaccine in many parts of the world), salmonella, and hepatitis B.

 

To sum up, then: the advantage of subunit vaccines is that they are almost certainly lower risk vaccines, and in fact are probably the safest of the three kinds of vaccines.

They are also probably the least effective of the vaccine types, because (as we have seen) each vaccine seems to protect against only a very few variants of HIV, namely, those variants that have a gp120 (or 160) molecule which is virtually identical to the gp120 that was used for the vaccine. Furthermore, the immune response stimulated by these subunit vaccines seems to last only several months to two years at the most. This means that revaccination would be necessary quite often. That would be a significant logistical challenge for already overburdened public health services in any community, whether developed or not.

 

Each of these three types of vaccines, then, has its own virtues and dangers. Volunteers should be made fully aware of the risks involved with the vaccine type they are using. If there is any meaningful risk at all of becoming infected with HIV and developing AIDS, participants will have to weigh that risk against the potential benefits of the vaccine. Becoming infected with HIV and developing AIDS is a harm for which it would be almost impossible to compensate a person.

11.7 Being monitored

Another of the "inconveniences and hazards reasonably to be expected" by vaccine trial participants, particularly in phase III studies, is that they will need to be monitored by researchers for a number of years, perhaps for many years, perhaps for the rest of their lives. In addition to preliminary screening, educating and counseling prior to vaccination, subjects will also need to be tested regularly after vaccination, and be given checkups and counseling regularly, perhaps as often as once every two or three months, or even more often. Checkups, testing and counseling will need to continue for some years, during the entire course of the trial.

How long might phase III trials last? The answer to that question, about when the endpoint of the study would be, depends entirely on the "criterion of effectiveness" in that study. If the criterion of effectiveness in phase III trials is only a) the emergence of certain immune system markers, then that could be achieved in a matter of months, so the trial would be short. This criterion (immunogenicity), however, is applicable in phase II trials only, and would not be an adequate criterion of efficacy for phase III trials by anyone's account. So immune system markers would not be used as a criterion, at least not in these first generation vaccine trials.

If the criterion of effectiveness is prevention of infection, then trials lasting three to five years may be adequate. But if the one of the criteria of effectiveness is prevention of disease (or postponement or amelioration of disease), which is the traditional criterion of effectiveness in most vaccines - and this doubtless will be one of the criteria, especially in the early trials - then phase III trials will need to last much longer. How long will subjects need to be followed before we know they will not get AIDS? Ten or fifteen years? Twenty years? For life?

If participants will need to be followed for that long, that means they will need to be checking in with researchers for that number of years, or for the rest of their lives. This may amount to only an inconvenience for many participants, but it may amount to a genuine hardship for others. Some may find their lives restricted in some significant ways due to this requirement that they be available. Volunteers should be made aware of this requirement and its possible attendant inconveniences and hardships.

11.8 Feeling safe

One of the biggest risks for participants in vaccine trials is the danger that they will now feel more secure and protected, as a result of having been "vaccinated," and will thus feel freer to engage in risky behaviors. This is no small risk, especially since it operates below the level of conscious motivations. Researchers must be particularly emphatic when they counsel trial participants to avoid risky behaviors. They must make it clear that the candidate vaccine being tested is not a proven protection against anything, and that in fact participants may be at greater risk of infection than if they had not chosen to participate in the study (due to possible factors such as antibody enhanced infectivity). Counselors must also make it abundantly clear that the study is double-blinded and placebo controlled, which means that a substantial percentage of the volunteers will be receiving a completely inert substance that provides absolutely no protection of any kind. Counseling that makes these facts clear must be thorough and forceful. (Such counseling must also, of course, be culturally and linguistically appropriate.)

In spite of such counseling, however, some subjects may choose to secretly get tested for HIV antibodies in some other testing facility in order to determine whether they were one of the subjects who was lucky enough to get the candidate vaccine. That is, they might choose to "unblind" themselves in the study. This unblinding could have two unfortunate consequences, one for the individuals and one for the study itself: a) the unfortunate consequence for the individual subject is that they may now feel even safer than when they first joined the study and may therefore feel much freer to engage in risky sexual or needle-sharing behaviors. Such behavior change could put that subject at even higher risk of HIV infection than the risk with which they had lived before joining the study. Such increased risk behavior on the part of those who had unblinded themselves might also b) have an unfortunate effect on the statistical structure of the study itself, as explained by one group of researchers:

 

If volunteers who discover that they have received the vaccine are falsely reassured and then engage in high-risk behavior at a greater rate than do placebo recipients, a less-than-optimal vaccine candidate (one with an efficacy of 50% to 60%) may be judged completely ineffective because of selective high-risk behavior among decoded vaccine recipients.... The unblinding of a few vaccine trial participants has already occurred in phase I trials.



Those who design these trials will want to devise ways of dealing with the problem of unblinding. They will want to be able to a) identify those who are most likely to try to unblind themselves (for example, by noticing, during pre-trial counseling, that these prospective subjects' main motivation for wanting to join the trial is to be protected by the "vaccine"), so that these persons might not be selected to participate in the trial; b) somehow motivate persons who do join the trial to not want to unblind themselves; c) somehow identify - probably by self-report - those who do unblind themselves, either deliberately or inadvertently; d) provide regular blinded HIV testing for those subjects who are worried that they may have exposed themselves to infection, so that they can be notified if their test is positive for the virus (not just for antibodies to the virus); and e) have open and frank discussion about some of the consequences of unblinding.

In addition, some trial subgroups, such as commercial sex workers and injection drug users, may be more likely than most to be arrested, charged with a crime, and incarcerated. It is not unusual for penal systems to require mandatory HIV screening for all inmates, so it is not unlikely that the problem of unblinding will be a very real one in these trials, even if it only happens inadvertently. Possible consequences for the statistical significance of the trials will need to be determined.

Probably the best way to minimize the likelihood of such untoward consequences is to insure that counseling about the risks of the study is thorough and clear, that subjects are made fully aware that this candidate vaccine is only possibly effective, that even if it is effective in some persons it may be completely ineffective in others, and that it must not be relied on at all for protection against anything. In fact, in some cases,

 

even vaccines that produce very good immunity may be overwhelmed by vast numbers of invading organisms and a "breakthrough" infection may result. This is not uncommonly seen, for example, with measles vaccination in developing countries where "breakthrough" infections occur quite frequently in the overcrowded conditions of socio-economically deprived communities, even though the vaccine itself usually provides excellent immunity.

 

Counseling that makes subjects aware of these realities may be effective in discouraging the feeling that volunteers in the study are somehow safer and can engage in risky behaviors.

Furthermore, in addition to the verbal counseling, condoms should probably also be provided free of charge to all participants, and depending on the setting of the study and the populations being studied, clean needles for drug injection should perhaps be provided as well.

Research sponsors must be particularly cognizant of their obligation to observe one of the most basic principles of medical ethics: "first do no harm" (primum non nocere). This means: at least don't make things worse. In vaccine trials that use killed virus vaccines or attenuated virus vaccines, there could be, as we saw above, some additional risk that participants may accidentally receive a vaccine with virus in it that is capable of causing disease. If this were to happen, then the vaccine trial would have made things worse for those particular subjects. Furthermore, if this tragic accident were to come to pass, and these trial participants continued to engage in risky behaviors (unprotected sex, sharing needles, and so on), then they would be putting not only themselves at risk, but they would be spreading to others the virus with which they had been accidentally infected. Vaccine trial participants would, in that case, be worsening the spread of the epidemic, and trial sponsors would bear some of the responsibility for that.

For all these reasons, there is a strong duty to provide clear, thorough and effective counseling to all participants. (See chapter 12 below for the ethical problem of counseling with dual intent.)

11.9 Immunosuppression

Another danger, even of the safer gp 120 subunit vaccines, is that the vaccine itself could have a compromising effect on the immune system of the vaccinated person. Two mechanisms for such immunocompromise may occur, syncytia and decrease in CD4 availability.

a) Syncytia is the pathological bunching together of a cluster of individual T4 cells into one big clump. All the T4 cells in the clump are disabled, and therefore become ineffective in their role as chief directing cell in the overall immune system response. The causes of syncytia (from the Greek: syn = together, and cyt = cell) are not entirely clear, but seem to be related to the function of gp120 and its ability to bind to the CD4 molecule on T4 cells. Whatever the mechanism of the pathology, the result is a compromise in the overall functioning of the person's immune system.

b) A second possible mechanism of immunosuppression could be a decrease in the number of CD4 molecules available on the surface of T4 cells for performing their normal functions. Dr Barry Shoub, Director of the National Institute of Virology at the University of Witwatersrand in Johannesburg, explains:

 

Another difficulty with the use of gp120 as an antigen in a vaccine is the fact that its attachment to the CD4 receptor may compete with the physiological function where the macrophage attaches to the CD4 receptor of the [T4 ] lymphocyte to present new antigens to the immune system. An additional fear is the fact that the antibodies elicited by gp120 could attach to the surface proteins of the macrophage involved in this interaction with the CD4 antigen.

 

If either of these two developments did occur then the vaccinees' normal immune response would be compromised to some degree. This means that trial subjects would thus be more susceptible to opportunistic infections which would take advantage of their weakened immune response.

11.10 Autoimmunity

Autoimmune disorders are those in which the immune system has lost some of its ability to distinguish between self and non-self cells, and consequently begins to attack some of the host's own self cells.

Shoub and others believe that there is some risk that vaccinees could develop autoimmune disorders. If the CD4 function is compromised (as explained just above), or if the antibodies elicited by the vaccine were to attach to the surface proteins of the macrophages that come to dock with the CD4 molecule on the surface of the T4 cells, then there is a risk that the immune system could begin to target the host's own body. Autoimmune disorders can be quite serious, so if this risk is a probable one, potential volunteers will want to be aware of it.

11.11 Malignancies

HIV unfortunately belongs to the family of viruses (Retroviridae) which have been associated with the development of malignancies. Because of this, if vaccines are made from whole inactivated viruses, or from whole attenuated viruses (but not if they are made from subunits of the virus), there could be some risk of cells developing malignancies. Shoub explains:

 

This...would make the use of the virus itself unsuitable to develop as an attenuated vaccine..., or as a "killed" or inactivated whole virus vaccine (because the nucleic acid would still be present and the genetic information could hold a theoretical danger of being able to transform cells into malignancy).

 

Shoub seems to indicate that this outcome has not been empirically seen in any work with retroviruses, but that from his knowledge of retroviruses and malignancy, the risk does theoretically exist. What form such a malignancy might take, or what the likelihood is that it would actually develop, is not presently understood.

11.12 Neurological disease of unknown origin

"The most feared situation in any vaccine trial is that cases of severe neurological disease of unknown etiology will be reported among the participants in the first weeks after an injection." A few vaccines currently in use do , though rarely, seem to cause such neurological disorders. A sheep-brain-derived rabies vaccine sometimes (about one per 400 vaccinations) results in myeloencephalitis, and a live attenuated poliovirus vaccine very rarely (about one per one million vaccinations) results in "vaccine paralysis."

In addition, Guillain-Barré syndrome is a well known serious neurologic disorder that very infrequently follows an influenza immunization procedure. A close friend recently suffered a relatively severe attack of the syndrome shortly after receiving his annual influenza vaccine (flu vaccines are made from inactivated viruses). This disease is a demyelinating neuropathy which results in loss of sensation and muscle strength throughout much of the body, particularly the limbs. It has a one to eight percent fatality rate, but those who do survive it generally recover most of their sensation and muscle strength within six months to a year.

Neuropathies of unknown etiology are an uncommon consequence of some immunization procedures, and the frequency with which they could follow from an HIV vaccine procedure is completely unknown.

To date, there are no tests for determining whether a given vaccine does or does not carry a risk of neuropathy; furthermore, our understanding of the mechanisms by which these neuropathies occur is fairly rudimentary.

11.13 Learning your antibody status

Some have considered the knowledge of their positive antibody status to be a "burdensome knowledge that should not be imposed". A certain number of subjects in these trials (in either a vaccine or placebo arm) will probably become HIV infected in the course of their usual risky behaviors, and when they do, then for them "the necessary surveillance of HIV antibody status might...ultimately result in knowledge of HIV infection that the subjects would otherwise have avoided". For some this might not be considered a burden at all, and may even be considered a benefit. But others may consider it a burden imposed on them by their participation in the trial.

11.14 Unknown and unanticipated risks

In addition to the above risks, some of which may perhaps be known (from phase I and II trials) to have some given degree of likelihood, great or small, there is also the theoretical possibility of completely unanticipated hazards. Birth defects that were the tragic consequence of using Thalidomide were completely unanticipated by researchers and clinicians, and did not show up until years afterwards when some Thalidomide recipients became pregnant and gave birth to severely deformed children.

It is possible that such unanticipated hazards could have shown up earlier in subjects who participated in phase I and phase II trials, but they may not have. Phase I and II trials are typically only one to three years in duration, and typically have only a few hundred subjects, so long term consequences may not come to pass in that shorter time span, in that smaller number of subjects. It may be that some hazards will emerge only in phase III trials, with their larger number of participants and the longer duration of the trials.

There is also the possibility that some unanticipated risks could even affect the regular sex partners of some subjects, thereby putting third parties at risk. This would be ethically analogous to Thalidomide having severe deforming effects on the fetuses of some users, but having no ill effect on the users themselves. This eventuality could potentially increase the number of persons who are at some degree of risk in these trials, and would (if such risks were to actually come to pass) have the consequence of potential increased liabilities for sponsors. This is the problem of endangerment of third parties who have not been informed that they were even at risk, and hence have not been asked for nor given consent to accept such risk. Because of this possibility, research sponsors may want to provide counseling to family and regular partners of subjects who are participating in the trials.

These are some of the risks potentially to be encountered by human subjects in HIV vaccine trials, particularly in developing nations. These risks are real and some are probable. Some of these harms, of course, are clearly more likely to occur than others. For example, not being allowed to participate in future HIV vaccine protocols is very likely to happen; developing immune suppression as a result of a surface subunit vaccine is probably much less likely to occur. Besides the degree of likelihood that a given harm will occur, some of these harms are also clearly more grievous than others. For example, some forms of social discrimination to which subjects may be vulnerable (loss of jobs, housing, insurance, and so on) would be much more grievous harms for most people than learning their antibody status.

These are the two dimensions of any potential harm: its likelihood of occurring, and its experienced grievousness if it does occur. Potential harms will probably need to be evaluated in both dimensions. ERCs will probably want to see potential harms evaluated in terms of each of these dimensions, and some subjects may wish to be made aware of such information. (I have proposed the use of some new application forms for detailing the potential harms that prospective subjects may wish to be aware of, and that Ethics Review Committees will definitely want to be aware of when they perform their ethical review of proposed vaccine protocols; these application forms are included as Appendix V in this book.)

Those who hold the Antithesis position insist that volunteers must be made fully aware of all these risks (and perhaps others) before they are asked for their consent to participate in the trials.

* To see the citations associated with material in this chapter, see a published printed copy of the book, or contact the author directly.

 

 

Curriculum Vita | TK homepage | Public lectures | Jenner homepage |

Philosophy homepage | EVT homepage | EVT Introduction | EVT chapter 11

EVT chapter 14 | Lancet Review