Protection of human subjects' rights in HIV-preventive clinical trials in Africa and Asia: experiences and recommendations

Introduction The year 2001 marks the beginning of the third decade of our awareness of the AIDS pandemic and, while there have been some successes in preventing HIV transmission, the epidemic remains uncontrolled, with an estimated additional 5.3 million new infections and 3 million deaths worldwide during 2000 [1]. New HIV prevention technologies, along with the political will and resources to assure access to them, are desperately needed. Fortunately, resources for prevention research are less limited now than they were in the first two decades of the pandemic. Hundreds of millions of dollars for the development of HIV-preventive vaccines and tens of millions of dollars to develop topical microbicides that could prevent sexual transmission of HIV and other sexually transmitted diseases (STD) are now available. Dozens of phase I and phase II clinical trials of HIV vaccines and microbicides, each involving tens to hundreds of participants, and several phase III trials with hundreds to thousands of participants have been performed or are underway. In the near future, many thousands of persons at risk for HIV infection will be enrolled in clinical trials of these new potential HIV-preventive interventions. Most of these research subjects may already be marginalized by the very characteristics and behaviors that first put them at risk for HIV infection, including engaging in commercial sex work, injection drug use, sex between men, or residence in a developing country with high HIV infection rates [2]. They may also be relatively unsophisticated, less educated, have poor access to health care, and have substantial language and cultural barriers between them and the scientific investigators [3]. With the increasing pressure to rapidly develop these public health technologies, researchers must be committed to protecting the rights of potential study subjects and not let them be sacrificed to expediency. In this paper, we review current guidelines in assuring the rights of study participants and review the experiences in three randomized, placebo-controlled, double-blinded clinical trials of potential new HIV prevention technologies. The education of study participants to obtain informed consent, reducing and monitoring risky behavior during a trial, and the role of community participation are discussed. Recommendations are made based on the authors' own observations and do not represent official recommendations of their organizations. Guidance on protection of human research subjects comes from several sources. The Nuremberg Code [4] was established following the Second World War in response to criminal medical experimentation on prisoners. It focuses on the voluntary consent and safety of study participants. The World Medical Association adopted the first Declaration of Helsinki in 1964; it has been amended five times since then, most recently in October 2000 [5], partly to address the issue of use of placebos in medical research. Internationally recognized guidelines have also been developed by the Council for International Organizations of Medical Sciences in the International Guidelines for Ethical Review of Epidemiological Studies in 1991 [6], and the International Guidelines for Biomedical Research Involving Human Subjects in 1993 [7]. Specific guidance for HIV vaccine research was published by UNAIDS in 2000 [8-10]. For microbicides, guidance is also available in a report from a 1997 international symposium sponsored by the Women's Health Advocates on Microbicides and the Population Council [11], and from the International Working Group on Microbicides [12]. The US government sponsors substantial HIV vaccine and microbicide research, both domestic and international. Guidance for the ethical conduct of these studies is provided by the Belmont Report, published in 1979 [13]. The US National Bioethics Advisory Commission has issued recommendations on research involving human participants [14], and on research in developing countries [15]. Many other guidelines have been issued [16], including national guidelines in countries where HIV prevention research is performed [17,18]. Informed consent and comprehension Each of these aforementioned documents asserts that consent should be freely given by research participants. The Nuremberg Code established as its first principle that 'the voluntary consent of the human subject is absolutely essential.' How should researchers assure that consent is truly 'informed' so that potential participants understand the study well enough to decide whether to join? Researchers must avoid scenarios in which potential participants are asked by inexperienced members of the study staff of very different cultural background to sign technical consent forms in an unfamiliar language. People may join the study despite a poor understanding if there is undue pressure from authority figures, including the study staff members themselves, or if there are coercive promises of excessive payments or life-saving medical care not otherwise available [19]. Most potential participants in HIV prevention trials are unfamiliar with clinical trial methods. Many languages have no direct translation for words describing key study concepts such as placebo or randomization, which can lead to considerable confusion and controversy [20-22]. What have researchers done in HIV prevention clinical trials to educate potential study participants so that consent is more truly informed? In the COL-1492 gel (Columbia Laboratories, New York, New York, USA) phase III trial, which was completed in 2000, 991 female sex workers from Thailand and three African countries were randomly assigned to use either COL-1492 (Advantage S™), a vaginal gel containing nonoxynol-9, or Replens™ (Columbia Laboratories), a similar comparison gel without nonoxynol-9 [23,24]. We report on the experiences from the Durban, South Africa, study site. Participants at that site were drawn from a community of sex workers located at truck stops along the major trucking route from the port city of Durban to the commercial capital, Johannesburg. The women were largely migrants from rural areas of the province of Kwazulu Natal, while their male sex partners, both the commercial and non-commercial, were migrants from other provinces, making HIV prevention especially challenging [25]. Most of the women had very little or no formal education and, prior to participation in the trial, most had little or no awareness about STD or HIV. Before women were screened for study eligibility, a general HIV/STD education workshop was given. To facilitate participation and help protect the rights of potential participants, 'community liaison persons' were designated by the investigators. These liaison persons first reviewed the consent forms to help guide accurate translation from English to simple Zulu that could be understood by potential study participants. The forms were then back-translated from Zulu to English for verification. Many women were unable to sign their name and had to simply mark the consent form with an 'X'. In this setting, obtaining truly informed consent was very difficult. Research staff members, together with a community liaison person, read the informed consent form to potential participants and evaluated their understanding by asking a few questions on the information given. Through the use of innovative explanations, role plays, and analogies, women were able to understand the use of a placebo and randomization to some extent. However, an evaluation of the women's understanding of the trial objectives and procedures 3 months after enrollment indicated that women were not able to retain the initial information given to them during the informed consent procedure. Women with no education fared worse than those who had some education, and nearly all women believed incorrectly that both the products had a protective effect. This belief had further credence as women experienced fewer symptoms of STD than usual during the trial, not realizing that regular treatment of STD might have been the main reason for this. To try to overcome these difficulties, study objectives were reiterated at each follow-up visit, and study monitors re-assessed participants' understanding during the course of the trial. However, focus group discussions on completion of the trial also suggested that some women still did not understand the blinding process. In the AIDSVAX® B/E (VaxGen, Brisbane, California, USA) phase III vaccine trial, 2545 injecting drug users (IDU) in Bangkok, Thailand, are receiving either a recombinant gp120-based HIV vaccine or injections with a comparison product containing only adjuvant [26]. The study achieved full enrollment in 2000, and completion is expected in late 2003. Before the Bangkok vaccine trial began, investigators assessed potential participants' comprehension of an educational session introducing a hypothetical clinical trial as part of a willingness-to-participate survey [27]. After the education sessions, comprehension was high, but the results indicated that special attention should be given to the concepts of use of a placebo and partial efficacy, so that trial participants would understand that study participation would not guarantee protection from infection. With the knowledge gained from this survey, materials were prepared to educate AIDSVAX® B/E study participants. Although investigators wished to keep the study consent form simple and understandable, it grew to 14 pages in Thai to cover all of the issues required for informed consent. A video and a booklet were also prepared to inform potential trial participants fully and to ensure that they understood that trial participation might not protect them from infection. To be considered eligible, potential participants needed to pass a comprehension examination on two occasions at least 1 week apart. The questions were the same both times, but the question order was changed to discourage passing the test by simply memorization. The examination included 'must know' questions that had to be answered correctly for the volunteer to be eligible. Potential participants were encouraged to take the consent form with them and discuss participation with their friends or relatives before any decision was made. However, they were reminded that they should not discuss the consent form with someone whom they did not want to know that they were an IDU or with a potential participant in an HIV vaccine trial. In the Carraguard™ (Population Council, New York, New York, USA) study, 165 women recruited from family planning clinics in Chiang Rai, Thailand, are using either Carraguard™, a carrageenan-containing vaginal gel, or a comparison gel. The study is scheduled to complete follow-up in late 2001 and, in contrast to the first two trials discussed, the Chiang Rai study focuses on safety and acceptability and is not designed to determine effectiveness. Before the trial began, a comprehension and willingness-to-participate survey was also performed at the study site [28]. The women surveyed demonstrated a good level of knowledge of microbicide trial issues after a 30-45 min group information session. The most common incorrect answers were about staff blinding with regard to participant treatment arm assignment, the need to use the microbicide before each act of intercourse, and the concept of partial effectiveness. Again, this indicated critical areas to be emphasized during the actual trial. In both the Carraguard™ and the AIDSVAX® B/E trials, the English-language versions of the consent forms were written at an eighth-grade reading level. They were then translated into simple Thai and back-translated from Thai to English by an independent translator to verify the English-to-Thai translation. The Carraguard™ trial consent form was adapted by a Thai behavioral scientist to make participant education materials. A pictorial flip chart was used for study introduction, and women received a booklet when they were screened for enrollment to take home. The consent forms and informational materials were reviewed with a community advisory group (described later) to be sure that the translations and messages were clear. A study comprehension examination was also used. To be eligible, women needed to answer correctly 80% of 20 questions. Women were given a second chance at least 1 day later if they scored less than 80% on the first attempt. During the study there is continuing assessment of study participants' knowledge and additional education. After the first 25 women were enrolled, in-depth interviews were performed to verify study comprehension and to improve study introduction techniques for subsequent enrollment. Study volunteers' responses to questions about sex behaviors and study gel and condom use at monthly follow-up visits are used to assess comprehension. Study monitors also meet with participants when they visit the study site and ask participants about their understanding of the study. Finally, after enrollment was complete, a participant information meeting was held, which was attended by 60% of the participants. Recommendations on assuring informed consent • A comprehension assessment before embarking on an HIV prevention clinical trial is useful to help assure good comprehension and truly informed consent With potential areas in participant understanding can be and • who have with participants should be well in and and should not have substantial language or cultural barriers from participants. • forms should be and understood by potential participants. English-language can be using the in and general guidance on consent forms are available • such as flip or role should be prepared to help inform study participants. These along with the consent should be reviewed and translated by staff members experienced in HIV research with the participant and by community • Informed consent not at enrollment. Study participants should be that they are to the study at any and information should be provided at subsequent study • with any of a study, should be to participants' understanding of the study and to guide additional if needed. A comprehension examination is useful to knowledge for potential participants to decide whether to join the study [15]. • understanding should to be evaluated as the study formal comprehension may be useful in participants' and to areas for with the study group to comprehension and consent can be included in the study Study or not otherwise with the study but with a good understanding of research such as ethical review members, can also help to make this These educational take considerable and especially human resources However, in study participants is as to research as is in or Study subjects must be well informed to protect their understanding is also critical to the of a study, especially in a microbicide study where the participants must the product before each sex in a vaccine trial, participants are more to be to the and follow-up Finally, while guidelines community the will in where is in these the of the to to in a study must be and monitoring risky behavior The aforementioned guidelines also that research should not the potential of research. The Declaration of Helsinki that to the of the human subject should take the of and new potential vaccines and microbicides must have demonstrated safety in the in and as well as human safety before on to What other should be to ensure the safety of HIV prevention trial is an and potential of in HIV prevention To efficacy, of must HIV infection during the trial, more in the placebo group than in the treatment group if the treatment is to be to be guidelines that HIV prevention must be given to study participants, along with for those at risk for sexual transmission, or access to injection for IDU However, if these are very in HIV infection, the for the trial of the vaccine or may be and the while any risk with the study products will still have at it has been suggested that condom use not be in clinical trials of new HIV prevention technologies, condom study and product development in otherwise infections in at risk In most that very HIV prevention be provided potential risk is that participants may that they are from infection despite informed that the of the study product is and that some participants in most are receiving a placebo product that should no at a the level of HIV infection risk to trial participants should be no than if they had not in the trial, and prevention should ensure that at least HIV infection risk behaviors do not those by participants at study enrollment the Durban site of the COL-1492 trial, women enrolled in the trial less than condom use at the With condom and use with commercial but was still less than Women were not able to condom use for sexual act to cultural and as barriers to sexual and less for sex when were used. women did to condom use with commercial partners, of sex was not the with partners, especially to of some male may have used the women's participation in the microbicide trial to not using incorrectly that the was already using some condom use with regular the trial. from condom use, interviews with women on completion of the trial suggested that trial participation had had a on the women medical care, for to regular treatment for It must be while an in condom use with commercial was these were either by or by have limited and, in the trial, many women who condom use for less than use or incorrect In the AIDSVAX® B/E study, participants are informed by the study staff about the need for injection of injection and the of injection the from in The and safety monitoring which Thai and international members, study including monitoring behavior and the of an of risky behavior of participants. A report from the AIDSVAX® B/E trial has demonstrated that the of IDU behavior has at with that at drug use from to use from to from to and from to In condom use from to In the Carraguard™ study, women are by the study research to discuss HIV infection risk with their main and to HIV infection The of HIV is use is and and are provided at study The prevention is based on of HIV prevention research at the including prevention studies with women recruited from the same of family planning behavior are reviewed by the the condom use was by only of During condom use was for about 60% of sex with regular sex partners, while sex with any other was Recommendations on reducing and monitoring risky behavior • HIV prevention should be provided for participants in HIV prevention • at risk for sexual transmission should along with condom use and treatment for other STD that may to HIV infection, and and • In IDU should and have access to treatment and injection where • The results of studies in the study should be used to improve HIV prevention • such as involving community in prevention should be used when • To avoid potential of prevention may be provided by a group independent of the researchers However, in many independent are not available and the researchers are to the • of prevention should be using behavioral during study follow-up visits and other such as rates of other STD when • a of risk should be no worse during a study than they were at enrollment. risk to the a of may be a in participants and in prevention • A that members from the country should review risk and prevention must be or a study should be if there is an in risk the of community on ethical review has been and it is required for of studies involving the US of Health and Human In more with members of the community of research participants is in more recently guidelines The UNAIDS Guidance Ethical in HIV that ensure ethical and scientific of the research, its to the and its by the community should be in an and in the and and of results of HIV vaccine Although the COL-1492 trial was not in a community a researchers the study as the community of sex workers at each truck using community liaison persons was to women into the trial. was also needed to the high follow-up rates the results of the COL-1492 study demonstrated a infection in the treatment arm than in the Replens™ comparison group This was with the community liaison persons prior to to the study participants. This was very in the of the results and the and of all was no to the of IDU in Bangkok for the AIDSVAX® B/E trial staff to with at the beginning of the trial. the trial a community was established made of trial participants and community members who could their with the members of the study staff issues have been and by the such as assuring informed consent and monitoring of risky and it has provided information to the study staff in their to facilitate the of the participants. The Rai Research Advisory was established by the investigators before beginning enrollment in the Thai Carraguard™ study The advisory group times before beginning the study, and the research considerable the community about microbicides and clinical research. This was to be as the group made in the study the consent and the participant information In it the study staff in to members of the community from which study participants would be The Advisory Group is of the of the study, and members have with the participants and their Recommendations on community • such as with a community advisory group with members the community from which study participants will be can help with the conduct a study and protect the rights of study • a group should review the study and, while major in study may not be their for making the study more to participants should be • The group should review the study consent forms and informational materials. for making those materials more to study participants should be • community advisory group members may meet with the study participants, be informed about study and with the of the study participants. In some if the study participants in or behaviors such as injecting drug use or sex such may not be In any study participants should be of who will be and must be to whether to With the substantial in resources for the development of HIV-preventive vaccines and topical microbicides, many thousands of will be enrolled in clinical trials of these new products in the near is public pressure to develop these new as as and many of these potential study participants may be in or or may in developing in the given from two microbicide clinical trials and vaccine clinical trial in and and considerable are required of researchers to protect the human rights of these study subjects to published to inform and study participants to ensure that they understand the study to truly informed consent and to protect from excessive risk is as to research as in the areas of and should be used to assess participants' comprehension and risky to inform participants are needed and, if participants unable to an understanding of a study should not be Studies in which risky behavior of participants should be of the of potential participants can be very in the of the study, the of the consent forms and information the of the study to the and the protection of the rights of the participants. The their in the Thailand and Carraguard™ including the COL-1492 Study the Population Council, Chiang Rai Health Chiang Rai the and the US for and They also and for their critical of the