Two letters published today in The Lancet argue that researchers should consult participating communities before designing drug trials, regardless of the social standing of those communities.
The letters focus on trials involving injecting drug users and sex workers. They were written in response to the journal's 26 March editorial about setbacks facing researchers testing the HIV/AIDS drug tenofovir.
The editorial underlined the difficulties for researchers undertaking trials in developing countries, especially with respect to providing extensive post-trial care for participants.
Although ethical concerns need to be taken into account, said the editorial, it seemed unfeasible that post-trial care of costly drugs could be guaranteed. And national legislation may conflict with ethical guidelines.
This week's letters, however, argue that ethical considerations must remain central to trial design.
Trials of tenofovir in sex workers were stopped in Cambodia last year and in Cameroon earlier this year after activists alleged the studies were unethical (see Cambodia's prime minister halts AIDS drug trial and Cameroon suspends trial of AIDS drug after protests).
The main accusations were that participants were not fully informed about potential risks and were not provided with adequate care once the trial finished.
This March, Nigeria halted another trial of the drug, citing technical problems with record-keeping (see UK's Wellcome Trust issues ethics guidelines).
The tenofovir studies were testing whether the drug could protect people at high risk of HIV infection — such as injecting drug users and sex workers — from infection.
A trial of the drug's effect on injecting drug users, planned for Thailand this year, has also drawn strong criticism from activists because it will not provide participants with clean syringes and needles.
The Thai trial's sponsor, the US Centers for Disease Control and Prevention, defends this decision by saying that providing drug users with needles and syringes would be illegal under US law.
In their letter to The Lancet, however, Seree Jintarkanon of the Thai Drug Users Network, and colleagues argue that, "the fact that the trial sponsor — the US government — opposes distribution of injecting equipment does not alter this ethical requirement".
Thai participants should also be guaranteed the highest quality of care after the trial ends, say the authors, adding that, at present the trial does not make provision for this.
Another letter in The Lancet this week, by Bebe Loff from the Hong Kong-based Network of Sex Work Projects and colleagues discusses the rights of sex workers taking part in drug trials.
The letters note the vulnerability of drug users and sex workers. Both groups live on the margins of society, with poor access to healthcare and are often unable to seek justice when their rights are infringed.
Although both groups have public representation in the form of networks and support groups, trial researchers do not consult these community networks until after finalising the design of their drug trials.
The correspondents say this implies that the consultation is perfunctory and not intended to greatly influence the trial design.
Jintarkanon and co-authors say that what is needed is an "institutional mechanism" that would ensure more significant community participation, a proposal currently championed by the Thai Red Cross.
The Lancet's 26 March editorial pointed out that despite the publication earlier this year of the UK-based Wellcome Trust's ethical guidelines for research in developing countries, and a discussion paper on existing guidelines of the UK Nuffield Institute for Bioethics, there is no consensus on standards of care after a trial finishes.
What is crucial, argue both letters in The Lancet, is that participants, especially those in vulnerable social situations, be treated ethically and with respect.
Like all trial participants, drug users and sex workers should be allowed a say in the decision-making process of the trials they participate in, say the authors.
*Free registration is needed to read these articles.