Drug users and clinical trials: a United States perspective

In the interests of extending the EATG's cogent critique, I offer the following observations.

The category of ”drug user“ contains a diverse range of people and practices; advocacy for greater inclusion of drug users within clinical research must ultimately work towards specifying which kinds of drugs and which patterns of drug use are of primary concern. For example, certain forms of HIV research, including cohort studies, typically include ”injection drug users“ as a category in describing study participants.

However, this classification is potentially misleading, reflecting an epidemiologic characterization of mode of HIV transmission rather than an assessment of current drug use. Thus, ”injection drug users“ may refer to current injectors, former users, people maintained on substitution therapy (such as methadone or buprenorphine), or those who move repeatedly through these different groups over time.

Different forms of drug use also pose different questions for research purposes. For example, any and all drugs may be of interest for basic drug-drug interaction studies with antiretrovirals. Studies of strategies to improve clinical care –for example, adherence interventions or directly-observed therapy– would presumably take a more narrow focus, investigating effects on people with chronic drug and alcohol addictions rather than recreational users of marijuana or ecstasy.

Shifting criteria
The EATG position paper notes that clinical trial enrollment criteria has shifted over the last decade from a blanket exclusion on people with ”substance abuse“ histories towards a focus on the discretion of individual investigators to determine whether a potential subject's drug or alcohol use would limit their ability to fulfill the requirements of the study. While this trend is welcome, it remains unclear whether more drug users now participate in clinical research than in the past. Notably, the current standard language shifts the onus from trial sponsors (largely industry and government) to individual researchers.

The EATG makes the case that investigators maintain a bias that drug users cannot be adherent to HIV treatment, and thus do not make suitable candidates for clinical research. In support of this argument, the position paper cites several studies affirming that HIV+ drug users demonstrate relatively similar adherence rates and clinical outcomes as other groups of PWHIV. However, these findings must be placed in the context of a large body of research that contradicts those conclusions. Opponents to the EATG position would have no difficulty citing numerous studies that link drug use to poorer adherence, poorer virologic and immunologic responses to HIV treatment, and worse clinical outcomes than non-drug users.

This apparent confusion in the research literature most likely reflects methodological issues in the assessment of a range of confounding conditions with established effects on adherence and clinical outcomes. Drug users have high rates of mental illness, often untreated or inadequately managed; in turn, the deleterious effects of mental illness on HIV treatment adherence have been well documented. Similarly, many illicit drug users suffer from unstable housing situations and involvement with the criminal justice system, including repeated periods of incarceration, with detrimental effects on HIV care and treatment. Finally, the overwhelming majority of HIV+ injection drug users are co-infected with hepatitis C, which increases the likelihood of experiencing hepatotoxicity from antiretrovirals and may thus limit the tolerability of treatment. Indeed, the overall health status of drug users, including high rates of anemia, may exacerbate the range of side effects associated with HIV treatment and thus discourage adherence.

From this perspective, the common notion of poor adherence among drug users as a class should be rejected. A more nuanced understanding would begin with the recognition that many drug users can successfully maintain high rates of adherence to complex treatments. Furthermore, perceptions of poor adherence in this group may mistake drug use as the proximal cause for worse outcomes, overlooking the multiple factors described above which mediate adherence.

A recent article by Harvard Medical School researchers conceptualizes this line of reasoning by invoking the concept of the ”stereotype.“ The authors interviewed HIV+ drug users (defined as people who had used any illicit drug, excluding marijuana, within the previous three months) about their experiences with adherence to HIV treatment and its relationship with their drug use. Their findings do not exclude a role for drug use in adherence problems –at least for some of the study participants, for some of the time– but effectively counter the implicit propositions which form the stereotype of the non-adherent drug user, which the authors list as:

1. Drug users lead ”chaotic“ lives;
2. Active drug users are always using drugs;
3. Being under the influence of drugs precludes taking medications as prescribed; and
4. Drug users are intrinsically different from non-users in the lives they lead and the problems they face.

(Ware et al.,. Adherence, stereotyping, and unequal HIV treatment for active users of illegal drugs. Soc Sci Med 2005 61(3):565-76.)

Taking this argument to its logical conclusion, advocacy for greater inclusion of drug users in clinical research will fail precisely to the degree that such efforts do not ”disassemble the concept of 'drug user' as it has typically been used in clinical research.“ (op cit.) This apparent paradox suggests that inclusion of drug users is contingent on investigators no longer viewing these individuals through the stereotypical lens of ”drug user.“

I would propose, therefore, that advocacy efforts for drug user inclusion in clinical research focus on influencing the views and practices of investigators, rather than targeting exclusion criteria in trial protocols. We need to develop a better understanding of how investigator judgment and discretion operates in actual scenarios. In particular, we need to know how researchers evaluate whether to exclude drug users from consideration in clinical trials based on ”active alcohol or substance use which, in the assessment of the investigator, could compromise the subject's safety or compliance with the study protocol procedures" –to cite representative language.

A focus on investigators does not exclude a role for industry and government research sponsors, who should query investigators about their evaluation criteria and decision- making processes, and support the development and dissemination of more detailed guidance or simple algorithms for assisting investigators in evaluating the likelihood of protocol compliance.

Finally, the EATG position paper highlights persistent gaps in data on interactions between antiretrovirals and illicit drugs, and calls for more research in this area. Advocacy with pharmaceutical companies has not, to date, been successful. However, new strategies may advance this goal:

1. Establish and fund a dedicated research center to conduct ARV/illicit drug interaction studies.
2. Use existing HIV cohorts with substantive representation of drug users for pharmacovigilance to detect signals of potential interactions that merit further investigation.
3. Improve the predictive value of in vitro and animal models for testing drug interaction potential.

The EATG's position paper offers a timely intervention into the politics and practice of clinical research. HIV+ drug users of all types across the globe will benefit from renewed advocacy.

Daniel Raymond, Policy Director for the Harm Reduction Coalition in New York and a member of the AIDS Treatment Activists Coalition's Drug Development Committee.

EATN - European AIDS Treatment News, Volume 14, II – Autumn 2005

More...