Therapeutic ethics and communities at risk in the presence of potential mutation to resistant strains to HIV antiviral medications
Samuel R. Friedman, Mark A. Wainberg* and Ernest Drucker┼
AIDS 1998, 12:2089-2093
From the National Development and Research Institutes, Inc., New York, New York, USA, the *McGill University AIDS (:enter, Jewish General Hospital, Montreal, Quebec, Canada, and the Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, and The Lindesmith Center, New York, New York, USA.
Sponsorship: Support was provided by the National Institute on Drug Abuse grant DA10411 'Drug; Use and HIV Risk among Youth'; Health Canada; the Medical Research Council of Canada; and the Lindesmith Center, Open Society Institute (New York).
Requests for reprints to: Dr Samuel R. Friedman, National Development and Research Institutes, Inc., New York, NY 10048, USA.
New combination therapies for HIV/AIDS demand up-to-date knowledge of rapidly developing complex therapeutic regimes These often arduous regimens require careful adherence by patients. regular monitoring, and rapid response to adverse reactions. The risk of failure is high, since days or weeks of hissed or incomplete medication can render antiviral medications ineffective for the individual patient through the rapid development of drug resistance.
Although little is vet known about; how transmissible such resistant strains really are. there has been considerable public discussion of whether physicians should refuse to prescribe combination therapy to patients whose life circumstances make it difficult to follow prescribed regimens and thus make them likely incubators of resistant strains (1-4]. The ethical issue of denial of treatment is usually posited as a conflict between public health and the individual (4]. From this perspective, therapy may potentially benefit individuals. but prescribing to non-adherent individuals may pose a risk to public health. Physicians, obliged to attend to the interests of both the individual and public health. are forced to mediate between these interests.
In this review: we reformulate these issues in terms of communities at risk', emphasizing populations in Which successful utilization of therapy might be lower due to drug use or other causes. We then review current knowledge about effects of therapies in reducing viral load and perhaps transmission, the extent of primary infection by resistant strains of HIV, and transmissibility of resistant strains. Finally. we discuss how I community at risk approach may resolve some ethical and practical issues raised by new . therapies.
Reformulation of communities at risk
Viewing these public health issues only in terms of individual patients' adherence diverts us from finding, effective solutions to them. It is more effective to consider a. group whose full participation is critical to successful clinical HIV/AIDS treatment: the 'community at risk' (here considered to include people .at behavioral risk plus their families, support networks and friends). 'Community at risk' is a public health concept often used to understand disease acquisition but not normally used to understand therapeutic intervention. For HIV and some other pathogens. 'community at risk ' also identities persons who may be at relatively immediate risk if resistant strains do develop. These communities are in excellent positions to influence patients' behaviors so as to (1) increase their probability of participating in successful treatment, (ii) perhaps reduce HIV transmission by reducing viral load, and (iii) reduce risk behaviors that might transmit resistant strains.
Some communities at risk may be at enhanced risk for treatment failure precisely because they are not seen as legitimate participants in decisions about their own medical care. For these groups, including injecting drug users (IDU)and. at some places and times, homosexual men, the concept of 'unlikely-to-adhere' patients can facilitate medical decision-making based on stereotype or stigma rather than science. This may support reduced participation, which in turn can harm individuals, communities at risk, and the broader public health. Conversely, when organized communities at risk help to define ethical and practical issues, support treatment participation, and exert positive peer influence on each other, risk behaviors can be reduced, therapeutic adherence increased, and public health policy improved. Thus, governments, medical providers and other relevant bodies have the obligation to provide environments and resources to facilitate their active and organized participation in protecting their own health and that of the broader public.
History of communities at risk within the HIV epidemic
The international history of the HIV epidemic, and that of the related outbreak of multidrug-resistant tuberculosis (MDR-TB) in New York City and elsewhere 15-111, provide evidence that (i) prevention is facilitated by including communities at risk in discussing and implementing programs, and (ii) effective medical treatment can be provided to `unreliable' groups like IDU without leading to runaway epidemics powered by mutant strains. Gay communities in many countries have been essential in devising and implementing programs and public policies for preventing and treating HIV in their at-risk community 112,131. Where permitted (as in Australia and sonic western European countries), IDU have organized similar (although smaller) organizations that actively help formulate national AIDS and hepatitis policy, develop and lead risk-reduction campaigns (such as syringe exchange programs), and assist IDU to obtain and adhere to therapy  Furthermore, behavioral risk reduction amongst IDU has been shown to function through their social networks and small-group discussions, a process that started well before formal Hl%,' prevention programs [15-17].
The case of MDR-TB
The history of the New York City MDR-T13 outbreak demonstrates the value of working with communities at risk. Failure to consult. with IDU early in this outbreak created obstacles to patients' entering or complying with therapy 1181. Later. however. outreach through needle-exchange programs and other userfriendly street-based harm-reduction projects that worked with IDU as individuals and as advisory committees was used to locate T13 cases and to provide directly observed therapy. Together with stricter institutional infection control measures. this quelled the New York City T13 Outbreak.
Specific issues that have an impact upon policy discussions and therapeutic decision-making as the HIV epidemic develops include (i) the impact of therapy on levels of virus in sexual fluids and blood. (ii) the extent of primary infection by drug resistant HIV strains, and (iii) the transmissibility of such strains 1191.
Impact of therapy on viral load in blood and sexual fluids
Treatment of infected subjects with antiviral drugs can diminish the amount of HIV detectable m blood 2(1,211. However, with some therapies, such as zidovudinc (ZDV) monotherapy. this drop in viral burden as often followed by a resurgence in the amount of virus detectable in blood and the development of HIV resistance to the compounds utilized [221. On a positive note, combinations of three different antiviral drugs in previously drug--naive patients have yielded profound drops in viral burden that are sustained for C> months to 1 year or longer, and triple drug therapy has lowered the amount of virus in plasina to levels below our ability to detect it. Two of the best examples of such combinations included indinavir, a protease inhibitor, together with ZDV-lamivudinr [23,24] or alternatively with ZDV-didanosine-nevirapine [251. In both situations, involving over 111(1 subjects in each study, viruses isolated from the cells of patients who were adherent to their antiviral regimens were shown to be wild-type in regard to both their phenotypes and genotypes for periods of 1-2 years after treatment initiation .125-27]. Thus, these antiviral regimens blocked viral replication sufficiently SO that the mutations that would normally encode drug resistance could not occur. (These findings, however also demonstrate that it is premature to conclude that therapy might eradicate HIV from infected hosts.)
Unfortunately, positive results were not obtained with patients who did not properly adhere to their antiviral drug regimens or who received less effective drug combinations, such as ZDV-nevirapine. For example, non-adherence with regard to didanosinc in the combination of ZDV-didanosine-nevirapine led both to higher levels of viral RNA in blood and also to the development of resistance against both nevirapine and ZDV  Hence, suboptimal therapy can promote the occurrence of drug-resistant strains. Furthermore, resistance to drugs despite adherence to triple combination protocols has also been demonstrated in patients who had previously been treated with antiviral compounds.
Several studies have documented that combination therapy can also reduce viral burden in genital fluids, the other major vehicle for HIV transmission. One report showed that high viral RNA load was commonly present in the semen of HIV-infected men at all stages of disease, and that viral burden in semen increased together with disease progression. Treatment with a protease inhibitor, indinavir, together with a non-nucleoside reverse transcriptase inhibitor, Sustiva (DMP-266), led to marked reductions in viral load in both blood and semen . A second study yielded similar results in 44 patients treated with a variety of antiviral regimens, and also showed that patients who had non-detectable levels of viral RNA in their blood also had non-quantifiable levels of viral RNA in their semen  In a related study, treatment of bacterial urethritis with antibiotics led to marginal reductions in HIV RNA copy number in seminal plasma 
Future studies should assess such material from patients at a variety of clinical sites under a variety of living conditions, and might show that successful antiviral therapy could cause HIV-infected individuals to become less infectious. We would need to be careful that such findings, if generated, did not give rise to feelings of complacency and consequent increases in high-risk behaviors. It is also important to note that problems of compliance, toxicity, and resistance may impact on the duration of benefits derived from any drug used to treat HIV.
Primary infection by resistant strains
Studies in several western countries have revealed that 5-13%, of viruses responsible for new cases of HIV infection are attributable to HIV strains containing the cocion 215 mutation in reverse transcriptase associated with resistance to ZDV 131] This reflects the wide spread (and often poorly organized) use of ZDV to treat HIV infection since the 1980s. It is noteworthy, however. that the rates at which this mutation and primary resistance to ZDV have been detected did not increase dramatically between 1989 and 1995 1311. Conceivably, current methods to detect HIV resistance at baseline a re not adequately sensitive, but it is also possible that drug-resistant viruses may be disadvantaged in growth potential in newly-infected individuals or that wild-type drug-sensitive variants are transmitted preferentially. More work is needed in this important area.
The identification of resistance in previously untreated patients may have practical significance. if either (1) drugs used in treatment were ineffective against drug resistant viruses that had been acquired either sexually or through sharing of drug-injecting equipment b\ individuals undergoing primary infection. or (ii) if the presence of resistance-conferring genotypes or phenotypes at low levels nonetheless potentiated the rapid selection of resistance once treatment was initiated. Individuals who take ZDV for prolonged periods will probably develop resistance to this compound. and ZDV may not be effective in individuals who harbor ZDV-resistant strains [32,33]. Whether this is true for other types of drug-resistant HIV. and whether viruses resistant to protease inhibitors have impaired fitness, are subjects that require further research.
Transmissibility of drug-resistant strains
There is no direct evidence that drug-resistant HIV variants are more transmissible than wild-types. and the opposite can potentially be inferred from the work cited above on ZDV resistance . In fact. ZDVresistant viruses grow slower than wild-type viruses [34[, and viruses containing the M184V mutation, which are resistant to 3TC. may be less tit than wildtype viruses in peripheral blood mononuclear cells , although not necessarily in T-cell lines 
In summary, the new therapies, when effective, confer considerable clinical and public health benefits. There is little evidence to suggest that ineffective therapy leads to widespread transmission of resistant strains.
The following ethical principles may be suggested: First, no physician should refuse effective therapy to a patient w ho wants it. This is Unethical, and may also promote increased black market activity in medicines. Physicians should discuss with patients the pros and cons of starting such therapy, including whether their lives are too disorganized to allow therapeutic adherence and the risks this poses to their and others' health. Such discussion and counseling requires physicians to Understand the realities of each patient's life: `accepted wisdom' about the lives of categories of people (such as IDU or homosexual men) is not an acceptable basis for clinical care. Furthermore, physicians' clinical cxperiencc may be misleading because patients often present images of their lives that reflect what they think physicians expect to hear X37].
Practitioners have an Obligation to remain up-to-date on the extent to which resistant strains do emerge in their patients and elm the extent it) which resistant strains are transmitted to others. They should also consider the extent to which failure to provide therapy to patients might endanger others. since such therapy may well reduce patients' infectivity to others. For newly developed therapies, clinicians, patients, and communities at risk niav have to make such decisions under considerable uncertainty.
Second. community groups. ,11c11 as gay or bisexual men, IDU, and neighborhood association,. have been key actors in developing behavioral norms for risk reduction [12-14] They are also crucial for developing effective norms to minimize therapy misuse.
Specific ethical norms, based on concepts Of mutual responsibility, Mutual assistance, Mutual respect, and 'doing no harm' might develop around the following points: (i) To make effective therapy available to the sick. there is an obligation to develop medical regimens that can be feasibly adhered to. For HIV, this means developing affordable medications that can be used by homeless gay men or IDU who lack access to refrigerators or watches. (ii) Government and medical institutions have an obligation to make adherence with therapy easier for recipients: file example, by making medications and monitoring available at trill\' convenient times and locations, and by assuring that cost does not restrict access. Positive examples include community directly observed therapy programs: on-fete-street medical vans that provide diagnosis and therapy to participants in syringe exchanges, Outreach projects. and soup kitchens: and integrating medical services with outpatient treatment for drug abuse. This obligation includes maintaining continuity of availability of medication lender unanticipated circumstances such as incarceration or hospitalization. (iii) To the extent that recipients of therapy are potential incubators for resistant strains, they have an obligation to themselves and others to adhere to therapeutic regimens (in the absence of severe side-effects). Recipients also have an obligation to retrain from risk behaviors that might transmit resistant strains to others. Such obligations remain valid even it: indeed. especially it, therapy fails to benefit the patient. (iv) Recipient, and community groups have an obligation to support individuals in reducing risk behaviors and in maintaining adherence to therapy. (v) Society and institutions have an obligation to assist recipients to avoid transmission behavior. Access to condoms and syringes is essential including in jails and prisons. This obligation is a duty owed both to the patient and his/her community at risk. (vi) Just as health-care providers and institutions Have an obligation to involve drug users. gays, and others at enhanced risk in developing ways to meet these obligation,, better. community groups have an obligation to solicit suggestions and critiques both from actual (and potential) recipients of these medications and from institutional sources.
Research is needed concerning the practical difficulties and outcomes of implementing these suggestions. In spite of the application of this (or any other) approach. resistant strains will arise and perhaps be transmitted Ongoing research is needed to detect such strains rapidly, determine their transmissibility . an.) understand social locations, institutional barriers. Mr behavioral cIrcumstances that increase the emergence of resistant strains. Quickly mobilizable resources m,1\ be needed to contain any outbreaks of resistant strains.
Every system of norms exists in a social i,11 context. Situations may be found in which it ,reins warranted for either government agencies or communities ,u risk to apply negative sanctions to recipients Mr providers of medications. Such systems of negative sanctions should be developed jointly by (i) experts in the social and biological development of resistant strains and (11, members of community groups. Together, they can both avoid self-defeating measure s ante balance individual rights and the risk to communities. Here. It is important to realize that both the threat to individual rights and the risk of becoming infected with a resistant strain are greatest for members of the relevant community at risk.
Finally, the overall goal must be increased enrollment in effective treatment both to protect individual patients and to reduce viral loads and transmission. Working with communities at risk can increase treatment enrollment, increase adherence. and reduce transmission risk behaviors by patients in whom therapy fails to control the disease.
The authors would like to acknowledge assistance provided by Sally Cooper, Sherry I Deren Maureen Miller and Rod Sorge.
1. Senak M: Predicting antiviral compliance: physicians' responsibilities vs patients' rights. 1 Int Assoc Physic ians .111)., Care 1997, 3:.15-48.
2. Sontag D, Richardson L: Doctors withhold HIV pill regimen from some. New Turk lime.,, 2 March I997A1.
3. Bayer R: Ethical challenges posed by new advances in the antiretroviral treatment of HIV disease. In the Implications lei Antiretroviral Treatment,: Intiirmal ( unsult.itiun. Edited by van Praag E, Fernyak S, Katz AiM. Geneva: Wrrrld He,illh Organization; 1997:95-101.
4. Baver R, Stryker I: Ethical challenges posed by clinical progress in AIDS. Am 1 Public Health, 1997, 87:1599- 16()=.
5. Frieden TR, Sterling T, Pahlos4Nlendez A, Kilburn IO. Cautlucn GM, Dooley SW: The emergence of drug-resistant tuberculo sis in New York City. N En,gl I tiled 1993, 328:521-526.
6. Chaulk C, Moore-Rice K, Rizzo R, Chaisson R: Eleven years of community-based directly observed therapy for tuberculosis. 1ANIA 1995, 274:9-15-951.
7. Longuet P, Pierre I. Lacassin F, et al.: A limited mullidrugresistant Mycobacterium tuberculosis outbreak: screening of contact hospitalized patients (pts). 35th Intent ienc e• Conference on Antimicrobial Agent, and (hemotherapk. San Francisco, September 1995 labstract'.
8. Bifani P, Plikaylis B, Kapur V, er a l.: Origin and interstate spread of a New York City multidrug-resistant Mycobacterium tuberculosis clone family. IAXIA 1996, 275:-452-4.57.
9. Des larlais DC, Hagan H, Friedman SR: Epidemiolog~ and emerging public health perspectives. In Substance Abuse. A Comprehensive Textbook. Third Edn. Edited by Lcrwinson 1H, Ruiz P, Millman RB, Langrod JG. Baltimore: Williams \ Wilkins; 1997:591-597.
10. Perlman DC, Perkins MP, Solomon N, Kir(hems L, Des larlais DC, Paone D: Tuberculosis screening at a syringe exchange program. Am 1 Public Health 1997, 87:862-86:3.
11. Perlman D, Salomon N, Garcia de Soria V, et al.: Tuberculosis directly observed preventive therapy for active drug users in two settings. College on the Problems of Drug Depenelencc annual,Lteeting. Nashville, June 199- abstractl.
12. Altman D: The primacy of politics: organizing around AIDS. AIDS 1991, 5 Isuppl 2):52:31-5238.
13. Coates TI, Aggleton P, Gutzwiller F, et ,)l.: HIV prevention in developed countries. Lancet 1990, 348:1 14 3-1 148.
14. Friedman SR, de Jong W, Wodak :V Community development as a response to HIV among drug injectors. AIDS 199 3. 7 Isuppl 1):5263-5269.
15. Friedman SR, Des Jarlais DC, Sotheran IL, Garber I. Cohen H, Smith D: AIDS and self-organization among intravenous drug users. Int 1 Addict 1987, 22:2(11-219.
16. Neaigus A, Friedman SR, Curtis R, et al.: The relevance of drug injectors' social and risk networks for understanding and preventing HIV infection. Sue Sci ,\fed 1994, 38:6--78.
17. Des Jarlais DC, Friedmann P, Hagan H, Friedman SR: The protective effect of AIDS-related behavioral change among injection drug users: a cross-national study. :lm 1 Public Health 1996, 86:1780-1785.
18. Curtis R, Friedman SR, Neaigus A, Jose 13, Goldstein M, Des Jarlais DC: Implications of directly observed therapy in tuberculosis control measures among IDUs. Public Health Rep 1994, 109:319-327.
19. Wainberg MA, Friedland G: Public health implications of antiretroviral therapy and HIV drug resistance. 1-3N1:1 1998, 279:1977-1983.
20. Ho D, Neumann A, Perelscm A, Chen y1 ', Leonard I. Markowitz ,'v1: Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995, 373:12 3--126.
21. Wei X, Ghosh S, Taylor M, el al.: Viral dynamics in human immunodeficiency virus type t infection. Nature 1995 373:117-122.
22. Fiolrxlniv tv1, Katzenstein D, Israelski D, Nivog,m I Reduction in plasma human immunodeficiency virus ribonucleic acid after dideoxynucleoside therapy as determined by the polvmerase chain reaction. I C7in Invest 1991 , 88:1 - ;-~ I
2.3. Hammer SM, Squires KE, Hughes ,MD, ct ,J.: A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection anti CD4 cell counts of 200 per cubic millimeter or less. I'm 337:725- -33.
24. Gulick RNA, Nlellors IW, Havlir D, et al : Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency with infection and prior antiretroviral therapy. \' Engl /.tiled 1997, 337:; 34-739.
25. Montaner J, Reiss P, Cooper D, et al.: A randomized, doubleblinded trial comparing combinations of nevirapine, didanosine and zidovudine for HIV-infected patients: The INCAS Trial. I3.MA 1998, 279:930-937.
26. Wong J, Hezareh M, Gunthard H, et al.: Recovery of replica. tion-competent HIV despite prolonged suppression of plasma viremia. Science 1997, 278:1291-1295.
27. Finzi D, Hermankova M, Pierson T, et il : Idenlificaliun of a reservoir for HIV-1 in patients on highly active retroviral ther apy. Science 1997, 278:1295-1300.
28. Gupta P, Mellors J, Kingsley L, et al.: High viral load in semen of human immunodeficiency virus type t-infected men at all stages of disease and its reduction by therapy with protease and nonnucleoside reverse transcriptase inhibitors. I 1 rr,,; 1997, 71:6271-6275.
29. Vernazza PL, Gilliam BL, Flepp M, vI il : Effect of antiviral treatment on the shedding of HIV-1 in semen. 1u 11:1249-1254.
30. Cohen M, Hoffman I, Royce R, et al.: Reduction of concentration of HIV-1 in semen after treatment of urethritis: implications for prevention of sexual transmission of HIV-l. I,uu -r 1997, 349:1868-1873.
31. Yerly S, Rakik A, Kinloch-de-Loes S. el ,il.: Prevalence de la transmission de virus resistant a la zidovudine en Suisse. Schrveiz,tiled Wochenschr 1996, 126:113-15- 1848.
32. D'Aquila RT, Johnson VA, Welles SL, et ,rl.: Zidovudine resistance and HIV-1 disease progression during antiretroviral therapy. Ann Intern Med 1995, 122:-3(11-4118.
33. Montaner 1, Schechter M, Rachlis A, et ,il.: Didano%ine compared with continued zidovudine therapy for HIV-infected patients with 200 to 500 CD4 cells/mm': a double-blind, randomized, controlled trial. Ann Intern ,•)eel 10L)-) 12:3:5n I-5' 1 .
34. Goudsmit J, de Ronde A, Ho D, et al.: Human immunodeficiency virus fitness in vivo: calculations based on a single zidovudine-resistance mutation at codon 215 of reverse transcriptase. 1 Virol 1996, 70:5662-5664.
35. Larder B, Kemp S, Harrigan P: Potential mechanism for sustained antiretroviral efficacy of AZT-3TC combination therapy. Science 1995, 269:696-699.
36. Back N, Niihuis M, Keulen W, et al.: Reduced replication of 3TC-resistant HIV-1 variants in primary cells due to a proces sivity defect of the reverse transcriptase enzyme. 1%113() 1 1996, 15:-3040-4049.
37, Davies I: The ~Nfyth of Addiction. Reading: I lamrrml: 1 91)4 .