Who benefits when a person living with HIV takes medications to slow the virus’ course? At the XVI International AIDS Conference in Toronto in August, Dr. Julio Montaner promoted a new strategy for HIV prevention. His mathematical model suggests that by creating universal access to HIV treatment, HIV’s contagion would be tamped down until the virus is virtually eliminated worldwide.
Montaner’s model assumes “that all HIV-infected people would be given therapy in the first year and that, after the first year, there would be no new HIV infections.” He bases this optimistic outcome on research that links HIV treatment with reduced HIV case rates in some regions. Montaner actually debuted his strategy earlier in the month, in the medical journal Lancet. There he noted high ratios of new infections to existing infections in countries with low HIV treatment coverage. With universal treatment, Montaner predicts that in 45 years, worldwide HIV infections could drop by 70%, to less than one infected person per 1,000 people living in any region.
This strategy wouldn’t come cheap. Even if medicines were discounted to just $1 per person per day, the plan would cost over $300 billion dollars before it achieved this success.
Montaner’s proposal grabbed headlines around the world, bringing the spotlight back onto the issue of healthcare access. Free universal HIV treatment access would certainly save lives. The UNAIDS and World Health Organization’s “3 by 5” campaign, launched in 2003, sought to bring HIV treatment to three million people by 2005.
While some progress was made, the campaign never fully achieved its objectives. In developing nations today, just 24% of people living with HIV can access treatment, even if they have progressed to clinical AIDS.
That means that three-out-of-four people living with HIV worldwide are forced to unnecessarily suffer the fate that U.S. patients faced only in the earliest days of HIV, when no treatment existed to slow the virus’ reproduction.
While Montaner’s proposal makes humanitarian sense for those living with HIV, would the “treatment as prevention” rationale for the model make sense here in the United States, where treatments are already widely available? Would free treatment mean that significantly more people living with HIV here would undertake treatment? And, if treatment stops HIV transmission, does that mean that persons living with HIV should initiate their treatment right away, in the very first year of their infection?
Does Lowering HIV Levels in the Bloodstream Reduce HIV Transmission Efficiency?
This is the central premise of Montaner’s proposed “treatment as prevention” strategy. Yet even today, 25 years into the HIV pandemic, the link between HIV levels in the blood and levels released through sexual fluids is not fully proven. Several studies have established an association between high HIV blood levels and higher odds of sexual transmission of HIV, but most do not prove causation.
In almost all cases, another factor might have explained the reduced rates of HIV transmission documented. Scientists call these unaccountable factors “confounding variables,” because they make a sticky business of trying to draw any clear conclusions between cause and effect.
In his Lancet article, Montaner summarizes several of the studies that tried to prove the link between HIV blood levels and transmission. He first notes that pregnant women living with HIV who take potent HIV treatments significantly reduce their chances of infecting their child during delivery. Indeed, HIV treatment for expectant mothers have dropped the odds of passing HIV to a child from roughly 25% to just 2%.
But childbirth represents an entirely different type of HIV transmission dynamic than does sexual exposure. The high risk in childbirth is due to direct and prolonged contact with the mother’s blood, not with sexual fluids.
The Ugandan Study
Montaner also cites a famous study by Thomas Quinn, director of the Johns Hopkins Center for Global Health, that showed persons living with HIV in Uganda did not transmit their virus to sexual partners if they had a very low amount of HIV in their bloodstream (under 1,500 copies per ml). This study has been endlessly quoted as proof that HIV treatment halts HIV transmission in its tracks.
In Quinn’s study, however, none of the infected individuals were taking HIV treatments. Those with low viral loads simply had an unusually robust natural immune response to their infection. In addition, the patients Quinn studied were counseled extensively to avoid risk behaviors with their partners. They were also treated for other sexually transmitted diseases, which would otherwise “amplify” HIV’s transmission efficiency. So multiple causes may have contributed to the outcome.
All that should be inferred from Quinn’s study is that those people lucky enough to have such a powerful immune response to HIV are probably unlikely to pass the virus to anyone else. Unfortunately, these folks are the exception to HIV’s normal course. In fact, those able to keep their virus undetectable without medicines represent just 0.33% of all persons living with HIV. Researchers at Massachusetts General Hospital are now proposing to study the many ways that the so-called “elite controllers’” bodies deal differently with the virus.
The link between HIV transmission and naturally high or naturally low levels of HIV in the bloodstream is well established. We know from Maria Wawer’s more recent Ugandan study of monogamous couples that newly HIV-infected men were 12 times more likely to infect their partners than were those with longer standing infection. During the first two months of infection, HIV levels in the bloodstream are at their peak because the immune system has not yet mounted an effective counter attack. Very high levels of HIV in the bloodstream are more likely to result in infectious amounts of HIV being released through sexual fluids.
It is much more difficult to establish whether viral load lowered through treatment has the same dampening effect on HIV transmission as does naturally low viral load. HIV medications face challenges reaching all sites in the body. This is why speculating about treated persons from studies done with untreated persons introduces some guesswork.
Other Countries
Most of the other studies Montaner cited in the Lancet are “ecological” in their methodology, meaning they evaluated the community-wide rates of HIV prevalence after treatment finally became commonly available.
True, in most instances, overall HIV prevalence dropped when treatments became part of a country’s public health system. Yet it’s also true that people in the developing nations studied were often poorly educated about HIV’s transmission dynamics and pathology. When treatments were rolled out to these communities, HIV awareness and prevention efforts were also often initiated or widened, making it difficult to distinguish which factor actually led to the decline in new infections, the community education, or the treatments.
More recently, two studies have tried to eliminate these “confounding variables,” and prove once and for all that treatment, not other risk reduction or awareness efforts, directly causes lower HIV rates.
In one study, when free treatment became available in Taiwan, new HIV infections declined by 53%. Montaner argues that there was no reduction in community risk behaviors that would account for the decline in HIV rates. Syphilis and gonorrhea case rates were unchanged during the study period.
However, syphilis is a poor marker for tracking HIV-specific risk behaviors. Though these diseases are spread through sexual contact, as is HIV, their different transmission efficiencies can create a disconnect between unprotected sexual encounters and the resulting infections with one, but not the other, pathogen.
In San Francisco two years ago, a startling surge in syphilis rates did not lead to the expected tsunami wave of new HIV infections. That’s because syphilis is so contagious that people can easily catch it during foreplay even if they use condoms during the most risky moments of sex. This contagion difference also exists, to a lesser extent, with gonorrhea.
So folks in San Francisco may have been having more sexual encounters or more partners, but donning condoms when it came to anal or vaginal sex that could have spread HIV. And people in Taiwan may have similarly reduced their riskiest of sexual practices, and that may have contributed to the overall reduction in HIV rates, even while these other STD rates could go unchanged. Such a community change could lead to reduced HIV prevalence without the credit going to HIV treatments.
The Missing Link
The only study to truly link HIV viral load levels to transmission odds with a specific sexual partner comes from Thailand. Researcher Sodsai Tovanabutra found that successfully treated HIV-positive husbands did not transmit the virus to their uninfected wives. The cause-and-effect proof was tracked specifically, with each couple’s risk ranked against the viral load recorded in the positive partner. For those men living with HIV who could not control their virus despite treatment, every tenfold increase in viral load was associated with an 81% increase in the odds he would transmit HIV. Nearly all infected partners in Tovanabutra’s study had HIV 1 subtype E.
If HIV Treatment Does Stop HIV Transmission, Should it Be Used for Prevention?
The Myth
If future studies confirm the transmission dampening effect of treatment for those infected with HIV 1 subtype B, the predominant strain in North America, then the rationale for Montaner’s treatment as prevention strategy would be affirmed, rather than just optimistically inserted into a mathematical model.
Here in the United States, though, the practice could also reinforce a dangerous urban myth. People successfully managing their HIV treatment sometimes refer to themselves as “HIV neutral.” They recognize that they still have virus in their bodies (they are “positive”), but they believe that treatment makes them incapable of transmitting it to anyone else (so “neutral” to others).
Yet HIV rarely follows such predictable associations. A University of Pittsburgh study found that while people with high levels of HIV in their bloodstream consistently “shed” HIV through their sexual fluids, even those with undetectable viral load sometimes released virus.
HIV does not exist as one uniform community in a person’s body. There is often a “disconnect” between the virus that is suppressed in the bloodstream and HIV colonies that have grown resistant to treatment, hiding out “compartmentalized” in the testes. The Pittsburgh researchers found that some people with HIV suppressed in their bloodstream still transmit HIV originating from resistant colonies in their genitals. They can transmit the virus because the resistant population is high in the testes, increasing the odds that it will spill out in the sexual fluids. This is especially bad news for the partner who may acquire not just HIV, but the treatment-resistant form that had developed in their sexual partner.
Even if suppressed HIV did always stop HIV transmission, this pattern would only protect one’s partners on the nights that a person’s virus levels remained suppressed. A viral load test essentially takes a snapshot of the HIV population as it travels the body’s highways, the veins. It cannot present a real time picture of HIV’s movements. A Johns Hopkins study last year closely followed 10 patients with consistently undetectable HIV levels (verified in numerous check ups over a three month period). By the end of three months, HIV had spiked up to detectable levels for nine of the 10 persons at least once during the study. Undetectable viral load at last check up is good news for the person living with HIV, but not a guarantee for tonight’s encounter.
Maybe Montaner’s premise is right, and HIV treatment would impact HIV transmission in general, as long as people on treatment did not overcompensate by increasing their practice of unprotected sex because they assumed they could not transmit the virus. And as long as they stay on their treatment faithfully, not missing doses, as this can allow the virus to mutate around the medications, and then pass to another person as resistant HIV.
What is the Primary Purpose of HIV Treatment?
The Problem With Forced Treatment
Let’s put all of these nuanced questions about treatment and transmission aside and simply assume that HIV treatment is one day finally proven to always stop HIV transmission. Wouldn’t the next logical step be to require all diagnosed persons living with HIV to undertake treatment?
Treatment would then essentially accomplish the same goals as would a worldwide vaccine. I can imagine that many people would conclude that anyone unwilling to take on treatment was needlessly allowing the chain of potential contagion to extend. Policymakers might even feel that such personal decisions would need to be outlawed, for the good of public health.
Yet another moral question would also call for an answer: Is it ethical to prescribe medications to persons living with HIV not primarily to improve their own health, but specifically to protect future sexual partners?
In other disease models, patients are at times forced to undertake compulsory treatment for the sake of public health. People living with tuberculosis (TB), for example, are highly contagious to those who merely breathe the same surrounding air, and so are often required to take on long treatment cycles. Those who do not take their treatment properly can be confined to establishments that dole out “directly observed therapy.” This helps both the patient and the community, as delayed treatment is the primary cause of the two million TB deaths each year.
But HIV is different. First, it’s almost exclusively transmitted through sexual intercourse, shared IV needles, or childbirth. That means that uninfected people can take steps to protect themselves from HIV, unlike a person who unknowingly shares a cubicle with a colleague who has tuberculosis. That shifts much of the burden for controlling the epidemic onto the HIV-negative community.
Second, immediate compulsory treatment is improper with HIV because our nation’s HIV treatment guidelines recommend postponing treatment until the life saving benefits of antiretroviral medications are truly needed, not rushing to treat during the period when a person’s own immune system can control the virus well. While every person’s progression differs, treatment may not be indicated for the average person until eight years into one’s infection. The Department of Health and Human Services recommends this later starting point for treatment (based on viral load and CD4 count) because the medications bring their own complications, such as potential cardiovascular abnormalities or renal failure.
If HIV treatment was ever made compulsory for the sake of prevention, people living with HIV might “burn through” the best treatments early in their infection, and then not have anything to fall back on when their immune prognosis becomes dire. Such a model would consign those living with HIV to years more side effects and complications, and possibly a shorter life expectancy, all in the name of protecting HIV-negative people from the virus.
Biology or Politics in Public Health?
My gut instinct is that Montaner’s proposal to use treatment as prevention is based in savvy politics as much as it is in science. The world’s wallet has been slow to open to help those already living with HIV. In the cruel calculus of humanitarian aid, HIV-infected people in third world nations may be “written off” in favor of other more “desirable” groups seeking support.
By suggesting that treatment can protect future HIV infections, maybe Montaner’s strategy can melt some icy hearts. In the United States, little was done to address HIV as long as it seemed to affect only gay men and Haitians. It took an “innocent victim” to justify our nation’s Ryan White CARE Act. Montaner’s team might recognize that even today people care more about the currently uninfected than about HIV-positive people in third world countries.
But medications should be dispensed primarily to benefit people living with a disease, not packaged in a rationale to defend the rest of us from “their” illness. Such a philosophy can easily morph into a replay of the early days of AIDS, when people living with HIV were said to be “dirty,” and when some pundits recommended moving all HIV-positive people to an isolated island.
Even if medical rather than behavioral science were the best bulwark against HIV, universal and perpetual treatment is not the most efficient option for stemming the worldwide epidemic.
Circumcision has been proven repeatedly to decrease a man’s susceptibility to HIV by at least 60% in acts of penetrating sex. This protective effect would also lower the odds of an HIV-infected man passing the virus to partners, since less men would have any virus in their bodies to pass along. Unlike the treatment as prevention strategy, circumcision would be a one-time medical procedure and would not depend on timely HIV diagnosis, or the complications of daily treatment seen in the treatment-as-prevention model.
Of course, neither strategy will protect every individual. That’s why we still need one-on-one prevention education. As a start, the recent proposal from the CDC (Centers for Disease Control and Prevention) to increase voluntary HIV testing by making it a part of regular medical check ups will help identify more persons living with HIV.
Yet diagnosis is not a panacea for prevention. Most PLWH (people living with HIV) try to protect their sexual partners. Yet some lack the skill sets or support networks to negotiate safer sex or to disclose to sexual partners. Others struggle with issues (homelessness, substance abuse, living in a domestic abuse situation, etc.) that can push safer sex or disclosure further down their list of priorities.
And even if all of the persons living with HIV who know their status could be guided to always protect their partners, there would always be a portion of newly infected people who would not yet be diagnosed.
Most people don’t visit their healthcare provider every day. Some will also continue to receive a false negative result during the window period of uncertainty following an infection. Newly infected persons living with HIV are, recall, 12 to 20 times more likely to release infectious levels of virus in a sexual encounter. So prevention must continue to focus on presumed HIV-negative individuals as well, to guide them to adopt safer sex practices.
Back to Basics
Effective HIV prevention requires trained professionals delivering behavioral messages repeatedly to at-risk populations around the nation. Just as Montaner’s study hopes to spark interest in funding universal treatment to slow the world’s epidemic, so here in the United States Johns Hopkins researcher David Holtgrave is calling for greater investment in prevention efforts.
The CDC’s 2001 Strategic Plan called for halving the number of new infections by 2005; yet the numbers remain unchanged, or possibly even on the rise. Holtgrave estimates that the rate of new infections in the U.S. could be halved if the five million persons most at-risk received prevention education. This effort would require increasing the annual prevention budget from $720 million by 57%.
Ideally, governments would see the wisdom of investing in treatment solely for the benefit of persons living with HIV, while also investing in prevention to protect against further infections. Both investments would pay off.
Stephen Fallon is the President of Skills4, Inc., a healthcare and disease-prevention consulting firm that specializes in gay lifestyle and health issues by providing workshops, technical assistance, and grant writing services to community organizations and health departments. Visit his website at www.skills4.org. If you need sources for any medical information cited in his columns, e-mail him at sfallon@skills4.org.
References:
- Montaner presents theory at International AIDS Conference: Montaner J. “Re-evaluating the cost-effectiveness of HAART - The case for expanding treatment access to curb the growth of the epidemic.” XVI International AIDS Conference, August 16th, 2006, Toronto, CA. Abstract WEPL01.
- Strategy debuted in the Lancet: Montaner JSG et al. “The case for expanding access to highly active antiretroviral therapy to curb the growth of the HIV epidemic.” The Lancet 368: 531-536, 2006.
- Uganda study was with treatment naïve patients counseled to avoid risks with partners: Quinn T, et al. “Viral load and heterosexual transmission of human immunodeficiency virus type 1.” New England Journal of Medicine 342(13): 921-29 (2000).
- Newly infected heterosexual Ugandans were 12x more likely to infect their wives: Wawer M. “Rates of HIV-1 Transmission per Coital Act, by Stage of HIV-1 Infection, in Rakai, Uganda. The Journal of Infectious Diseases.” Journal of Infectious Diseases, May 1, 2005;191:1391-1393,1403-1409.
- HIV levels in semen typically 20x higher during first two months of infection: Pilcher C., et al. "Estimating transmission probabilities over time in acute HIV infection from biological data," 9th Conference on Retroviruses and Opportunistic Infections, Feb. 24-28, 2002; Poster 366-M.
- STDs amplify HIV’s transmission rate 2x-3x higher: Sexton J, Garnett G, Rottingen J. "Metaanalysis and Metaregression in Interpreting Study Variability in the Impact of Sexually Transmitted Diseases on Susceptibility to HIV Infection." Sexually Transmitted Diseases June 2005; Vol. 32; No. 6: P. 351-357.
- Gonnococcal urethritis amplifies HIV transmission 10x: Cohen MS, Hoffman IF, Royce RA, et al. “Reduction of concentration of HIV-1 in semen after treatment of urethritis: implications for prevention of sexual transmission of HIV-1: AIDSCAP Malawi Research Group.” Lancet. 1997;349:1868-1873.
- Free treatment cuts new HIV infections in Taiwan: Jung Der-Wang, et al. Journal of Infectious Diseases 2004;190:879-885.
- Disconnect between new STD rates and HIV rates means STD risks do not always equal HIV risks: "San Francisco Serosorting May Explain Odd HIV Data: STDs Have Risen, but Not New HIV Infections.” AIDS Alert May 1, 2004 Vol. 19, No. 5, P. 55.
- First true link demonstrates decreasing VL = decreasing odds of transmission for HIV 1-E: Tovanabutra S, et al. “Male Viral Load and Heterosexual Transmission of HIV-1 Subtype E in Northern Thailand. Journal of Acquired Immune Deficiency Syndromes. 29(3):275-283, March 1, 2002.
- Spikes occur in patients who generally have UDVL: Nettles R, "Intermittent HIV-1 Viremia (Blips) and Drug Resistance in Patients Receiving HAART," Journal of the American Medical Association 2005;293(7):817-829.
- Disconnect between blood and semen HIV viral load levels: Gupta P. “HIV-1 shedding pattern in semen correlates with the compartmentalization of viral quasispecies between blood and semen.” Conference on Retroviruses Opportunistic Infections 2000 Jan 30-Feb 2; 7:96 (abstract no. 114).
- Suppressed plasma viremia despite resistant HIV in testes: Eron, J. et al. “Resistance of HIV-1 to antiretroviral agents in blood and seminal plasma: implications for transmission.” AIDS 12(15): 181-189. October 22, 1998.
- Directly observed TB therapy: Salomon J, et al. "Cost-Effectiveness of Treating Multidrug-Resistant Tuberculosis," Public Library of Science Medicine 2006;3(7):e241.
- Protective effect of male circumcision: Gray R, Wawer M, Thoma M, et al. “Male circumcision and risks of female HIV and sexually transmitted infections acquisition in Rakai, Uganda.” 13th Conference on Retroviruses and Opportunistic Infections; February 5-8, 2006; Denver, Colorado. Abstract 128. And Quinn TC. “Circumcision and HIV transmission: the cutting edge.” 13th CROI. Abstract 120.
- Halve U.S. HIV infections by increasing reach of prevention education: Holtgrave DR. Curran JW. What Works, and What Remains to Be Done, in HIV Prevention in the United States. Annual Review of Public Health, in press.
|
