Rates of bacterial sexually transmitted infection (STI) among men who have sex with men (MSM) in the United States are probably at an all-time high. In King County, WA, the rate of syphilis among MSM is now almost 3 times that calculated in 1981 at the outset of the HIV epidemic, with an estimated 3% of HIV-infected MSM in the area acquiring syphilis each year.1 (To our knowledge, a similarly high incidence of syphilis has never been documented in any defined population.) King County is not unique; researchers and public health officials have observed similar trends nationally and internationally,2–4 and a recent analysis showed that the national incidence of primary and secondary syphilis in 15- to 24-year-old MSM rose from near zero in 1994 to 404 cases per 100,000 seven years later.5 Trends in gonorrhea are similar. The explosive increase in bacterial STI among MSM is, in many ways, a consequence of our success fighting HIV. Antiretroviral therapy has rendered that infection treatable, and many MSM have adopted seroadaptive behaviors that likely diminish their HIV risk while driving up their risk of bacterial STI.2,3,6 Although the population-level effects of HIV preexposure prophylaxis (PrEP) are still unknown, the intervention has the potential to further promote divergent trends in HIV and bacterial STI.
So the sexual health landscape for MSM is complex. There are reasons for optimism, including efficacious new interventions, reduced HIV associated mortality, and declining HIV incidence in some geographic areas and population groups.7 But there are also causes for concern. In the United States, new HIV infections are increasingly concentrated among southern African American MSM,8 and HIV threatens to replicate the national experience with syphilis and gonorrhea in heterosexuals in whom uneven prevention success has exacerbated profound racial and geographic disparities. Moreover, we in the STI prevention community increasingly feel that we face an out of control STI epidemic in MSM, and that we do not have a viable plan to deal with it. In this context, we need creative thinking.
In this issue of the journal, Bolan and colleagues present a study that reflects creative thinking and that suggests a novel approach to STI prevention: STI PrEP.9 They conducted a small randomized trial designed to assess the feasibility of doxycycline prophylaxis to prevent bacterial STI, particularly syphilis, among HIV-infected MSM at high risk for recurrent STI. The trial is interesting and provocative and challenges the STD community to think about whether, when, and how to use prophylactic antimicrobials to prevent bacterial STIs among MSM.
The investigators enrolled 30 HIV-infected MSM who were engaged with HIV care and had experienced at least 2 episodes of syphilis since being diagnosed as having HIV. The study randomly assigned participants to receive 36 weeks of either doxycycline 100 mg daily or contingency management, a financial incentive to stay free of STI. At baseline and 12-week intervals for 48 weeks, the participants were clinically evaluated and tested for rectal, pharyngeal, and urethral gonorrhea; rectal and urethral chlamydial infection; and syphilis. At each study visit, participants completed a computer self-interview to evaluate their sexual behavior. To monitor therapeutic compliance, doxycycline blood levels were measured at 3 follow-up visits in persons assigned to receive the drug. The primary study outcome was incident bacterial STI, including syphilis, gonorrhea, or chlamydial infection.
Seventy-seven percent of participants completed the study, and more than 90% of doxycycline treated persons had drug levels showing recent medication ingestion during at least 2 of 3 study visits. The trial observed an almost 75% reduction in incident STI measured over 48 weeks, a finding that was driven by lower rates of chlamydial infection and syphilis (but not gonorrhea) in participants receiving doxycycline. Reported sexual risk behaviors did not vary between the 2 study arms. The overall validity of these outcomes is indirectly supported by the differential benefit observed for the 3 STIs, with doxycycline apparently failing only for the infection for which tetracycline resistance is prevalent (although gonococcal antimicrobial susceptibility was not tested). The study suggests that PrEP against bacterial STI, which has strong biological plausibility, might be effective and convincingly demonstrates that a larger trial is feasible.
The implications of this study are controversial. Medical providers in general, and perhaps infectious diseases specialists in particular, do not typically like to treat patients with long-term antimicrobials, so for many of us, there is an almost reflexive objection to the idea of STI PrEP. However, doxycycline is commonly administered for weeks to months for significant but non–life-threatening conditions like acne and rosacea, and for weeks to years to prevent malaria. Trimethoprim-sulfamethoxazole and macrolide antibiotics are routinely used to prevent Pneumocystis jiroveci pneumonia, toxoplasmosis, and mycobacterial infections in people with AIDS. Given these precedents, it seems that the relevant question is when, if ever, long-term antimicrobial PrEP with doxycycline might be indicated in MSM at high risk for STI. Addressing this question requires balancing the benefits and risks of the intervention both for individuals at risk and for the population as a whole.
How important are the infections we hope to prevent? Is the goal to prevent infection, or only serious clinical outcomes? The most frequent complications and serious outcomes of bacterial STIs result from infection of the female genital tract and fetuses. Although the potentially serious outcomes in MSM are not to be minimized, in general serious morbidity from bacterial STIs is infrequent in males. Prospective cohort data suggest that at least 20% of HIV-infected MSM likely acquire a bacterial STI annually, most frequently rectal chlamydial infection.10 If prophylaxis prevented 75% of these STIs, 6.7 men would need to be treated for a year to avert one new infection (i.e., number needed to treat [NNT] = 6.7). However, because serious morbidity is uncommon in men, simply averting STIs, a large proportion of which would be asymptomatic, may not represent a large benefit. If King County’s experience is typical, approximately 3% of HIV-infected MSM acquire syphilis annually.1 This is about half the risk of malaria observed in placebo-treated participants in at least one malaria prophylaxis study,11 though perhaps comparable to the risk experienced by some lower-risk populations in which such prophylaxis is commonly used. Malaria is substantially more life-threatening than syphilis, so preventing malaria is probably more beneficial for individual patients than preventing syphilis. An estimated 2% to 8% of noncongenital syphilis infections result in early complications, mostly visual symptoms or hearing loss.12,13 These overt complications usually resolve with therapy, but some do not and syphilis may be a cause of more subtle neurological morbidity, an area of active research.14 Assuming prophylaxis is highly targeted to a population with an annual syphilis risk of 5%, that complications occur in 5% of cases, and that prophylaxis is 75% effective in preventing complications, the NNT would approximate 533 people to prevent one case of complicated syphilis. Similar considerations would apply to gonorrhea or chlamydial infection and complications such as epididymitis, proctocolitis, disseminated gonococcal infection, and reactive arthritis. These are very rough, “back of the envelope” estimates, but they suggest that the NNT for STI PrEP is low if the objective is to prevent any STI, but quite high if the goal is to prevent significant morbidity. An HIV-uninfected person taking doxycycline might also lower his risk for acquiring HIV infection, but the magnitude of that benefit is difficult to estimate.
Prophylaxis might also have a public health benefit in preventing HIV infection, but the magnitude of the effect is difficult to estimate. Chesson et al.15 calculated that 3% to 5% of all HIV infections among African Americans in the United States were attributable to syphilis in 2000, but we are unaware of contemporary estimates of the faction of HIV cases in MSM attributable to bacterial STI, some proportion of which might be averted with STI PrEP. Treating people with the primary objective of advancing the public’s health as opposed to the individual patient’s health presents difficult ethical issues. We do not typically ask patients to accept the expense, inconvenience, and side effects of medical therapy with the primary objective of improving the population’s health. That said, many patients in fact are open to considering public health as one factor in their medical decisions.16 Another public benefit might be reduced health care costs; the cost of diagnosing, treating, and following patients with syphilis is not inconsiderable.
Is long-term doxycycline safe? Bolan’s small trial provides few data on this issue, but doxycycline is an old drug and there is extensive experience with both short- and long-term administration. Its use is associated with phototoxicity, gastrointestinal symptoms, and, more rarely, esophageal ulceration and perhaps inflammatory bowel disease.17 It is not associated with the development of Clostrium difficile infection and may actually be protective against C. difficile.18 Recent data suggest that prolonged use induces changes in the gut microbiota and possibly abnormal weight gain.19 Thus, although severe, clearly defined complications are rare, mild symptoms are relatively common, and long-term use of the drug may have important consequences that are not well defined. Because both complicated STI and major toxicity associated with doxycycline are uncommon, a randomized trial probably could not fully address the balance of risk and benefits associated with long-term prophylaxis.
The population-level risks of bacterial STI prophylaxis are likewise hard to define. Long-term use of doxycycline is likely to promote resistance in patients’ endogenous microbial flora and could lead to more widespread doxycycline resistance in MSM. Although we agree with the authors that this issue may not be critical for currently identified STI pathogens, the effect on bacteria like Staphylococcus aureus, Streptococcus pneumoniae, enterococci, coliforms, and other bacteria may be important. On the other hand, depending on how narrowly clinicians target their use of STI PrEP, the population impact might be limited.
So where do we go from here? The investigators’ goal in undertaking the current study was to define the feasibility and potential value of a larger trial. Sexually transmitted infection PrEP has high biological plausibility, and the trial’s success in enrolling patients and achieving a high level of patient adherence is convincing that such a trial is feasible. However, to us it seems premature to commit the substantial resources such a trial would require; a number of questions need to be addressed first. At a minimum, these should include literature reviews and consultations to estimate the potential individual-level benefits and risks associated with STI PrEP, including but not limited to microbiome effects and the individual health implications of those effects; dynamic mathematical modeling to estimate the potential population-level risk and benefits of the intervention, including cost effectiveness; and community consultations to assess the candidate population’s willingness to accept risk related to STI PrEP and the value its members place on averting STI, particularly asymptomatic or readily treatable STIs.
Although we have misgivings about STI PrEP as an intervention at this time, Bolan and his colleagues deserve credit and our collective thanks for creative thinking. STI among MSM is among the most pressing issues facing our field. We need to think more deeply about what our goals are in terms of STI prevention in MSM, and we need to try new things. Creativity is in short supply, and we need to be open to new ideas.
1. Public Health–Seattle & King County. 2012 King County Sexually Transmitted Diseases Epidemiology Report. Seattle, WA, USA: Public Health–Seattle & King County, 2013.
2. CDC. Sexually Transmitted Disease Surveillance 2012. Atlanta: CDC, November, 2013.
3. Savage EJ, Marsh K, Duffell S, et al. Rapid increase in gonorrhoea and syphilis diagnoses in England in 2011. Euro Surveill 2012; 17.
4. Savage EJ, Hughes G, Ison C, et al. European Surveillance of Sexually Transmitted Infections. Syphilis and gonorrhoea in men who have sex with men: A European overview. Euro Surveill 2009; 14.
5. Peterman TA, Su J, Bernstein KT, et al. Syphilis in the United States: On the rise? Exp Rev Anti Infect Ther. In press.
6. Golden MR, Stekler J, Hughes JP, et al. HIV serosorting in men who have sex with men: Is it safe? J Acquir Immune Defic Syndr 2008; 49: 212–218.
7. Buskin SE, Jaenicke T. HIV/AIDS Epidemiology Profile for Community Planning. Seattle, WA: Public Health–Seattle & KIng County, Washington State Department of Health, 2013.
8. CDC. HIV Surveillance in Men how Have Sex with Men. Atlanta: CDC, 2013.
9. Bolan RK, Beymer MR, Weiss RE, et al. Doxycycline prophylaxis to reduce incident syphilis among HIV-infected men who have sex With men who continue to engage in high-risk sex: a randomized, controlled pilot study. Sex Transm Dis 2015; 42: 98–103.
10. Mayer KH, Bush T, Henry K, et al. Ongoing sexually transmitted disease acquisition and risk-taking behavior among US HIV-infected patients in primary care: Implications for prevention interventions. Sex Transm Dis 2012; 39: 1–7.
11. Ohrt C, Richie TL, Widjaja H, et al. Mefloquine compared with doxycycline for the prophylaxis of malaria in Indonesian soldiers. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 1997; 126: 963–972.
12. Dombrowski JC, Pederson R, Marra CM, et al. Complicated syphilis may be more common than previous estimates suggest. Sex Transm Dis 2014; 41: S105.
13. CDC. Symptomatic early neurosyphilis among HIV-positive men who have sex with men—Four cities, United States, January 2002-June 2004. MMWR Morb Mortal Wkly Rep 2007; 56: 625–628.
14. Marra CM, Deutsch R, Collier AC, et al. Neurocognitive impairment in HIV-infected individuals with previous syphilis. Int J STD AIDS 2013; 24: 351–355.
15. Chesson HW, Pinkerton SD, Voigt R, et al. HIV infections and associated costs attributable to syphilis coinfection among African Americans. Am J Public Health 2003; 93: 943–8.
16. Dombrowski JC, Harrington RD, Fleming M, et al. Letter to the editor: Treatment as prevention: Are HIV clinic patients interested in starting antiretroviral therapy to decrease HIV transmission? AIDS Patient Care STDS 2010; 24: 747–750.
17. Lee TW, Russell L, Deng M, et al. Association of doxycycline use with the development of gastroenteritis, irritable bowel syndrome and inflammatory bowel disease in Australians deployed abroad. Intern Med J 2013; 43: 919–926.
18. Turner RB, Smith CB, Martello JL, et al. Role of doxycycline in Clostridium difficile
infection acquisition. Ann Pharmacother 2014; 48: 772–776.
19. Angelakis E, Million M, Kankoe S, et al. Abnormal weight gain and gut microbiota modifications are side effects of long-term doxycycline and hydroxychloroquine treatment. Antimicrob Agents Chemother 2014; 58: 3342–3347.