Another STI associated with HIV-1 acquisition

In this issue, Napierala Mavedzenge and colleagues [1] report a two-fold increased risk of HIV-1 acquisition among Zimbabwean and Ugandan women who were infected with Mycoplasma genitalium, and attribute approximately 9% of HIV-1 acquisitions to infection with this emerging sexually transmitted infection (STI). They conducted a nested case–control study in which they matched HIV-1 seroconversions (cases) to women of similar age and risk who were HIV-negative at the same point in time (controls) and identified M. genitalium infections at the visit 3 months before HIV detection. With this rigorous study design, they were able to clearly show that M. genitalium infection preceded the acquisition of HIV-1. Whereas evidence suggesting a link between M. genitalium and HIV infection has been accumulating for some time now [2], this is the first report from a longitudinal study and thus presents the most compelling evidence to date that M. genitalium may be involved in HIV-1 acquisition. Nevertheless, this finding is hardly surprising. Increased risk for HIV acquisition in longitudinal studies has been previously shown for individuals infected with essentially every other STI [3–6]. Given this, it would have been more surprising if M. genitalium was not associated with HIV acquisition. The larger question lies in what the implications of this new evidence are for HIV prevention and STI care. Notably, the prevalence of M. genitalium in this sample was higher than that of any other bacterial STI. At the visit prior to the detection of HIV-1, M. genitalium was detected in 9.4% of women, whereas Neisseria gonorrhoeae was only detected in 5.8% of women, and Chlamydia trachomatis in only 4.7%. This higher prevalence and incidence of M. genitalium relative to N. gonorrhoeae and C. trachomatis has been observed in other African settings, although not all [7–10]. Despite this relative importance, M. genitalium has received little attention, partly due to its later emergence on the scene. Detected for the first time in 1980 [11], systematic studies were not possible until after the development of nucleic acid amplification tests (NAATs) in the early 1990s [12]. Culture and staining methods of detection are not effective given its fastidious nature and lack of a cell wall [13], making diagnosis difficult. As Napierala Mavedzenge and colleagues appropriately note, this is the first report from a longitudinal study and the finding needs to be confirmed in other studies. However, assuming this association with HIV-1 acquisition is confirmed, they propose that M. genitalium screening and treatment among women at high risk may be warranted as part of an HIV prevention strategy. Whereas this is a logical conclusion, there are a number of challenges associated with this approach. Successful screening programs have two distinct requirements. The first of these is the existence of an accessible sensitive and specific diagnostic test. Whereas commercial diagnostic tests for M. genitalium are available in some parts of the world (although not currently in the US), the assays are sophisticated NAATs which require expensive equipment and reagents, as well as highly trained personnel and are typically only available in large urban medical centers. Rapid point-of-care diagnostics for M. genitalium do not exist. This is a shared challenge for the diagnosis of other STIs and is the reason that syndromic management of STI is widespread in resource-poor settings. Given the currently available diagnostic assays for M. genitalium, the requirement for rapid results that would allow treatment on the basis of a positive diagnostic test is not met. The second requirement of a successful screening and treatment program is the existence of an effective treatment regimen for the infection. As Napierala Mavedzenge and colleagues note, the tetracyclines that form the backbone for syndromic treatment regimens have poor efficacy against M. genitalium[14]. Whereas M. genitalium-associated male urethritis responds better to azithromycin than doxycycline [15,16], there are increasing concerns about the development of azithromycin resistance. The most recent trial among men demonstrated a 33% failure rate for azithromycin in the treatment of M. genitalium infection [16]. Moxifloxacin currently remains effective against M. genitalium[17], but it is expensive and anecdotal reports of treatment failure after moxifloxacin have begun to emerge. There is serious question as to the accessibility of an effective antibiotic for M. genitalium. Perhaps more important than concerns about whether requirements are met for an effective screening and treatment program, however, is the fact that previous efforts to treat STIs to reduce risk for HIV acquisition have not been sufficiently successful. Introduction of syndromic management for bacterial STI was only of benefit in one trial and that could not be replicated [18]. Furthermore, Napierala Mavedzenge calculated that the proportion of HIV-1 infections attributable to M. genitalium was only 8.7%, as compared to the 72.6% attributed to HSV-2 infection, an incurable STI whose treatment to diminish HIV risk has been similarly ineffective [19,20]. In mature epidemics, only a small minority of new HIV infections are associated with curable STIs [21] and, whereas STI control as an end in itself is an important endeavor, it is more suitable in the context of HIV prevention as an adjunct to larger combination prevention activities [22]. A further limitation of this and other epidemiologic studies to identify STI pathogens associated with HIV-1 acquisition is the virtually impossible task of disentangling the potential biologic effects of M. genitalium infection from the risk behavior that resulted in exposure to both HIV and M. genitalium. Napierala Mavedzenge and colleagues showed that nearly one-quarter of HIV-1 acquisitions could be attributable to partner risk, a far larger proportion than were attributable to M. genitalium infection. Association does not equal causation and it is quite possible that the same risk behaviors that led to M. genitalium infection were also responsible for subsequent HIV infection. In this era of biomedical prevention successes such as male circumcision, treatment as prevention, and pre-exposure prophylaxis (PrEP), emphasis on interventions to reduce risk behavior have waned. However, risk behavior remains an important and potentially modifiable contributor to the HIV epidemic worldwide and efforts to reduce risk behavior should continue to be combined with evidence-based biomedical strategies. Identifying new factors that influence susceptibility to HIV-1 is important if we are to expand our repertoire of successful interventions. Napierala Mavedzenge and colleagues have made an important contribution to this endeavor by demonstrating for the first time that M. genitalium plays a role in HIV-1 acquisition. Nevertheless, screening and treating M. genitalium is an extremely difficult and costly undertaking, and unlikely to have a significant impact on reducing HIV-1 incidence. Our HIV prevention resources should be focused on implementing those proven interventions that together will have the largest impact. Acknowledgement L.E.M. would like to thank Jared M. Baeten for helpful discussions on this topic. Conflicts of interest L.E.M. has received study drugs from Pfizer, Inc. and conference support from Bio-Rad.