Ebola Virus RNA in the Semen of Male Survivors of Ebola Virus Disease: The Uncertain Gravitas of a Privileged Persistence

A short-term consultancy for the World Health Organization (Geneva, Switzerland) began after the author agreed to write this commentary. (See the brief report by Sow et al on pages 1475–6.) The 2014–2016 outbreak of Ebola virus disease (EVD) in West Africa has been devastating in scope and severity, resulting in >11 000 deaths and leaving in its wake >10 000 EVD survivors [1]. A steep learning curve over this time has included a growing understanding of the survivor's predicament, as grateful emergence from Ebola treatment units (ETUs) is tempered by the realization of EVD's aftermath [1–5]. The recognition that Ebola virus may persist long after clearance from the blood, particularly in immune-privileged tissues, has been a key aspect of this ongoing tutorial. Reports of uveitis [6] and meningoencephalitis (Michael Jacobs, personal email communication) associated with viral persistence suggest that organ- or tissue-specific inflammatory syndromes may be of consequence to the individual survivor. In what we hope is the last mile of this outbreak, the public health consequence and the residual risk of sexual transmission of persistent virus in the semen of male EVD survivors are increasingly under the microscope [7]. Prior to this outbreak, Ebola virus had only been detected in the semen of a few male survivors by cell culture (maximum duration 82 days after disease onset) [8] and by detection of viral RNA by reverse transcription polymerase chain reaction (RT-PCR; maximum duration, 101 days after disease onset) [9]. The clinical and epidemiologic gravitas of these limited findings was uncertain, and no cases of sexual transmission of Ebola virus had been clearly documented. That changed when Mate et al provided the first compelling molecular evidence of human-to-human sexual transmission from a Liberian EVD survivor with EBOV RNA detectable in semen 199 days after disease onset [10, 11]. Virus was not detected by cell culture in semen specimens from this survivor. In a coincident first lesson, Deen et al reported the persistence of Ebola virus RNA in initial semen samples from 9 of 9 Sierra Leonean survivors (100%) 2–3 months after disease onset, 26 of 40 (65%) 4–6 months after onset, and 11 of 43 (26%) 7–9 months after onset [12]. These findings and their implications for sexual transmission have recently been comprehensively reviewed [13, 14]. There are many more than 1200 EVD survivors in Guinea. In this issue of The Journal of Infectious Diseases, Sow et al provide the first reports of viral RNA persistence in 68 male survivors of EVD from Conakry and Macenta, Guinea, in whom 8 of 68 men (11.8%) representing 10 of 98 semen samples (10.2%) had detectable Ebola virus RNA in semen. The prevalence of positive RT-PCR results for all samples appeared to decrease over time, with positive results for 4 of 14 specimens (28.5%) collected at 1–3 months, 3 of 18 (16.5%) collected at 4–6 months, 2 of 31 (6.5%) collected at 7–9 months, 1 of 29 (3.5%) collected at 10–12 months, and 0 of 6 (0%) collected at >12 months. One subject with longitudinal follow-up testing showed increasing cycle threshold values over time. Otherwise, the data reported are initial single samples. The authors acknowledge limitations, including the lack of data from viral culture and from genomic sequencing, noting that both are pending. In addition, the use of different PCR methods makes direct comparison between the Conakry and Macenta findings difficult, particularly among small numbers of patients. Nonetheless, these data confirm in Guinean EVD survivors the persistence of viral RNA in the semen. As is typical of the last 2 years, more questions than answers remain. First, what is the true prevalence of persistent viral RNA in EVD survivors? In this study, far fewer Guinean men had detectable RNA in semen when compared to Sierra Leonean survivors at similar time points (28.5% vs 100% at 1–3 months and 16.5% vs 65% at 4–6 months, respectively). This could be a true biological difference or related to sampling selection, sample handling, and variable sensitivity of the particular PCR targets. Understanding of the characteristics of EBOV-specific RT-PCR as applied to semen specimens is limited but undergoing more-careful characterization [15, 16]. We will learn more from ongoing research in all 3 countries, as well as from operational data that emerge from country-wide semen testing programs now offered to EVD survivors outside of study settings. If these findings reflect a biological difference between these populations, interesting questions about the predictors and pathogenesis of semen persistence remain (discussed below). Second, what are the implications of persistent viral RNA for semen infectivity? The relationship between the detection of viral RNA, the isolation of virus in cell culture, and the risk of sexual transmission is yet to be determined. Pending characterization by semen culture, results in these Guinean men and the referenced Sierra Leonean survivors should help clarify infectivity, at least by that standard. There is a cautionary hint of principle offered by the Liberian transmission [10, 11] that culture negativity may not rule out the possibility of transmission. In at least one EVD survivor tested at multiple time points, viral RNA was detected for many months after results of semen culture were negative (unpublished clinical data), suggesting ongoing viral replication. Perhaps viral isolation in cell culture is not a sensitive enough gold standard to rule out viability and infectivity. Additional downstream investigations that could inform this uncertainty include the detection of ongoing viral replication by (1) assessing EBOV messenger RNA expression in semen samples; (2) assessing sequential virus evolution in semen, using comparative whole-genome sequencing; or (3) performing in vivo infectivity assays that involve inoculation of samples into susceptible immunodeficient animal models. Third, what are the public health consequences for human-to-human sexual transmission and the re-ignition of outbreaks? Both research studies and national programs are currently attending to this potential reservoir in male EVD survivors. There is room for equipoise as we assess, manage, and mitigate this risk. While a few other sexual transmission events have been suspected (although not confirmed), they have been rare events. In the Kenema and Kailahun eastern districts that were once ground zero for the Sierra Leone outbreak, there have not been suspected sexual transmission events related to persistent virus in the semen of a male survivor, despite a large potential reservoir [17]. The absence of documented sexual transmission in prior outbreaks (admittedly, rare events may have been missed) suggests the same. Nonetheless, the last year has included repeated reminders that these rare events are consequential and that staying at zero transmissions in the last mile requires a careful attention to an uncertain residual risk [18]. Fourth, are there clinical consequences of semen virus persistence for the individual EVD survivor, and is semen persistence associated with either organ-specific or systemic inflammation? Testicular pain [4] and orchitis [19] have been reported in this and prior outbreaks; thus far, any connection between semen persistence and organ-specific clinical syndromes (orchitis, epididymitis, and infertility) in this outbreak remains unclear. In the semen or other compartments, the relationship between viral persistence, immune activation, and systemic inflammation also may inform understanding of the pathogenesis of common post-EVD clinical sequelae. Fifth, how does the host-pathogen interaction determine the immunopathogenesis and clinical predictors of viral persistence? These questions reflect broader unknowns about the persistence of pathogens in immune-privileged sanctuary sites. Immune privilege in the testis is a complex, multifactorial phenomenon only partially explained by the geographic segregation of the blood-testis barrier. Layered structural, cellular, and molecular mechanisms outside the scope of this piece underlie the relative privilege of this and other viral sanctuaries [20, 21]. These observations in EVD survivors beg a new look at the host-pathogen interaction on these immunologic playing fields, with questions to be asked of the host, of the virus, and of their initial interaction during acute EVD. With regard to the host, do qualitative or quantitative aspects of the host immune response during acute EVD determine whether, how much, and for how long Ebola virus persists? Are there host genetic factors that contribute to persistence? With regard to the virus, is there something unique about the virus that seeds and then persists in these sites? How is the virus evolving in this space? Comparison of blood specimens, initial semen specimens, and EBOV whole-genome sequences over time might be informative, although they are only available from a few survivors. With regard to the acute infection, could known factors that determine the outcome of acute EVD (age, viral load, and severity of organ dysfunction) [22, 23] also predict the prevalence and kinetics of viral persistence? Do therapeutic interventions during acute EVD influence these same factors? Details of acute illnesses in these survivors are not provided, but correlation to viral loads or nadir cycle threshold values and whether these patients received experimental interventional therapy during acute EVD is of great interest, especially in a Guinean setting in which some patients may have received oral antivirals as part of standard ETU care [24]. Furthermore, during convalescence, can and should the semen be cleared by therapeutic intervention in men with persistently EBOV RNA-positive semen? This study confirms the standard persistence of EBOV RNA in the semen of Guinean survivors and reveals that much remains to be understood. Ongoing research and programmatic data from all 3 countries will clarify public health and individual consequences. From Liberia, the National Institutes of Health PREVAIL (Partnership on Research in Ebola Virus in Liberia) III study will provide a first controlled, longitudinal look at the natural history of survival after EVD and an important clinical and immunologic context. While we stay tuned, an informed and respectful attention to safe sex practices [25] in EVD survivors remains one of their many urgent and important care needs. As well, a vigilant and nimble capacity in all three countries to rapidly detect and effectively respond to further flare-ups of EVD related to sexual transmission from a survivor remains critical. Potential conflict of interest. Author certifies no potential conflicts of interest. The author has submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.