Multisite Mpox Infection and Viral Dynamics Among Persons With HIV in Metro Atlanta

Abstract

The 2022 mpox outbreak primarily involved sexual transmission among men who have sex with men and disproportionately affected persons with human immunodeficiency virus (HIV). We examined viral dynamics and clinical features in a cohort evaluated for mpox infection at a comprehensive HIV clinic in Atlanta, Georgia. Viral DNA was found in 8 oropharyngeal and 5 anorectal specimens among 10 mpox cases confirmed by lesion swab polymerase chain reaction. Within-participant anatomic site of lowest cycle threshold (Ct) value varied, and lower Ct values were found in oropharyngeal and anorectal swabs when corresponding symptoms were present. This provides insight into mpox infection across multiple anatomic sites among people with HIV.

The 2022 global mpox outbreak primarily involved sexual transmission among men who have sex with men and disproportionately impacted persons with human immunodeficiency virus (HIV) [1]. Moreover, Black people with HIV (PWH) represented the majority of severe and fatal mpox cases [2, 3]. Although historically associated with diffuse rash, the 2022 mpox outbreak commonly involved a localized anogenital rash [4, 5]. Co-occurring syndromes included proctitis, tonsillitis, and pharyngitis, and some were asymptomatic [4, 5].

While viral DNA detection from cutaneous lesions is the gold standard for diagnosis, the presence of replication-competent virus at mucosal sites in the absence of obvious skin lesions has been described [6]. Several cohort studies have included oropharyngeal or anorectal swabs, but so far none among PWH in the southern United States (US) [4, 5, 7–10]. Understanding mpox viral dynamics across anatomic sites may inform mechanisms of transmission and testing strategies, especially among those without skin lesions.

In this study we describe clinical and virological characteristics of mpox in a cohort of PWH at a comprehensive HIV care center in Atlanta, Georgia. Assessment of viral load, genetic characteristics, and infectivity of Monkeypox virus (MPXV) from cutaneous, pharyngeal, and anorectal sites provides insight into viral dynamics in a highly affected population in the southern US during the 2022 mpox outbreak.

METHODS

Study Design and Participants

We performed an observational cohort study at the Ponce Center, one of the largest comprehensive HIV care centers in the US serving >6000 PWH in metro Atlanta and surrounding areas. Participants presenting with clinical suspicion for mpox were enrolled between 8 September and 11 November 2022. A cutaneous lesion swab for clinical standard-of-care polymerase chain reaction (PCR) (performed in a national commercial laboratory) was collected concurrently with research specimens from a cutaneous lesion, oropharyngeal, and anorectal sites, which were frozen dry. A clinician abstracted medical record data.

This study was approved by the Emory Institutional Review Board (IRB00009146) as part of the Center for AIDS Research specimen repository study. Written informed consent was obtained from all participants.

Assays and Sequencing

A nonvariola orthopoxvirus PCR was used to detect mpox DNA (Supplementary Table 1) [11]. A focus-forming infectivity assay was performed, and full-genome sequencing was attempted for all PCR-positive specimens (see Supplementary Methods).

RESULTS

Participant Characteristics

Twenty PWH were enrolled in our study, of whom 10 were considered mpox cases (Supplementary Table 2): 9 positive on the clinical test and 1 without a clinical test due to an administrative error who was treated empirically and produced 2 positive research swabs. All participants were male, median age was 40 years (interquartile range [IQR], 33–45 years), and 16 (80%) identified as non-Hispanic Black. Fifteen (75%) participants had CD4⁺ lymphocyte counts ≥200 cells/μL and 18 (90%) had been prescribed antiretroviral therapy (ART) during the prior year. Although ART adherence was not systematically assessed, 12 (60%) had detectable HIV viral loads (≥20 copies/mL). Among mpox cases (n = 10), 1 was coinfected with gonorrhea. Among those without mpox (n = 10), an alternate sexually transmitted infection (STI) diagnosis was made in 5 individuals (2 gonorrhea, 2 chlamydia, 1 herpes simplex virus [HSV]). Among cases for whom the presence or absence of symptoms were documented, 7 of 8 (87.5%) had systemic symptoms, 8 of 10 (60.0%) had pharyngitis, and 4 of 7 (57.1%) had anorectal symptoms. The median time from mpox symptom onset to presentation was 5 days (range, 2–14 days), and median time since last sexual encounter was 14 days (range, 7–28 days).

MPXV DNA at Mucosal Sites

Mucosal site PCR testing revealed viral DNA in oropharyngeal swabs for 8 of 10 cases with a median cycle threshold (Ct) value of 31.2 (IQR, 27.9–33.4) and anorectal swabs in 5 of 10 cases with a median Ct value of 29.9 (IQR, 26.1–34.2) (Figure 1A). The median Ct value for MPXV DNA detected from cutaneous lesions was 27.4 (IQR, 23.2–32.3). All but 1 case had viral DNA detected at multiple sites and 3 participants had viral DNA present in all 3 specimens. No MPXV DNA was detected in research specimens from the 10 patients who had a negative clinical specimen.

Figure 1.

Viral load assessment for participants enrolled in the study. A, 10 of 20 enrolled participants were diagnosed with mpox. Cutaneous lesion, oral, and rectal swabs were collected and tested for Monkeypox virus DNA. Polymerase chain reaction (PCR)–positive swabs were tested in a cell culture infectivity assay. B, Within-participant comparison shows the highest viral load varied across the cohort. PCR-negative specimens were not tested in the cell culture assay. Comparison of cycle threshold value at each anatomic site based on the presence or absence of pharyngitis (C) or rectal symptoms (D) suggests that higher viral loads at the corresponding anatomic site may be associated with the presence of symptoms. *One participant had the presence of a perianal lesion documented in the encounter physical examination, but whether rectal symptoms were present is unknown. Abbreviations: Ct, cycle threshold; FFA, focus-forming infectivity assay; NT, not tested in focus-forming infectivity assay; NVAR, nonvariola orthopoxvirus; PCR, polymerase chain reaction.

Five of 38 (1/19 clinical, 4/19 research) cutaneous lesion swabs produced invalid results due to the absence of RNaseP, likely reflecting poor sample collection technique. RNaseP was absent in only 1 of 19 anorectal specimens and 0 of 19 oropharyngeal specimens.

Relative viral load differed across cases. For each case, the site with the lowest Ct value was equally distributed among cutaneous lesions (n = 3), oropharynx (n = 3), and anorectum (n = 4) (Figure 1B ). Six of 10 participants with mpox reported pharyngitis: All had detectable viral DNA in the oropharynx, as well as 2 of 4 participants who did not report pharyngeal symptoms (Figure 1C ). The median viral load in the oropharynx for participants with pharyngitis was higher than those without. All 4 individuals who reported rectal symptoms had detectable mpox DNA at the anorectum (median Ct value, 23.7; Figure 1D ). No MPXV DNA was detected in the anorectal swab of the 3 participants who reported absence of anorectal symptoms.

The proportion of participants with CD4⁺ lymphocyte count >200 cells/μL was not significantly different between individuals with and without mpox (90% vs 60%). The proportion of participants with HIV and mpox who had undetectable HIV RNA levels did not differ in the groups with and without mpox (50% vs 70%). Among those with confirmed mpox, there was no significant association between absolute CD4⁺ lymphocyte count and mpox viral load, nor between HIV viral load and MPXV viral load (Supplementary Figure 1A and 1B).

No notable trend was observed when comparing anatomic site Ct values and time since mpox-related symptom onset (Supplementary Figure 1C).

Infectivity Assays

Focus-forming units were observed with eluate from 5 of 5 rectal swabs (100%), 5 of 8 (62.5%) oral swabs, and 6 of 9 (66.7%) cutaneous lesion swabs (Supplementary Figures 2 and 3, Supplementary Table 3).

Full MPXV Genome Sequencing

Near-full-length MPXV genome sequences were obtained from 15 specimens across 8 cases, including all 3 anatomic sites from 3 separate cases (Supplementary Table 4). In a phylogenetic analysis that included reference MPXV sequences collected throughout the US, the consensus sequences from 6 of our participants clustered together, while the other 2 intermingled with reference sequences from other locations (Supplementary Figure 4). For participants 5 and 25, sequences obtained from different anatomic sites clustered together (Figure 2B ). Interestingly, for participants 9 and 16, the sequence from the oropharynx clustered separately from sequences from the skin and rectum, though many nodes in this region of the tree had low bootstrap support (Supplementary Figure 4). Within-sample virus diversity, measured by average Shannon entropy, was higher in the oropharyngeal samples from participants 9 and 16 (Figure 2C ).

Figure 2.

Infectivity and sequencing studies show replication-competent virus at all sample sites and greater genetic diversity in oral specimens. A, Example of infectivity assay for 3 specimens from participant 16, all of which showed cytopathic effect in 100% of 8 assay replicates (infectivity assay results for all specimens are shown in Supplementary Figure 3). B, Phylogenetic tree shows clustering by participant, except that oral specimen for participants 9 and 16 clustered away from the corresponding lesion and rectal specimens. Ultra-fast bootstrap of 95% or higher is indicated with a black circle at the cluster node. C, Average Shannon entropy from each sample suggests greater viral diversity in oral swabs.

DISCUSSION

We characterized mpox viral dynamics using swabs obtained from cutaneous lesions and mucosal (pharyngeal and rectal) sites among a cohort of PWH in Atlanta, Georgia. Given the wide range of clinical presentations, studying mpox viral dynamics across different tissue sites may inform understanding of disease pathogenesis, transmission mechanisms, and testing strategies for future mpox outbreaks.

Our population was majority non-Hispanic Black race, and all were men with HIV. Presentation with pharyngitis and anorectal symptoms were common. We found mpox coinfection with Neisseria gonorrhoeae, similar to other studies that reported coinfections with other STIs, including N gonorrhoeae, Chlamydia trachomatis, syphilis, and HSV [5, 12]. This reinforces that diagnosis of 1 STI should not dissuade from suspecting and testing for mpox in individuals at risk.

Among 10 individuals with an mpox-positive cutaneous lesion swab, we detected MPXV DNA in the oropharynx of 8 and the anorectum of 5 individuals. Notably, 13.2% of the cutaneous lesion swabs were invalid due to the absence of RNaseP. Although PCR testing of cutaneous lesions is reported to have 91%–100% sensitivity [13], our findings suggest that lesion swabs may be subject to poor collection techniques in real-world scenarios. We observed only a 2.6% invalid rate among mucosal swabs, which may be less prone to these limitations. Mucosal site testing may therefore increase the yield of mpox testing, especially in cases where cutaneous swabs are not reliable or when there are no cutaneous lesions amenable for testing.

Disparate results of anorectal swab testing in individuals with and without rectal symptoms raises interest regarding symptom-directed mucosal testing as a strategy in certain clinical scenarios, as has been suggested in other studies [7]. As our study only included participants with cutaneous lesions suspicious for mpox, the role of mucosal testing in persons without cutaneous lesions is unclear. At least 3 studies have performed retrospective testing of swabs collected for STI screening, and mpox was detected in 2%–8% of asymptomatic patients [6–8]. The presence of MPXV DNA and even replication-competent virus in some cases without obvious cutaneous signs or symptoms of mpox supports the need to further characterize the utility of screening mucosal sites in asymptomatic individuals. The potential benefits of multisite mucosal testing in both symptomatic and asymptomatic individuals and the cost-benefit of such approaches need further study.

Infectivity studies recovered replication-competent virus from all 3 sample types, underscoring the potential for transmission from both cutaneous and mucosal sites. Interestingly, infectious virus was recovered from oral swabs in 2 individuals with PCR-positive but infectivity-negative lesion swabs. While potentially biased by sampling error, this observation highlights the need to examine viral kinetics at various exposure sites in future work—particularly whether shedding of transmissible virus can linger in mucosal sites after cutaneous lesions have resolved.

Near-full MPXV genome sequencing from multiple anatomic sites revealed 2 individuals in whom the oral specimen harbored a more diverse virus population than the corresponding anorectal and lesion specimens, suggesting compartmentalized transmission and/or evolution. The observation of separate phylogenetic clustering also raises the possibility of coinfection, though this is difficult to ascertain given the limited diversity in the MPXV genome and sparse sampling in our local region. Both participants had pharyngitis and anorectal symptoms in addition to genital and cutaneous lesions, though specific sexual practices were not systematically evaluated. A prior study showed evidence of compartmentalized intrahost evolution over time in individuals with advanced HIV [14]. On a population level, our observation that most sequences from this cohort clustered separately from other US reference sequences suggests a local transmission network. Given the clusters of rapid sexual transmission of HIV-1 recently described in Atlanta, it is possible that mpox transmission could occur in a similar manner [15]. While no conclusion can be reached due to our small sample size, this observation raises interest in further studying the interplay between viral evolution, transmission, and specific mucosal environments, particularly in the setting of underlying immunocompromise from coexistent HIV infection.

Limitations of our study include the small, all-male cohort of PWH from a single center. Only participants with cutaneous lesions suspicious for mpox were included, so our study may have omitted participants with mpox without skin lesions or who were asymptomatic. Ct value was used as a surrogate estimate for MPXV viral load; however, this value can be affected by quality of the sample and time since symptom onset, which in turn may be impacted by recall bias. We studied mucosal sample types like those used in our local practice screening for gonococcal and chlamydial infections, but other sample types likely warrant study including blood, saliva, and conjunctival swabs. We only collected samples from participants during a single visit, and sequential sampling is needed for insight into viral kinetics at different anatomic sites. Finally, we only included outpatients, which may not reflect severe cases requiring hospitalization.

Our study is unique in its focus on non-Hispanic Black men living with HIV in Atlanta, Georgia, a vulnerable population disproportionately affected by the 2022 mpox outbreak. It is also one of few studies in the US to prospectively collect and study samples from multiple anatomic sites, whereas many other studies retrospectively screened swabs for MPXV that had been originally collected for other STI testing.

Our analysis demonstrates presence of MPXV DNA and replication-competent virus in the oropharynx in most participants and the anorectum in half the participants, regardless of symptoms and concurrent HIV virological control. Higher MPXV viral loads were seen at mucosal sites with corresponding symptoms. Further studies specifically among PWH as well as studies to examine the generalizability and cost-benefit of multisite testing and screening, can inform and improve the response to future mpox outbreaks.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Notes

Acknowledgments. We dedicate this work to the memory of our colleague, friend, and co-author of this manuscript Dr Albert M. Anderson, who passed away suddenly during its preparation. This study was supported in part by the Emory Integrated Genomics Core, which is subsidized by the Emory University School of Medicine and is one of the Emory Integrated Core Facilities.

Data availability. All mpox genome sequence data generated in this study are available at the National Center for Biotechnology Information under BioProject.

Financial support. This work was supported by the Center for AIDS Research at Emory University (P30AI050409) and the National Institute of Biomedical Imaging and Bioengineering of the National Institutes of Health (NIH) (contract number 75N92022D00015-75N92022F00002, innovation funnel validation center for RADx independent test assessment program for monkeypox). Additional support was provided by the Georgia Clinical and Translational Science Alliance of the NIH (award number UL1TR002378). B. K. T. receives funding support from the NIH Building Interdisciplinary Research Careers in Women’s Health program. A. W. F. is supported by an institutional training grant from the National Institute of Allergy and Infectious Diseases (grant number T32AI157855). G. L. D. is supported by an institutional training grant from the National Heart, Lung, and Blood Institute (T32HL069769).

Supplement sponsorship. This article appears as part of the supplement “Mpox: Challenges and Opportunities Following the Global 2022 Outbreak,” sponsored by the Centers for Disease Control and Prevention (Atlanta, GA).