Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Lance Okeke

HIV/AIDS Medicine · Duke University

View

Dr. Raphael J. Landovitz

HIV/AIDS Medicine · UCLA

View

Dr. Aniruddha Hazra

HIV/AIDS Medicine · UCMC

View

Dr. Shital Patel

HIV/AIDS Medicine · BCM

View

Dr. Monica Gandhi

HIV/AIDS Medicine · UCSF

View

Dr. Ashwin Balagopal

HIV/AIDS Medicine · Johns Hopkins

View

Dr. Ofole Mgbako

HIV/AIDS Medicine · NYU Langone

View

Dr. Joshua Barocas

HIV/AIDS Medicine · CU Anschutz Medical Campus

View

Dr. Chris Beyrer

Global Health · Duke University

View

Dr. Keith Kaye

General ID · RU

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Mpox in People With HIV: Predictors of Diagnosis, Outcomes, and Vaccine Effectiveness in a Multisite Cohort.

    Feb 9, 2026DOI: 10.1093/cid/ciae464Journal ArticleCID

    Written by: Michalina MontañoAdrienne E ShapiroBridget M WhitneyLaura BamfordGreer BurkholderEdward R CachayKaterina A ChristopoulosHeidi M CraneJoseph A C DelaneyJoseph J EronRob J FredericksenPeter W HuntJeffrey M JacobsonJeanne C KerulyH Nina KimKenneth H MayerRichard D MooreSonia NapravnikApril PettitMichael S SaagGeorge A YendewaMari M KitahataRachel A Bender Ignacio

    Find these authors on Roon

    RB
    Dr. Rachel Bender Ignacio
    HIV/AIDS Medicine · UW / Fred Hutch

    METHODS

    CNICS is a prospective cohort of adults with HIV in routine clinical care at 10 academic medical centers across the United States [9]. We included all PWH enrolled in CNICS and engaged in care defined as having had at least 1 visit to a participating HIV clinic on or after 1 January 2022 at 9 CNICS sites with available data: Case Western Reserve University, Fenway Health, Johns Hopkins University, University of Alabama at Birmingham, University of California-San Diego, University of California-San Francisco, University of North Carolina, University of Washington, and Vanderbilt University. We identified all participants diagnosed with mpox between 1 June 2022 and 31 May 2023. Data collection methods for the CNICS cohort have been described elsewhere [9]. Briefly, data are collected via electronic medical records including statewide vaccination registry data, other institutional data systems, and patient-reported outcomes (PROs), undergo rigorous data-quality assessment, and are harmonized in a quarterly-updated central data repository.CNICS participants were identified as having mpox if they had either a positive mpox polymerase chain reaction (PCR) test result, a documented mpox diagnosis, or prescription for tecovirimat without accompanying negative mpox PCR result. All other PWH in CNICS were considered not to have been diagnosed with mpox. We examined protective vaccination against mpox, defined as having received 1 or more Modified Vaccina Ankara-Bavarian Nordic (MVA-BN) or ACAM2000 vaccine doses ≥14 days before mpox diagnosis for people diagnosed with mpox, or at any time before 31 May 2023 for people without mpox. Among people with mpox, we compared those who did or did not receive a tecovirimat prescription. Mpox-related hospitalization was defined as hospitalization within 14 days after mpox diagnosis.We examined age (<40/≥40 years), sex at birth, race and ethnicity, geographic region, trans identity, CD4 count (cells/μL), HIV viral load (VL; copies/mL), antiretroviral therapy (ART) status, alcohol use, depression, and anxiety. Because hyperglycemia and liver disease impact immune regulation, hepatitis C diagnosis, hepatic fibrosis scores, and diabetes were included as covariates. We used the most proximal value before mpox diagnosis and on or before most recent clinic visit for people without mpox for all covariates. Trans identity was defined either as self-report of being transgender or self-reporting a gender identity that did not match sex at birth. We defined HIV VL <50 copies/mL as undetectable, and participants with an active ART prescription as on ART. CD4 count was included as a categorical variable (<200, 200–349, 350–499, ≥ 500). Smallpox/mpox vaccines received before 2022 were included as covariates in models with mpox diagnosis as the outcome but not in models in which vaccination was the outcome of interest. We calculated risk scores for hepatic fibrosis (FIB-4) using laboratory data [10, 11]. Depression, anxiety, and alcohol use were obtained from PROs administered at routine clinic visits every ~6–9 months. We defined at-risk alcohol use as having an Alcohol Use Disorders Identification Test-Consumption (AUDIT-C) score of ≥5 for males and ≥4 for females, and binge drinking as frequency of consuming ≥4 (females) or ≥5 (males) drinks on 1 occasion [12]. Depression and anxiety were assessed using the Patient Health Questionnaire (PHQ)-9 and PHQ-5, respectively. We identified recent (prior 12 months) bacterial STIs caused by and by positive nucleic acid amplification test at any anatomic site [13], and syphilis occurrences by a first nontreponemal titer of 1:16 or greater, a nontreponemal titer of ≥1:4 after a nonreactive nontreponemal test, or a 4-fold or greater increase in nontreponemal titer from a prior documented test [14].
    Neisseria gonorrhoeae
    Chlamydia trachomatis
    We calculated incidence rates of mpox diagnosis and mpox vaccination among PWH in CNICS between 1 June 2022 and 31 May 2023. Time at risk began at either 1 June 2022 or date of cohort entry (if after 1 June 2022). For mpox incidence, time was censored at date of mpox diagnosis (the first date of either positive mpox PCR result, mpox clinical diagnosis, or tecovirimat prescription) or death. For vaccination incidence, time was censored at date of first dose of MVA-BN given between 1 June 2022 and 31 May 2023, or death. We used generalized linear models with Poisson distribution and robust variance to examine characteristics associated with mpox diagnosis and vaccination against mpox among CNICS cohort participants. We adjusted models using disease risk scores (DRS) to account for potential confounding [15, 16]. DRS included all the following except the predictor of interest for each individual regression: age, CNICS site, race/ethnicity, bacterial STI within the past 12 months (binary), diabetes, and history of hepatitis C virus. Missing PRO data precluded their use in DRS. We conducted subgroup analysis to calculate the VE of mpox/smallpox vaccines against mpox in PWH (1 – adjusted relative risk [aRR]) across CD4 count and HIV VL strata. To exclude vaccinations given as postexposure prophylaxis, we conducted sensitivity analyses examining VE restricted to vaccine received 21 or more days before diagnosis. Among people with mpox, we used robust Poisson regression adjusted for age and CNICS site to identify characteristics associated with tecovirimat prescription and mpox-related hospitalization. All analyses were conducted in Stata version 18 (StataCorp, College Station, TX). CNICS research has been approved by institutional review boards at each site.

    RESULTS

    We identified 413 people with mpox among 19 177 CNICS participants engaged in HIV care during the study period. Median age among people with mpox was 39 years, and all 413 people with mpox were assigned male sex at birth (Table 1). People without mpox had a median age of 52 years and 79.7% were male at birth. Most people with and without mpox were on ART (92% and 97%, respectively) and had undetectable HIV VL (76% and 86%, respectively). A substantially higher proportion of people with mpox had a recent bacterial STI (50.6% vs 11.3%).
    Incidence of mpox during the study period was 2.2/100 person-years (100 PY). Incidence was highest among Hispanic/Latine participants (4.9/100 PY, P < .001), and those with a recent bacterial STI (10.0/100 PY, P < .001; Figure 1A–C). In adjusted analysis, younger age and White and Hispanic/Latine identities were associated with higher risk of mpox acquisition (Table 2). We were unable to assess the adjusted relationship between sex at birth and mpox diagnosis because all cases were assigned male at birth. Detectable HIV VL and not having an active ART prescription were associated with mpox diagnosis (aRR, 1.59; 95% confidence interval aRR, 2.15; 95% CI: 1.39–3.31, respectively), although CD4 count was not. Including CD4 category and HIV VL in the same model did not substantially change findings. PWH with a recent bacterial STI had 5-fold higher risk of mpox diagnosis compared to those without evidence of a recent bacterial STI (aRR, 5.23; 95% CI: 4.27–6.41). Persons with diabetes and those with higher FIB-4 scores had a lower risk of mpox diagnosis.
    A total of 87 people with mpox and 2625 people without mpox received ≥1 dose of MVA-BN vaccine between 1 June 2022 and 31 May 2023. Of these, 51 people received their first dose on or before their cohort entry date and were not included in incidence calculations for vaccine receipt. Incidence of vaccination against mpox between 1 June 2022 and 31 May 2023 was 15.9/100 PY. Incidence of vaccination was higher for Hispanic (30.0/100 PY) and non-Hispanic White PWH (21.0/100 PY) compared to non-Hispanic Black PWH (6.0/100 PY, P < .001; Figure 1D), and for those with a recent bacterial STI compared with those without a recent bacterial STI (40.4/100 PY vs 13.0/100 PY, P < .001; Figure 1E). Those with CD4 <200 had lowest incidence of vaccination (9.9/100 PY) compared with other CD4 categories (14.1–16.4/100 PY, P < .001; Figure 1F). In adjusted analysis, male sex at birth was a strong predictor of vaccination against mpox (aRR, 16.07; 95% CI: 11.01–23.44; Table 3). Transgender people (aRR, 1.25; 95% CI: 1.05–1.49) and those with a recent bacterial STI (aRR, 1.88; 95% CI: 1.75–2.03) were also more likely to receive vaccines during the study period compared to cisgender PWH and those without recent STIs. Those with lower CD4 counts and those with detectable VL were less likely to be vaccinated than those with CD4 ≥500 and those with undetectable VL, but PWH without an active ART prescription were more likely to have been vaccinated than those on ART.
    Among 52 people with mpox who received vaccine doses before mpox diagnosis, 19 received a single dose ≥14 days before diagnosis, 13 of whom received vaccine ≥21 days before diagnosis. Zero people diagnosed with mpox had received 2 or more smallpox/mpox vaccines. Receiving ≥1 dose of vaccine against mpox was highly effective overall (≥14-day VE 84%; 95% CI: 75–90; Figure 2). In sensitivity analysis restricting vaccine receipt to ≥21 days before mpox diagnosis, VE was 90% (95% CI: 89–94). Adjusted relative risk reduction associated with vaccination among PWH differed across CD4 and HIV VL categories (Table 4). VE was higher for those with undetectable HIV VL (86%; 95% CI: 76–92) and those with higher CD4 count (CD4 ≥500 VE, 86%; 95% CI: 74–93; CD4, 350–400 VE, 89%; 95% CI: 55–95), than with CD4 200–349 (VE, 72%; 95% CI: 6–91) and those with detectable VL (VE, 65%; 95% CI: 11–86); the estimate for VE for CD4 <200 had a wide CI (−1.43 to .87).
    Of 413 people with mpox, 168 (40.7%) were prescribed tecovirimat and 34 (8.2%) were hospitalized within 2 weeks of mpox diagnosis (Table 1). Of 34 PWH who were hospitalized, only 2 received tecovirimat before the admit date, at 1 and 3 days prior. In adjusted models examining characteristics associated with tecovirimat prescription, having CD4 <200 cells/μL was associated with receiving tecovirimat (aRR, 2.02; 95% CI: 1.51–2.70, compared to CD4 ≥500; Table 5). Having a detectable VL and not having an active ART prescription were associated with 2-fold and 4-fold increased risk of hospitalization, respectively, compared to undetectable VL and being on ART (Table 6). Lower CD4 count was also associated with higher risk of hospitalization (CD4, 200–349: aRR, 3.20; 95% CI: 1.44–7.09; CD4 <200: aRR, 2.58; 95% CI: .78–8.49, compared to CD4 ≥500). No one who received ≥1 vaccine dose ≥14 days before diagnosis was hospitalized; only 2 of the hospitalized patients were vaccinated (1 and 7 days before mpox diagnosis).

    DISCUSSION

    In a large, geographically diverse cohort of PWH in routine HIV care, we found a high incidence of mpox (2.2/100PY), with those aged <40 years, of Hispanic/Latine identity, not virally suppressed, having a recent bacterial STI, or unvaccinated most likely to be diagnosed. Being off ART, virally unsuppressed, or having a CD4 count <350 were the strongest predictors of hospitalization. PWH with CD4 <200 were approximately twice as likely to have received tecovirimat as those with a CD4 count between 200 and 350, despite both groups having comparably high risk of hospitalization in our study and others [17]. Importantly, we demonstrated high effectiveness of mpox/smallpox vaccines against mpox, although disparities in vaccine receipt by race/ethnicity and clinical characteristics related to care engagement and immune status were identified.
    We observed relatively lower mpox incidence among Black PWH in CNICS, despite the disproportionate burden of HIV in US Black populations and the overrepresentation of Black persons among mpox cases overall, especially among Black sexual and gender minorities [18]. High incidence of mpox among Hispanic/Latine participants may relate to assortative mixing in sexual networks, disparities in healthcare access, or lack of dissemination of information to monolingual Spanish-speaking communities. Populations with less access to information and resources may have been less able to protect themselves, including through behavior change and vaccination. Although mpox and mpox vaccination were both high among Hispanic/Latine PWH, vaccine disparity for Black PWH was marked, with the vaccination rate being less than one third that of PWH with other racial/ethnic identities.
    Our study confirmed that STI diagnoses within the prior year was highly predictive of mpox diagnosis [19]. The increased likelihood of mpox diagnoses among persons with detectable HIV VL and no current ART could be due to immunologic susceptibility, although the same constellation of social and medical vulnerabilities that lead to detectable viremia and ART interruption could also increase likelihood of exposure [20]. People with recent STIs were appropriately more frequently vaccinated (~30% with recent STI vs 10% without, cumulatively), although at least two thirds of people eligible for vaccine by this Centers for Disease Control and Prevention (CDC) criterion were not vaccinated [21]. Persons with CD4 ≥350 were the most likely to be vaccinated, with persons <200 being just over half as likely to be vaccinated when adjusting for other factors. It is possible that persons with viremia despite ART prescription were also those less engaged in care and had fewer opportunities for vaccination; conversely, providers often defer vaccines for those with low CD4 count for concerns of safety or lack of efficacy [22, 23]. The gap in vaccinating persons with CD4 <200 raises questions about whether some PWH or their providers had safety concerns, even though MVA-BN is a replication-deficient vaccine and is safe to administer in advanced immunosuppression [24].
    The effectiveness of the MVA-BN vaccine in PWH was as good or better than estimates in the general population; prior observational studies demonstrate VE between 65 and 85 [25–27]. VE of a single vaccine dose in the CNICS population ranged from 87% to 89% among people with a CD4 count ≥ 350 and suppressed VL, to 65%–74% in those with unsuppressed viremia or lower CD4 counts. We saw a dose-response trend by immune strata and virologic status, providing face validity to our findings. When limiting analysis to vaccine doses given >21 days before diagnosis, VE was 90%; three quarters of vaccines received by people with mpox were within 21 days before diagnosis. No one who received 2 vaccines was diagnosed with mpox. Although less effective than vaccines given as primary prophylaxis, vaccines given during the incubation window continued to prevent symptomatic disease. No PWH with mpox who had been vaccinated 14 or more days before diagnosis were hospitalized, indicating protection against severe disease even among PWH. Taken together with our evidence that people with lower CD4 and unsuppressed viral load were both least likely to be vaccinated and at highest risk of severe disease, we urge a prioritization of these populations for MVA-BN vaccination efforts, especially focusing on vaccine access and confidence for populations with earned mistrust in our institutions.
    Tecovirimat was provided to persons with CD4 <200 according to prior CDC guidance at twice the rate of those with CD4 ≥500. People with detectable viremia or not on ART were not more likely to receive tecovirimat despite increased risk of hospitalization, which could indicate either lower engagement in care or else lack of clinician recognition that untreated HIV is as functionally immunocompromising, if not more, than depressed CD4 count. We were unable to evaluate whether tecovirimat mediated the risk between low CD4 count and hospitalization as only 2 persons received tecovirimat prior to hospitalization.
    Limitations of our study include lower ascertainment of select laboratory results and medical records not accessible by CNICS sites. Ascertainment of mpox diagnoses external to CNICS sites may have been incomplete. Clinical data lack some elements of social disparities intertwined with detectable viral loads and care engagement, which may have left residual confounding in the association with HIV viremia and unfavorable outcomes despite adjustment. We lacked vaccine registry access in 1 state; this site was excluded from analyses requiring vaccine predictors or outcomes. Given the short study period relative to the frequency of PRO administration (every 6–9 months), mental health and alcohol use data were only available for approximately half the cohort and should be interpreted with caution; other substance use was not analyzed. This study benefitted from robust routine data collection at diverse clinical sites in the United States and highly curated and validated data, from which associations between HIV and many other comorbidities, including coronavirus disease 2019, have been studied [28, 29].
    In conclusion, in this multisite cohort of more than 19 000 PWH, we answered several outstanding questions related to mpox in PWH. Most importantly, we demonstrated that the MVA-BN smallpox/mpox vaccine is highly effective at preventing symptomatic mpox disease and hospitalization. Although there were decrements in VE among those with low CD4 and unsuppressed viremia, these are the groups who have the highest risk of severe outcomes and should be prioritized for vaccination given the overall strong effectiveness, especially in preventing hospitalization. Current CDC guidance recommends only providing tecovirimat to PWH with “advanced or poorly controlled” HIV [except through the STOMP study (NCT05534984)], which means that many people at highest risk for severe mpox may not be treated, as we observed among people with CD4 200–500 or viremia. Consideration of earlier tecovirimat treatment for unvaccinated PWH not on ART and at higher CD4 thresholds is warranted. In this syndemic of HIV and mpox, the burden of disease and vaccine disparities further weighs on those already facing systemic barriers to engaging with HIV care. Urgent efforts should focus on MVA-BN vaccination for persons facing health disparities, including and especially those with low CD4 or uncontrolled HIV, and raising clinician awareness of the high effectiveness this vaccine despite immunosuppression.
    RBDr. Rachel Bender Ignacio

    Dr. Rachel Bender Ignacio

    Infectious Disease · HIV & AIDS Medicine
    Fred Hutchinson/University of Washington/Seattle Children’s Cancer Consortium
    rachelb