Mpox in people with advanced HIV infection: a global case series

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ “This is an Accepted Manuscript of an article published by Elsevier in The Lancet on March 18, 2023, available at: https://doi.org/10.1016/S0140- 6736(23)00273-8 .” 1 ARTICLE TYPE 1 Original research 2 3 TITLE: 4 Mpox in persons with advanced HIV infection: a global case series 5 6 AUTHORS 7 Oriol Mitjà, PhD*1 Andrea Alemany, MD*1 Michael Marks, PhD*2,3,4 Jezer I. Lezama Mora, MD5 Juan 8 Carlos Rodríguez-Aldama, MD6 Mayara Secco Torres Silva, MD7 Ever Arturo Corral Herrera, MD8 Prof 9 Brenda Crabtree-Ramirez, MD8 Prof José Luis Blanco, PhD9,10,11 Nicolo Girometti, MD12 Valentina 10 Mazzotta, MD13 Aniruddha Hazra, MD14 Macarena Silva, MD15 Juan José Montenegro-Idrogo, MD16,17 11 Prof Kelly Gebo, PhD18 Prof Jade Ghosn, PhD19,20 María Fernanda Peña Vázquez, MD21 Eduardo Matos 12 Prado, PhD22 Uche Unigwe, MD23 Judit Villar-García, PhD24 Noah Wald-Dickler, MD 25,26 Jason Zucker, 13 PhD 27 Roger Paredes, PhD28 Prof Alexandra Calmy, PhD29 Laura Waters, PhD 30 Cristina Galvan-Casas, 14 MD1, 31 Prof Sharon Walmsley, MD MSc32 Prof Chloe M Orkin, MD MSc33 on behalf of SHARE-NET 15 writing group 16 17 *These authors equally contributed to this work. 18 AFFILIATIONS 19 1Skin Neglected Tropical diseases and Sexually Transmitted Infections section, Fight Infectious 20 Diseases Foundation, University Hospital Germans Trias i Pujol, Badalona, Spain 21 2Clinical Research Department, London School of Hygiene & Tropical Medicine, London, UK 22 3 Hospital for Tropical Diseases, University College London Hospitals. London, UK 23 4 Division of Infection and Immunity, University College London, London, UK 24 5Clinica Especializada Condesa Cuauhtémoc, Mexico City, Mexico 25 6Clinica Especializada Condesa Iztapalapa, Mexico City, Mexico 26 2 7Instituto Nacional de Infectología Evandro Chagas, Fundação Oswaldo Cruz (INI-Fiocruz), Rio de 27 Janeiro, Brazil 28 8Departamento de Infectología, Instituto Nacional de Ciencias Médicas y Nutrición, Salvador Zubirán, 29 México City, México 30 9Infectious Diseases Department, Hospital Clínic de Barcelona, Barcelona University, Barcelona, Spain 31 10Centro de Investigación Biomédica en Red de Enfermedades Infecciosas (CIBERINFEC), Instituto de 32 Salud Carlos III, Madrid, Spain 33 11Institut d’Investigacions Mèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain 34 12Chelsea and Westminster Hospital NHS Foundation Trust – Department of HIV and Genitourinary 35 Medicine, UK 36 13National Institute for Infectious Disease, Lazzaro Spallanzani, Rome, Italy 37 14Section of Infectious Diseases and Global Health, University of Chicago Medicine, Chicago, IL, USA 38 15Hospital San Borja Arriarán, Santiago de Chile, Chile 39 16Hospital Nacional Dos de Mayo, Lima, Perú 40 17Centro de Investigaciones Tecnológicas Biomédicas y Medioambientales (CITBM), Lima, Perú 41 18Johns Hopkins University School of Medicine, Baltimore, MD, USA 42 19Assistance Publique – Hôpitaux de Paris Nord, Department of Infectious Diseases, Bichat University 43 Hospital, Paris, France 44 20Université Paris Cité, INSERM UMR 1137 IAME, Paris, France 45 21Hospital Enrique Garcés del Sur, Quito, Ecuador 46 22Hospital Nacional Arzobispo Loayza, Lima, Perú 47 23Infectious Disease Unit Department of Medicine, University of Nigeria Teaching Hospital, Enugu, 48 Nigeria 49 24Infectious Disease Unit, Hospital del Mar. Department of Medicine and Life Sciences, Universitat 50 Pompeu Fabra. IMIM (Hospital del Mar Medical Research Institute), Barcelona, Spain. 51 25 Los Angeles County + University of Southern California Medical Center, Los Angeles, California 52 26Keck School of Medicine of the University of Southern California, Los Angeles, California 53 27Division of Infectious Diseases, Columbia University Irving Medical Center, New York, NY 10032, USA 54 28Infectious Disease Department, Fight Infectious Diseases Foundation, University Hospital Germans 55 Trias i Pujol, Badalona, Spain 56 29HIV/AIDS Unit, Division of Infectious Diseases – Geneva University Hospitals, Geneva, Switzerland 57 30Central & North West London NHS Trust, London, UK 58 31Hospital Universitario de Móstoles, Madrid, Spain 59 32University Health Network, University of Toronto, Toronto, Canada 60 3 33Blizard Institute and SHARE Collaborative, Queen Mary University of London, Department of 61 Infection and Immunity, Barts Health NHS Trust, London, UK 62 63 64 CORRESPONDING AUTHOR 65 Professor Chloe Orkin 66 Department of Immunobiology, The Blizard Building, 4 Newark Street, 67 E1 2AT, London, UK 68 [email protected] 69 70 WORD COUNT 71 Abstract: 328 72 Research in context: 607 73 Body text: 3520 74 75 KEY WORDS 76 Monkeypox; Monkeypox virus; MPXV; HIV; immunosuppression; mpox; AIDS, CD4, IRIS 77 4 ABSTRACT 350 78 Background. People living with HIV account for 38-50% of those affected in the 2022 multi- 79 country mpox outbreak. Most reported cases had high CD4 counts and similar outcomes to 80 those without HIV. Emerging data suggest worse clinical outcomes and higher mortality in 81 more advanced HIV. We describe the clinical characteristics and outcomes of mpox in a 82 cohort of persons with HIV and low CD4 cell counts (CD4 <350 cells/mm3). 83 Methods. A network of clinicians from 19 countries provided data from confirmed mpox 84 cases between May 11 and Jan 18 , 2023, in persons with HIV infection and CD4 counts <350 85 cells/ mm3. We describe their clinical presentation, complications, and causes of death. 86 Analyses were descriptive. 87 Findings. We include data of 382 cases: 367 cisgender men, 4 cisgender and 10 transgender 88 women with a median age of 35 years. At mpox diagnosis, 349 (91.4%) were known to be 89 living with HIV; 228/349 (65.3%) adherent to antiretroviral therapy (ART); 32/382 (8.4%) had 90 a concurrent opportunistic illness. The median CD4 count was 211 cells/mm3 (IQR 117-291); 91 with 85 (22.3%) individuals with CD4 counts <100 cells/mm3, and 94 (24.6%) 100-200 92 cells/mm3. Of those with HIV viral load data available, 193/354 (54.5%) were undetectable. 93 Severe complications were more common in persons with CD4<100 compared to those with 94 >300 cells/mm3, including necrotising skin lesions (54.1% vs. 6.7%), lung involvement (29.4% 95 vs. 0%) occasionally with nodules, and secondary infections and sepsis (43.5% vs. 9.3%). 96 Overall, 107/382 (28.0%) were hospitalised of whom 27/107 (25.2%) died. All deaths 97 occurred in those with CD4 counts <200 cells/mm3. Amongst those with CD4 counts <200 98 cells/mm3 more deaths occurred in those who also had high a HIV viral load. An immune 99 reconstitution inflammatory syndrome to mpox was suspected in 21/85 (24.7%) persons 100 5 initiated or re-initiated on ART of whom 12/21 (57.1%) died. Sixty-two (62/382, 16.2%) 101 received tecovirimat and 7 (2.2%) cidofovir or brincidofovir; three had laboratory 102 confirmation of tecovirimat resistance. 103 Interpretation. A severe necrotizing form of mpox in the context of advanced 104 immunosuppression appears to behave like an AIDS-defining condition, with a high 105 prevalence of fulminant dermatological and systemic manifestations and death. 106 Funding. None 107 108 6 RESEARCH IN CONTEXT 635 109 Evidence before this study 110 In 2022, mpox, a disease caused by an orthopox virus referred to as monkeypox virus (MPXV) 111 has caused outbreaks in 110 countries. Two distinct clades of MPXV, Clade I and Clade II, have 112 existed in different geographic regions. The subclade IIb, identified in the 2022 outbreak, 113 originates from subclade IIa mpox, which is considered a self-limiting disease. Unlike the 114 previous epidemiological descriptions in West Africa, mpox transmission in this outbreak has 115 been closely associated with the sexual networks of gay-and-bisexual men-who-have-sex- 116 with-men (GBMSM), in which a high proportion of people are living with HIV. Some evidence 117 suggests greater disease severity in people with advanced HIV. This finding warrants careful 118 evaluation of the interplay between HIV, immune status and clinical manifestations of mpox. 119 We searched PubMed for the terms “monkeypox, mpox AND (HIV)” from inception to Dec 31, 120 2022. Publications were predominantly letters, perspectives, case reports, and public health 121 agency reports. In the multi-country outbreak, scientific publications of case series have 122 described similar clinical outcomes in people living with HIV to those in people without HIV 123 infection. However, most cases series included people with HIV and high CD4 counts (> 500 124 cells/mm3) and suppressed HIV viral loads. In contrast, a Nigerian case series in the 2017-18 125 outbreak reported that more severe outcomes in hospitalized people were observed in 126 persons living with HIV especially those who were viraemic and immunosuppressed. In a 127 Center for Disease Control (CDC) report on 758 mpox cases in persons living with HIV during 128 the multi-country outbreak (median CD4 639 cells/mm3; 3% < 200/mm3) 10% (68/758) were 129 hospitalized for a median duration of 2 days (0-7). Worse rectal symptoms were described in 130 those with HIV. A second CDC report identified adverse outcomes in 47 people with HIV and 131 7 mpox who had low CD4 counts, 12(26%) died and 5 deaths were attributed to mpox. Based 132 on these data, individuals with HIV and advanced disease have been identified as cases 133 requiring expert clinical advice, close surveillance and prioritised for anti-viral treatments 134 such as tecovirimat, and preventive vaccines (where available). 135 136 Added value of this study 137 This mpox case series is the largest in those with advanced HIV disease. We characterized 138 382 persons with HIV and CD4 < 350 cells/mm3. Individuals with lower CD4 counts 139 presented with widespread, large, necrotising, and coalescing skin lesions. Some individuals 140 also developed lung nodules without an alternative confirmed or suspected diagnosis. 141 Severe secondary bacterial infections were common. Frequent and severe oral, ano-genital, 142 and ocular presentations and complications are described. Immune reconstitution 143 inflammatory syndrome (IRIS) was suspected in a quarter of those starting or re-initiating 144 antiretroviral therapy (ART) after mpox diagnosis, 57% of whom died. The greatest disease 145 severity, hospitalisation, and mortality was observed in individuals with both low CD4 count 146 and high HIV viral load. This international case series includes 27 of the 60 persons reported 147 to have died of mpox in the multi-country outbreaks, all 27 are persons with HIV and CD4 < 148 200 cells/mm3. 149 150 Implications of all the available evidence 151 Our findings support the consideration of a severe, disseminated, and necrotising form of 152 mpox infection as an AIDS-defining condition in CDC and WHO HIV disease classifications. This 153 is based on the observation of protracted illness with fulminant disseminated necrotising 154 8 cutaneous lesions, systemic complications, and mortality in those with CD4 cell counts <200 155 cells/mm3. Clinicians should also be aware that starting ART in people with advanced HIV and 156 mpox may contribute to deterioration and possible death, possibly as part of an immune 157 reconstitution syndrome. Our data reinforces the importance of HIV testing in mpox cases. 158 Our findings support the recommendations that all at-risk persons with HIV with a CD4 count 159 <200 cells/mm3 should be prioritised for preventive mpox vaccination. There should also be 160 consideration for use of potential mpox antivirals where available despite lacking data on 161 their effectiveness and a concerted global effort to ensure access in countries without access 162 to antivirals and vaccines. 163 15 skin microtrauma or rubbing) were manifested by ulcers exhibiting a lineal distribution or 300 overlying bony prominences . In cases where epithelialization had occurred, tissue 301 destruction resulted in disfiguring scarring (Supplementary Figure 5 &6, and Dermatology 302 Atlas). 303 304 In total, 35/382 (9.2%) people presented with respiratory complications (Supplementary 305 Table 1). Of these, eleven individuals (2.9%) presented with numerous bilateral diffuse 306 pulmonary nodules; 4/11 diagnosed with an X-ray only, and 7/11 were further characterized 307 on CT scanning. All the radiographic images were reviewed by two specialist radiologists who 308 concurred that these nodular lesions were unusual and characterised by well-defined 309 borders, absence of cavitation and no adjacent areas of ground glass shadowing; most of the 310 nodules were peri-vascular suggesting hematogenous spread and, generally, ranged in size 311 from of 5 to 20 mm. In all three patients with nodules in whom BAL or lung biopsy were 312 performed, a positive MPXV PCR result was obtained (with negative microbiological results 313 for P. jirovecci and M. tuberculosis) (Figure 1A). 8/11 had a CD4 count below 100 cells/mm3. 314 315 12/35 (34.3%) individuals with respiratory complications had dyspnoea of whom - two had 316 normal chest X-rays, and ten had no available radiology report. Additionally, 6/35 (17.1%) 317 individuals presented with pleural effusion (1 with a positive MPXV PCR on BAL), and 3/35 318 (8.6%) with ground glass changes (2 with suspected opportunistic infections; 1 with a positive 319 MPXV PCR on BAL). 320 321 16 Twelve (3.1%) individuals were reported to have neurological involvement (Supplementary 322 Table 2), including one (0.3%) case classified as encephalitis with orbital, frontal and temporal 323 oedema on CT scan, and with a positive MPXV PCR result, and negative HSV-1 and 2, and VZV 324 results in cerebro-spinal fluid (CSF). Of the nine (2.4%) cases with altered mental status or 325 confusion six had normal CSF and/or radiological findings and three did not undergo imaging 326 or CSF examination. In five cases confusion was attributed to sepsis, in one to respiratory 327 failure in one to hepatic encephalopathy, and the cause was undetermined in two cases. 328 Neurological symptoms were almost exclusively described in persons with HIV with CD4 less 329 than 100 cells/mm3. 330 331 In 76/382 (19.9%) individuals, secondary bacterial infections were diagnosed, including 332 cellulitis, abscesses, and sepsis. Among 17 patients with sepsis, 8 had positive blood cultures 333 and the following pathogens were identified: three Pseudomonas aeruginosa, two ESBL E. 334 coli, two polymicrobial, and one Shigella flexneri. Additionally, 12 patients had a positive 335 result from an abscess or deep wound sample culture: three Pseudomonas aeruginosa, two 336 Klebsiella pneumoniae, two ESBL E. coli, three methicillin-sensitive and two methicillin- 337 resistant Staphylococcus Aureus. 338 339 All complications were more common in individuals with CD4<100 compared to individuals 340 with CD4>300 cells/mm3. This included dermatological (57.6% vs. 9.3%), respiratory (29.4% 341 vs. 0%), CNS (10.6% vs. 1.3%), bacterial infection (43.5% vs. 9.3%), ocular (15.3% vs. 1.3%), 342 gastrointestinal (27.1% vs. 6.7%), rectal complications(56.5% vs. 28.0%), oropharyngeal 343 (34.1% vs. 24.0%), and genito-urinary 34.1% vs. 9.3%) (Figure 2A). 344 17 345 Overall, 107/382 (28.0%) individuals were hospitalised; of these, 7 (1.8%) survived an 346 admission to intensive care and 27 (7.1%) died (Table 3). Among the 27 people who died, 347 the median CD4 count was 35 cells/mm3 (IQR 24-100), and the median HIV viral load was 5 348 log copies/ml (IQR 4-5), only one patient was HIV virologically suppressed. Among those 349 who died, severe necrotising or haemorrhagic skin lesions occurred in 25/27 (92.6%), 350 bloodstream or deep tissue bacterial infections (24/27; 88.9%), respiratory symptoms and 351 respiratory failure (23/27; 85.2%), neurological (8/27; 29.6%), rectal (21/27; 77.8%), and 352 oropharyngeal (18/27; 66.7%) involvement were described (Table 3). Ocular disease 353 occurred in (13/27; 48.1%), 8 of whom had peri-orbital cellulitis. The reported cause of 354 death was septic shock and multi-organ failure in 20/27 (74.15%), respiratory failure 4/27 355 (14.8%), disseminated mpox in 2/27 (7.4%) and cardiac arrest in 1/27 (3.7%). 356 357 Rate of hospitalisation and ICU admission increased with declining CD4 counts and rising viral 358 loads (Figure 2). No deaths occurred in individuals with CD4 counts >200cells/mm3. Mortality 359 was incrementally higher among people in the lowest CD4 strata (CD4<100 27.1% vs. CD4 360 100-200 4.3% vs. CD4>200 0%; Figure 2B) and amongst those with the highest viral loads (HIV 361 VL log≥4 16.2% vs. HIV VL undetectable 0.5%; Figure 2C). In those with CD4 count <100 362 cells/mm3 (n=85) and available HIV VL, the death rate was 7.1% (1/14) for individuals with VL 363 <50 copies/ml and 29.8% (14/47) for those with HIV VL ≥4 log copies/ml. 364 365 Among 85 persons started or restarted on ART, in 21 (24.7%) the managing clinician 366 suspected IRIS as a cause for clinical deterioration (Supplementary Table 3). Of these, 6 367 18 (28.6%) were newly diagnosed and 15 (71.4%) were known to be living with HIV but either 368 not receiving or not adherent to ART. All had CD4 count <200 cells/mm3. The median time 369 from onset of mpox symptoms to the start of ART was 21 days (range 0-73), and from the 370 start of ART to worsening of mpox symptoms was 14 days (range 3-64). Nine of 21 (42.9%) 371 were treated for IRIS with steroids, and 10 (47.6%) received supportive care. Of those with 372 suspected IRIS, 3/21 (14.3%) were admitted to the ICU, 5 (23.8%) were hospitalised in a 373 general ward, and 12 (57.1%) died. 374 375 Forty-three (41.7%) of the 103 hospitalized patients and twenty-one (7.5%) of the 279 376 outpatients received antivirals to treat mpox. Sixty-two (62/382, 16.2%) individuals received 377 tecovirimat (5 received both oral and IV) and 7 (1.8%) cidofovir or brincidofovir. All patients 378 receiving mpox-specific antiviral therapy were treated in Europe or the USA, except two 379 who received tecovirimat in Brazil. Laboratory confirmation of tecovirimat resistance 380 (presence of FL13L mutations by sequencing) was detected in 3/5 people tested, who had 381 severe immunocompromise, disseminated and progressive mpox infection despite 382 prolonged treatment (>14 days) with tecovirimat and finally died. Sampling for resistance 383 testing was conducted after at least one course of tecovirimat had been completed. Nobody 384 who died had received mpox-vaccination prior to or during 2022. 385 386 DISCUSSION 387 Our large case series describes a severe, disseminated form of mpox infection with 15% 388 mortality in individuals with advanced HIV-related disease characterised by CD4 counts below 389 200 cells/mm3. This fulminant form of mpox is characterized by massive necrotising skin, 390 19 genital and non-genital cutaneous and mucosal lesions, sometimes accompanied by lung 391 involvement with multifocal nodular opacities or respiratory failure, severe cutaneous and 392 bloodstream secondary bacterial infections. The severity of oral and anogenital complications 393 are more marked than previously described.4–10,19 As described in the CDC classification, 394 disseminated forms of coccidiodomycosis, histoplamosis and mycobacterium avium complex 395 are considered to be AIDS-defining illnesses.20 Very similarly, we describe that people with 396 the lowest CD4 counts (<100 cells/mm3) and highest HIV viral loads (> 4log c/ml) had 397 disseminated forms of mpox strongly suggesting that this severe necrotising form of mpox 398 with systemic involvement is also an AIDS-defining condition (Supplement Table 4).20 We 399 describe in detail the clinical course of 27 people with CD4 counts < 200 cells/mm3 who died, 400 representing more than 40% of all mpox deaths reported in 2022. We also wish to raise 401 awareness of the 57% mortality rate in those in whom IRIS was suspected following ART 402 initiation/re-initiation. 403 404 This data builds on the observations of the altered natural history and course of mpox that is 405 emerging. To date most information about the intersection of HIV and mpox reports on those 406 with well-controlled HIV infection.4–11 During the 2022 multi-country outbreak even the 407 largest series include very few people living with HIV and CD4 counts < 350/mm3 (12%) or 408 <200/mm3 (3%).4,21,22 We hypothesized that mpox may have a different clinical presentation 409 in individuals with advanced immunosuppression, as can be the case with some pathogens. 410 Although the self-limiting clinical course in individuals with well-controlled HIV is very similar 411 to that of individuals without HIV, our series provides evidence that the disease is very 412 different in those with advanced HIV. The protracted duration and larger number of skin 413 20 lesions in these individuals also raises the possibility of a more prolonged period of infectivity, 414 but further studies are needed to investigate this. 415 416 Prior work has shown that people living with HIV with high CD4+ T-cell counts (>350 cells/ 417 mm3) mount a poxvirus-specific T-cell response that is similar to those without HIV infection,23 418 but there are no data on immunological responses in those with low CD4 counts (<350 cells/ 419 mm3). In our series low CD4 cell count especially when < 200/mm3 was strongly associated 420 with increasing severity of mpox disease relative to those with CD4 200-350/mm3 and 421 compared to previous reports where CD4 > 500/mm3. Data from animal model MPXV studies 422 have shown that CD4 depletion before immunisation of non-human primates decreased the 423 development of protective B-cell responses and antibodies, and increased infection 424 severity after monkeypox virus challenge, which is consistent with our findings.24 Moreover, 425 in our series, the effect of a low CD4 count on severity or death varied with the HIV plasma 426 viral load, with higher viral loads associated with increased frequency of severe illness in any 427 given CD4 group. Prior work has shown that replicating HIV virions target antigen-specific T 428 cells that are activated to combat other pathogens, resulting in impaired T-cell responses.25– 429 27 Thus, it is possible that a substantial fraction of MPXV-specific CD4 T cells might die or be 430 impaired due to either complete or abortive HIV infection. Others have shown impaired 431 immune responses to hepatitis B and other vaccines in those with low CD4 and unsuppressed 432 HIV virus replication, providing further evidence that HIV replication may interfere with 433 immune response to other pathogens.27,28 Based on our findings we believe that a severe 434 necrotising form of mpox with systemic manifestations exists. This form of mpox affected 435 those with CD4 counts <200 cells/mm3 - the precise CD4 threshold considered to be AIDS- 436 21 defining in international guidelines (Supplement table 4). Given the 15% mortality in this 437 group, strong consideration should be given to designating this disseminated form of severe 438 mpox as a new AIDS-defining condition in definitions and guidelines. 439 440 Several limitations of our research need to be highlighted. Our data are derived from an 441 observational retrospective convenience case series from countries with high numbers of 442 mpox infections. We were, therefore, unable to assess how well our cohort represents the 443 entire population of people living with HIV and who developed mpox infection. Study sites 444 may have been more likely to include individuals with more severe outcomes which may have 445 biased the relationships we saw between both CD4 counts and HIV viral load and clinical 446 outcomes. Individuals included in this case series had symptoms that led them to seek 447 medical care; therefore, persons who were asymptomatic, had milder symptoms, or lacked 448 access to medical care could have been missed and thus we may have overestimated illness 449 severity.29 Also outcomes may have been missed if people reattended with severe disease at 450 different sites that were not part of the case series. Due to data collection from multiple sites 451 some characteristics may have been collected in a heterogeneous manner, and laboratory 452 techniques also differed from site to site. Many of the cases had a concomitant opportunistic 453 infection and it may be difficult to ascertain many of the outcomes to mpox relative to the 454 other pathogens. For example, some healthcare settings included did not have access to 455 certain radiological and microbiological investigations so we cannot be certain of the role of 456 mpox as opposed to other opportunistic infections. However, we suggest that the coincident 457 perivascular non-cavitating lung nodular pattern described by two radiologists and seen in 458 eleven patients, without a documented suspicion or microbiological evidence of a co- 459 22 opportunistic pathogen, may be a manifestation of mpox disease and further research is 460 warranted. 461 Many of the deaths were associated with multiorgan system failure, and the relative 462 contribution of the mpox to death unclear. We have raised the possibility of IRIS reactions 463 with the initiation of antiretroviral therapy but in the absence of a strict definition and in the 464 absence of details on confounding conditions that may have contributed to adverse outcomes 465 the data are uncertain. Data are currently not available from randomised controlled clinical 466 trials on the impact of MPXV antivirals or preventive vaccines on the course of mpox, and 467 with the limited use in this series their role cannot be evaluated. 468 469 Physicians caring for persons with mpox and advanced HIV disease must be made aware of 470 the severe outcomes and high mortality that can occur especially when cutaneous and 471 bloodstream bacterial super-infection set in. Clinical trials tailored to this group are needed 472 to evaluate the impact of antiviral agents and preventive vaccines to modify disease 473 outcomes. In the absence of these data, persons with HIV, low CD4 counts who need to be 474 hospitalised with mpox should be considered for expanded access to these therapies where 475 available. We have raised the risk of deterioration after initiation of antiretroviral therapy 476 that carried a 57% mortality rate. This needs to be considered when caring for persons with 477 HIV and advanced disease with mpox who are not on therapy. Further research into the role 478 of IRIS is necessary to better understand the role of potential interventions, such as early 479 versus delayed initiation of ART, the concomitant use of steroids or other immunomodulatory 480 strategies leading to a reduction in the frequency of IRIS. 481 482 23 Our data also reinforces the recommendation that HIV testing (in addition to other sexual 483 transmitted pathogens) be performed in every case of mpox. Further those with HIV 484 infection and high risk for mpox infection should be prioritised for preventive vaccine. Two 485 thirds of the deaths we have reported have occurred in Latin America. Our findings are 486 particularly pertinent for countries with low levels of HIV diagnosis and/or without universal 487 free access to ART and intensive care units, where the interaction of uncontrolled HIV 488 infection and mpox is more prevalent. In these countries, a concerted effort to provide 489 urgent access to mpox antivirals and vaccines is of vital importance. 490 491 CONTRIBUTORS 492 OM and CMO conceived and designed the study. CMO, and OM co-ordinated the global 493 collaboration. AA and CGC managed the global data collection. CMO, OM, AA, MM, CGC 494 developed the case report form. MM and OM analysed and interpreted and vouch for the 495 data. All authors except MM, and CMO submitted cases. OM, CMO, AA, CGC and MM drafted 496 the first draft of the manuscript. CGC and JV prepared the image library. SW edited the final 497 draft. All authors reviewed the manuscript and vouch for the accuracy and completeness of 498 the data. All authors were responsible for the final decision to submit for publication and have 499 seen and approved the manuscript. CMO, MM and OM had full access to all data. 500 501 DATA SHARING 502 De-identified participant data collected, including individual participant data, and will be 503 made available from the corresponding author on reasonable request. 504 505 24 DECLARATION OF INTERESTS 506 We declare no competing interests. 507 508 ACKNOWLEDGMENTS 509 We thank Roser Escrig for her assistance in administering data sharing agreements, and 510 Gerard Carot-Sans for providing comments to the manuscript. We would like to thank the 511 Swiss HIV Cohort Study network, who made possible and facilitated data collection. We wish 512 to acknowledge Dr Maria Beiras-Garcia, Department of imaging and Radiology, Complejo 513 Hospitalario de Pontevedra, Pontevedra, Spain for reviewing and commenting on the 514 radiological images. Thank you to the individuals with mpox and families of deceased 515 individuals who graciously consented to contribute data to further knowledge of this disease. 516 31 with HIV infection CD4 count (cells/mm3), median (IQR) 211(117-291) 47(27-77) 156 (125- 184) 259 (221- 280) 326 (316- 338) CD4 count among 27 people who died, (cells/mm3), median (IQR) 35 (IQR 24- 100) HIV viral load strata RNA copies/ml) Not available 28(7.3%) 11 (12.9%) 4 (4.3%) 10 (7.8%) 3 (4%) <50 193 (50.5%) 14 (16.5%) 50(53.2%) 80 (62.5%) 49 (65.3%) 50-200 26(6.8%) 3 (3.5%) 6 (6.4%) 8 (6.3%) 9 (12%) 201-log4 30 (7.9%) 10 (11.8%) 6 (6.4%) 10 (7.8%) 4 (5.3%) ≥log4 105 (27.5%) 47 (55.3%) 28 (29.8%) 20 (15.6%) 10 (13.3%) History of mpox vaccination Vaccination before 2022 16(4.2%) 2 (2.4%) 4 (4.3%) 7 (5.7%) 3 (4%) Thirdgeneration vaccine for preexposure 21(5.9%) 4 (4.7%) 3 (3.2%) 9 (7.0%) 5 (6.7%) Thirdgeneration vaccine postexposure 5 (1.3%) 1 (1.2%) 0 (0%) 3 (2.3%) 1 (1.3%) Concurrent opportunistic infection Oesophageal candidiasis 4 (1%) 3 (3.5%) 1 (1.1%) 0 (0%) 0 (0%) CMV endorgan disease 1 (0.3%) 1 (1.2%) 0 (0%) 0 (0%) 0 (0%) Disseminated herpes simplex 1 (0.3%) 1 (1.2%) 0 (0%) 0 (0%) 0 (0%) Histoplasmos is 2 (0.5%) 1 (1.2%) 0 (0%) 1 (0.8%) 0 (0%) Isosporosis 1 (0.3%) 1 (1.2%) 0 (0%) 0 (0%) 0 (0%) Kaposi Sarcoma 4 (1%) 2 (2.4%) 0 (0%) 2 (0.8%) 0 (0%) Disseminated Mycobacteriu m Avium Intracellulare 3 (0.8%) 2 (2.4%) 1 (1.1%) 0 (0%) 0 (0%) 32 Pneumocystis jirovecci pneumonia 6 (1.6%) 5 (5.9%) 1 (1.1%) 0 (0%) 0 (0%) Toxoplasmosi s 2 (0.5%) 1 (1.2%) 0 (0%) 1 (0%) 0 (0%) Tuberculosis 8 (2.1%) 5 (5.9%) 1 (1.1%) 2 (0.8%) 0 (0%) PLWH = People living with HIV; ARV =Antiretroviral therapy; Third generation vaccine= MVA -BVN 631 *For the purpose of the table, seven individuals were classified as CD4 <100 cells/mm3 despite not having formal 632 CD4 counts: three individuals from Peru did not have information on CD4 counts due to lack of testing reagents 633 but had CDC Stage C disease on the basis of an opportunistic infection; four patients from Nigeria were tested 634 using a qualitative CD4 count test (Visitect CD4 Lateral Flow Assay, visually interpreted result of above or below 635 200 CD4 cells/ mm3) with a result of <200 CD4 cells/mm3. 636 ** This non-binary individual was assigned male at birth. 637 638 33 TABLE 2. CLINICAL DATA 639 TOTAL n (%) N = 382 CD4 <100* N = 85 CD4 100- 200 N =94 CD4 201- 300 N =128 CD4 >300 N = 75 Gastrointestinal Overall 55 (14.4%) 23 (27.1%) 14 (14.9%) 13 (6.4%) 5 (6.7%) Diarrhoea 38 (9.9%) 15 (17.6%) 9 (9.6%) 10 (7.8%) 4 (5.3%) Gastrointestinal Bleeding 20 (5.2%) 6 (7.1%) 7 (7.4%) 4 (3.1%) 3 (4%) Obstruction 6 (1.6%) 1 (1.2%) 2 (2.1%) 2 (1.6%) 1 (1.3%) Oesophagitis 11 (2.9%) 7 (8.2%) 3 (3.2%) 1 (0.8%) 0 (0%) Highest Care-level Outpatient 275 (72.0%) 32(37.6%) 69 (73.4%) 111 (86.7%) 63 (84.0%) Hospitalization in general ward 73 (19.1%) 26 (30.5%) 19 (20.2%) 16 (12.5%) 12 (16.0%) ICU-level# 34 (8.9%) 27 (31.8%) 6 (6.4%) 1 (0.8%) 0 0 Ultimate Outcome Death # 27 (7.1%) 23 (27.1%) 4 (4.3%) 0 0 Organ Support Need for ventilation 21(5.5%) 16 (18.8%) 4 (4.3%) 1 (0.8%) 0 Indication for ventilation Respiratory failure 17 (4.5%) 14 (16.5%) 2 (2.1%) 1 (0.8%) 0 (0%) Sedation 1 (0.3%) 0 ( 0%) 1 (1.1%) 0 (0%) 0 (0%) Low GCS/Coma 3 (0.8%) 2 (2.4%) 1 (1.1%) 0 (0%) 0 (0%) Need for Inotropes 16 (4.2%) 13 (15.3%) 3 (3.2%) 0 (0%) 0 (0%) Antimicrobial treatment Antibiotics 144 (37.7%) 52 (61.2%) 34 (36.2%) 38 (29.7%) 20 (26.7%) Tecovirimat (oral) 52(13.6%) 21(24.7%) 11(11.7%) 15 (11.7%) 5 (1.5%) Tecovirimat (intravenous) 15 (3.9%) 13 (15.3%) 1 (1.1%) 1 (0.8%) 0 (0%) IVIG 6 (1.6%) 6 (7.1%%) 0 (0%) 0 (0%) 0 (0%) Cidofovir/Brincidof ovir 7 (1.8%) 5 (5.9%) 2 (2.1%) 0 (0%) 0 (0%) Genotypic resistance to Tecovirimat Samples sequenced 5 4 1 0 0 Presence of F13L mutations 3 3 0 0 0 34 conferring resistance Immune restitution inflammatory syndrome (IRIS) Antiretroviral started or restarted 85 (22.3%) 40 (47.1%) 23 (24.5%) 15 (11.70%) 7 (9.3%) Deterioration consistent with IRIS 21 (5.5) 15(17.6%) 6 (6.4%) 0 0 IRIS treatment provided 19 (5.0%) Steroids 9 NSAIDS 1 Supportive care 9 14 (16.5%) Steroids 8 NSAIDS 1 Supportive care 5 5 (5.3%) Steroids 1 NSAIDS 0 Supportiv e care 4 NA NA *For the purpose of the table, seven individuals were classified as CD4 <100 cells/mm3 despite not having formal 640 CD4 counts: three individuals from Peru did not have information on CD4 counts due to lack of testing reagents 641 but had CDC Stage C disease on the basis of an opportunistic infection; four patients from Nigeria were tested 642 using a qualitative CD4 count test (Visitect CD4 Lateral Flow Assay, visually interpreted result of above or below 643 200 CD4 cells/mm3) with a result of <200 CD4 cells/mm3. 644 **The categories within a group of organ involvements are not mutually exclusive; therefore, an individual may 645 present with multiple manifestations of the group. 646 #All individuals who died received ICU-level care. 647 ⱡ Among the 12 patients with dyspnoea, two had a normal chest X-ray, and ten either had no radiology 648 examinations done or the report was unavailable. 649 650 Further detail on respiratory and IRIS cases in supplementary tables 1 and 3. 651 35 TABLE 3. Detailed information about fatal cases 652 653 Patient Age Region where medical care was provided CD4 (cells/mm3) HIV Viral Load (copies/ml) HIV and ART Status Opportunistic Infections Peak number of lesions Necrotising skin lesions Bacterial Infections (Culture result when available) Respiratory Complications Ventilatory Support provided * CNS Complications Ocular Complications Rectal Complications Oropharyngeal Complications MPX Antiviral Therapy Started ARV on admission Suspected IRIS (treatment) Days from symptom onset to death Cause of Death 1 35 America s 32 Log 4 Known HIV but not adherent to ART Oesophagea l candidiasis 300 Yes Pyomysitis/absces s (skin biopsy: K. Pneumoniae and P. aeruginosa) None No None None Pain Tonsilitis None - Not available No No 51 Shock and Multi-organ failure 2 31 Europe 24 Log5 New diagnosis None 100 Yes None Diffuse perivascular nodules (MPXV positive BAL specimen) IMV None None None None Oral and IV TPOXX, and IVIG Yes Yes (nsaids) 196 Shock and Multi-organ failure 3 33 America s 17 Log 5 Known HIV not on ART PJP* 100 Yes Sepsis (blood: ESBL E. Coli) Ground-glass opacification IMV None Periorbi tal cellulitis Pain Lymphade nopathy Oral and IV TPOXX, Cidofovir, and IVIG Yes Yes (steroids) 63 Shock and Multi-organ failure 4 46 America s 57 Log 4 Known HIV not on ART None 100 Yes Sepsis Ground-glass opacification (MPXV positive BAL specimen) IMV Confu sion None Procti tis Throat Pain Oral and IV TPOXX, and IVIG Yes Yes (steroids) 87 Shock and Multi-organ failure 5 30 America s 121 Log 5 New diagnosis None 250 Yes Non-genital cellulitis and sepsis Diffuse perivascular nodules IMV None Periorbi tal cellulitis Procti tis Tonsilitis None - Not available Yes Yes (Supportiv e care) 47 Shock and Multi-organ failure 6 44 America s 106 Log 5 Known HIV not on ART None 100 Yes Non-genital cellulitis and sepsis Diffuse perivascular nodules IMV Encep halitis Periorbi tal cellulitis Procti tis Throat Pain None - Not available Yes Yes (Supportiv e care) 49 Shock and Multi-organ failure 7 37 America s 25 <50 Known HIV not on ART Oesophagea l candidiasis 150 Yes Necrotising Cellulitis and sepsis (Swab: K. pneumoniae, E. faecalis) Diffuse perivascular nodules IMV Confu sion None Procti tis Tonsilitis None - Not available Yes Yes (Supportiv e care) 38 Respiratory Failure 8 34 Europe 13 Log 5 Known HIV not on ART None 25 Yes Perianal and rectal abscesses and sepsis (Blood: ESBL E. Coli) Diffuse perivascular nodules and pleural effusion (MPXV PCR positive in transthoracic lung biopsy) NIMV None Keratitis Perfo ration Lymphade nopathy Oral and IV TPOXX, and Cidofovir Yes Yes (steroids) 117 Shock and Multi-organ failure 36 9 41 America s 7 Log 5 Known HIV not on ART None 200 Yes Sepsis Diffuse perivascular nodules IVM None None None None None – Not available No No 15 Respiratory Failure 10 41 America s 171 Log 6 Known HIV not on ART PJP* 30 Yes Sepsis Ground-glass opacification and large lung cavity No None None Procti tis Tonsilitis None – Not available No No 18 Shock and Multi-organ failure 11 32 America s Unk now n Log 5 Known HIV not on ART PJP* 20 Haem orrha gic Sepsis Consolidation No Confu sion Periorbi tal cellulitis Procti tis Lymphade nopathy None – Not available No No 39 Shock and Multi-organ failure 12 23 America s Unk now n Unk now n New diagnosis PJP* 15 Yes Genital cellulitis Consolidation No None None Pain Throat Pain None – Not available No No 18 Respiratory Failure 13 32 America s Unk now n Unk now n Known HIV not on ART TB* 10 Yes None Diffuse perivascular nodules IMV None None Procti tis Throat Pain None – Not available No No 32 Shock and Multi-organ failure 14 35 Europe 33 Log 6 New diagnosis Visceral leishmaniasi s 150 Yes Necrotising Cellulitis (Blood: P. aeruginosa) None IMV None None Pain Lymphade nopathy IV TPOXX Yes Yes (steroids) 87 Respiratory Failure 15 46 America s 6 Log 5 Known HIV but not adherent to ART Kaposi Sarcoma 50 Yes Sepsis Pleural Effusion IMV Confu sion None Procti tis None None - Not available No No 40 Shock and Multi-organ failure 16 40 Africa 99 Unk now n Known HIV not on ART TB (pulmonary) * 200 Yes Sepsis Consolidation, multilobar No None None Pain None None - Not available Yes Unknown 26 Shock and Multi-organ failure 17 47 America s 32 Log 4 New diagnosis None Not kno wn Yes Gluteal abscess & Sepsis (Blood: P. aeruginosa) None No None Keratitis Pain None Oral TPOXX, Cidofovir and IVIG Yes No 94 Shock and Multi-organ failure 18 41 America s 10 Log 5 Known HIV not on ART None Not kno wn Yes Sepsis (Blood: P. aeruginosa, Clostridum sporogenes) Shortness of Breath IMV None Periorbi tal cellulitis Pain Tonsilitis Oral and IV TPOXX, Cidofovir, and IVIG Yes Yes (steroids) 85 Disseminate d mpox 19 32 America s 58 Log 5 Known HIV not on ART Oesophagea l candidiasis Not kno wn No Non-genital cellulitis Pleural effusion No Confu sion Periorbital oedema None Tonsilitis Oral and IV TPOXX, and Brincidofovir Yes No 78 Cardiac Arrest 20 34 America s 115 Log5 Known HIV not on ART None 35 Yes Sepsis Shortness of Breath No None None Pain None None - Not available No No 26 Shock and Multi-organ failure 21 38 America s 70 Log5 Known HIV not on ART None 32 No Sepsis Shortness of Breath IMV None Conjunc tivitis Pain None None - Not available Yes Yes (Supportiv e care) 71 Shock and Multi-organ failure 22 32 America s 23 Log5 Known HIV not on ART CMV retinitis 23 Yes Sepsis None No None None Pain None None - Not available Yes Unknown 78 Shock and Multi-organ failure 23 48 Africa 90 Unk now n Known HIV but not adherent to ART TB (disseminat ed) 100 0 Yes Sepsis (Swab: K. pneumoniae, P aeruginosa) Shortness of breath NIMV None Keratitis None Throat Pain None – Not available Yes No 25 Shock and Multi-organ failure 24 45 Africa 110 Unk now n Known HIV not on ART None 100 0 Yes Sepsis Shortness of Breath NIMV None Periorbi tal cellulitis None None None – Not available No No 4 Shock and Multi-organ failure 37 25 28 Africa 99 Unk now n Known HIV not on ART TB (pulmonary) 100 0 Yes None Shortness of breath No Confu sion Keratitis None Throat Pain None – Not available No No 4 Shock and Multi-organ failure 26 29 America s 35 Log5 New diagnosis None 50 Yes Necrotising Cellulitis & sepsis (Blood: Polymicrobial) Pleural effusion IMV Confu sion Periorbi tal cellulitis Procti tis Tonsilitis Oral and IV TPOXX, and IVIG Yes Yes (supportiv e care) 83 Disseminate d mpox 27 33 America s 35 Log4 Known HIV not on ART None 50 Yes Non-genital cellulitis Pleural effusion & ulcerative lesions on the trachea (MPXV PCR positive on BAL specimen) IMV None None Procti tis and Bowe l obstr uctio n Throat pain IV TPOXX Yes Yes (steroids) 46 Shock and Multi-organ failure 654 Legend: All deceased individuals were cis male, except for patient 16 and 24 that were cis female. 655 None had received smallpox vaccination before 2022 or had been vaccinated as pre-exposure or post-exposure since May 2022. * Since 656 ventilation was often unavailable, answering "no" does not necessarily mean it was not needed. 657 IMV – Invasive Mechanical Ventilation, NIV – Non-Invasive Ventilation 658 *Clinical suspicion, not microbiological confirmation 659