Clinical prognostic value of TTV and HCMV but not EBV for outcomes in hospitalized HIV-positive patients

Abstract

Opportunistic infections caused by viruses, bacteria, fungi, and parasites, are commonly reported in hospitalized human immunodeficiency virus (HIV)-positive patients, but their detrimental contribution to disease severity remains under explored. In this study, we examined the coinfection profiles of 126 HIV-positive patients with suspected respiratory, bloodstream, or neurological infections. Lower respiratory tract (LRT) samples, cerebrospinal fluid, and blood samples collected within the first seven days of admission were subjected to metagenomic next-generation sequencing (mNGS). Additionally, a multiplex polymerase chain reaction (PCR) detection kit to identify ten commonly known respiratory pathogens was applied to the LRT samples. Of 126 HIV-positive patients, 111 (88.1 %) were coinfected with at least one known virus. Epstein-Barr virus (EBV) (71/111, 64.0 %), human cytomegalovirus (HCMV) (64/111, 57.7 %), and torque teno virus (TTV) (63/111, 56.8 %) were the three most prevalent coinfected viruses. Fungal coinfections (58/126, 46.0 %) and bacterial coinfections (47/126, 37.3 %) were less frequent than viral coinfections. Higher viral loads of coinfection were associated with fungal coinfections (odds ratio [OR] = 2.573, 95 % confidence interval [CI]: 1.150-5.757, P = 0.0214) and lower CD4⁺/CD8⁺ T cell ratios (OR = 0.048, 95 % CI: 0.005-0.429, P = 0.0067). Importantly, patients with higher loads of HCMV and TTV, but not EBV, exhibited worse clinical outcomes. Specifically, patients with HCMV reads per million (RPM) > 0 and TTV RPM > 5 exhibited significantly higher risks of poor prognosis and intensive care unit (ICU) admission. In contrast, EBV RPM showed no association with clinical outcomes in this context. In conclusion, HCMV and TTV may serve as prognostic biomarkers linked to poorer outcomes in HIV-positive patients. Detection of HCMV and TTV could predict clinical outcomes and improve patient management strategies.

Keywords: Epstein-Barr virus (EBV); Human cytomegalovirus (HCMV); Human immunodeficiency virus (HIV); Pathogen-coinfected profile; Prognostic value; Torque teno virus (TTV).

Fig. 1

Flowchart of participant enrollment, sample collection, and testing methods. Abbreviations: HIV, human immunodeficiency virus; LRT, lower respiratory tract; CSF, cerebrospinal fluid.

Fig. 2

Coinfection virome spectrum of HIV-positive patients in blood, LRT, and CSF. A) The spectrum of total viral coinfection. B) Combinations of coinfection virus. Solid red circles and connecting lines represent viral coinfection patterns, blue bars represent the corresponding number of patients in this coinfection combination, Roman numerals indicate the total number of coinfection virus. C) Viral spectrum in blood. D) Viral spectrum in LRT. E) Viral spectrum in CSF. Abbreviations: LRT, lower respiratory tract; CSF, cerebrospinal fluid; HIV, human immunodeficiency virus; EBV, Epstein-Barr virus; HCMV, human cytomegalovirus; TTV, torque teno virus; HSV, herpes simplex virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; HHV, human herpesvirus; JCPyV, JC polyomavirus; BKPyV, BK polyomavirus; HPV, human papillomavirus; TLMV, TTV-like mini virus; RhV, rhinovirus; VZV, varicella-zoster virus; IFV, influenza virus; HAdV, human adenovirus; HPyV, human polyomavirus; MPXV, mpox virus; MCV, molluscum contagiosum virus; B19, human parvovirus B19; hMPV, human metapneumovirus; hCoV-229E, human coronavirus 229E.

Fig. 3

Association of bacterial and fungal coinfections with viral replication in HIV-positive patients. A) Bacteria spectrum. B) Fungi spectrum. C) Bacterial, fungal, and viral coinfection profile. Solid red circles and connecting lines represent bacterial, fungal and viral coinfection patterns, blue or black bars represent the corresponding number of patients in this coinfection combination. D–G) Effect of bacteria, and fungi on total virus, EBV, HCMV, and TTV coinfection. Black open circles represent outliers. Bacteria-pos: n = 32; Fungi-pos: n = 51; Bacteria-pos plus fungi-pos: n = 29; Bacteria-neg plus fungi-neg: n = 59. Abbreviations: Pos, positive; Neg, negative; HIV, human immunodeficiency virus; EBV, Epstein-Barr virus; HCMV, human cytomegalovirus; TTV, torque teno virus; M. tuberculosis, Mycobacterium tuberculosis; S. aureus, Staphylococcus aureus; E. faecium, Enterococcus faecium; K. pneumoniae, Klebsiella pneumoniae; P. aeruginosa, Pseudomonas aeruginosa; S. pneumoniae, Streptococcus pneumoniae; L. pneumophila, Legionella pneumophila; E. coli, Escherichia coli; B. cepacia, Burkholderia cepacia; C. striatum, Corynebacterium striatum; C. gilardii, Cupriavidus gilardii; S. haemolyticus, Staphylococcus haemolyticus; E. faecalis, Enterococcus faecalis; A. xylosoxidans, Achromobacter xylosoxidans; E. avium, Enterococcus avium; S. paucimobilis, Sphingomonas paucimobilis; S. agalactiae, Streptococcus agalactiae; A. baumannii, Acinetobacter baumannii; T. whipplei, Tropheryma whipplei; P. jirovecii, Pneumocystis jirovecii; T. marneffei, Talaromyces marneffei; C. neoformans, Cryptococcu neoformans; A. fumigatus, Aspergillus fumigatus; P. kudriavzevii, Pichia kudriavzevii.

Fig. 4

Correlation analysis of DNA virome, core signature viruses (HCMV, TTV, EBV), and immune cell counts in the blood. A) Correlation analysis of total virus, HCMV, TTV, EBV, and different immune cells. Negative values highlighted in the red box indicate that the corresponding virus exhibits a negative correlation with the immune status. B–D) Correlation between CD4⁺ T cell and RPM of total virus, HCMV, and TTV, respectively. E–G) Correlation between CD4⁺/CD8⁺ T cell ratio and RPM of total virus, HCMV, and TTV, respectively. The shaded pink area in panels B–G indicates the statistical 95% CI. Abbreviations: DNA, deoxyribonucleic acid; EBV, Epstein-Barr virus; HCMV, human cytomegalovirus; TTV, torque teno virus; RPM, reads per million; WBC, white blood cell; NEU, neutrophil; EOS, eosinophil; CI, confidence interval.

Fig. 5

Multiple logistic regression analysis of risk factors for viral coinfection and their association with clinical outcomes. A) Multiple logistic regression, influencing factors for high-risk virome. B) ROC curve of Multiple logistic regression model. C–F) Comparison of total virus, EBV, HCMV, and TTV RPM in patients with improved group, poor prognosis, and ICU admission. Black open circles represent outliers. Improved patients: n = 71; Poor prognosis patients: n = 23; ICU admission patients: n = 32. Abbreviations: ROC, receiver operating characteristic; AUC, area under curve; OR, odds ratio; CI, confidence interval; EBV, Epstein-Barr virus; HCMV, human cytomegalovirus; TTV, torque teno virus; RPM, reads per million; ICU, intensive care unit; ART, antiretroviral therapy; LRT, lower respiratory tract; CSF, cerebrospinal fluid.