Abstract
Background
Opportunistic infections remain a leading cause of morbidity in people living with HIV (PLWH). Talaromyces marneffei (T. marneffei) accounts for up to 15% of HIV-related hospitalizations in endemic regions. Metagenomic next-generation sequencing (mNGS) offers rapid pathogen detection; however, its utility in diagnosing HIV-associated coinfections is uncertain.
Methods
This retrospective study enrolled 56 hospitalized PLWH with coinfections at the Third Affiliated Hospital of Sun Yat-sen University from March 2022 to October 2024. All patients underwent pathogen detection using both mNGS and CTM, with their diagnostic performance compared. Clinical data, treatment adjustments, and outcomes were analyzed.
Results
mNGS demonstrated significantly higher detection rate (84.4%, 54/64; 95% CI: 73.1–92.2%) than CTM (28.1%, 18/64; 95% CI: 17.6–40.8%; p < 0.0001), especially for T. marneffei detection (100% vs. 45.5%, p < 0.0001). mNGS identified T. marneffei in 39.3% (n = 22/56) of patients, including two rare cases (urinary and intracranial infections) missed by CTM. mNGS revealed mixed infections in 82.1% (46/56) of patients, substantially higher than the 5.4% detected by CTM. Notably, mNGS-guided therapy adjustments occurred in 74.1% of cases, compared with 22.2% for CTM (p < 0.001), correlating with clinical improvement in 90% (36/40) of adjusted regimens.
Conclusion
Our data demonstrated that mNGS had a higher positive detection rate than CTM for detecting coinfections among PLWH, especially for T. marneffei and mixed infections. These results highlight the clinical value of mNGS as a complementary tool for pathogen identification in this vulnerable population.
Keywords: diagnosis, HIV/AIDS, metagenomic next-generation sequencing, mixed infection, Talaromyces marneffei
1. Introduction
People living with HIV (PLWH) are highly susceptible to opportunistic infections by diverse pathogens due to compromised immunity (1–3). Despite advances in antiretroviral therapy, delayed diagnosis of opportunistic infections remains a critical obstacle to effective patient management. Conventional testing methods (CTM), including culture and smear microscopy, are limited by their restricted pathogen detection range, time-consuming procedures, and susceptibility to prior antibiotic use (4). Therefore, the development of novel diagnostic strategies is urgently needed in this vulnerable population.
The past decade has witnessed burgeoning interest in the application of metagenomic next-generation sequencing (mNGS) for microbial identification due to its unbiased approach, rapidity, and high sensitivity (5–8). By comparing sequencing data with specialized databases, mNGS theoretically enables clinicians to identify a wide range of microorganisms (9–13). In recent years, mNGS has been increasingly utilized in diagnosing infectious diseases and detecting resistance mutations of pathogens (11, 12, 14, 15). Invasive talaromycosis accounts for up to 15% of HIV-related hospital admissions in endemic regions such as Guangxi and Guangdong provinces, with the highest mortality rate among AIDS-associated complications (16). Recognized as a medium priority pathogen in the 2022 World Health Organization (WHO) Fungal Priority Pathogens List, T. marneffei poses diagnostic challenges due to nonspecific presentations (17). While several case reports have described mNGS detection of T. marneffei (18–22), the utility of mNGS for diagnosing T. marneffei infections among PLWH remains unclear.
In this study, we aimed to compare the pathogen diagnostic performance of mNGS with that of CTM, particularly in cases of T. marneffei and mixed infections. We further characterized the pathogen spectrum in PLWH in Southern China and assessed mNGS-guided therapeutic adjustments in hospitalized PLWH.
2. Materials and methods
2.1. Study design and subjects
This study enrolled 56 patients infected with HIV who were hospitalized at the Third Affiliated Hospital of Sun Yat-sen University from March 2022 to October 2024. The inclusion criteria for this study were as follows: (1) Patients diagnosed with HIV/AIDS by Western blot; (2) patients aged over 18 years; (3) patients whose primary diagnosis at discharge was an HIV-related infection; (4) patients who had completed conventional pathogen testing (smear, culture, or molecular methods) and mNGS testing; (5) patients with complete medical records, including laboratory testing data, documented treatment, and clinical outcome records. The exclusion criteria were: (1) patients with a combined diagnosis of a malignant tumor; (2) patients undergoing immunosuppressive therapy within the past 3 months; (3) patients with repeat hospital admissions within the study period.
In this study, detailed demographic information, clinical data, and pathogen detection results of the patients were obtained and documented from the electronic medical record system. Final diagnosis, serving as the reference standard, was determined by agreement of two independent infectious disease specialists based on a comprehensive review of all patient information. The Diagnosis of T. marneffei was established using a composite reference standard (CRS). This required the presence of criterion (i) and at least one additional criterion from (ii), (iii), or (iv) (23, 24): (i) Clinical manifestations such as fever, rash, weight loss, gastrointestinal abnormality, lymphadenopathy, hepatomegaly, splenomegaly, anemia, thrombocytopenia, and leukopenia. (ii) Positive cultures from specimens like blood, bronchoalveolar lavage fluid (BALF), cerebrospinal fluid (CSF), pus, or urine. (iii) Histopathological evidence of infection, with characteristic yeast cells identified in tissue biopsies via Periodic Acid-Schiff (PAS) or Wright’s staining. (iv) Detection of T. marneffei DNA by mNGS in blood, BALF, CNS, pus, or urine. For the primary analysis, we evaluated mNGS performance against this CRS. To address potential incorporation bias, we performed a pre-specified sensitivity analysis excluding criterion (iv), redefining the reference standard as clinical manifestations plus conventional testing methods only.
The criteria for adjusting medication regimens included the addition or discontinuation of antibacterial or antiviral agents, based on the results of pathogen examination and/or antimicrobial susceptibility testing. Therapeutic improvement was defined as the complete resolution of clinical symptoms (e.g., normalization of body temperature, cessation of productive cough) and laboratory markers, culminating in discharge with restored functional status. Treatment failure was defined as death or discharge against medical advice. The research protocol was approved by the ethics committee of the Third Affiliated Hospital of Sun Yat-sen University (No. II2025-035-01). Written informed consent was obtained from each participant.
2.2. Conventional pathogen testing
Peripheral blood samples, BALF, CSF, pus, and urine were collected from patients based on their symptoms and suspected sites of infection. Conventional diagnostic methods included bacterial culture, Gram staining, acid-fast staining, and reverse transcription polymerase chain reaction (RT-PCR). GeneXpert and T-SPOT assays were employed for the detection of Mycobacterium tuberculosis. Cryptococcus smear examination and Cryptococcus antigen testing were utilized for the identification of Cryptococcus species. PCR-based methods were applied for the detection of cytomegalovirus (CMV), Epstein–Barr virus (EBV), severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), influenza viruses, and hepatitis viruses.
2.3. mNGS platform
All mNGS procedures adhered to the standardized protocols established by the Department of Laboratory Medicine of the Chinese Medical Association for NGS. Genomic DNA isolation, enzymatic fragmentation, end repair, A-tailing, adapter ligation, and purification were performed using an automated NGS workstation. Library preparation and sequencing analysis were conducted on the Illumina Nextseq 550 DX platform, generating approximately 20 million raw paired-end reads per specimen. Raw data were cleaned by removing low-quality reads, adapter sequences, and fragments shorter than 50 bp and by filtering out human genome sequences (GRCh38.p13) using Bowtie2. Final pathogen identification was achieved by comparing filtered sequences with the proprietary Pathogen Genomic Database (DaAn Gene, Guangzhou, China). Criteria for positive mNGS results: (1) The species-specific strictly mapped read number (SMRN) per million ratios for T. marneffei was ≥1; (2) Other pathogens were identified as previously described (25).
2.4. Statistical analysis
Continuous variables were expressed as mean ± standard deviation (SD) for normally distributed data and as median (interquartile range, IQR) for non-normally distributed data. Comparisons of continuous variables between two groups were performed using the independent samples t-test for normally distributed data and the Mann–Whitney U test for non-normally distributed data. Categorical variables were analyzed using the chi-square test or Fisher’s exact test. Exact 95% confidence intervals for proportions were calculated using the Clopper-Pearson method implemented in R. All statistical analyses were conducted using SPSS version 23.0. Figures were generated using GraphPad Prism version 8.0 and R version 4.2.0.
3. Result
3.1. Clinical characteristics
Between March 2022 and October 2024, a total of 286 PLWH were admitted to the Third Affiliated Hospital of Sun Yat-sen University. This study enrolled 56 patients with AIDS-related infections who underwent both CTM and mNGS testing (Figure 1). The characteristics of the participants are detailed in Table 1. The median age of the participants was 38.5 years, and the majority were male (91.0%). The primary symptoms observed were cough (75.0%), fever (69.6%), dyspnea (51.7%), lymphadenopathy (66.0%), weight loss (57.1%), rash (26.7%), hematological alterations (anemia, leukopenia, or thrombocytopenia, 26.7%), neurological manifestations (headache or altered consciousness, 25.0%), and diarrhea (21.4%). The participants exhibited elevated 1,3-β-D-glucan levels, with a median of 216.3 pg/mL (IQR: 78.5–563.0 pg/mL), and increased lactate dehydrogenase levels, with a median of 321 U/L (IQR: 230–461 U/L). A significant decrease was observed in both CD4+ T cell counts and CD4/CD8 T cell ratios among the participants, consistent with the typical features of HIV infection.
Figure 1.
Study flowchart.
Table 1.
Demography and symptomatology of participants.
| Variable | Data |
|---|---|
| Age (years) | 38.5 (30.3–54.8) |
| Male (%) | 51 (91.0) |
| Symptoms | |
| Cough | 75.0% |
| Fever | 69.6% |
| Dyspnea | 51.7% |
| Lymphadenopathy | 66.0% |
| Weight loss | 57.1% |
| Rash | 26.7% |
| Hematological alterations | 26.7% |
| Neurological symptoms | 25.0% |
| Diarrhea | 21.4% |
| Variable | |
| WBC (×109/L) | 5.71 (3.69–7.71) |
| Hemoglobin (g/L) | 116 (96–130) |
| Neutrophil (×109/L) | 3.97 (2.43–6.03) |
| Platelet (×109/L) | 218 (135–293) |
| AST (U/L) | 44 (26–73) |
| ALT (U/L) | 33 (17–47) |
| eGFR | 116.4 (96.3–129.2) |
| Lactate dehydrogenase (U/L) | 321 (230–461) |
| 1, 3β-D-glucan assays test (pg/mL) | 216.3 (78.5–563.0) |
| CRP (mg/L) | 33.9 (17.2–75.6) |
| Procalcitonin (ng/mL) | 0.12 (0.05–0.51) |
| CD4 (cell/mm3) | 32 (10–77) |
| CD8 (cell/mm3) | 418 (208–664) |
| CD4/8 ratio | 0.08 (0.04–0.19) |
| IgG (g/L) | 19.9 (14.4–29.3) |
| IgA (g/L) | 4.2 (2.8–6.8) |
| IgM (g/L) | 1.5 (1.1–1.8) |
Data are presented as median (inter quartile range) or the number of patients (percent).
3.2. Detection performances of CTM and mNGS
A total of 64 samples from 56 PLWH were analyzed by mNGS, including peripheral blood specimens (n = 26), BALF (n = 29), CSF (n = 6), pus (n = 2), and urine (n = 1) (Figure 2A). The overall pathogen detection rate was significantly higher with mNGS (84.4%, 54/64; 95% CI: 73.1–92.2%) than with CTM (28.1%, 18/64; 95% CI: 17.6–40.8%; p < 0.0001). This superior positivity rate of mNGS was consistently observed across different sample types (Figure 2B). We further compared the detection results of mNGS and CTM according to presenting symptoms (Figure 2C). Irrespective of the presenting symptoms, mNGS demonstrated higher positivity compared to CTM. Besides, in patients presenting with hematological symptoms and rash, the positivity of CTM was relatively higher.
Figure 2.
Detection performance of mNGS and CTM: (A) Types of specimens for mNGS testing; (B) positivity of mNGS and CTM across different specimen types; (C) positivity of mNGS and CTM in patients presenting with different symptoms. mNGS, metagenomic next-generation sequencing; CTM, conventional testing methods.
As illustrated in Figure 3A, patients with positive mNGS results exhibited significantly lower CD4+ T cell counts (p = 0.0147) and CD4/CD8 T cell ratios (p = 0.0091) compared to those with negative mNGS results. These findings highlight the potential influence of host immune status on the diagnostic efficiency of mNGS.
Figure 3.
Detection performance of mNGS in mixed infections and T. marneffei: (A) Absolute CD4+ T cell counts and CD4+/CD8+ ratios in patients with positive and negative mNGS results; (B) proportions of mixed infections detected by mNGS and CTM; (C) clinical manifestations of T. marneffei infections in the study cohort; (D) detection performance of mNGS in T. marneffei infections. mNGS, metagenomic next-generation sequencing. *p < 0.05, **p < 0.001.
Due to their compromised immune systems, individuals with HIV are more susceptible to mixed infections. Among 64 samples from 56 PLWH, mNGS detected mixed infections in 51 samples from 46 individuals (82.1%) (Figure 3B), while conventional testing methods detected mixed infections in only 3 samples from 3 individuals (5.4%). The most common type of mixed infection was a combination of viral, bacterial, and fungal infections (n = 23), followed by coinfections of fungi and viruses (n = 22). These findings highlight the complexity of infections in individuals with HIV/AIDS and underscore the utility of mNGS in accurately identifying multiple pathogens.
3.3. Talaromyces marneffei infections detected by mNGS
Talaromyces marneffei was detected in 39.3% (22/56, 95% CI: 26.5–53.3%) of PLWH in our cohort (Supplementary Table S1). Talaromyces marneffei infection usually presents with atypical symptoms including lymphadenopathy, fever, weight loss, rash, dyspnea, and cough (Figure 3C). Importantly, mNGS demonstrated 100% positivity rate (22/22, 95% CI: 84.6–100.00%) across diverse specimen types: peripheral blood (n = 13), BALF (n = 8), CSF (n = 1), and urine (n = 1). In contrast, CTM showed markedly lower positivity (45.5%, 10/22, 95% CI: 24.4–67.8%, p < 0.0001), detecting the pathogen in 53.8% of blood cultures (7/13) and 37.5% of BALF cultures (3/8) (Figure 3D). To validate the diagnostic performance of mNGS independently, we conducted a sensitivity analysis using CTM positivity as the sole microbiological reference standard. Among 56 patients, 10 were confirmed to have T. marneffei infection. mNGS demonstrated a sensitivity of 100.0% (10/10) and a specificity of 73.9% (34/46; 95% CI: 58.9–85.7%). These results confirm that mNGS successfully identified all culture-positive cases. The 12 mNGS-positive/CTM-negative cases presented with compatible clinical syndromes (fever, weight loss, lymphadenopathy) and responded to antifungal therapy, suggesting true infections missed by conventional methods (Supplementary Table S1). Importantly, mNGS identified T. marneffei in two diagnostically challenging cases: (1) a patient with an 8-month history of urinary frequency and urgency, in whom conventional diagnostics failed to yield a definitive diagnosis, and (2) a patient presenting with 3 months of dizziness and ataxia, where the pathogen was detected by cerebrospinal fluid mNGS. Following the mNGS-guided diagnosis, both patients received targeted antifungal therapy with liposomal amphotericin B, leading to resolution of urinary and neurological symptoms within 2–3 weeks.
3.4. Pathogen profiles based on mNGS
In this study, a total of 50 pathogens were detected using mNGS among the participants, including bacteria (n = 21), fungi (n = 13), viruses (n = 15), and protozoa (n = 1). The most common pathogenic bacteria were Haemophilus parainfluenzae (n = 6), Klebsiella pneumoniae (n = 6), and Streptococcus pneumoniae (n = 6). The predominant fungi were Pneumocystis jirovecii (n = 26), followed by T. marneffei (n = 23) and Candida tropicalis (n = 6). The most frequently detected viruses were Human betaherpesvirus 5 (CMV, n = 45) and Human gammaherpesvirus 4 (EBV, n = 41), with Torque teno virus (n = 36) also commonly detected.
As illustrated in Figure 4 and Supplementary Table S2, mNGS identified 40 pathogens in BALF samples and 23 pathogens in blood. Furthermore, mNGS was employed to detect Cryptococcus in two CSF samples. In addition, T. marneffei, CMV, and EBV were identified in three separate CSF samples using mNGS. Notably, the presence of T. marneffei was also confirmed in urine samples using mNGS. Meanwhile, Entamoeba histolytica was detected in pus aspirates from liver abscesses, and Mycobacterium tuberculosis was identified in pus aspirates from lymph node abscesses via mNGS.
Figure 4.
Pathogen spectrum identified by mNGS. (A) Pathogen spectrum detected by mNGS in bronchoalveolar lavage fluid (BALF). (B) Pathogen spectrum detected by mNGS in blood specimens. (C) Comprehensive pathogen spectrum detected by mNGS across all specimen types. mNGS, metagenomic next-generation sequencing.
Compared to mNGS, conventional detection methods identified a significantly lower number of pathogens. Talaromyces marneffei was the most frequently isolated pathogen. Other pathogens detected by traditional methods included a single instance of Cryptococcus identified through India ink staining, a case of tuberculosis confirmed by Xpert and T-SPOT assays, a positive result for CMV DNA, and a microscopic identification of Entamoeba histolytica. Overall, conventional pathogen detection methods exhibited limited ability to detect several fungi, including Pneumocystis jirovecii, T. marneffei, and Aspergillus. These results highlight that the mNGS platform demonstrated superior capability in revealing a wider range of pathogens compared to the culture-based platform in PLWH.
3.5. Influence of mNGS on diagnosis, antimicrobial therapy adjustment, and patient outcomes
The reporting time for mNGS was shorter than that of CTM (median: 1 day [IQR 1–2] vs. 4 days [IQR 3–6]; p < 0.0001). However, since specimen submission for mNGS testing was initiated later than CTM in clinical practice, no significant difference was observed in confirmation time (from hospital admission to definitive diagnosis) between the two methods (median: 4 days [IQR 2–8] vs. 5 days [IQR 4–7]; p = 0.315). Clinicians were significantly more likely to modify treatment regimens based on mNGS findings compared to CTM results (74.1% vs. 22.2%, p = 0.0009), indicating that mNGS information can effectively inform clinical decision-making (Figure 5A).
Figure 5.
mNGS impact on diagnosis, antimicrobial therapy, and outcomes: (A) Time from hospital admission to pathogen diagnosis via mNGS or CTM; (B) proportion of treatment regimen adjustments based on mNGS and CTM results; (C) Sankey diagram illustrating pathogen detection and clinical decision pathways. mNGS, metagenomic next-generation sequencing; CTM, conventional testing methods. ***p < 0.0001.
Empirical antimicrobial therapy was initiated at admission and adjusted based on mNGS or CTM results. Regimen modifications occurred in 70.6% (12/17) of mNGS+/CTM+ cases and 75.7% (28/37) of mNGS+/CTM− cases (Figure 5B). Empirical therapy was maintained throughout for mNGS−/CTM− (n = 9) and mNGS−/CTM+ (n = 1) cases. Treatment guided by mNGS results achieved 90% effectiveness (36/40), while empirical therapy alone resulted in 83.3% improvement (20/24).
Figure 5C shows the dynamic trajectories of 64 samples across five stages: sample, CD4+ T cell stratification, diagnostic concordance, treatment, and outcome. 93.2% (41/44, 95% CI: 81.3–98.6%) of severely immunocompromised individuals (CD4+ T cell <50 cells/mm3) were positive by mNGS or CTM, compared to 70% (14/20, 95% CI: 45.7–88.1%) in those with CD4+ T cell ≥50 cells/mm3 (p = 0.022). All mNGS−/CTM− patients received empirical therapy. Empirical antifungal escalation was similar in mNGS+/CTM− and mNGS+/CTM+ patients (12/17; 95% CI: 44.0–89.7%vs. 28/37; 95% CI: 58.8–88.2%, p = 0.467). One mNGS−/CTM+ patient, with a history of tuberculosis, received empirical anti-tuberculosis therapy before Xpert results. Most cases (36/40, 95% CI: 76.3–97.2%) achieved clinical remission after treatment adjustments. Treatment failures were mainly observed in severely immunodeficient patients with severe pneumonia. Empirical therapy failures were due to central nervous system infections or undetermined pathogens.
4. Discussion
In this study, we assessed the diagnostic utility of mNGS in PLWH. mNGS demonstrated high positivity rate and a broad pathogen detection spectrum compared to CTM, exhibiting good performance in identifying T. marneffei and mixed infections. This study revealed that the positivity rate of mNGS was significantly higher than that of CTM (84.4% vs. 28.1%, p < 0.0001) and comparable to previous studies (6, 26). Consistent with the literature (26, 27), our data demonstrated that host immunocompetence critically determines the diagnostic efficacy of mNGS, showing enhanced detection positivity in patients with lower CD4+ T cell counts.
Approximately 30% of T. marneffei remain undiagnosed or misdiagnosed attributed to the atypical clinical manifestations (28, 29). Central umbilicated rash is pathognomonic for T. marneffei, but is absent in up to 60% of patients (30). In our cohort, fewer than 50% presented with suggestive skin lesions, while nonspecific symptoms such as fever, cough, lymphadenopathy, and weight loss were more common, overlapping with other opportunistic infections. In our study, the detection rate of T. marneffei by mNGS was significantly higher than that by CTM, particularly in BALF samples (100% vs. 37.5%, p = 0.026), consistent with prior studies (31). While inclusion of mNGS in our reference standard may overestimate specificity, mNGS maintained perfect sensitivity (100%) and NPV (100%) in sensitivity analysis, confirming its value for excluding disease. The lower specificity (73.9%) and PPV (45.5%) likely reflect imperfect gold standards rather than false positives, supported by clinical findings and treatment response in mNGS-positive/CTM-negative cases. Notably, mNGS facilitated the diagnosis of two atypical T. marneffei cases involving the urinary tract and central nervous system, both of which had remained undiagnosed for several months following initial clinical assessment. Our findings emphasize the pressing need for the incorporation of mNGS into T. marneffei control programs within endemic regions.
mNGS also demonstrated a broader capacity for identifying potential pathogens. In this study, a therapy-refractory case initially misdiagnosed as a bacterial liver abscess was rapidly and accurately diagnosed as Entamoeba histolytica through mNGS analysis of pus samples. These findings demonstrate the enhanced detection capability of mNGS for fastidious pathogens and agents of neglected tropical diseases. Furthermore, mNGS offers a distinct advantage in diagnosing mixed infections due to its unbiased detection capabilities (26, 32). In our study, most patients (82.1%, n = 46/56) exhibited mixed infections, with the most common combinations being bacterial-fungal-viral and fungal-viral infections. This pattern may be explained by the high prevalence of respiratory infections in our cohort, with CMV and Pneumocystis jirovecii frequently detected in immunocompromised individuals with pulmonary infections (6, 20, 33, 34).
mNGS is limited in its ability to discriminate colonization from infection (35, 36). mNGS results require cautious interpretation to establish definitive diagnoses and avoid unnecessary treatment. In practice, clinical microbiologists categorized detected organisms by mNGS as pathogens of concern, suspected pathogens, or colonizing flora based on specimen type and clinical context (9); clinicians subsequently integrated these classifications with clinical manifestations, laboratory findings, and radiological evidence to guide treatment decisions. In our PLWH cohort, detection of T. marneffei or Pneumocystis jirovecii with compatible features led to antifungal therapy. By contrast, viral detections such as CMV were more difficult to interpret, illustrating that therapeutic decisions are inherently pathogen-specific.
This study has several limitations that should be acknowledged. First, the single-center design and relatively small sample size may introduce selection bias. Future prospective, multicenter studies with larger cohorts are warranted to directly compare the diagnostic accuracy of mNGS and CTM in PLWH. Second, the non-simultaneous application of mNGS and CTM precluded valid comparisons of their individual contributions to reduce the time to definitive diagnosis.
5. Conclusion
Our data demonstrated that mNGS had a higher positive detection rate than CTM for detecting coinfections among PLWH, especially for T. marneffei and mixed infections. These results highlight the clinical value of mNGS as a complementary tool for pathogen identification in this vulnerable population.
Acknowledgments
We especially thank all the PLWH participants from the Third Affiliated Hospital of Sun Yat-sen University.
Funding Statement
The author(s) declared that financial support was received for this work and/or its publication. This study was supported by the Prevention and Control of Emerging and Major Infectious Diseases-National Science and Technology Major Project (2025ZD01904600, 2025ZD01904602, 2025ZD01904900, 2025ZD01904902); National Key Research and Development Program of China (2024YFA1108802, 2024YFA1108803); National Natural Science Foundation of China (81901940, 82070612); Natural Science Foundation of Guangdong Province (2021A1515010306, 2025A1515012430); Guangzhou Science and Technology Program Key Projects (2023B03J0154, 2023B01J1007).
Footnotes
Edited by: Amit Ranjan, Columbia University, United States
Reviewed by: Caihong Bi, Emory University, United States
Yong Xiong, Wuhan University, China
Data availability statement
The data presented in the study are deposited in the CNGBdb repository, accession number CNP0009096 (https://doi.org/10.26036/CNP0009096).
Ethics statement
The studies involving humans were approved by the ethics committee of the Third Affiliated Hospital of Sun Yat-sen University. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
Author contributions
XL: Conceptualization, Writing – review & editing, Methodology, Writing – original draft, Visualization, Formal analysis. DC: Methodology, Data curation, Writing – original draft. YX: Methodology, Data curation, Formal analysis, Writing – review & editing. ZL: Data curation, Writing – review & editing. XY: Writing – review & editing, Data curation. YeZ: Writing – review & editing, Data curation. LL: Data curation, Writing – review & editing. YuZ: Data curation, Writing – review & editing. YiZ: Data curation, Writing – review & editing. ZH: Methodology, Conceptualization, Formal analysis, Supervision, Writing – review & editing. BL: Funding acquisition, Supervision, Conceptualization, Writing – review & editing.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmed.2026.1800314/full#supplementary-material
References
- 1.Bielick C, Strumpf A, Ghosal S, McMurry T, McManus KA. National hospitalization rates and in-hospital mortality rates of HIV-related opportunistic infections in the United States, 2011-2018. Clin Infect Dis. (2024) 79:487–97. doi: 10.1093/cid/ciae051, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hoving JC, Brown GD, Gómez BL, Govender NP, Limper AH, May RC, et al. AIDS-related mycoses: updated progress and future priorities. Trends Microbiol. (2020) 28:425–8. doi: 10.1016/j.tim.2020.01.009, [DOI] [PubMed] [Google Scholar]
- 3.Richards GA, Zamparini J, Kalla I, Laher A, Murray LW, Shaddock EJ, et al. Critical illness due to infection in people living with HIV. Lancet HIV. (2024) 11:e406–18. doi: 10.1016/s2352-3018(24)00096-1, [DOI] [PubMed] [Google Scholar]
- 4.Grčević D, Batinić D, Ascensao JL, Marušić M. Pre-treatment of transplant bone marrow cells with hydrocortisone and cyclosporin a alleviates graft-versus-host reaction in a murine allogeneic host–donor combination. Bone Marrow Transplant. (1999) 23:1145–52. doi: 10.1038/sj.bmt.1701777, [DOI] [PubMed] [Google Scholar]
- 5.Blauwkamp TA, Thair S, Rosen MJ, Blair L, Lindner MS, Vilfan ID, et al. Analytical and clinical validation of a microbial cell-free DNA sequencing test for infectious disease. Nat Microbiol. (2019) 4:663–74. doi: 10.1038/s41564-018-0349-6 [DOI] [PubMed] [Google Scholar]
- 6.Miao Q, Ma Y, Wang Q, Pan J, Zhang Y, Jin W, et al. Microbiological diagnostic performance of metagenomic next-generation sequencing when applied to clinical practice. Clin Infect Dis. (2018) 67:S231–40. doi: 10.1093/cid/ciy693, [DOI] [PubMed] [Google Scholar]
- 7.Goldberg B, Sichtig H, Geyer C, Ledeboer N, Weinstock GM. Making the leap from research laboratory to clinic: challenges and opportunities for next-generation sequencing in infectious disease diagnostics. MBio. (2015) 6:e01888–15. doi: 10.1128/mbio.01888-15, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zhang Y, Ai J-W, Cui P, Zhang W-H, Wu H-L, Ye M-Z. A cluster of cases of pneumocystis pneumonia identified by shotgun metagenomics approach. J Infect. (2019) 78:158–69. doi: 10.1016/j.jinf.2018.08.013, [DOI] [PubMed] [Google Scholar]
- 9.Gu W, Miller S, Chiu CY. Clinical metagenomic next-generation sequencing for pathogen detection. Annu Rev Pathol Mech Dis. (2019) 14:319–38. doi: 10.1146/annurev-pathmechdis-012418-012751, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wu X, Li Y, Zhang M, Li M, Zhang R, Lu X, et al. Etiology of severe community-acquired pneumonia in adults based on metagenomic next-generation sequencing: a prospective multicenter study. Infect Dis Ther. (2020) 9:1003–15. doi: 10.1007/s40121-020-00353-y, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ávila-Ríos S, Parkin N, Swanstrom R, Paredes R, Shafer R, Ji H, et al. Next-generation sequencing for HIV drug resistance testing: laboratory, clinical, and implementation considerations. Viruses. (2020) 12:617. doi: 10.3390/v12060617, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kullar R, Chisari E, Snyder J, Cooper C, Parvizi J, Sniffen J. Next-generation sequencing supports targeted antibiotic treatment for culture negative orthopedic infections. Clin Infect Dis. (2022) 76:359–64. doi: 10.1093/cid/ciac733, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hilton SK, Castro-Nallar E, Pérez-Losada M, Toma I, McCaffrey TA, Hoffman EP, et al. Metataxonomic and metagenomic approaches vs. culture-based techniques for clinical pathology. Front Microbiol. (2016) 7:484. doi: 10.3389/fmicb.2016.00484, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Xie G, Zhao B, Wang X, Bao L, Xu Y, Ren X, et al. Exploring the clinical utility of metagenomic next-generation sequencing in the diagnosis of pulmonary infection. Infect Dis Ther. (2021) 10:1419–35. doi: 10.1007/s40121-021-00476-w, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kalbitz S, Ermisch J, Kellner N, Nickel O, Borte S, Marx K, et al. Metagenomic next-generation sequencing as a diagnostic tool in the clinical routine of an infectious diseases department: a retrospective cohort study. Infection. (2024) 52:1595–600. doi: 10.1007/s15010-024-02300-2, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jiang J, Meng S, Huang S, Ruan Y, Lu X, Li JZ, et al. Effects of Talaromyces marneffei infection on mortality of HIV/AIDS patients in southern China: a retrospective cohort study. Clin Microbiol Infect. (2019) 25:233–41. doi: 10.1016/j.cmi.2018.04.018, [DOI] [PubMed] [Google Scholar]
- 17.Sati H, Alastruey-Izquierdo A, Perfect J, Govender NP, Harrison TS, Chiller T, et al. HIV and fungal priority pathogens. Lancet HIV. (2023) 10:e750–4. doi: 10.1016/S2352-3018(23)00174-1, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Zhao J, Su H-T, Zhang D, Li X-T, Lu M-Y, Du J, et al. The role of mNGS in the diagnosis of talaromycosis and case series. BMC Infect Dis. (2025) 25:191. doi: 10.1186/s12879-025-10529-x, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chen Q, Qiu Y, Zeng W, Wei X, Zhang J. Metagenomic next-generation sequencing for the early diagnosis of talaromycosis in HIV-uninfected patients: five cases report. BMC Infect Dis. (2021) 21:865. doi: 10.1186/s12879-021-06551-4, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Tan Y, Chen Z, Zeng Z, Wu S, Liu J, Zou S, et al. Microbiomes detected by bronchoalveolar lavage fluid metagenomic next-generation sequencing among HIV-infected and uninfected patients with pulmonary infection. Microbiol Spectrum. (2023) 11:e0000523. doi: 10.1128/spectrum.00005-23, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shi Y, Yang J, Qi Y, Xu J, Shi Y, Shi T, et al. Detection of Bartonella vinsonii subsp. berkhoffii in an HIV patient using metagenomic next-generation sequencing. Emerg Microbes Infect. (2022) 11:1764–7. doi: 10.1080/22221751.2022.2094287, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Li X, Li Z, Ye J, Wu Y. Diagnostic performance of metagenomic next-generation sequencing for Pneumocystis jirovecii pneumonia. BMC Infect Dis. (2023) 23:455. doi: 10.1186/s12879-023-08440-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Yaokai Chen, Taisheng Li, Wenhong Zhang. AIDS and hepatitis C professional group, Society of Infectious Diseases, Chinese Medical Association: expert consensus on the diagnosis and treatment of AIDS complicated with Penicilliosis Marneffei (2024 updated version). Chin J AIDS & STD (2024), 30: 563–572. doi: 10.13419/j.cnki.aids.2024.06.02(In Chinese). [DOI] [Google Scholar]
- 24.Ye S, Qin J, Miao X, Lang G, Wang M, Chen G, et al. Voriconazole is inferior to amphotericin B deoxycholate as the initial induction therapy for HIV-associated Talaromyces marneffei fungemia: a multicenter retrospective study. PLoS Negl Trop Dis. (2025) 19:e0013012. doi: 10.1371/journal.pntd.0013012, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Peng J-M, Du B, Qin H-Y, Wang Q, Shi Y. Metagenomic next-generation sequencing for the diagnosis of suspected pneumonia in immunocompromised patients. J Infect. (2021) 82:22–7. doi: 10.1016/j.jinf.2021.01.029, [DOI] [PubMed] [Google Scholar]
- 26.Xie Y, Dai B, Zhou X, Liu H, Wu W, Yu F, et al. Diagnostic value of metagenomic next-generation sequencing for multi-pathogenic pneumonia in HIV-infected patients. Infect Drug Resist. (2023) 16:607–18. doi: 10.2147/idr.s394265, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Sun T, Wu X, Cai Y, Zhai T, Huang L, Zhang Y, et al. Metagenomic next-generation sequencing for pathogenic diagnosis and antibiotic management of severe community-acquired pneumonia in immunocompromised adults. Front Cell Infect Microbiol. (2021) 11:661589. doi: 10.3389/fcimb.2021.661589, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Mai T, Chan JFW, Vo Trieu Ly, Ngo Thi Hoa, Thuc H, Lan H, et al. Superiority of a novel Mp1p antigen detection enzyme immunoassay compared to standard BACTEC blood culture in the diagnosis of talaromycosis. Clin Infect Dis. (2020) 73:e330–6. doi: 10.1093/cid/ciaa826 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kawila R, Chaiwarith R, Supparatpinyo K. Clinical and laboratory characteristics of Penicilliosis marneffei among patients with and without HIV infection in northern Thailand: a retrospective study. BMC Infect Dis. (2013) 13:464. doi: 10.1186/1471-2334-13-464, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Chen J, Song W, Liu L, Tang Y, Qi T, Zhang R, et al. Clinical characteristics and prognosis of Penicilliosis among human immunodeficiency virus–infected patients in eastern China. Am J Trop Med Hyg. (2017) 96:1350–4. doi: 10.4269/ajtmh.16-0521, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Mao Y, Shen H, Yang C, Jia Q, Li J, Chen Y, et al. Clinical performance of metagenomic next-generation sequencing for the rapid diagnosis of talaromycosis in HIV-infected patients. Front Cell Infect Microbiol. (2022) 12:962441. doi: 10.3389/fcimb.2022.962441 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Wang J, Han Y, Feng J. Metagenomic next-generation sequencing for mixed pulmonary infection diagnosis. BMC Pulm Med. (2019) 19:252. doi: 10.1186/s12890-019-1022-4, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Chen Q, Chen Z, Tan Y, Wu S, Zou S, Liu J, et al. Blood microbiota in HIV-infected and HIV-uninfected patients with suspected sepsis detected by metagenomic next-generation sequencing. BMC Infect Dis. (2024) 24:1210. doi: 10.1186/s12879-024-10009-8, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Jiang J, Bai L, Yang W, Peng W, An J, Wu Y, et al. Metagenomic next-generation sequencing for the diagnosis of pneumocystis jirovecii pneumonia in non-HIV-infected patients: a retrospective study. Infect Dis Ther. (2021) 10:1733–45. doi: 10.1007/s40121-021-00482-y, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Simner PJ, Miller S, Carroll KC. Understanding the promises and hurdles of metagenomic next-generation sequencing as a diagnostic tool for infectious diseases. Clin Infect Dis. (2017) 66:778–88. doi: 10.1093/cid/cix881, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Chen J, Zhang R, Liu L, Qi T, Wang Z, Song W, et al. Clinical usefulness of metagenomic next-generation sequencing for the diagnosis of central nervous system infection in people living with HIV. Int J Infect Dis. (2021) 107:139–44. doi: 10.1016/j.ijid.2021.04.057, [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data presented in the study are deposited in the CNGBdb repository, accession number CNP0009096 (https://doi.org/10.26036/CNP0009096).