Herpesvirus-associated diseases: biomarkers and advancements in clinical research

Abstract

Herpesviruses are responsible for a broad spectrum of diseases that impact various tissues and organs, which could lead to complex pathological manifestations and severe complications. These infections represent a significant burden on global public health. Increasing researches have focused on identifying biomarkers associated with herpesvirus-related diseases and the underlying mechanisms. Studies using clinical samples may better reflect the molecular basis of disease progression, offering new insights into the illustration of disease development. This review provides a comprehensive summary of recent progress in biomarker discovery across a range of disease conditions caused by herpesviruses, highlighting the latest experimental findings and clinical observations, with a particular focus on the potential applications of these biomarkers in clinical settings. Additionally, we discuss the major challenges in the current research of herpesvirus-associated diseases, intending to foster the discovery of novel diagnostic and therapeutic strategies and deepen our understanding of herpesvirus pathogenesis.

Background

Herpesvirus is a pervasive double-stranded DNA enveloped viruses, causing lifelong infections and a spectrum of diseases [1]. Eight herpesviruses were identified as pathogens for humans: the α-herpesvirus subfamily (herpes simplex virus 1 [HSV-1], herpes simplex virus 2[HSV-2], and varicella-zoster virus [VZV]), the β-herpesvirus subfamily (human cytomegalovirus [HCMV], HHV-6, and HHV-7), and the γ-herpesvirus subfamily (Epstein-Barr virus [EBV] and HHV8) [2, 3] (Table 1).

Table 1.

List of human herpesviruses and their related diseases

Virus species	Diseases	Symptoms
HHV-1/Human Herpesvirus 1/Herpes Simplex Virus 1 (HSV-1)	Encephalitis, Keratitis	Headache, fever, altered consciousness, focal neurological deficits
HHV-2/Human Herpesvirus 2/Herpes Simplex Virus 2 (HSV-2)	Genital herpes, Aseptic meningitis	Painful blisters, headache, fever, local neuralgia
HHV-3/Varicella Zoster Virus (VZV)	Chickenpox, Shingles	Painful, Itchy rash, blisters, fever.
HHV-4/Epstein-Barr Virus (EBV)	Nasopharyngeal Cancer, Gastric Cacer, Burkitt’s Lymphoma, Burkitt’s Lymphoma, EBV mononucleosis, Oral hairy leucoplakia	Nasal congestion, lymphadenopathy, fever, weight loss, difficulty swallowing, upper abdominal pain, splenomegaly, hepatomegaly
HHV-5/Human Cytomegalovirus(HCMV)	Congenital infection, Glioblastoma, HCMV infection in transplantation (solid organ transplant (SOT) and hematopoietic stem cell transplant (HSCT))	Congenital abnormalities, neurological symptoms, muscle weakness
HHV-6/Human Herpesvirus 6 (HHV-6 A and HHV6B)	Human herpesvirus 6 encephalitis, Human herpesvirus 6 myelitis, HHV6 infection in SOT and HSCT	Headache, epileptic seizures, acute cerebellar ataxia, mild hemiparesis, neurological symptoms
HHV-7/Human Herpesvirus 7	Encephalitis	Headache, fever, altered consciousness, focal neurological deficits
HHV-8/Kaposi’s Sarcoma-associated Herpesvirus (KSHV)	Kaposi’s sarcoma, Multicentric Castleman’s Disease, Primary Effusion Lymphoma	Purple or brown nodules or plaques on the skin or mucous membranes

Human herpesviruses (HHVs) are responsible for a wide array of diseases, ranging from common conditions such as herpes simplex, chickenpox, herpes zoster (HZ), and infectious mononucleosis to more severe disorders, including herpes simplex encephalitis (HSE), nasopharyngeal carcinoma, lymphomas, and congenital anomalies. These impacts are particularly pronounced in pediatric populations. Despite their wide prevalence, many individuals exposed to one or more HHVs remain asymptomatic, as the virus enters a latent phase that may persist stably for a lifetime. Under conditions of acute stress or immunosuppression—such as HIV infection, organ transplantation, or post-traumatic immunosuppression—primary infection and reactivation in these immunocompromised patients of the virus can occur, leading to a broad spectrum of complex clinical manifestations [4].

Numerous studies on human herpesvirus-associated diseases have focused on analyzing clinical samples from symptomatic or infected individuals. Blood samples, such as serum or plasma, and cerebrospinal fluid (CSF) are among the most commonly examined specimens. The concept of cell-free nucleic acids emerged in the late 20th century and has since gained increasing attention. DNA, mRNA, and microRNAs released into the bloodstream have been extensively investigated as potential biomarkers, particularly in the context of tumor detection and diagnosis [5].

In recent years, extensive research has concentrated on human herpesvirus-associated diseases, with a particular focus on identifying potential biomarkers. These investigations, conducted for purposes such as screening, diagnosis, monitoring disease progression, and prognostic evaluation, involve rigorous research and validation of biomarkers through cohort studies using clinical samples. Beyond cell-free nucleic acids in the blood, studies now encompass various sample types, including CSF, saliva, tears, amniotic fluid, and biopsies, either individually or in combination. Additionally, proteins, antibodies, cytokines, and other biomolecules are being actively explored for their potential as diagnostic and prognostic biomarkers.

Due to the limitations of animal model studies on HHVs, clinical samples have become essential for advancing our understanding of herpesvirus pathogenesis, disease burden, diagnosis, prognosis, and management. This article provides a systematic review of these studies, compiling and summarizing cohort investigations involving patients with herpesvirus-associated infections. The review aims to offer valuable insights for the development of novel diagnostic and therapeutic strategies, while also enhancing our understanding of the mechanisms underlying herpesvirus pathogenesis.

HSV-1/2

HSV-1 and HSV-2 are both classified within the alpha-herpesvirus subfamily. It is estimated that approximately 90% of the global population is infected with one or both HSV types. Currently, there is no cure for these infections. After the primary infection, HSV-1 predominantly establishes latency in the trigeminal ganglia, while HSV-2 resides in the sacral sensory ganglia. Reactivation typically occurs years later with immunocompromised, resulting in mucocutaneous lesions, including cold sores and genital herpes [6]. In some cases, HSV infection may lead to inflammation of the central nervous system (CNS), manifesting as encephalitis or meningitis. HSV-1 also plays a significant role in ocular diseases, particularly herpes simplex keratitis (HSK), which is the leading cause of infectious blindness and severe visual impairment in developed countries [7, 8].

Encephalitis and meningitis

HSV-1 can cause sporadic HSE, a devastating condition with a high rate of complications and mortality, even with the availability of antiviral treatments. In non-neonatal cases, HSV-1 is the primary etiologic agent, accounting for the vast majority of cases, while HSV-2 contributes to less than 10%. A prospective study involving 349 patients with encephalitis identified HSV-1 as the causative pathogen in 25% of cases, making it the leading etiological factor [9] (Table 2).

Table 2.

Summary of cohort study information based on HSV clinical sample studies

Author	Disease	Biomarker	No. of controls	No. of patients	Age	Sample
Ahmed et al. [11]	Herpes simplex encephalitis	14-3-3 family proteins calreticulin	-	13	25–87 years	CSF
Lind et al. [12]	Herpes simplex encephalitis and meningitis	HSE: CXCL8, 9, 10, 11, CCL8 HSM: CXCL11, CCL8	-	14	27–89 years	Serum, CSF
Armangue et al. [13]	Herpes simplex encephalitis	-	51	49	5–68 years	CSF, Serum
Le Maréchal et al. [9]	Herpes simplex encephalitis	-	-	349	19–60 years	Blood, CSF
Ramirez et al. [14]	Herpes simplex encephalitis	-	-	33	0–77 years	CSF
Miller et al. [15]	Herpes simplex meningitis	-	-	33	19–55 years	CSF
Kim et al. [16]	Herpes epithelial keratitis	23 miRNAs	7	8	42–65 years	Tear
Yang et al. [17]	Herpes epithelial keratitis	IL1A, IL12B, DEFB4A, CAMP	3	3	21–50 years	Tear
Johnston et al. [18]	Genital herpes	-	-	82	16–64 years	Oral and genital swabs

In neonatal herpes encephalitis, both HSV-1 and HSV-2 can be causative agents. Aseptic herpes simplex meningitis (HSM), distinct from HSV encephalitis, is generally caused by HSV-2 in immunocompetent adults, although HSV-1 can also be a factor. HSM typically presents with mild symptoms, such as fever, headache, neck stiffness, and photosensitivity. However, in some cases, patients may experience more severe symptoms, including ascending myelitis [10] (Table 2).

HSV-1 and HSV-2 can target the CNS, triggering an inflammatory response within the CNS. Ahmed et al. conducted a study utilizing CSF proteomics to investigate herpesvirus infections in CNS [11]. This study processed patient samples in a high-throughput manner to explore the development of host CSF responses in alpha-herpesvirus infection. Specifically, the study focused on the proteomic characterization of the CSF in patients with meningitis or encephalitis caused by HSV-1. A comparative analysis was performed by contrasting the CSF proteome of HSV-1-infected patients with those infected by HSV-2 or VZV. Notable differences were observed in the CSF proteome of HSV-1-infected individuals, including elevated levels of 14-3-3 family proteins and calreticulin. Additionally, organic substance metabolic process, cellular nitrogen compound metabolic process and small molecule metabolic process were down-regulated in HSV-1 infections. The study aimed to identify potential biomarkers for HSV-1 encephalitis (Table 2).

In 2018, Liza Lind and colleagues conducted a study on chemokines and cytokines in the CSF and serum of patients with HSE and HSM to elucidate the varying degrees of CNS inflammation exhibited by closely related viruses. The results revealed elevated levels of most tested chemokines and cytokines in the CSF of both HSE and HSM patients. Notably, despite HSE symptoms being more severe, chemokine levels in the CSF of HSM patients surpassed those in HSE patients. Only five chemokines (CXCL8, 9, 10, 11, and CCL8) exhibited higher levels in CSF compared to serum, contributing to a positive CSF-serum chemokine gradient. CXCL8, CXCL9, and CXCL10 were highly expressed in both HSE and HSM, while CXCL11 and CCL8 were exclusive to HSM, differentiating between the two conditions. Particularly, CXCL11 was solely expressed in CSF above serum levels in HSM patients whereas this was not the case in patients with HSE. No correlation was found between CSF and serum levels of chemokines, suggesting that there is no chemokine influx or efflux between them [12] (Table 2).

In the same study, Armangue et al. reported a prospective observational study and a retrospective study, detailing the frequency, clinical characteristics, risk factors, and prognosis of autoimmune encephalitis following HSE in two patient cohorts, A and B. Cohort A comprised 51 prospectively studied patients with HSE, while Cohort B included 49 patients with new or exacerbated symptoms after HSE, explored retrospectively. Autoantibody analysis revealed autoimmune encephalitis in 27% of patients with HSE, associated with neuronal antibody production and typically manifesting within two months post-treatment. The importance of timely diagnosis has been emphasized through cohort studies which have found that the severity of symptoms is age-related, with younger children having a poorer prognosis [13] (Table 2).

Ramirez et al. investigated the correlation between HSV-1/2 viral load in CSF and the severity of HSE in a cohort of 33 patients, primarily neonates and children [14]. However, the study revealed that the initial HSV viral load did not predict neuroradiological disease or clinical outcomes in HSE patients during hospitalization, at discharge, or in the long-term follow-up exceeding 3 months (Table 2).

Stephanie Miller and colleagues conducted a retrospective observational cohort study involving 28 patients diagnosed with HSV-2 meningitis, confirmed in cerebrospinal fluid by polymerase chain reaction (PCR). The long-term follow-up, averaging 3.4 years, showed no occurrence of seizures, neurologic disability, or death among patients, suggesting to some extent that symptomatic recurrence of meningitis is not universal [15] (Table 2).

In patients with HSV-infected encephalitis or meningitis, detection of key factors in the cerebrospinal fluid can help in the early diagnosis of the patient’s condition, as well as in predicting the progression of the patient’s disease and reducing the risk of serious complications.

Keratitis

HSK is the most common form of ocular involvement by HSV-1, representing a major cause of corneal scarring, opacification, and, in severe cases, blindness. HSK can be classified into epithelial, stromal, and endothelial types, depending on the anatomical layer affected. Despite ongoing research, the molecular mechanisms driving this condition are not yet fully understood [19]. Current evidence suggests that most ocular HSV manifestations arise from the reactivation of latent infections.

In a recent investigation, Kim et al. examined tear samples obtained from 8 patients diagnosed with HSK and 7 age-matched controls. The primary aim of the study was to explore the mechanisms of action of miRNA molecules in HSK. The researchers employed real-time fluorescence quantitative polymerase chain reaction to assess the expression of 43 different miRNAs in the tear fluid of HSK patients, comparing the findings with those of controls. The investigation revealed that, among the miRNAs, 23 exhibited up-regulation relative to the control group, miR-15b-5p, miR-16-5p, miR-20b-5p, miR-21-5p, miR-23b-3p, miR-253p, miR-29a-3p, miR-30a-3p, miR-30d-5p, miR-92a-3p, miR-124-3p, miR-127-3p, miR-132-3p, miR-142-3p, miR-145-5p, miR-146a-5p, miR-146b-5p, miR-155-5p, miR-182-5p, miR-183-5p, miR-221-3p, miR-223-3p and miR-338-5p. Consequently, these findings suggest a potential role for these miRNAs in herpesvirus infections, particularly in association with host immunity. The researchers also aimed to assess the severity of HSK by measuring the area of dendritic or geographic ulcers, attempting to correlate ulcer size with the expression levels of the identified miRNAs. However, their analysis found no significant correlations between the miRNA expression and the ulcer area [16] (Table 2).

In addition to examining miRNA at the transcriptomic level, Hua Yang et al. utilized proteomic approaches to compare and analyze total protein profiles in tears from patients with HSV-1 epithelial keratitis. The study aimed to identify potential candidate biomarkers for HSV-1 epithelial keratitis. A total of three patients with HSV-1 epithelial keratitis and three healthy controls were included, leading to the identification of 1,275 proteins in tear fluid. Notably, 326 proteins were specific to the tears of patients with HSK. Bioinformatics analysis revealed the potential involvement of tear proteins in metabolic processes, antigen presentation, inflammatory response, TNF-mediated, and T-cell receptor pathways. The studies concluded that immune cells and inflammatory cytokines played a role in HSK. Specifically, IL1A, IL12B, DEFB4A, and CAMP, associated with inflammatory response and viral infection suppression, exhibited significantly higher expression in patients with HSK compared to healthy controls, suggesting that these four factors may serve as biomarkers for HSK [17] (Table 2).

Genital HSV infections

HSV-1 is the leading cause of first-episode genital herpes in many countries [18, 20]. Examining HSV-1/2 viruses beyond their impact on various inflammatory diseases is crucial. Johnston et al. investigated the pattern of viral shedding in the oral and genital tracts of HSV-1-infected patients [18]. The study also characterized the trajectory of HSV-specific antibodies and T-cell responses. Eighty-two participants were enrolled with a follow-up period of up to 2 years. Monitoring HSV-1 virulence involved serum collection for antibody analysis and DNA extraction for PCR. Results indicated more frequent genital HSV-1 shedding after the initial infection, particularly in patients experiencing their first infection. Furthermore, shedding rates declined rapidly within the first year after infection (Table 2).

Conclusion

HSV-1/2, as members of the alpha-herpesvirus subfamily, are associated with a diverse range of clinical manifestations affecting the CNS, ocular tissues, and genital mucosa. Recent clinical cohort studies and biomarker-based investigations have advanced our understanding of the pathogenesis and diagnostic potential of HSV-related diseases. In the CNS, HSV-1 remains the predominant cause of sporadic encephalitis, while HSV-2 is more frequently linked to aseptic meningitis. Biomarker analyses, particularly in cerebrospinal fluid, have revealed distinct chemokine and protein expression patterns, such as elevated 14-3-3 family proteins, CXCLs, and calreticulin, which may serve as indicators of disease severity and etiology. In HSK, both transcriptomic and proteomic studies of tear fluid have identified potential diagnostic biomarkers, including multiple dysregulated miRNAs and immune-associated proteins such as IL1A and CAMP. Furthermore, investigations of genital HSV-1 infections underscore the dynamics of viral shedding and host immune responses during and after primary infection. Collectively, these findings highlight the importance of site-specific biomarker discovery in improving early diagnosis, disease stratification, and management of HSV infections across diverse clinical presentations.

VZV

VZV, an alpha-herpesvirus related to HSV-1 and HSV-2, initially causes chickenpox during primary infection. Following this phase, the virus establishes latency in the dorsal root ganglia or cranial nerve ganglia, where it can reactivate later in life, resulting in HZ (shingles), characterized by a painful, localized rash and blistering [21]. Chickenpox typically presents as a widespread rash during the primary infection, while HZ exhibits more localized symptoms in the dermatome corresponding to the reactivated nerve. Complications of VZV reactivation include herpes zoster ophthalmicus, which is particularly common in older adults and can lead to severe corneal involvement. Additionally, motor nerve paralysis may occur when the virus spreads to the anterior horn cells of the spinal cord or visceral nerve fibers. In some cases, this infiltration can cause facial nerve paralysis when the virus affects the facial nerve [22].

Herpes zoster

HZ is characterized by the persistence of a latent VZV reservoir in ganglia along the neuraxis, which can reactivate later in life. Despite the availability of antiviral treatments, the clinical manifestations of VZV reactivation vary widely, ranging from uncomplicated herpes zoster to life-threatening conditions such as meningoencephalitis [23]. Acute pain, particularly in the form of postherpetic neuralgia (PHN), and extensive cranial nerve involvement are common. Given the variability in clinical severity and the potential for serious complications, there is a need for reliable biomarkers to facilitate early diagnosis, risk stratification, outcome prediction, and a deeper understanding of the mechanisms underlying clinical variability.

Kuhn et al. conducted a mass spectrometry analysis of cerebrospinal fluid from 45 patients with VZV, categorizing them based on three distinct VZV reactivation patterns: segmental HZ, facial nerve HZ, and HZ meningitis and/or encephalitis. The study was conducted to explore metabolite biomarkers associated with varicella-zoster virus reactivation in the CNS. The sum of hexoses and the amino acids arginine, serine, and tryptophan exhibited a negative correlation with leukocyte counts in all samples. Increased expression of metabolites linked to VZV meningoencephalitis suggested associations with neuroinflammatory/immune activation, neuronal signaling, cellular stress, turnover, and death (e.g., autophagy and apoptosis) [24]. These findings imply that these metabolites may serve as indicators of processes associated with end-organ damage (Table 3).

Table 3.

Summary of cohort study information based on VZV clinical sample studies

Author	Disease	Biomarker	No.of controls	No.of patients	Age	Sample
Kuhn et al. [24]	Herpes zoster	Amino acids arginine, series, tryptophan	15	45	13–89 years	CSF
Khazan et al. [25]	Herpes zoster	IL-6, IL-18, ferritin, CRP, MLT, Homocysteine	47	43	≥ 20 years	Serum
Oskay et al. [26]	Herpes zoster	Hcy, CRP	53	53	54–88 years	Blood
Park et al. [27]	Herpes zoster	-	27	30	≥ 18 years	Saliva, Plasma
Nithyanandam et al. [28]	Ocular herpes zoster	-	-	64	6–75 years	-
Niederer et al. [29]	Ocular herpes zoster	-	-	869	52.9–75.4 years	-

In the same year, Khazan et al. published an article, which investigated the impact of oxidative status on patients with HZ compared to oxidative stress biomarker levels in control subjects. This case-control study involved measuring serum levels of total antioxidant capacity (TAC), total oxidative status, oxidative stress index, glutathione, superoxide dismutase, and total polyphenol content (TPC) in 43 patients with HZ and 47 age-matched controls. The biomarker patterns were compared, revealing that HZ patients exhibited significantly lower TAC and reduced TPC levels [25]. These findings suggest an imbalance in oxidative stress in patients with HZ, indicating this status as a potential predictive factor.

In a subsequent study, researchers conducted a related investigation measuring melatonin, indole dioxygenase, IL-6, IL-18, ferritin, CRP, and total homocysteine levels during HZ in same patients. This aimed to further assess oxidative and inflammatory stress biomarker levels in HZ patients. HZ patients demonstrated significantly higher serum levels of IDO, IL-18, IL-6, ferritin, hsCRP, and tHcy, along with significantly lower MLT levels. These differences correlated with the severity of rash and pain, indicating increased inflammation-related oxidative stress in HZ patients, which is consistent with the findings of previous studies.

Oskay et al. conducted a study examining the levels of major antioxidants in the blood of patients with HZ, specifically analyzing serum levels of uric acid (UA), total bilirubin (TBil), albumin (ALB), and vitamin D in 53 patients with HZ and 53 age- and sex-matched healthy controls. Inflammatory markers, including homocysteine (Hcy) and C-reactive protein (CRP), were also examined. The study revealed significantly lower levels of serum UA, TBiL, and ALB in HZ patients compared to the control group (p < 0. 001), however, the difference in vitamin D level was not statistically significant. Meanwhile, Hcy and CRP levels were significantly higher in HZ patients, positively correlating with disease activity [26]. These results suggest an association between low antioxidant levels and uncontrolled varicella-zoster virus reactivation, acute neurologic injury, and PHN.

In addition to assessing biomarkers in cerebrospinal fluid and serum, saliva can also be studied as an assay substance. According to a study conducted by Park et al., real-time fluorescence quantitative PCR detected varicella-zoster virus DNA in saliva. This study involved 52 patients with HZ, simulated HZ patients (n = 30), and healthy college students (n = 27). Positive salivary VZV DNA was identified, leading to subsequent measurements of VZV-specific cellular-mediated immunity (CMI) or the persistence of salivary varicella-zoster virus DNA. This approach seeks to identify biomarkers predictive of PHN development. In a study involving saliva collection from 70 patients with HZ, the persistence of salivary VZV DNA was found to correlate with lower VZV-specific CMI responses at the onset of HZ. Moreover, weak VZV-specific CMI responses were associated with the subsequent development of PHN. This immunological sluggishness may permit uncontrolled viral replication during reactivation, prolonging viral activity and increasing the likelihood of PHN development [27] (Table 3).

Although progress has been made in identifying potential biomarkers through metabolomic, inflammatory biomarker analysis, and nucleic acid detection approaches, no highly sensitive and specific biomarker has yet been validated for widespread clinical use in the early diagnosis or risk prediction of VZV-related conditions. Future research should focus on larger, multicenter studies to validate these findings and incorporate multi-omics approaches, including genomics, proteomics, and metabolomics, to uncover key molecular mechanisms underlying VZV infection and reactivation. These efforts will pave the way for personalized prevention and treatment strategies.

Ocular herpes zoster

Herpes zoster ophthalmicus (HZO) is caused by the reactivation of VZV along the ophthalmic branch of the trigeminal nerve, accounting for approximately 10–20% of all HZ cases [30]. Clinically, HZO presents with corneal hyperesthesia, visual impairment, corneal stromal opacification, and in severe cases, corneal ulceration, iridocyclitis, and secondary glaucoma [31].

Nithyanandam et al. conducted a prospective longitudinal observational study to assess the prognostic visual acuity in HZO and the factors affecting the prognosis of the vision. This study involved 64 patients with HZO within 72 h of onset of rash who was followed up for 6 months, and the main outcome measure was best-corrected visual acuity. Patients underwent detailed ophthalmologic and dermatologic examinations at presentation and follow-ups at weeks 1, 2, and 4, and months 3 and 6. In this study, the peak of vision loss occurred at 1 - 2 weeks of follow-up, and 9 / 64 showed moderate to severe vision loss, and multifactorial analysis identified uveitis as the primary predictor of vision loss in HZO [28]. In a recent retrospective study by Niederer et al., the incidence of moderate and severe vision loss after ocular HZO was determined, and associated factors were identified. The study included 869 patients with acute HZ, and the primary outcome measure was the proportion of individuals with moderate and/or severe loss of vision following an acute episode of HZO, and secondary outcome measures included causes and factors associated with permanent loss of vision owing to HZO [29]. It revealed that approximately 1 in 10 individuals with HZO may experience moderate or severe vision loss, primarily due to corneal scarring. The study concluded that older age, immunosuppression, and uveitis were identified as factors associated with severe permanent vision loss secondary to HZO (Table 3).

Conclusion

VZV, a neurotropic alpha-herpesvirus, presents a dual clinical challenge: acute reactivation-associated disease and long-term complications such as PHN and vision loss. Recent biomarker-driven investigations have provided important insights into the immunopathogenesis and heterogeneity of VZV-related syndromes. Metabolomic profiling of cerebrospinal fluid in patients with HZ meningoencephalitis has revealed specific amino acid alterations linked to neuroinflammation and neuronal injury. Parallel studies have demonstrated that oxidative and inflammatory stress responses, as reflected by serum biomarkers such as IL-6, IL-18, homocysteine, and CRP, correlate with disease severity and may contribute to complications such as PHN. Salivary VZV DNA persistence and diminished virus-specific cellular immunity further support a model in which host immune failure underlies prolonged viral activity and neuropathic sequelae. In ocular herpes zoster, longitudinal cohort analyses have identified uveitis, age, and immunosuppression as independent predictors of permanent vision loss, highlighting the need for early prognostic stratification and aggressive ophthalmologic management. Despite these advances, no single biomarker has yet achieved sufficient sensitivity or specificity for routine clinical application. Future directions should prioritize multi-omics integration, longitudinal immune profiling, and validation across diverse cohorts to enable early risk prediction and the development of targeted therapeutic strategies in VZV-associated diseases.

EBV

EBV is a gamma herpesvirus, primarily infecting B cells and establishing lifelong latent infections. Over 95% of healthy adults globally harbor a latent EBV infection [32]. EBV is implicated in the etiology of several malignancies, including nasopharyngeal carcinoma, gastric cancer, Burkitt’s lymphoma, and Hodgkin’s lymphoma, with a marked geographic distribution in prevalence and incidence (Table 4).

Table 4.

Summary of cohort study information based on EBV clinical sample studies

Author	Disease	Biomarker	No. of controls	No. of patients	Age	Sample
Chan et al. [34]	Nasopharyngeal carcinoma	EBV DNA	-	-	40–62 years	Plasma
Lam et al. [35]	Nasopharyngeal carcinoma	EBV DNA	-	-	25–79 years	Plasma
Paudel et al. [36]	Nasopharyngeal carcinoma	EBNA1 IgA	SCHS Discovery: 20 SCHS Validation: 22 SCS Validation: 37	20 22 37	45–74 years 45–74 years 45–64 years	Sera
Coghill et al. [37]	Nasopharyngeal carcinoma	BXLF1, LF2, BZLF1, BRLF1, EAd, BGLF2, BPLF1, BFRF1, BORF1	Case-control Study: 175 Prospective CSP Cohort: 117 Prospective TFS Cohort: 77	175 37 26	-	Blood
Li et al. [38]	Nasopharyngeal carcinoma	P85-Ab	Biomarker identification: 39 Assay development: 221 Assay validation: 529	12 42 71	-	Serum
Jiang et al. [39]	Nasopharyngeal carcinoma	BART 2-5p	Discovery stage: 24 Validation Cohort 1: 118 Validation Cohort 2: 284 Validation Cohort 3: 88	24 148 103 22	58.7 ± 9.3 years 51.3 ± 14.5 years 49.0 ± 8.4 years 49.5 ± 8.3 years	Serum
Xing et al. [40]	Nasopharyngeal carcinoma	MIC-1	VCA-IgA-positive healthy donors: 72 Normal subjects with negative VCA-IgA: 219	190	-	Plasma
Zheng et al. [41]	Nasopharyngeal carcinoma	EBV DNA methylation	Discovery Cohort: 16 Validation Cohort 1: 379 Validation Cohort 2: 53	20 422 55	-	Saliva
Chen et al. [5]	Nasopharyngeal carcinoma	EBV DNA	1217	767	35–52 years	Plasma
Chan et al. [34]	Nasopharyngeal carcinoma	EBV DNA	-	-	≥ 18 years	Plasma
Tan et al. [43]	Nasopharyngeal carcinoma	BamHI-W 76 bp, VCA IgA, EA IgG, EBNA1 99 bp	106	187	-	Plasma
Qiu et al. [45]	Gastric cancer	EBV DNA	2620	140	Controls: 58.2 ± 11.8 years Patients: 55.5 ± 11.7 years	Plasma
Sundar et al. [46]	Gastric cancer	PD-L1, CD8A, GZMA, PRF1, PD-1	193	71	24–84 years	EBVaGC samples
Song et al. [47]	Gastric cancer	anti-LF2 IgG, anti-BORF2 IgG, anti-BALF2 IgG	Latvian Cohort: 34 Korea and Poland Cohort: 65	28 24	Mean age: 63 years Mean age: 57 years	Blood
Bai et al. [48]	Gastric cancer	CTLA- 4, TMB, SMARCA4	Training cohort: 100 Validation cohort: 53	24 23	-	Tissue samples
Guo et al. [49]	Gastric cancer	BMRF2	-	-	-	Stomach samples

Nasopharyngeal carcinoma

Nasopharyngeal carcinoma (NPC), characterized by local invasion and early distant metastasis, is a highly malignant tumor closely associated with EBV [33]. The prognosis for patients with metastatic NPC is poor, making early diagnosis critical for improving treatment outcomes and overall prognosis. Screening asymptomatic individuals for EBV DNA or antibodies is especially important in regions with a high incidence of NPC, as early detection can significantly enhance treatment efficacy and outcomes. Following diagnosis, stratifying patients based on their prognosis is crucial for developing personalized treatment strategies. Identifying reliable biomarkers for the early clinical diagnosis and prognosis of NPC is urgently needed, along with exploring novel and effective therapeutic approaches to improve patient outcomes.

Current screening methods for NPC are considered inadequate. Traditional screening methods for NPC include EBV serologic testing, nasopharyngeal swab EBV DNA testing, endoscopy, and magnetic resonance imaging, but these methods suffer from insufficient sensitivity, a high false positive rate, complexity, and high cost. Additional research is crucial to develop biomarkers with higher positive predictive values, which would enable earlier diagnosis and better identification of individuals at high risk. Blood samples remain the standard source for biomarker discovery in NPC diagnosis.

Plasma EBV DNA has emerged as an established biomarker for NPC. A prospective study by Chan et al. investigated the utility of EBV DNA in plasma samples for screening early-stage NPC in asymptomatic populations. The study analyzed EBV DNA in plasma samples from asymptomatic participants, with those initially testing positive retested after 4 weeks. Consistently positive individuals underwent nasal endoscopy and magnetic resonance imaging for NPC diagnosis. The sensitivity and specificity of EBV DNA in plasma samples for NPC screening were 97.1% (95.5–98.7) and 98.6% (98.6–98.7), respectively, indicating the effectiveness of EBV DNA analysis in screening early asymptomatic NPC [34]. Lam et al. conducted a study sequencing the abundance and size of EBV DNA molecules in plasma from patients with or without NPC, demonstrating the feasibility of screening for NPC by detecting plasma EBV DNA. The establishment of cutoff values in an exploratory dataset and subsequent testing in a validation sample set supported the efficacy of this approach [35]. Sequencing-based analyses revealed higher plasma EBV DNA levels and longer length of plasma viral fragment in NPC patients compared to non-NPC subjects. These findings suggest the potential development of a highly accurate blood test for NPC screening, enhancing positive predictions.

EBV antibodies have emerged as pivotal biomarkers for early diagnosis, extensively explored by Paudel et al. In this study, healthy human sera from a Singaporean cohort, later developing NPC, were screened for potential use as biomarkers for EBV antibodies. The findings were validated in an independent cohort in Shanghai, China. This serologic inquiry delineated specific EBV antibodies capable of distinguishing individuals at risk for developing NPC. Notably, IgA against EBV nuclear antigen 1 (EBNA1) exhibited 100% sensitivity and 100% specificity in differentiating NPC cases from matched controls up to four years in both the Singapore and Shanghai cohorts. This underscores the potential utility of IgA against EBNA1 as a robust biomarker for identifying individuals at risk for NPC [36]. Coghill et al. study complemented this research by seeking novel EBV antibodies for risk stratification in the early detection of NPC. They employed customized protein microarrays targeting 199 sequences of 86 EBV proteins, evaluating 607 subjects for anti-EBV IgG and IgA antibody responses [37]. Analyzing the differences in response patterns between NPC cases and controls, they developed an antibody-based risk score for predicting NPC. The risk prediction analyses identified specific antibody targets—BXLF1, LF2, BZLF1, BRLF1, EAd, BGLF2, BPLF1, BFRF1, and BORF1—that effectively differentiated NPC status. The resulting risk score demonstrated superior accuracy in predicting NPC in the general population of Taiwan compared to the currently used viral capsid antigen/EBNA1 IgA biomarker alone. Moreover, this EBV-based risk score enhanced the prediction of NPC in genetically high-risk families. In the pursuit of advancing NPC screening, Li et al. validated the performance of a novel biomarker, anti-BNLF2B total antibody P85-Ab. This was achieved through a large-scale prospective screening procedure, comparing it with the standard bis-antibody-based screening method (EBNA1-IgA and EBV-specific VCA-IgA) [38]. The results indicated that P85-Ab holds promise as a novel biomarker for NPC screening. P85-Ab showed higher sensitivity than the two-antibody method (97.9% vs. 72.3%; ratio, 1.4 [95% CI, 1.1 to 1.6]), higher specificity (98.3% vs. 97.0%; ratio, 1.01 [95% CI, 1.01 to 1.02]), and a higher positive predictive value (10.0% vs. 0.4.3%; ratio, 2.3 [95% CI, 1.8 to 2.8]. This suggests a significant stride in enhancing the accuracy and efficacy of NPC screening procedures.

Jiang et al. and Xing et al. investigated miRNAs and cytokines, respectively, as potential biomarkers. Jiang et al. quantified 17 BART microRNAs encoded by the EBV BamHI region, focusing on BART microRNA cycling, particularly BART 2-5p (with an area under the curve higher than 0.8), using previous microarray and sequencing data. During the validation phase, the sensitivity, specificity and AUC of BART 2-5p was 93.9%, 89.8%, 0.972 (95%CI: 0.954–0.989), respectively [39]. Analysis of EBV BART in serum samples from NPC patients revealed the potential of BART 2-5p as an early diagnostic biomarker, with circulating levels rising even before clinical diagnosis. Xing et al. employed ELISA to analyze plasma macrophage inhibitory cytokine-1 (MIC-1) levels in 190 NPC patients, 72 VCA-IgA-positive healthy donors (VP), and 219 VCA-IgA-negative normal subjects (VN), aiming to assess the diagnostic value of plasma MIC-1 in differentiating NPC patients and explore its complementary role with widely used EBV-associated markers, specifically the EBV capsid antigen-specific IgA [40]. The results demonstrated significantly higher plasma MIC-1 levels in NPC patients compared to VN and VP patients. MIC-1 was found to complement VCA-IgA titer and EBV DNA copy number detection in NPC testing, improving the identification of EBV DNA-negative NPC patients and differentiating between NPC and VCA-IgA-positive healthy controls.

Saliva samples, offering a practical advantage over blood samples in terms of ease of collection, were explored by Zheng et al. Using 987 saliva samples, they obtained the methylation profile of EBV DNA through capture sequencing to detect differences in EBV DNA methylation for the diagnosis of NPC [41]. Zheng et al. found that significantly elevated levels of DNA methylation in patients with NPC compared to healthy individuals in two cohorts of saliva samples. These results were fully validated, highlighting the potential application of this method in the early detection of NPC.

EBV DNA and EBV antibodies, among others, may also serve as potential prognostic markers. Chen et al. identified quantitative measurement of plasma EBV DNA by real-time PCR as a robust prognostic marker for patients with NPC in 1984 patients with non-disseminated NPC. Blood samples collected within 3 months of the end of radiotherapy and thereafter every 3 to 12 months underwent cell free EBV (cfEBV) DNA analysis. Patient follow-up continued until disease recurrence was detected or a median time of 60 months. The investigation aimed to assess the diagnostic efficacy of clinical tests, based on conventional surveillance modalities (imaging scans and pathologic examinations), for disease recurrence, calculating sensitivity, specificity, and accuracy [5]. During the follow-up period, patients with detectable cfEBV DNA exhibited a significantly higher recurrence rate (63.8%) compared to patients with undetectable cfEBV DNA (8.6%). The sensitivity, specificity, and accuracy of cfEBV DNA for local recurrence were 68.8%, 80.0%, and 78.2%, respectively; for regional recurrence, 80.2%, 80.0%, and 85.9%, respectively; and for distant metastasis, 91.1%, 80.0%, and 92.8%, respectively. Notably, the sensitivity of cfEBV DNA for detecting extrapulmonary metastasis was higher (94.9-96.5%) compared to that of lung metastasis (78.4%). These results suggest that plasma cfEBV DNA in patients with NPC serves as an early indication of tumor recurrence, especially extrapulmonary metastasis. However, up to 40% of patients experiencing disease recurrence may have undetectable plasma EBV DNA after treatment. Chan et al. collected plasma samples from 769 patients with stage IIB-IVB NPC 6–8 weeks after radiotherapy for sequencing-based quantification and size analysis of plasma EBV DNA to explore whether this approach could more accurately predict the prognosis of NPC patients. PCR-based analysis predicted local and distant recurrence with a sensitivity of 42.3% and 85.3%, respectively. In comparison, sequencing-based analysis demonstrated a significant improvement in predicting local and distant recurrence with sensitivities of 88.5% and 97.1%, respectively, compared to traditional PCR methods [42]. Tan et al. systematically evaluated six identified biomarkers in NPC cases, including two EBV DNAs (BamHI-W 76 bp and EBNA1 99 bp) and four anti-EBV antibodies (early antigen [EA] IgA, EA IgG, EBNA-1 IgA, and VCA IgA), as well as four new biomarkers, including one EBV DNA (BamHI-W 121 bp) and three miRNAs (ebv-miR -BART7-3p, hsa-miR-29a-3p, and hsa-miR-103a-3p) [43]. BamHI-W 76 bp remained the most sensitive plasma biomarker, and the combination of BamHI-W 76 bp with VCA IgA or EA IgG showed potential for improving specificity or sensitivity in detecting NPC. EBNA1 99 bp was identified as capable of discerning NPC patients with poor prognosis in both early and advanced stages of the disease.

In these studies, NPC screening has garnered significant attention. Traditional methods, such as imaging and biopsies, often fall short in early detection, prompting researchers to seek sensitive and specific biomarkers. One of the most promising biomarkers is circulating EBV DNA in plasma, which has demonstrated sensitivities of 97.1% and specificities of 98.6% in asymptomatic individuals. It is also the most extensively studied biomarker. Additionally, antibodies against EBNA1 show promise, with 100% sensitivity and specificity in predicting NPC up to four years before diagnosis [36].

Emerging biomarkers, including microRNAs and cytokines like MIC-1, are also being explored for their diagnostic potential [39]. Furthermore, studies on prognostic biomarkers have highlighted their role in risk stratification, helping to identify patients at higher risk for recurrence and metastasis. Saliva samples have revealed DNA methylation patterns that can distinguish NPC patients from healthy individuals, offering a non-invasive screening method. Overall, integrating various biomarkers—such as EBV DNA, antibodies, microRNAs, and methylation patterns—could significantly enhance diagnostic accuracy and facilitate early intervention, ultimately improving patient outcomes in NPC management (Table 4).

Gastric cancer

EBV-associated gastric cancer (GC), denoted as EBVaGC, constitutes approximately 5–10% of global GC cases. This subtype possesses distinct clinicopathologic and molecular characteristics, representing a unique molecular entity within the spectrum of GC. Notably, EBVaGC exhibits a lower overall mortality rate, a higher incidence in men compared to women, and a notable increase in intratumor or peritumoral immune cell infiltration in comparison to EBV-negative GC. Despite being a prevalent infection in over 95% of adults, the full impact of EBV positivity on GC remains incompletely understood [44].

Current biomarkers of EBVaGC have primarily concentrated on prognosis-related indicators, encompassing treatment outcomes and recurrence. Qiu et al. conducted a study involving 2760 GC patients from Sun Yat-sen University Cancer Center. Immunohistochemical examination for Epstein-Barr encoding region (EBER) was performed, and EBV-DNA load was assessed in plasma and tissue samples of EBVaGC patients at baseline [45]. Subsequent dynamic monitoring of plasma EBV-DNA load in EBVaGC patients revealed its predictive value for disease recurrence and poor response to chemotherapy. This underscores the significance of dynamic monitoring of plasma EBV-DNA load as an easily accessible biomarker for monitoring EBVaGC.

Similarly, Sundar et al. analyzed the transcripts of immune genes associated with EBVaGC and explored their correlation with clinical prognosis. Transcriptomic analysis through immunohistochemistry was conducted on EBVaGC samples from surgically resected primary tumors and control subjects. Examined genes included PD-L1 and other immune-related genes associated with cytolytic activity, cytokines, and immune checkpoints within the tumor, such as CD8A, GZMA, PRF1 and PD-1 [46]. Compared to controls, EBVaGC demonstrated elevated expression of all examined immune genes (p < 0.01). PD-L1 immunohistochemical expression aligned with PD-L1 transcript expression, and the pattern of tumor-infiltrating lymphocytes varied between low PD-L1 and high PD-L1 groups. Notably, a substantial proportion of EBVaGC exhibited low expression of PD-L1 and other immune genes, corresponding to lower immunohistochemistry scores and a poorer prognosis.

Specific anti-EBV antibodies play a crucial role in clinical diagnosis, epidemiological studies, and immune-based precision therapy for EBV-positive GC. Song et al. conducted an immunoassay on plasma samples from 28 EBV-positive and 34 EBV-negative Latvian patients with GC, assaying 85 EBV proteins using a polymicrobial nucleic acid programmable protein array (EBV-nappa). Through this approach, they identified EBaGC-specific antibody responses, offering a potential noninvasive method for detecting EBV-positive GC and shedding light on its role in carcinogenesis. Nine IgG antibodies demonstrated the capability to distinguish tumor EBV status, with seven of them subsequently tested by ELISA. The top three antibodies exhibited an area under the operating characteristic curve ranging from 0.81 to 0.85 for distinguishing tumor EBV status [47]. Notably, the EBV-associated GC-specific humoral response is specific to the cleavage cycle, particularly the early-stage antigens, in contrast to humoral responses observed in other EBV-associated malignancies such as NPC and lymphoma, which are specific to the late-stage antigens. This distinction is essential in understanding the unique nature of EBV-associated GC compared to other malignancies.

The precise impact of EBV infection on the efficacy of immune checkpoint blockade (ICB) in GC, as well as the underlying mechanisms, remains unclear. Bai et al. addressed this gap by establishing a next-generation sequencing (NGS)-based EBV detection algorithm and validating it in two independent GC cohorts. Their study aimed to investigate the efficacy and potential biomarkers of immune checkpoint blockade in EBVaGC identified through NGS. EBVaGC patients with elevated cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) levels demonstrated poor responses to programmed death 1/ligand 1 (PD-1/L1) monotherapy, as well as anti-CTLA-combination therapies. Significant differences were observed in tumor mutational load (TMB) levels and SMARCA4 mutation frequency between the ICB-responsive group and the non-responsive group. Ultimately, CTLA-4, TMB, and SMARCA4 mutations were identified as potential predictive biomarkers for ICB efficacy in EBVaGC, providing valuable insights for optimizing ICB therapy for EBV-associated GC [48].

Guo et al. investigated the prevalence, characteristics, and EBV mRNA profile of EBVaGC in Gansu using clinical samples. They collected and analyzed 270 stomach samples from patients with GC to assess the presence of EBV DNA and EBER through nested PCR and in situ hybridization, respectively. The study’s results revealed a relatively low prevalence of EBVaGC in Gansu Province, Northwest China (6.7%). EBVaGC in this region exhibits unique clinicopathological features, including older age, fewer lymph node metastases, and no discernible gender bias. This emphasizes that geographic differences manifest distinct characteristics in the relationship between EBV and GC. Notably, PCR detected a higher rate of EBV infection compared to EBER-ISH, suggesting that some EBV infections, overlooked by traditional detection methods, may express unique transcripts, playing a significant role in the pathogenesis of GC [49].

Recent research on EBVaGC has identified several promising biomarkers, with a strong focus on plasma EBV-DNA load, immune checkpoint markers, and EBV-specific antibodies. Among these, plasma EBV-DNA load has garnered significant attention due to its predictive value for disease recurrence and chemotherapy response, making it a highly accessible biomarker for clinical monitoring. Additionally, immune gene markers, particularly PD-L1, have been extensively studied for their role in tumor immunology, with elevated PD-L1 expression correlating with prognosis and immune checkpoint therapy outcomes. Research also underscores the importance of EBV-specific antibodies for non-invasive diagnosis and distinguishing EBV-positive GC. Notably, antibodies targeting EBV early-stage antigens show strong specificity, differentiating EBVaGC from other EBV-related malignancies. Finally, CTLA-4, TMB, and SMARCA4 mutations have emerged as key biomarkers for predicting the efficacy of immune checkpoint blockade therapy. In summary, these biomarkers hold significant potential for improving diagnosis, prognostication, and personalized treatment strategies in EBVaGC [48] (Table 4).

Many laboratories utilize the detection of IgG and IgM antibodies against the EBNA-1 and VCA antigens to differentiate between past and current EBV infections [50]. The presence of IgG antibodies against EBNA-1 typically indicates a resolved infection, as these antibodies persist for life following initial exposure to the virus. In contrast, IgM antibodies against VCA are indicative of a recent or acute infection, usually present during the early phase of the disease. The combination of these serological markers is essential for accurately determining the stage of EBV infection, which plays a crucial role in clinical management and understanding the epidemiological dynamics of the virus.

Conlusion

EBV, a ubiquitous gamma-herpesvirus with strong oncogenic potential, is implicated in multiple malignancies including NPC and EBVaGC. Over the past decade, substantial advances have been made in the identification of diagnostic and prognostic biomarkers that enhance early detection and personalized clinical management. In NPC, circulating EBV DNA remains the most established biomarker, supported by complementary serological markers such as EBNA1 IgA and VCA IgA. Recent studies further highlight the value of antibody-based risk scores, BART microRNAs, plasma cytokines (e.g., MIC-1), and EBV DNA methylation profiles—particularly in non-invasive samples like saliva—for refined risk stratification and screening.

In EBVaGC, plasma EBV DNA dynamics, immune checkpoint markers (e.g., PD-L1), and EBV-specific antibodies have emerged as key prognostic tools. In addition, genomic and transcriptomic features—including CTLA-4 expression, TMB, and SMARCA4 mutations—offer promising predictive markers for immunotherapy response. Importantly, differences in humoral response profiles across EBV-driven cancers underscore the need for disease-specific biomarker strategies. Moreover, standard serological markers such as EBNA1 IgG and VCA IgM remain crucial for staging primary versus latent infection, with implications for epidemiological monitoring and clinical decision-making.

Collectively, these biomarker advancements not only improve early diagnosis and prognosis of EBV-associated malignancies but also provide a framework for individualized therapeutic approaches. Continued integration of serological, molecular, and multi-omics data will be essential for translating biomarker discovery into precision oncology for EBV-related cancers.

HCMV

HCMV infects the majority of the global population, with seroprevalence rates ranging from 60 to 90% worldwide [49, 51]. Although typically asymptomatic in immunocompetent individuals, HCMV represents a significant opportunistic infection risk in fetuses and organ transplant recipients. Infected infants can develop congenital diseases, while in transplant recipients, HCMV is associated with immune rejection, leading to organ damage. These outcomes pose serious health risks to both fetuses and immunocompromised individuals (Table 5).

Table 5.

Summary of cohort study information based on HCMV clinical samples studies

Author	Disease	Biomarker	No.of controls	No.of patients	Age	Sample
Vorontsov et al. [52]	Congenital CMV related fetal brain injury	Chemerin, Gal-3BP	26	17	Fetus	Amniotic fluid
Ouellette et al. [54]	Sensorineural hearing loss	16-gene classifier signature: CD40, MYST2, LOC286135, JMJD2A, RABGAP1, RAB9B, AK3L1, MATR3, ARHGEF9, C10orf59, LOC645431, MPDU1, PAXIP1, CLEC4G, GLCCI1, LEO1	31	49	11–24 days	Blood
Fourgeaud et al. [60]	Sensorineural hearing loss	CMV DNA	-	-	7–26 days	Saliva
Shlonsky et al. [55]	Sensorineural hearing loss	CMV DNA	-	-	0–28 days	Saliva
Lilleri et al. [56]	Congenital cytomegalovirus infection	CMV DNA	-	-	0–28 days	Saliva
Sigdel et al. [58]	HCMV infection in transplantation	SERPINA12, CP, complement activation, proteins enriched in the humoral and innateimmune responses	31	31	CMV DNAemia: 22–77 years No CMV DNAemia: 30–74 years	Plasma
Camargo et al. [59]	HCMV infection in transplantation	Nonprotective signature (NPS; IL-2-IF-γ + TNF-α-MIP-1β+) Protective signature (PS; IL-2+ IF-γ + TNF-α + MIP-1β+)	19	Spontaneous Controllers: 16 Noncontrollers: 21	48–63 years	Blood
Ahn et al. [61]	HCMV infection in transplantation	Increase in pathways forinterferon signaling and cytotoxic T cell function	31	31	CMV DNAemia median: 55 years No CMV DNAemia median: 53 years	Blood
Vietzen et al. [62]	HCMV infection in transplantation	gB-specific Abs	114	HCMV-seropositive (R+): 35 Seronegative recipients of positive organs (D+/R-): 28	18–66 years	Plasma
Zamora et al. [63]	HCMV infection in transplantation	-	33	28	2–53 years	Serum

Congenital cytomegalovirus infection

HCMV is highly prevalent in the population but typically remains asymptomatic in healthy individuals. However, maternal infection during pregnancy can lead to congenital cytomegalovirus (cCMV) through vertical transmission. cCMV is one of the most common congenital infections globally and is associated with severe outcomes, including stillbirth and long-term sequelae in surviving infants. These sequelae may include brain damage, neurodevelopmental disorders, and sensorineural hearing loss (SNHL). Early diagnosis and appropriate monitoring are essential to mitigate these risks, highlighting the importance of preventive measures and awareness during pregnancy.

Prenatal biomarkers for the severity of cCMV infection are currently lacking. Vorontsov et al. conducted a whole proteomic analysis of mid-gestation amniotic fluid samples, comparing amniotic fluid from fetuses with severe cCMV to that of fetuses with asymptomatic cCMV infection. This search for differential proteins aimed to serve as prognostic biomarkers for determining cCMV-associated fetal brain injury. In this cohort, proteins associated with inflammatory and neurologic disease pathways were identified. Two of these proteins, the immunomodulatory proteins retinoic acid receptor marker (chemerin) and galactose lectin 3-binding protein (Gal-3BP), were highly predictive of cCMV severity in an independent validation cohort. They effectively distinguished between severe (n = 17) and asymptomatic (n = 26) fetal cCMV, with 100%-93.8% positive predictive value, and 92.9%-92.6% negative predictive value (for chemerin and Gal-3BP, respectively). The results suggest that these two proteins can guide early prognostic stratification and individualized treatment of cCMV-infected fetuses. Additionally, they offer valuable insights into inflammatory pathways and therapeutic targets involved in the progression of CMV-associated fetal brain injury [52].

CMV infections stands out as the primary contributor to SNHL among children [53]. In the investigation of SNHL caused by cCMV, Ouellette et al. assessed the blood transcriptional profiles of 80 infants with cCMV (comprising 49 symptomatic and 31 asymptomatic cases). Over 3 years, they monitored the emergence of late-onset SNHL, aiming to discern disparities in gene expression profiles between symptomatic and asymptomatic cCMV-infected infants and their correlation with late-onset SNHL. Their endeavor was to delineate variations in gene expression profiles between infants presenting with symptomatic and asymptomatic cCMV infection, exploring their connection with delayed SNHL [54]. By investigating transcriptional profiles and subsequently validating in a separate cohort, Ouellette et al. identified a set of 16 genes related to the development of delayed-onset SNHL, achieving 92% accuracy, which underscores the utility of these genes as biomarkers for hearing loss in cCMV infection.

Current methods for diagnosing cCMV in newborns include targeted testing and universal screening. Targeted screening involves testing newborns identified as having risk factors for cCMV. Fourgeaud et al. assessed the feasibility and efficacy of targeted cCMV screening in newborns who failed hearing tests in France. Saliva samples from these infants were analyzed using PCR, and hearing loss was confirmed by an otolaryngologist [60]. The use of PCR on saliva samples successfully confirmed CMV infection, demonstrating the practicality of targeted cCMV screening. This approach enables the early initiation of antiviral therapy within the first month of life for newborns with cCMV and hearing loss, thereby improving the prognosis. While universal screening, which involves testing all newborns, increases the detection of asymptomatic cCMV infections, it also significantly raises healthcare costs. Therefore, more refined and cost-effective screening methods are urgently needed to meet clinical demands. Shlonsky et al. assessed the utility of real-time PCR for screening newborns with cCMV using mixed saliva, comparing CMV DNA detection in individual and mixed saliva samples [55]. The study collected 1,000 newborn saliva specimens, with real-time PCR used to analyze 100 sample pools, each containing 10 samples, and 1,000 individual samples for CMV DNA. Urine CMV PCR was then performed to validate the diagnosis post a positive saliva CMV DNA test. Results indicated high specificity for both mixed and individual saliva sampling (99.9% and 98.1%, respectively). Despite a positive predictive value of 85.7% for mixed samples and 24.0% for individual saliva samples, neonatal saliva collection emerges as a reliable method for identifying asymptomatic cCMV infections. Combining samples for testing not only enhances laboratory workflow but also curtails costs.

Lilleri et al. conducted a prospective study to explore the incidence, outcome, and risk factors associated with cCMV infection in newborns born to mothers with preconception immunity. Additionally, they assessed the potential necessity and efficacy of health recommendations in this population. The results revealed a high incidence of cCMV infection in neonates born to mothers with preconception immunity in Northern Italy. The prevalence of cCMV in neonates/fetuses born to mothers with preconception immunity was 0.19. Diagnosis of cCMV involved saliva screening, followed by confirmation through urine (or dried blood spot) testing. Notably, 28 out of 45 salivary HCMV DNA-positive neonates (62%) did not exhibit pre-existing infections. This highlights the possibility of a false-positive test for cCMV with a saliva-only test, emphasizing the importance of using it as a preliminary screening with subsequent confirmation in a urine sample. The risk of cCMV infection in fetuses of immunized mothers was at least 10-fold lower than that observed in previous studies involving seronegative mothers [56]. This supports the protective role of maternal preconception immunization in preventing cCMV and underscores the potential effectiveness of HCMV vaccination strategies. Current vaccine development strategies for CMV include mRNA vaccines, which aim to elicit strong immune responses against CMV envelope glycoproteins, and subunit vaccines that focus on purified viral proteins to stimulate immunity. The integration of new biomarkers, such as viral DNA load and specific immune response markers, is essential for enhancing diagnostic accuracy and monitoring vaccine efficacy, ultimately informing new health recommendations to protect pregnant women and their infants from congenital CMV infection.

Biomarker research on cCMV infection has focused on identifying predictors for severity, hearing loss, and the potential for therapeutic interventions. Notably, chemerin and Gal-3BP have emerged as promising biomarkers for stratifying cCMV severity. These proteins, associated with inflammatory and neurologic pathways, demonstrated high predictive accuracy (93.8-100%) in distinguishing severe from asymptomatic cCMV cases, offering the potential for early prognostic stratification. Gene expression profiles have also shown great utility, particularly in identifying 16 genes related to delayed-onset SNHL in cCMV-infected infants, achieving 92% accuracy. Moreover, targeted screening methods, including saliva CMV PCR, have demonstrated effective in diagnosing cCMV in newborns with hearing loss, enabling timely antiviral treatment initiation. However, universal screening remains costly, highlighting the need for more efficient, cost-effective diagnostic approaches. Studies also underline the protective role of preconception maternal immunity, which reduces cCMV risk in neonates, suggesting the potential benefits of HCMV vaccination (Table 5).

HCMV infection in transplantation

Reactivation of the virus occurs in individuals during periods of decreased immune function, such as after transplantation, including both SOT and HSCT, when immunosuppression is prolonged. In contrast, CMV infection, whether de novo or reactivated after allogeneic transplantation, is thought to induce broad immune effects, which include an increased susceptibility to graft rejection, significantly contributing to chronic graft injury and reduced graft survival [57].

To further comprehend the evolution and pathogenesis of CMV infection in immunocompromised hosts, Sigdel et al. conducted a proteomic study involving 168 plasma samples from 62 propensity score-matched kidney transplant recipients (31 patients with CMV DNAemia, 31 patients without CMV DNAemia) [58]. Blood samples were collected at regimen times of 3 and 12 months post-transplantation. The samples were segregated according to the proteomic profile of their CMV DNA status. A subset of 17 plasma proteins affecting humoral and innate immune pathways was identified, predicting CMV onset at 3 months post-transplant. Additionally, Camargo et al. conducted an in-depth phenotyping of CMV-specific T-cells to predict CMV after allogeneic HSCT [59]. Using flow cytometry, they investigated in vitro CD81 T-cell cytokine production against the CMV-pp65 peptide in three clinically distinct subgroups of seropositive HSCT patients. The study revealed that two CMV-specific CD81 T-cell functional subsets were strongly associated with CMV infection risk. CMV-specific CD81 T-cell cytokine profiles demonstrated a robust predictive value for the risk of CMV reactivation, offering valuable insights for guiding clinical decision-making in HSCT recipients.

Ahn et al. assessed the impact of CMV infection on whole-blood leukocyte gene expression over time by analyzing blood samples collected at multiple time points over 12 months post-transplant. The study focused on a matched cohort of 62 kidney transplant recipients, comparing those with or without CMV DNAemia. The investigation aimed to elucidate the mechanisms underlying patients’ susceptibility to CMV reactivation and the increased risk of rejection in transplant recipients [61]. The analysis of a time-series gene set comprising differentially expressed genes unveiled the dynamics of genes and pathways involved in the immune response to CMV DNAemia in kidney transplant patients. At long-term time points, hundreds of genes exhibited differential expression, including genes enriched for pathways crucial for macrophage, interferon, and IL-8 signaling. Notably, there were significant time trends indicating increased interferon signaling and cytotoxic T cell function pathways. Understanding the transcriptional changes induced by CMV DNAemia may provide insights into the mechanism underlying the elevated risk of rejection in transplant recipients and suggest protective strategies against the negative immune effects of CMV.

In the context of lung transplant recipients (LTRs), Vietzen et al. investigated the impact of HCMV infections. The study analyzed antibody responses against HCMV glycoprotein B (gB) and the pentameric complex (PC), along with antibody-dependent cellular cytotoxicity (ADCC) responses in HCMV-seropositive (R⁺) LTRs and seronegative recipients of positive organs(D⁺/R^-). Within one year post-transplantation, significantly higher levels of gB-specific antibodies were observed in R⁺ LTRs compared to controls. Additionally, R⁺ patients exhibited higher levels of ADCC, as measured by FEA and CD107. The study also identified HCMV-specific antibodies in 23 D ⁺ R^- patients, suggesting that transplantation triggers robust ADCC, particularly driven by gB-specific antibodies. To assess the potential reduction of CMV infections after HSCT, Zamora et al. reanalyzed randomized controlled specimens for intravenous immunoglobulin (IVIG) prophylaxis of CMV from a previous study [63]. This reanalysis focused on PC-entry neutralizing antibodies (nAb) and quantitative CMV in 61 CMV donor-positive/receiver-negative (D⁺/R^-) HSCT patients (33 controls, 28 with CMV IVIG) [62]. The results showed higher neutralizing antibody titers in patients receiving CMV IVIG, correlating with lower infection rates. This finding tentatively supports the notion that CMV IVIG prophylaxis may modestly enhance PC-entry nAB activity in D⁺/R^- HSCT recipients.

CMV reactivation is a critical concern in transplant recipients, increasing the risk of graft rejection and chronic injury. Key biomarkers with clinical value include plasma proteins, which predict CMV onset post-transplantation, and CMV-specific CD8⁺ T-cell cytokine profiles, which reliably forecast reactivation risk in HSCT patients. These immune markers are essential for guiding early intervention and personalized treatment. Additionally, gB-specific antibodies and ADCC responses in lung transplant recipients indicate robust immune responses that could help control CMV infections. Prophylactic use of CMV IVIG in HSCT recipients has shown promise, with higher neutralizing antibody levels related to reduced infection rates, suggesting potential for improving CMV management in high-risk patients.

The application of IVIG in the treatment of cCMV infections in pregnant women is of great significance. Kagan et al. found that IVIG treatment can significantly reduce the risk of mother-to-child transmission when started in a timely manner and with an appropriate dosage during the early stage of primary HCMV infection [64]. Specifically, the mother-to-child transmission rate in this study was 6.5%, while the transmission rate in the historical data of the untreated control group was as high as 35.2%. The study also found that the level of the anti-IgM index is an important indicator for predicting mother-to-child transmission, indicating that the intensity of the immune response in the early stage of infection may affect virus transmission. These results suggest that IVIG, as a therapeutic means, has potential application value in the early stage of HCMV infection, especially in reducing mother-to-child transmission.

In the case of congenital infection in pregnant women, the use and efficacy of IVIG is an important research area. The quality of HCMV-specific antibodies, especially the binding capacity of high-affinity IgG, is closely related to the reduction of the risk of congenital HCMV infection [65]. However, in some cases, the use of IVIG may not fully enhance these key protective antibody responses. Therefore, further research is needed to explore how to optimize IVIG or other immunotherapies to enhance these protective antibody responses and more effectively prevent cCMV infection (Table 5).

HCMV infection in AIDS

The immunosuppressed state of HIV/AIDS patients poses a significant clinical concern, and HCMV is closely associated with a variety of complications in these individuals. The immune system damage caused by HIV infection facilitates CMV replication, which can lead to various diseases, including retinitis and gastrointestinal disorders that severely impact patients’ quality of life and prognosis. Research into the mechanisms by which HCMV regulates the cell cycle provides new insights into the survival strategies of HCMV in immunosuppressed environments [66].

In the study by Zhao et al., researchers explored the relationship between the detection of HCMV DNA in urine and the occurrence of end-organ diseases (EODs) in stage 2/3 HIV-1-infected patients. The study found that the presence of HCMV DNA in urine was significantly associated with an increased incidence of pulmonary and cardiovascular EODs. Specifically, patients with positive HCMV DNA in urine had an adjusted hazard ratio of 1.939 for developing pulmonary EODs. Furthermore, longitudinal analysis indicated that anti-HCMV treatment reduced the incidence of these EODs among patients. These findings suggest that early detection of HCMV DNA could aid in predicting and managing EODs in HIV-1-infected individuals, warranting its implementation as a routine test [67]. Additionally, a recent study by Chatterjee et al. investigated HCMV-induced EODs in HIV-infected individuals and identified specific immunological markers that can differentiate between HCMV-induced retinitis and gastroenteritis. Through serum ELISA and quantitative HCMV DNA testing, the researchers observed significant differences in the expression of the CXCL9, 10, and 11-CXCR3 chemokine pathways. Furthermore, the gL gene sequences of HCMV exhibited distinct phylogenetic branches, indicating adaptive genetic variation in different tissues. These results underscore the potential of immunological markers for identifying various HCMV-related diseases in HIV-infected patients [68].

Glioblastoma multiforme

The relationship between HCMV and glioblastoma multiforme (GBM) is an area of significant interest. Studies have demonstrated that HCMV can replicate within GBM tumor cells and may influence their self-renewal capabilities. The biological characteristics of HCMV in GBM suggest its potential as a biomarker for tumor development [66].

HCMV is considered a “co-carcinogenic” virus rather than a direct carcinogen, as it affects the malignant characteristics of cells through various mechanisms, including the regulation of proliferation, angiogenesis, anti-apoptosis, and evasion of immune surveillance. Consequently, HCMV could function as a valuable biomarker for GBM in studies of tumor biology. However, the choice of detection methods for HCMV significantly impacts the reliability of its use as a biomarker. Immunohistochemistry (IHC) technology demonstrates high sensitivity in detecting HCMV proteins, while PCR methods often face challenges in identifying HCMV nucleic acids. Future research should focus on optimizing detection methods to enhance our understanding of HCMV’s role in GBM and explore its potential as a therapeutic target [69].

Conclusion

HCMV infection presents a substantial health burden in both congenital and immunocompromised settings. In cCMV, early biomarkers such as chemerin, Gal-3BP, and gene expression signatures have shown strong predictive value for disease severity and delayed-onset SNHL, offering promising avenues for early risk stratification and individualized intervention. Saliva-based PCR diagnostics and maternal immunity studies further emphasize the clinical utility of non-invasive screening and the potential of maternal immunization or IVIG prophylaxis to reduce vertical transmission.

In transplant recipients, HCMV reactivation is a leading cause of graft dysfunction and mortality. Proteomic and immunologic biomarkers—including plasma immune proteins and CMV-specific CD8⁺ T cell cytokine profiles—enable preemptive identification of high-risk individuals and inform personalized immunotherapeutic strategies. Moreover, gB-specific antibody and ADCC responses in lung transplant recipients, along with enhanced PC-entry neutralizing activity via CMV IVIG, highlight adaptive immune responses that may be harnessed for protection.

Among HIV/AIDS patients, HCMV remains a major co-pathogen, with urine CMV DNA serving as a valuable predictor of EODs. Distinct chemokine signatures and viral genomic variation suggest tissue-specific tropism and disease phenotypes, underlining the need for precision diagnostics. In GBM, HCMV is increasingly recognized as a potential oncomodulatory virus. Its protein expression within tumor tissue and roles in immune evasion and tumor progression position it as both a biomarker and a therapeutic target, although methodological limitations in detection warrant further standardization.

Together, these findings underscore the multifaceted clinical relevance of HCMV and highlight the growing importance of integrating molecular biomarkers into diagnostics, prognostication, and treatment planning across diverse patient populations.

HHV-6, 7

HHV-6 is a member of the β-herpesvirus family that replicates in activated CD4⁺ T lymphocytes. There are two subtypes, HHV-6A and HHV-6B, with the majority of confirmed primary infections and reactivation events caused by HHV-6B. Limited information is available regarding the epidemiology and clinical impact of HHV-6A. HHV-6 reactivation syndrome occurs in approximately 50% of HSCT recipients and 20-62% of SOT recipients [70]. HHV-7 also belongs to the β-herpesvirus family and shares genetic homology with HHV-6 and CMV. HHV-7 infection is common in children, with most acquiring it around 2 years of age. Approximately 75% of children are seropositive by the age of 5 years, and the prevalence of HHV-7 antibodies increases with age to 80-98% in adults, with passively acquired antibodies present in almost all newborns [71].

Transplant patient

HHV-6 can cause severe diseases, especially in immunocompromised individuals such as HSCT and SOT recipients, as well as patients with AIDS [72]. While the opportunistic pathogenic role of HHV-6 is well established, the role of HHV-7 remains less clearly defined. Affected areas include the CNS, bone marrow, lungs, gastrointestinal tract, skin, and liver. In HSCT recipients, HHV-6 can cause both subcutaneous effusions, a benign condition linked to primary infection, and severe encephalitis associated with viral reactivation. Diagnostic methods, primarily based on serology and direct antigen detection, have limitations, with viral DNA quantification in the blood being the most prominent technique. Many questions remain regarding HHV-6A, HHV-6B, and HHV-7, particularly concerning their clinical implications and treatment options in immunocompromised patients [72].

In 1993, researchers identified a unique feature of HHV6: the integration of HHV-6A and HHV-6B genomes into the human chromosome (ciHHV-6) [73]. However, this phenomenon, commonly observed in transplant settings, can lead to clinical confusion, as it may be misinterpreted as an active infection, potentially resulting in unnecessary and harmful treatments. The initial detection of ciHHV-6 in transplant patients was documented in a study involving 205 transplant recipients (both SOT and HSCT), revealing a ciHHV-6 prevalence of 0.9% [74]. Additionally, a study conducted in the UK that assessed 500 blood donors for HHV-6 viral load found a high viral copy number in the 0.8% donors and a low viral load prevalence in the 1.4% donors [75]. In another study, Sang-Oh Lee et al. performed a quantitative HHV-6 PCR assay on whole blood samples from 47 renal transplant recipients, identifying only one case of ciHHV-6 within the cohort [76].

Berzero et al. conducted a retrospective study encompassing the patient cohort that exhibited positive results for HHV-6 DNA in CSF and/or blood within the Fondazione IRCCS Policlinico San Matteo Virology Database from January 2008 to September 2018. The evaluation focused on identifying the number of patients meeting the criteria for HHV-6 encephalitis in both immunocompetent and immunodeficient groups. In the immunodeficient group, 45 cases were ultimately identified, comprising 30 children and adolescents (age < 18 years) (30/45, 67%) and 15 adults (age ≥ 18 years) (15/45, 33%). The study further assessed the number of patients meeting the criteria for febrile convulsions/encephalitis in primary HHV-6 infection and those meeting the criteria for HHV-6 encephalitis resulting from viral reactivation in both the immunocompetent and immunodeficient groups. Hair follicle testing was conducted to confirm HHV-6 integration into human chromosomes among patients exhibiting positive results for HHV-6 DNA in whole blood and CSF, with blood HHV-6 viral loads exceeding 106 copies/mL. Notably, 11 patients with positive hair follicle HHV-6 DNA tests were identified as having ciHHV-6. The findings underscore several key observations: (i) within immunocompetent hosts, CNS infection with HHV-6 is confined to infants and children under 3 years of age during their initial infection; (ii) recipients of HSCT may develop HHV-6 encephalitis due to viral reactivation, potentially lacking typical clinical imaging features of limbic lobe encephalitis, especially when cranial MRI is performed soon after symptomatic onset; and (iii) the CSF/blood viral replication ratio proves instrumental in interpreting PCR test results, facilitating the accurate diagnosis of HHV-6 encephalitis in HSCT recipients [77] (Table 6).

Table 6.

Summary of cohort study information based on HHV-6/7 clinical sample studies

Author	Disease	No.of controls	No.of patients	Age	Sample
Berzero et al. [77]	Human herpesvirus 6 encephalitis	886	45	30/45 < 18 years 15/45 ≥ 18 years	CSF, Blood
Lee et al. [76]	Kidney transplantion	46	1	20–75 years	Blood
Zhou et al. [78]	Allogeneic hematopoietic cell transplant	496	242 CMV: 126 HHV-6: 56 CARV: 27 EBV: 26 HSV: 7	7 months to 73 years	Body fluid, Tissue specimen
Mehta et al. [79]	Inborn errors of immunity	463	30	Patients not requiring second HSCT: 0.1–19.3 years Patients requiring second HSCT: 0.09–17.1 years	Blood
Leong et al. [75]	-	496	4	17–59 years	Blood

In addition to the integration into the human chromosome, HHV-6 induces reactivation of the virus in immunodeficient individuals, posing challenges for effective treatment. The phenomenon of viral reactivation was investigated in a study involving patients who underwent HSCT. This investigation enrolled 30 patients, revealing that 13 individuals (43.3%) experienced viral reactivation, with HHV6 accounting for the majority (n = 6) of these cases [79].

HHV-6B DNA is commonly identified in bronchoalveolar lavage fluid (BALF) of immunocompromised patients with lower respiratory tract disease (LRTD). The role of HHV-6B as a pulmonary pathogen remains unclear, and the potential contribution of first-episode viral infections to noninfectious pulmonary complications is not well-defined. Joshua A. Hill et al. conducted a comprehensive study spanning the years 1992 to 2015, focusing on allogeneic HSCT recipients who underwent BAL to assess LRTD [80]. Testing for HHV-6B DNA in BALF using PCR, the researchers employed multivariate proportional risk modeling to evaluate the association between HHV-6B⁺ BALF and overall mortality, respiratory failure, and death. The study also explored the impact of anti-HHV-6B antiviral drugs on these outcomes. HHV-6B⁺ BALF was detected in 147 out of 553 individuals (27%). Subjects with HHV-6B⁺ BALF exhibited a significantly elevated risk of overall mortality and death from respiratory failure, independent of comorbid pathogens, compared to subjects with HHV-6B- BALF. The findings suggested that the detection of HHV-6B in BALF was correlated with high mortality in patients with LRTD following allogeneic HSCT. A parallel study during the same timeframe retrospectively analyzed 738 patients with allogeneic HSCT [78]. This study established a novel mouse model of bone marrow transplantation to investigate herpesvirus reactivation post-transplantation. However, further investigations were required to ascertain specific causality leading to organ damage. Studies have indicated that initial herpesvirus infection is associated with an increased risk of non-infectious lung injury, including interstitial pneumonia syndrome (IPS) and bronchiolitis obliterans syndrome (BOS), after HSCT, as supported by in vivo observations in mice (Table 6).

Current evidence indicates a high detection rate of HHV-6B DNA in BALF among allogeneic HSCT recipients with LRTD, with a significant association with increased overall mortality and respiratory failure. Although the exact pathogenic role of HHV-6B in pulmonary complications remains unclear, initial viral infections appear to be related to an elevated risk of non-infectious lung injuries, such as IPS and BOS. These findings highlight the potential involvement of HHV-6B in post-transplant pulmonary pathology and underscore the need for further mechanistic studies and validation in animal models to elucidate its causal role and identify therapeutic targets.

Conclusion

Although HHV-6/7 are clinically significant in immunocompromised populations—particularly in HSCT and SOT recipients—biomarker-based investigations of host responses to these viruses remain relatively underdeveloped. Current studies have predominantly focused on viral reactivation and ciHHV-6, with limited exploration of canonical host-derived biomarkers such as proteins, metabolites, and immunomodulatory molecules. For instance, in the context of HHV-6B-associated LRTD following HSCT, high viral DNA loads detected in BALF have been associated with increased mortality and respiratory failure. While this suggests a clinically meaningful association, such measurements are primarily pathogen-centric and fall short of the broader criteria for host-based biomarkers capable of stratifying risk or elucidating mechanistic pathways.

Furthermore, ciHHV-6, identified in 0.8–0.9% of transplant patients, can result in persistently high HHV-6 DNA levels, confounding the distinction between latent integration and active infection. Recent studies have highlighted the utility of additional diagnostic tools—including hair follicle DNA analysis and CSF-to-blood viral DNA ratios—to differentiate ciHHV-6 from HHV-6 encephalitis, particularly in cases lacking characteristic radiographic findings. Despite emerging clinical evidence linking HHV-6 to encephalitis, pulmonary injury, and post-transplant organ dysfunction, comprehensive investigations of host oxidative stress markers, inflammatory cytokines, and immune signaling pathways remain scarce. In contrast to HSV, VZV and CMV, HHV-6/7 research on host biomarkers is still at an early stage of development.

Taken together, the clinical heterogeneity and diagnostic ambiguity associated with HHV-6/7-related disease underscore the urgent need for systematic biomarker discovery. Future studies should extend beyond viral nucleic acid quantification to incorporate multi-omics strategies—such as immunophenotyping, proteomics, metabolomics, and single-cell sequencing—to elucidate the underlying molecular mechanisms and enable robust risk stratification. These approaches will be instrumental in advancing early diagnosis and guiding precision medicine strategies for immunocompromised patients affected by HHV-6/7.

HHV-8

HHV-8, also known as KSHV, is a carcinogenic virus closely associated with a variety of both malignant and non-malignant diseases. HHV-8 infection is particularly significant in immunocompromised populations, and the virus is linked to several clinical conditions, such as Kaposi’s sarcoma (KS), primary effusion lymphoma (PEL), and multicentric Castleman’s disease (MCD). The diagnosis, prognosis, and risk stratification of these diseases require a comprehensive consideration of factors such as the viral replication status, cytokine levels, and the host’s immune response. By identifying these biomarkers, a better understanding of disease progression and treatment outcomes can be achieved, ultimately optimizing treatment plans and improving patient prognosis.

Kaposi’s sarcoma

KS is a multifocal angiodysplastic soft-tissue sarcoma that primarily affects the skin and mucous membranes [81]. KS occurs in several forms, including classic, endemic, iatrogenic, and AIDS-associated variants. The prevalence of KS has risen significantly in Africa, particularly in sub-Saharan regions, largely due to the widespread incidence of AIDS, which predominantly affects children and young adults [82]. As KS cases have increased, clinical investigations have focused on identifying biomarkers associated with the disease. These biomarkers serve dual purposes: aiding in the diagnosis, subtyping, risk assessment, and prognosis of KS, while also providing insights into the underlying mechanisms of disease development and progression, thereby offering valuable information for the development of targeted therapies.

Broccolo et al. explored biomarkers associated with AIDS-related KS, focusing on the relationship between plasma HHV-8 DNA viral load and clinical status in AIDS-KS. The study involved 378 blood samples collected longitudinally from 62 patients with AIDS-KS receiving antiretroviral or antitumor therapy. Patient classification was based on clinical status: onset of disease (OD), disease progression (PD), stable or partial remission, and complete remission. Results indicated a strong association between plasma HHV-8 DNAemia and disease onset or progression, emphasizing the active role of replicating virus in clinically active AIDS-KS. Accurate assessment of plasma HHV-8 load proved valuable in monitoring AIDS-KS during antiretroviral or antineoplastic therapy [86]. In a separate study, Aka et al., investigated biomarkers for classical KS in plasma from 15 HIV-negative classical KS cases compared to plasma from 29 matched controls, utilizing a multiplexed set of immune markers. Among the 70 markers examined, CXCL10 (IP-10), sIL-1RII, sIL2RA, and CCL3 (MIP-1A) were strongly associated with KS. These findings, along with previous observations, suggest a tumor-promoting role for certain cytokines, especially CXCL10. However, the limited number of samples and case-control design in this study limit conclusive insights into KS risk or pathogenesis [83]. Therefore, larger and well-designed prospective studies are imperative for a more robust evaluation of the association between these markers and KS (Table 7).

Table 7.

Summary of cohort study information based on HHV-8 clinical sample studies

Author	Disease	Biomarker	No.of controls	No.of patients	Age	Sample
Aka et al. [83]	Kaposi’s sarcoma	CXCL10 (IP-10), sIL-1RII, sIL-2RA, CCL3 (MIP-1A)	29	15	-	Plasma
Privatt et al. [84]	Kaposi’s sarcoma	-	13	11	26–58 years	Plasma
Ngalamika et al. [85]	Kaposi’s sarcoma	IL-5, IL-6, IP-10	7	4	Poor Responders: 31–55 years Responders: 31–45.5 years	Plasma
Broccolo et al. [86]	Kaposi’s sarcoma	HIV-1 viral load, CD4 + T-cell counts	OD: 16 PD: 6	Stable or partial remission: 13 Complete remission: 27	34–51 years	Blood
Chugh et al. [87]	KSHV-associated malignancies	Circulating miRNAs	18	18	-	Plasma
El-Mallawany et al. [88]	Kaposi’s Sarcoma	KSHV viral load, IL-6, IL-10	-	25	1.7–17 years	Plasma
Horna et al. [91]	Multicentric castleman’s disease	CXCL13, VEGFR1, PGF	19	UCD: 24 iMCD: 8	-	Lymphnode, Spleen
Zhou et al. [92]	Multicentric castleman’s disease	LRP	-	39	25–67 years	Effusions, Peripheral Blood, Bone Marrow
Rasmussen et al. [93]	Multicentric castleman’s disease	CRP, HHV-8 DNA	26	73	-	Plasma
Lurain et al. [94]	Primary effusion lymphoma	-	-	Training cohort: 59 Validation cohort: 58	33–54 years 39–55 years	-
Kaji et al. [95]	Primary effusion lymphoma	CD19, CD20, CD79a	4	64	57–98 years	Effusion

Circulating miRNAs, particularly intra-exosomal miRNAs, offer the advantages of stability and ease of detection, making them potential minimally invasive and stable biomarkers. Chugh et al. investigated the sensitivity of plasma, pleural fluid, serum from patients with KSHV-associated malignancies, from two mouse models of KS, to host and viral circulating miRNAs [87]. Gene ontology analysis of signature exosomal miRNA targets identified several signaling pathways known to be crucial in the pathogenesis of KSHV. Functional analysis of endothelial cells exposed to patient-derived exosomes revealed increased cell migration and IL-6 secretion, suggesting that exosomes from KSHV-associated malignancies are functional and contain a unique subset of miRNAs, which may serve as potential disease biomarkers and contribute to the paracrine phenotype characterizing KS. Privatt et al. explored whether the plasma metabolome could differentiate between asymptomatic KSHV-infected individuals, with or without HIV co-infection, and those with symptomatic KS. Plasma samples from KSHV seropositive subjects from sub-Saharan Africa were categorized into three groups: (A) asymptomatic KSHV-only infections, (B) asymptomatic co-infected patients with KSHV and HIV, and (C) patients with clinically diagnosed KS symptoms [84]. Metabolite profiles for lipids and polarities were analyzed, revealing evident differences in both polar and nonpolar plasma metabolomics in both comparisons. Integration of metabolic results with previously reported KS transcriptomic data suggested dysregulation of the amino acid/urea cycle and purine metabolic pathways, consistent with viral infection in the progression of KS disease. Research indicates that metabolic changes are closely associated with KSHV infection and disease progression. These findings provide new perspectives for further investigation into the diagnosis and prognosis of KSHV-related diseases.

The treatment of KS involves either antiretroviral therapy (ART) alone or ART combined with cytotoxic chemotherapy. HIV-associated/epidemic Kaposi’s sarcoma (EpKS) is an AIDS-defining angio-proliferative malignancy. Ngalamika et al. compared plasma cytokine and chemokine levels in responders (patients with clinical regression of EpKS tumors) and poor responders (patients with progressive or non-responsive EpKS), along with KSHVnAb response, aiming to identify immune markers associated with ART treatment outcomes. At follow-up, IL-6 levels in responders were lower than in poor responders, and IP-10/CXCL-10 levels in responders were significantly lower than in poor responders. No significant difference was observed in KSHV nAb levels between responders and poor responders [85]. Elevated plasma levels of IL-5 may be a good prognostic marker for ART treatment in EpKS, while high plasma levels of IL-6 and IP-10 may indicate a poor prognosis. These findings provide potential biomarkers for identifying treatment responses in early EpKS patients, helping to avoid unnecessary chemotherapy and optimize treatment plans (Table 7).

The lytic activation of KSHV and the associated elevation of KSHV viral load may play a role in shaping the distinctive clinical manifestation observed in children with KS in KSHV-endemic areas of Africa. El-Mallawany et al. investigated the interplay between KSHV viral load, human IL-6, and IL-10 levels in 25 children with KS in Lilongwe, Malawi. The presence of detectable KSHV viral load significantly correlated with lymphadenopathy-associated KS (p = 0.004), while undetectable KSHV viral load was associated with a higher likelihood of developing hyperpigmented skin lesions (P = 0.01) [88]. Some children with KS exhibited both detectable KSHV viral load and elevated IL-6 levels (Table 7). In KS tumors, activated mast cells (MC) are notably enriched, contributing to the inflammatory microenvironment. Nevertheless, the mechanisms driving MC activation remain incompletely understood. Byakwaga et al. [89] aimed to elucidate whether immunoglobulin E (IgE), a potent MC activator, correlates with the incidence and severity of KS. In a cross-sectional study involving untreated HIV-infected adults with or without KS in Uganda, it was observed that the prevalence of KS in adults with KS exceeded that in adults without KS. Plasma IgE levels were found to be higher in subjects with KS, revealing a dose-response relationship between IgE levels and the presence and severity of KS, even after adjusting for age, sex, CD4⁺ T-cell count, and HIV RNA levels. Higher eosinophil counts were also associated with IgE levels, along with elevated plasma IL-33 concentrations in patients with KS [85, 89]. These findings suggest that IgE-driven atopic inflammation may contribute to the pathogenesis of KS. Consequently, therapies targeting IgE-mediated MC activation may represent a novel approach to the treatment or prevention of KS.

KS often co-occurring with HIV, has prompted extensive exploration of biomarkers for diagnosis, prognosis, and treatment monitoring. Among these, plasma HHV-8 DNA load is the most widely studied and clinically significant, strongly correlating with disease onset and progression, particularly in AIDS-related KS, and serving as a key marker for treatment response. Additionally, cytokines such as CXCL10 (IP-10) and IL-6 have emerged as critical players in tumor progression, with CXCL10 being identified as a potent tumor-promoting factor. Exosomal miRNAs are gaining attention as promising non-invasive biomarkers, offering valuable insights into KS pathogenesis and its dynamic progression. Elevated IgE levels have also been linked to KS severity, suggesting that targeting IgE-mediated MC activation may provide novel therapeutic strategies (Table 7).

Multicentric Castleman’s disease

HHV-8 plays a pivotal role in the pathogenesis of MCD, a lymphoproliferative disorder characterized by generalized lymphadenopathy, splenomegaly, and pronounced cytokine-driven inflammation. HHV-8-associated MCD typically arises in immunocompromised individuals and is associated with severe complications and poor clinical outcomes [90].

Horna et al. found that in lymph node samples from patients with HHV-8-MCD, gene expression related to the complement cascade was significantly upregulated, particularly key components of both the classical and alternative pathways. Additionally, the expression of the chemokine CXCL13 was increased within the lymph nodes, which may be associated with B cell recruitment and the inflammatory response. In terms of angiogenesis, both VEGFR1 and its ligand PGF were markedly upregulated, suggesting a potential role for these molecules in disease-associated vascular proliferation [91]. These specific molecular alterations not only offer potential biomarkers for the diagnosis of HHV-8-MCD but also provide important insights into its pathophysiological mechanisms and the development of novel therapeutic strategies.

In the study by Zhou et al., the identification of lambda-restricted plasmablasts (LRP) via flow cytometry has emerged as a significant diagnostic tool for KSHV-related MCD. LRP exhibit a consistent immunophenotypic profile characterized by high expression of CD38 and IRF4, along with lambda light chain restriction. This finding indicates that LRP are the extranodal counterparts of KSHV-infected plasmablasts, similar to those found in lymph nodes of MCD patients. The research results demonstrate that the detection of LRP can effectively differentiate KSHV-MCD from other KSHV-related diseases, such as KSHV inflammatory cytokine syndrome (KICS). Furthermore, LRP identification reveals a subset of KICS patients who closely resemble those with active KSHV-MCD in terms of clinical presentation [92].

Rasmussen et al. conducted an analysis of patients with MCD who were treated with rituximab. The study found that a serum CRP level above 20 mg/L was associated with a higher risk of disease progression, with a two-year progression-free survival rate of 63.7%. Additionally, patients with HHV-8 DNA viral load exceeding 3 log copies per mL had a two-year progression-free survival rate of 48.1%, indicating a close correlation between high viral load and an increased risk of relapse. The study also revealed the impact of HIV infection status on prognosis, with HIV-positive patients having a five-year progression-free survival rate of 62%, compared to only 26% for HIV-negative patients. This difference may be attributed to immune recovery in HIV-positive patients following antiviral treatment, which reduces the risk of HHV-8 infection [93]. The dynamic changes in these biomarkers provide valuable information, helping physicians better assess the patient’s disease status and adjust treatment plans accordingly to reduce the risk of disease progression.

In the study of diseases associated with HHV-8, serological testing is a critical diagnostic tool for determining infection status. These tests measure antibody levels against HHV-8 antigens, providing valuable information for clinicians in monitoring disease progression and treatment responses. Specifically, in the context of MCD, serological testing for HHV-8 plays a key role in differentiating between various subtypes. The differential diagnosis between HHV-8-associated Castleman's disease and idiopathic multicentric Castleman's disease (iMCD) depends largely on the detection of HHV-8. Research indicates that HHV-8 infection is commonly linked to classic Castleman's disease, which is characterized by excessive IL-6 production from HHV-8-infected plasmablasts or plasma cells—an occurrence less frequently observed in iMCD. Therefore, a negative serological result for HHV-8 can help exclude HHV-8-associated Castleman's disease and support a diagnosis of iMCD. While current serological tests are instrumental in both diagnosis and monitoring, further research is necessary to refine their sensitivity and specificity, thereby enhancing their clinical applicability (Table 7).

Primary effusion lymphoma

PEL is a rare B-cell lymphoma typically characterized by serous effusions without lymphadenopathy. HHV-8 infection plays a crucial role in the onset and progression of PEL. Due to the similarity of PEL’s clinical presentation to other more common diseases, its diagnosis is often challenging. Moreover, PEL has a poor prognosis, especially in the absence of effective treatment. Therefore, identifying diagnostic biomarkers for PEL, assessing prognostic factors, and performing risk stratification are crucial for developing individualized treatment plans and improving patient survival rates [90].

Lurain et al. developed and validated the Primary Effusion Lymphoma Prognostic Score (PEL-PS) in their study. The researchers discovered that this scoring system could effectively classify patients into two prognostic groups: those with zero adverse prognostic factors (PEL-PS 0), who exhibited a significantly better median overall survival (OS) compared to patients with one or two adverse prognostic factors (PEL-PS 1–2). Specifically, patients with a PEL-PS of 0 had a median OS of 10.6 years, while those with a PEL-PS of 1–2 had a median OS of only 0.8 years. This discovery provides clinicians with a valuable tool for assessing the prognosis of PEL patients and guiding the exploration of alternative treatment strategies for those with a worse prognosis [94].

In the study conducted by Kaji et al., the researchers aimed to gain deeper insights into the clinical characteristics, diagnostic methods, and prognostic factors of HHV-8-associated PEL. Through a retrospective analysis of 95 PEL patients from Japan, the researchers particularly focused on HHV-8-negative cases. They employed immunohistochemistry and polymerase chain reaction techniques to assess the HHV-8 status of the patients and provided a detailed analysis of their clinical features, pathological manifestations, and immunophenotypes. The results indicated that HHV-8-negative PEL predominantly affects HIV-negative elderly patients, characterized by lymphoma cells expressing mature B cell markers (such as CD19, CD20, and CD79a) while lacking plasma cell-related markers. In terms of prognosis, patients receiving chemotherapy showed a high overall response rate (95%) and complete remission rate (73%). The 2-year overall survival rate was 84.7%, with a progression-free survival rate of 73.8%, suggesting a better prognosis than HHV-8-positive PEL. Furthermore, the study identified several independent risk factors associated with prognosis, including an Eastern Cooperative Oncology Group performance status score ≥ 2, age ≥ 70 years, and the presence of ascites [95] (Table 7).

Conclusion

HHV-8, or KSHV, is etiologically related to several lymphoproliferative and oncogenic diseases, including KS, MCD, and PEL, particularly in immunocompromised individuals. Recent biomarker studies have deepened our understanding of the pathogenesis, diagnosis, prognosis, and therapeutic monitoring across these distinct disease entities.

Circulating HHV-8 DNA load remains the most extensively validated biomarker, correlating strongly with disease onset, progression, and therapeutic response in KS. Additionally, proinflammatory cytokines such as CXCL10 and IL-6 are demonstrated to have prognostic value. Circulating exosomal miRNAs and plasma metabolomic alterations further reflect underlying tumor biology, while elevated IgE levels point to the role of mast cell activation in KS pathogenesis, offering potential new therapeutic targets.

In HHV-8–associated MCD, dysregulated immune pathways—including complement cascade components, angiogenic mediators (VEGFR1/PGF), and chemokines like CXCL13—suggest a multifaceted inflammatory milieu contributing to disease progression. The identification of LRP has emerged as a distinguishing diagnostic marker for differentiating KSHV-MCD from KICS and other related disorders. Prognostic markers such as serum CRP levels and HHV-8 viral load are also valuable for disease monitoring and treatment stratification.

The development of the PEL-PS enables clinically meaningful stratification of patients by survival outcome. Furthermore, HHV-8-negative PEL, observed predominantly in HIV-negative elderly populations, displays a distinct immunophenotypic and clinical profile, reinforcing the need to consider PEL as a heterogeneous disease spectrum.

Collectively, HHV-8–associated malignancies are characterized by diverse immunopathological profiles and biomarker landscapes. Future large-scale, multicenter studies incorporating multi-omics approaches are warranted to validate these candidate biomarkers, which will be instrumental in refining diagnostic criteria, improving risk stratification, and advancing individualized therapeutic strategies for patients with HHV-8–associated diseases.

Conclusions

Current investigations of clinical samples from herpesvirus-associated diseases primarily focus on the discovery and validation of biomarkers, which involves a comparative analysis of the expression of these molecules between patient samples and control samples from healthy control samples. Various indicators, such as viral DNA load, miRNAs, viral-associated antibodies, and cytokines, are examined to identify markers differentially expressed between patients and healthy individuals, thereby establishing their potential as biomarkers. These candidates are then validated in larger cohorts to confirm their clinical utility [96].

Beyond the realm of biomarkers, additional studies utilizing clinical samples delve into diverse areas, such as direct examinations of the virus itself using virus-containing biospecimens, local epidemiologic investigations, and exploration of transcriptomic changes during the progression of virus-induced diseases [97]. Biomarkers play a crucial role in both screening and prognosis, which is particularly important given the latency nature of herpesviruses, that these viruses are widely distributed in the environment and frequently infect individuals asymptomatically[98]. Once infected, herpesviruses remain latent in the body, often undetected. However, certain conditions can trigger viral reactivation, leading to severe diseases, including malignant tumors. As a result, early detection of these latent infections before symptoms arise is essential for timely intervention and improved prognosis [4]. For patients already diagnosed, the emphasis shifts toward assessing prognosis and risk stratification, enabling the customization of treatment plans. A deeper understanding of the prognostic differences among various herpesviruses may reveal factors and mechanisms contributing to diverse clinical outcomes.

For all types of herpesviruses, the most traditional biomarker is circulating viral DNA [99]. Plasma DNA, easily obtainable and measurable through PCR, serves as a straightforward biomarker and is widely used for screening various virus-associated diseases and assessing prognosis [100–102]. However, the specificity of viral DNA alone often falls short of current clinical needs. Studies are increasingly focusing on the exploration and application of novel biomarkers, such as transcriptomics, antibodies, and cytokines. These efforts aim to improve the sensitivity and specificity of disease detection by utilizing a combination of multiple biomarkers.

Notably, the use of saliva as a diagnostic fluid has gained increasing attention for the detection of HHVs, particularly VZV, CMV, and EBV. Salivary antibody testing offers a non-invasive, cost-effective alternative to serum-based assays, especially in large-scale screening and in populations where blood collection is less feasible. Choi et al. demonstrated that virus-specific IgA and IgG antibodies to both VZV and CMV can be reliably detected in saliva, providing supplementary diagnostic information when combined with conventional serology, especially in early infection or reactivation stages [103]. Waters et al. showed that salivary CMV DNA positivity strongly correlates with systemic immune activation and elevated plasma gB-specific antibody levels in renal transplant recipients, suggesting that saliva can reflect both viral replication and host immune response [104]. For EBV, salivary IgA detection has shown potential in reflecting mucosal immunity and is under investigation as a biomarker for early diagnosis of EBV-associated malignancies such as NPC. Collectively, salivary antibody testing for VZV, CMV, and EBV holds promise as a minimally invasive tool for viral surveillance and immunomonitoring [105]. However, standardization of detection thresholds and further validation across diverse clinical settings are required to fully integrate these approaches into routine diagnostics.

Blood samples remain the most commonly used and versatile source for multifaceted analysis [106]. However, certain diseases may require more specific samples, such as nasopharyngeal brushes for NPC, biopsy tissues for stomach cancer and other tumors, or amniotic fluid for congenital diseases [107, 108]. In recent years, saliva has gained increasing attention as a new biological sample [109]. With its simplicity and non-invasiveness, saliva is particularly suitable for large-scale screenings and potential home-based testing scenarios. It has demonstrated clinical significance in various virus-related diseases [110].

The significance of biomarker research goes beyond its role in accurate disease diagnosis and prognosis, as it also serves as a critical clinical detection tool. Further exploration of the underlying mechanisms may reveal valuable insights into viral eradication processes and the pathogenesis of various diseases. Importantly, limited basic research on the pathogenesis of herpesvirus-associated diseases has combined with the information from clinical samples, which should be explored more in the future and these efforts will provide insightful perspectives and new targets for the prevention and treatment of herpesvirus-associated diseases.

Abbreviations

ADCC

Antibody-Dependent Cell-mediated Cytotoxicity

AIDS

Acquired Immune Deficiency Syndrome

AIDS-KS

AIDS-related Kaposi’s Sarcoma

ALB

Albumin

ART

Antiretroviral Therapy

BAL

Bronchoalveolar Lavage

BALF

Bronchoalveolar Lavage Fluid

BART

BamHI-A Rightward Transcript

BOS

Bronchiolitis Obliterans Syndrome

cCMV

Congenital Cytomegalovirus

CMI

Cell-Mediated Immunity

CMV

Cytomegalovirus

CNS

Central Nervous System

CRP

C-reactive Protein

CSF

Cerebrospinal Fluid

CTLA-4

Cytotoxic T Lymphocyte-associated Antigen 4

EA

Early Antigen

EBER

Epstein–Barr Virus-Encoded Small RNA

EBNA

Epstein–Barr Nuclear Antigen

EBV

Epstein–Barr Virus

ELISA

Enzyme-Linked Immunosorbent Assay

GBM

Glioblastoma Multiforme

GC

Gastric Cancer

HCMV

Human Cytomegalovirus

HHV

Human Herpesvirus

HIV

Human Immunodeficiency Virus

HSCT

Hematopoietic Stem Cell Transplantation

HSE

Herpes Simplex Encephalitis

HSM

Herpes Simplex Meningitis

HSV

Herpes Simplex Virus

HZ

Herpes Zoster

HZO

Herpes Zoster Ophthalmicus

IgE

Immunoglobulin E

IL

Interleukin

IPS

Interstitial Pneumonia Syndrome

KICS

KSHV Inflammatory Cytokine Syndrome

KS

Kaposi’s Sarcoma

KSHV

Kaposi’s Sarcoma-Associated Herpesvirus

LRP

Lambda-Restricted Plasmablasts

LRTD

Lower Respiratory Tract Disease

MC

Mast Cell

MCD

Multicentric Castleman’s Disease

MIC-1

Macrophage Inhibitory Cytokine-1

MRI

Magnetic Resonance Imaging

NGS

Next-Generation Sequencing

NPC

Nasopharyngeal Carcinoma

PCR

Polymerase Chain Reaction

PEL

Primary Effusion Lymphoma

PEL-PS

Primary Effusion Lymphoma Prognostic Score

PHN

Postherpetic Neuralgia

SOC

Solid Organ Transplant

TAC

Total Antioxidant Capacity

TMB

Tumor Mutational Burden

TPC

Total Polyphenol Content

UA

Uric Acid

VCA

Viral Capsid Antigen

VZV

Varicella-Zoster Virus

Author contributions

Conceptualization, D.Y., L.J., and Y.L.; investigation, L.L., M.F., X.L., and W.S.; writing—original draft preparation, L.L., M.F., and D.Y.; writing—review and editing, L.L., M.F., X.L., W.S., X.W., Y.X., L.M., K.Y., Y.L., L.J., and D.Y.; project administration, Y.L., L.J., and D.Y.; funding acquisition, D.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by a grant from the Chinese Institutes for Medical Research, Beijing (Grant number: CX23YQ04), a grant from the Chinese Institutes for Medical Research and Capital Medical University (Grant number: KCB2301), a grand from the Natural Science Foundation of Xiamen (Grant number: 3502Z202473020), and a grand from Scientific Research Foundation of State Key Laboratory of Vaccines for Infectious Diseases, Xiang An Biomedicine Laboratory(Grant number: 2023XAKJ0102052).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.