Mucormycosis, a lethal and opportunistic infection disease caused by fungi of the order Mucorales, aggressively invades human blood, organs, and tissues (Donnelly et al., 2020; Panda et al., 2024; Pappas et al., 2021). The Mucorales order comprises 55 genera and 261 species, with 38 recognized as pathogenic to humans. Rhizopus arrhizus is the most prevalent pathogenic genus globally, followed by Mucor and Rhizomucor, while Apophysomyces and Cunninghamella are less frequently implicated. These fungi are ubiquitous in the environment and exhibit a high propensity for colonizing the human respiratory tract (Liang et al., 2024; Roden et al., 2005). Although Mucorales colonization does not immediately provoke disease, it serves as a prerequisite for chronic and allergic mycoses, as well as localized airway infections in invasive fungal diseases. The diagnosis relies on histopathological analysis, conventional microbiological testing (CMT), and imaging, with histopathology or culture considered as the “gold standard” for diagnosis of mucormycosis (Donnelly et al., 2020; Pappas et al., 2021). Histopathological analysis of sterile specimens was critical for confirmation, but there exist difficulties in sampling (Hammer et al., 2018). For CMT, including culture and direct microscopic examination of specimens, also faces limitations in timely diagnosing mucormycosis (Skiada et al., 2018; Wang et al., 2024). While other microbiological testing (OMT) like the galactomannan (GM) antigen testing and (1–3)-β-D-glucan (G) testing have difficulties with accuracy (Lass-Flörl et al., 2021; Lmoth et al., 2021). On account of nonspecific symptoms and signs, early mucormycosis identification is still a challenge in clinic. Definitive diagnosis of mucormycosis, particularly distinguishing between colonization and active infection, remains a significant clinical hurdle (Sipsas et al., 2018; Skiada et al., 2018). However, there remains a paucity of studies focused on differentiating Mucorales infection from colonization.

mNGS is a unbiased sequencing of all nucleic acids (DNA/RNA) in clinical samples (blood, cerebrospinal fluid, respiratory secretions, etc.), and the identification of pathogens (bacteria, viruses, fungi, parasites) through bioinformatics comparison. It does not require pre-assumption of pathogens and is suitable for detecting unknown infections, mixed infections, or rare pathogens (Gu et al., 2019). While, nucleic acid testing by PCR for single agents to multiplexed PCR testing using syndromic panels generally include the most common pathogens associated with a defined clinical syndrome (Chiu and Miller, 2019). The application of metagenomic next-generation sequencing (mNGS) has gained prominence in the clinical diagnosis of infectious diseases, particularly when empirical anti-infective therapies prove ineffective or when CMT fails to identify the etiology. Compared to CMT, mNGS demonstrates superior diagnostic performance for invasive fungal infections, and multiple studies highlight its ability to detect fungal pathogens undiagnosable by traditional methods (Jia et al., 2023; Jiang et al., 2024; Liu et al., 2021; Wang et al., 2024; Zhang et al., 2024, 2024).

Furthermore, the utility of mNGS in differentiating fungal colonization from infection have been explored, primarily by establishing thresholds for pathogen-specific read counts. For instance, Liu et al. demonstrated that bronchoalveolar lavage fluid (BALF) mNGS could distinguish Pneumocystis jirovecii colonization from infection with an area under the curve (AUC) of 0.973, identifying an optimal threshold of 14 reads (Liu et al., 2021). Jia et al. reported a species-specific read number (SSRN) cut-off of 2.5 for diagnosing invasive pulmonary aspergillosis (IPA) versus non-IPA, with distinct thresholds of 1 and 4.5 for immunocompromised and diabetic IPA patients, respectively (Jia et al., 2023). Similarly, the study of Jiang et al. discovered an optimal mNGS RPTM (reads per ten million) cut-off value of 23 for discriminating between Aspergillus infection and colonization (Jiang et al., 2024). Despite these advancements, critical gaps persist in understanding the clinical characteristics of patients and the microbial compositional differences between those with Mucorales colonization and infection.

In this study, we evaluated the efficacy of mNGS, culture and OMT in distinguishing Mucorales infection from colonization. Furthermore, we delineated variations in antimicrobial management strategies, clinical indicators, and shifts in pulmonary microbial composition between these patient groups.

Mucormycosis, a disease with high morbidity and mortality rate, is difficult to diagnose and treat (Cornely et al., 2019). Although Mucorales infection and colonization have clear definitions (Cornely et al., 2014a, 2014b; Donnelly et al., 2020), timely and precise diagnosis of invasive Mucorales infection or colonization is still complicated and difficult in clinic. However, studies focus on distinguishing Mucorales infection from colonization are barely reported. This study was carried out to evaluate the efficacy of mNGS in differentiating Mucorales infection from colonization. Moreover, it also outlined the distribution characteristics of mucormycosis, clinical characteristics, immune changes, outcome, antibiotic adjustment of Mucorales infection and colonization patients, as well as the variations in sample microbiota.

The main reported pathogens in mucormycosis are Rhizopus, Mucor, and Lichtheimia, followed by Rhizomucor, Cunninghamella, Apophysomyces, and Saksenaea (Cornely et al., 2019, 2014a). Consistent with previous researches, our study identified 14 Mucorales species among the patients, with 13 species leading to Mucorales infection. Among them, Rhizopus microsporus, Rhizopus arrhizus, and Rhizomucor pusillus were the most prevalent in patients with Mucorales infection. Rhizopus delemar, while Rhizopus arrhizus, and Rhizomucor pusillus were the most common species in patients with Mucorales colonization ( Figure 1 ). Additionally, from Table 1 and spearson correlation analyses, we found that the Mucorales RPTM value, LOHS, hsCRP, immunocompromised, malignant blood tumor, and antifungal changed accounted for the Mucorales infection, which may be beyond the existed research findings (Cornely et al., 2019).

With the widespread application of mNGS, it offers a hypothesis-free, unbiased approach to pathogen detection, enabling the identification of novel or unexpected organisms, semi-quantitative analysis, and comprehensive genomic coverage. However, its limitations include high host background noise, substantial cost and turnaround time, incomplete reference databases, and susceptibility to environmental contamination (Gu et al., 2019). However, the benefits of using mNGS for pathogen detection have become increasingly apparent, especially for rare and emerging pathogens, such as mucormycosis, hyalohyphomycosis (Fusarium, Paecilomyces, Scedosporium, etc.), and phaeohyphomycosis (Alternaria, Bipolaris, Cladosporium, Rhinocladiella, etc.) (Ling et al., 2024; Xing et al., 2023; Wang et al., 2024; Li et al., 2024). A research has shown that mNGS of infected body fluids by Illumina sequencing has a combined sensitivity and specificity of 79% (95% CI 73.5–85.2%) and 91% (95% CI 87.3–93.8%) for bacteria and 91% (95% CI 84.2–100%) and 89% (95% CI 85.7–92.5%) for fungi, respectively (Gu et al., 2021). The above indicates that mNGS is a highly effective option even before OMT results are available. Early and precise detection of pathogen of severe or rare infectious patients is critical for clinicians to give a timely fast intervention and targeted therapy as quickly as possible. It suggests that the medical related organisms including Candida, Cryptococcus, Mucorales, and Aspergillus has increased in subjects with impaired immune function, and the thick cell wall of fungi is difficult to break to release nucleic acid which lead to false negative mNGS results (Bittinger et al., 2014). While the diagnostic performance of mNGS has improved with optimized extraction methods (Gu et al., 2021). Besides, the positive diagnostic threshold criteria for mNGS should be defined according to different host and pathogen status. Based on these, this study laid the foundation for the establishment of the positive threshold criteria according to different host and pathogen status in some ways.

Numerous studies have investigated the diagnostic ability of mNGS for Mucorales infection, but there remains little research on the distinction of Mucorales colonization and infection (Wang et al., 2024; Zhang et al., 2024). Meaningfully, our study laid the foundation for the establishment of the positive threshold criteria according to different host and pathogen status in some ways. We observed that mNGS displayed superior accuracy in diagnosing Mucorales infection and distinguishing it from colonization when compared to culture and OMT (P < 0.05). The optimal cut-off value of RPTM for mNGS was 51. At this threshold, mNGS achieved a sensitivity of 58.82% and a specificity of 90.00% for the final diagnosis ( Figure 2B ). Furthermore, multiple (≥10) nodules, pleural effusion and halo sign were reportedly associated with pulmonary mucormycosis (Chamilos et al., 2005; Legouge et al., 2014). However, we found that imaging has limitations in diagnosing mucormycosis in clinical, especially when it comes to co-infection of multiple pathogens. Indeed, this research can serve as a valuable reference for analyzing patients with Mucorales infection and colonization. Notably, even though mNGS serves as a precise pathogen infection test method and has potential diagnosis in clinic, the final diagnosis of the disease counts on clinical experts who integrate the patient’s symptoms, clinical laboratory test results, and etiological findings to make a comprehensive decision. And in the future, it is necessary for us to conduct prospective studies with a large amount of data about distinction of Mucorales infection and colonization.

mNGS had significant impact on treatment regimens, particularly in infectious disease (Zhang et al., 2024). Equally, in this study, 68.18%% and 70.00%% showed improvement among the patients who received only antifungal treatment, and antibacterial combined with antifungal treatment, respectively. This suggested that timely clinical intervention and targeted antifungal therapy for patient prognosis is of great importance. Although Shannon and Simpson indexes were higher in the infection group, no significant differences were observed in species abundance and diversity between the two groups ( Figure 6 ). Even the microbial diversity differences are minimal and not statistically significant, these microbiome findings are as exploratory and mainly hypothesis generating. Incidentally, Rhizomucor pusillus appeared more frequently in Mucorales infection group. Additionally, TTV, and Rhizomucor delemar were significantly more abundant in patients with Mucorales infection and colonization individually. TTV is a member of Anellovirida, which is commonly present in patients with various blood diseases, organ transplants, tumors, periodontitis, and even the healthy population (Maggi and Bendinelli, 2010; Nishizawa et al., 1997). In our study, nine patients were diagnosed with TTV infection, with five patients immunocompromised and three patients suffered from blood disease. However, whether the value of TTV in the infected group indeed existed or was influenced by confounding factors like patients’ immune status, further prospective clinical studies are needed to verify. And further exploration is necessary to deeply understand the potential interaction mechanism between TTV, Rhizomucor delemar and Mucorales infection. The disparity of the different results of microbiome analysis may because of the advanced age of our patients, their relatively lower mortality rate, their immune status, and no restrictions on the type of diseases they exhibited.

In this study, we conducted a comprehensively retrospective study to analyze the clinical characteristics, immune changes, outcome, antibacterial and antifungal adjustment, and microbiota changes in individuals with Mucorales infection and colonization. Furthermore, the efficacy of mNGS was evaluated to distinguish Mucorales infection and colonization. With meticulously designed and analyzed, the study also exists limitations. First, not all patients underwent all clinically laboratory tests, which results in a lack of corresponding comparative diagnostic performance results. The second problem relates to the single-center study. Finally, the sample size is indeed small, and the number of some sample types like CSF, pleural fluid, pus, etc. is little, which may cause a bias in the analysis outcomes.

In this investigation, the performance of mNGS in distinguishing Mucorales infection from colonization, with the differences in patients’ clinical characteristics, antibacterial and antifungal adjustment, and microbiota analysis, were analyzed. We found that mNGS has a high diagnostic efficacy for distinguishing Mucorales infection and colonization, which was better than culture and OMT used in this retrospective research. Moreover, mNGS played a more important role on the guidance of medication in patients with Mucorales infection. Malignant blood tumor, immunocompromised, LOHS, and hsCRP were significant different indicators between patients with Mucorales infection from colonization.