Invasive fungal infections (IFIs) are systemic infections resulting from the establishment of yeasts or molds in deep-seated tissues, blood, and sterile sites. Nearly 70% of these infections worldwide are attributed to invasive candidiasis, followed by cryptococcosis (20%) and aspergillosis (10%)¹. Endemic fungal strains, such as Blastomyces, Histoplasma, ParaCoccidioides, and Coccidioides, also contribute to severe systemic infections. IFIs affect approximately 1.9 million in- dividuals annually, and 3 million people globally suffer from chronic severe fungal infections^1,2.

Although invasive mold infections (IMI) are relatively less common than invasive yeast infections, the incidence of mold-related diseases and hospitalizations is continuously increasing^1,3. The disease burden of mold infections is in- creasing in Korea^4,5. IMI can be caused by Aspergillus spp. and other septated hyaline molds such as Penicillium, Paecilomyces, Acremonium, Scopulariopsis, Fusarium, and Scedosporium, Mucorales, and dematiaceous fungi^3,4. Mold infections are seldom controlled at the initial stage but can escalate in severity due to delayed diagnosis, complicating treatment and leading to higher mortality^1,3. This increase in mold-related diseases and hospitalizations highlights the urgent need for timely and accurate fungal species detection and identification.

This article reviews the epidemiology and impact of COVID-19 on these infections and current nonculture-based diag- nostic methods, such as serological and molecular methods, with the updated consensus definitions for Invasive Fungal Diseases (IFD) by the Cancer/Mycoses Special Interest Group (EORTC/MSG)6.

Invasive mold infection is exacerbated by aggressive medical treatments such as chemotherapy and transplants, invasive devices, such as central venous catheters, and the emergence of new at-risk populations, including individuals with in-fluenza and SARS-CoV-2-associated Aspergillosis^7-9. Invasive aspergillosis (IA) and invasive mucormycosis (IM) are the most prevalent IMIs. The mortality rate for candidemia exceeds 25%, while IA results in death in 40% to 90% of immuno- compromised patients. The rate approaches 85% for un- treated IA and nearly 100% for untreated IM, depending on the patient's underlying conditions^2,9.

Aspergillus fumigatus is the leading cause of culture-positive IA, followed by A. flavus, A. niger, and A. terreus. Infections from nonfumigatus Aspergillus species are more likely to become disseminated¹. Mucormycosis is primarily caused by fungi from the order Mucorales, which frequently results in respiratory infections that can extend to the sinuses, eyes, or central nervous system. IM is commonly caused by Rhizopus spp., while other species, including Mucor spp., Lichtheimia spp., Apophysomyces spp., Cunninghamella spp., and Saksenaea spp., also cause IMI^2,4.

Other hyaline molds, Fusarium spp., Scedosporium spp., Penicillium spp., and Acremonium spp., are also implicated in IMIs. Fusarium spp. typically disseminates in patients with a hematologic malignancy and hematopoietic stem cell trans- plant (HSCT) recipients but appears localized in solid organ transplant (SOT) recipients, leading to better outcomes than in neutropenic patients¹⁰. Scedosporium spp. can cause condi-tions ranging from mycetoma to lethal disseminated infections, especially in SOT recipients from nearly drowned donors¹¹. Dermatiaceous fungi such as Alternaria spp., Exophiala spp., and Curvularia spp. can cause diseases ranging from skin lesions to disseminated infection in high-risk patients¹⁰. SOT recipients frequently exhibit skin conditions, while HSCT re- cipients are prone to lung complications, which are particularly problematic in these patients^9,11. Indeed, Aspergillus spp., Fusarium spp., and Mucorales are known to cause healthcare-associated infections in chronic disease patients¹².

COVID-19 has impacted the incidence of fungal infections. COVID-19-associated pulmonary aspergillosis (CAPA) was first identified in China in early 2020, affecting about 10% of invasively ventilated COVID-19 patients and about 20.1% of those with acute respiratory failure requiring ventilation. A. fumigatus is the predominant pathogen causing CAPA linked to azole-resistant Aspergillus strains⁸. The patients most commonly affected are those with acute respiratory failure due to COVID-19, particularly patients treated with systemic corticosteroids or tocilizumab¹³. Unlike influenza-associated pulmonary aspergillosis, CAPA is usually diagnosed about 8 days after ICU admission and has a mortality that exceeds 40%. During India's second COVID-19 wave in early 2021, over 47,500 COVID-19-associated mucormycosis (CAM) cases were reported. The prevalence of CAM among hospitalized COVID-19 patients was 0.27% in India and about 1~2% in Europe^8,14. CAM commonly leads to rhino-orbital-cerebral and pulmonary diseases in COVID-19 patients with high mortality and extensive necrosis, often necessitating disfiguring surgical debridement. Major risk factors include uncontrolled diabetes and systemic corticosteroid overuse¹³.

Several criteria, such as those from EORTC/MSG, Bulpa, Blot, and the modified Aspergillosis in ICU (AspICU), have been recommended for the timely diagnosis and management of IFIs in various clinical settings^6,7,15. Definitions by the EORTC/ MSG for immunocompromised individuals were first proposed in 2002 and later updated in 2008 and 2020, incorporating host factors, clinical criteria, and mycological criteria6,^15,16. However, the AspICU criteria proposed for ICU patients do not include microbial biomarkers such as the galactomannan (GM) antigen and Aspergillus polymerase chain reaction (PCR)⁷.

According to the EORTC/MSG definition, a case confirmed by microscopy, histopathology, and culture from a sterile site is "proven". The diagnosis is "probable" if three criteria are met: the presence of at least 1 host factor, a clinical feature, and mycologic evidence. Conversely, a case that meets the appropriate host factors and clinical evidence but lacks mycological support is considered "possible"¹⁶. Due to increasing data on Aspergillus GM thresholds and the new PCR-based diagnostic tests, the 2020 update to the EORTC/MSG criteria includes modifications in mycological criteria and additional host factors, such as SOT and hematologic malignancies. This document underscores the importance of laboratory diagnosis in identifying IFD, highlighting the roles of GM testing and PCR assays for Aspergillus6 (Table 1). Adopting the 2020 EORTC/MSG criteria led to a significant 27.3% reduction in the diagnosis of probable IPA, contributing to a more uniform patient classification of treatment response, prognosis, and mortality. This consistency enhances the reliability of mortality predictions compared to the 2008 criteria¹⁷.

1. The conventional method: microscopy, histopathology, and culture-based tests

Histopathologic demonstration of tissue invasion by hyphae or mold recovery from a culture of a specimen obtained from a clinically sterile site is a "proven" IMD16. This category can apply to any patient, regardless of their immunocompromised status. However, these conventional methods have inherent limitations, including a prolonged turnaround time, the inva- sive nature of the samples, and lower sensitivity. The propor- tion of proven cases was relatively low, from 1.8% to 27.0. Blood culture sensitivity for aspergillosis was markedly low at 1~5%, in contrast to 50~95% for candidemia^1,19. Recovery of Aspergillus spp. from blood cultures rarely indicates endo- vascular disease and is almost always contamination6. A nationwide study in Korea found that only 8% of IPA cases were proven with these methods²⁰.

2. Galactomannan (GM) antigen assay

GM is a polysaccharide cell wall component released by Aspergillus spp. during growth. It can be detected in serum, bronchoalveolar lavage (BAL) fluid, cerebrospinal fluid (CSF), and others^15,21. The 2002 EORTC/MSG definition included Aspergillus antigen detection, such as GM, as a criterion for probable IA¹⁶. However, the Platelia Aspergillus EIA (Bio-Rad, Marnes-la-Coquette, France), an immune-enzymatic sandwich microplate assay using a rat antigalactofuranose monoclonal antibody (EB-A2) was first approved in 2003 and remains the only FDA-approved GM assay, despite the availability of multiple commercial assays. Initially, the cut-off for GM was between 1.0 and 1.5 to minimize false positives, but in 2008, this threshold was lowered to 0.5 to improve performance in adult hematology patients^21,22. The 2020 EORTC/MSG definition raised the cut-off back to 1.0 for individual plasma, serum, BAL fluid, or CSF samples, setting specific criteria for simultaneous positive results in serum and BAL fluid: ≥0.7 for serum and ≥0.8 for BAL⁶.

The GM assay has high sensitivity and moderate accuracy in diagnosing IA in immunocompromised patients. In patients with hematologic malignancies, the GM assay has higher sensitivity and specificity in BAL fluid than serum^15,21. Sensitivity is low in SOT recipients due to less frequent angioinvasion. False-positive results are usually reported in pediatric patients without clinical signs of IA. Galactofuranose in various fungi and certain substances has led to reports of false-positive results_²¹. These false-positive results can arise from colonization by fungi, such as the genera Fusarium, Histo- plasma, Penicillium, Alternaria, Paecilomyces, or Geotrichum, the administration of piperacillin-tazobactam or β-lactams, patients receiving total parenteral nutrition, such as Plasma-Lyte (Baxter, Deerfield, IL, USA), NP2 Enfant AP-HP parenteral nutrition (Fresenius, Kabi, Sevres, France), and intravenous gluconate solution^23,24.

Additionally, the cross-reactivity of BAL fluid samples with Mycoplasma pneumoniae or anesthetic drugs has been re- ported^24,25. Patients with mucositis or compromised intestinal barriers have false-positive GM results due to dietary factors. When using BAL fluid for testing, lung transplant recipients have a significantly higher rate of false-positive results due to extensive airway colonization^25,26. These inaccuracies lead to unnecessary diagnostic procedures, treatments, and radiation exposure from CT scans aimed at detecting IPA, adverse reactions from antifungal medications, and increased health- care costs due to unnecessary tests and treatments. False negatives or reduced detection has also been reported in patients with chronic granulomatous disease or hyperimmuno- globulin E (hyper-IgE) syndrome (formerly Job syndrome) despite their increased risk of IA²⁷.

3. (1,3)-β-D-glucan (BDG) assay

(1,3)-β-D-glucan (BDG), a polysaccharide component of fungal cell walls, is a pan-fungal antigen detected in various species, including Candida spp., Pneumocystis jivoveci, Asper- gillus spp., Acremonium spp., and Fusarium spp. with ex- ceptions such as Cryptococcus spp., Mucorales, and the yeast phase of Blastomyces dematitidis22. The commercial BDG assay, the Fungitell Assay (Associates of Cape Cod Inc., East Falmouth, MA, USA) has received FDA approval and numerous other BDG assays are available, each with various cut-off values: Fungitell assay, cut-off value 60~80 pg/mL; the Wako β-glucan test, cut-off value 11 pg/mL (Wako Pure Chemical, Osaka, Japan); and the Fungitec-G, cut-off value 20 pg/mL (Seikagaku, Tokyo, Japan)^28,29.

The revised 2020 EORTC/MSG criteria exclude BDG detec- tion as a microbiologic criterion for IFD since BDG detection is not specific to a single IMD⁶. False positives can occur in concurrent bacteremia such as Pseudomonas aeruginosa and Streptococcus species, certain β-lactam antibiotics, specific antitumor chemotherapy agents, various fractionated blood components, and hemodialysis membranes. However, BDG detection is considered suitable for diagnosing probable IFD in specific clinical situations, such as patients with hematologic malignancies, post-hematopoietic stem cell transplantation neutropenia, and selection of high-risk ICU patients with complications from gastrointestinal surgery suspected of having an infection^29,30. Additionally, due to its relatively high NPV, BDG testing is considered useful for excluding IFD in patients²⁸.

4. Polymerase chain reaction (PCR) assay

The PCR assay is recommended for Aspergillus spp. and Pneumocystis jirovecii infections, and serological testing will likely enhance the diagnosis. Molecular diagnostics represent the only nonclassical mycological approach for IM^4,31. The International Fungal PCR Initiative (FPCRI) standardized DNA extraction protocols from serum, plasma, and whole blood to improve diagnostic accuracy³². BAL fluid is preferred for detecting respiratory pathogens, while debrided tissue is favored for skin and soft tissue infections. In formalin-fixed paraffin-embedded (FFPE) tissue, amplification of fungal DNA by PCR combined with DNA sequencing can be used when molds are seen. However, caution is needed with FFPE tissue due to a low DNA yield and potential amplification interference from colonizers³¹.

The 2020 updates to the EORTC/MSG definitions now include Aspergillus-specific plasma, serum, whole blood, and BAL fluid PCR assays as a mycological criterion for probable IA⁶. The European Aspergillus PCR Initiative (EAPCRI) Working Group of the International Society of Human and Animal Mycology has validated Aspergillus PCR assays in plasma, serum, and whole blood, except BAL fluid^19,32. Aspergillus PCR assays have moderate accuracy for diagnosing IPA in immunocompromised patients. They require two positive results for confirmation to achieve higher specificity. In high-risk hematology patients, these assays in blood show a high NPV of 94% and a PPV of 70%, reducing the need for empir- ical antifungal therapy³³. Aspergillus PCR assays in patients not previously treated with antifungals show a sensitivity of 71.4% and a specificity of 92.3%, with a PPV of 62.5% and an NPV of 98.3%; however, in patients with Aspergillus-active prophylaxis, the PPV drops to 5.4% while the NPV increases to 100%, indicating that PCR screening is more effective in patients not on primary antimold prophylaxis³⁴.

PCR assays for Aspergillus DNA in specimens such as BAL fluid, CSF, and tissue are useful for diagnosing IA in SOT recipients or patients under long-term immunosuppressive therapy⁶. However, BAL fluid assays may not differentiate between colonization and active infection, with a PPV of about 72% in nonhematology patients^14,35. While Candida PCR assays lack standardization, Aspergillus PCR assays are well-standardized, aiding in result comparison and detecting azole resistance mutations. Still, the absence of an FDA-endorsed Aspergillus PCR assay limits its global use³¹. Despite the lack of significant differences in mortality between the redefined possible and probable IA groups, elevated DNA levels (greater than 150 copies/mL) in quantitative PCR have been associated with higher 90-day mortality rates^6,33,36. The false-positive results in Aspergillus PCR assays are often due to environmental contamination, cross-reactions with other fungi, such as Penicillium spp., or respiratory tract colon- ization³⁷.

The absence of serological tests to confirm the diagnosis of IM highlights the need for a molecular assay. The increasing use of Mucorales-specific PCR assays enhances accuracy in detecting DNA in blood and BAL fluid, using dual loci (ITS and TEF1a) for DNA barcoding^31,38. Mucorales DNA has been detected in blood 3 to 68 days before conventional methods, eight days earlier than histology or culture, and three days before imaging³¹. Quantitative PCR in serum samples has been proven to be of high prognostic value. Patients with decreasing fungal loads have better survival rates. However, more comprehensive studies are needed to clarify its sensitivity and role in patients treated with antifungals and those with rhino-orbital-cerebral mucormycosis^31,39.

IMIs are exacerbated by medical interventions, such as chemotherapy and transplants, invasive devices, and the emergence of new at-risk populations, including those with influenza and SARS-CoV-2-associated aspergillosis. Diag- nostic mycological approaches have evolved beyond trad- itional microscopy and culture to include serology-based and molecular approaches, offering rapid, sensitive, and specific detection. This fact is particularly vital for timely and accurate diagnosis, allowing the initiation of appropriate treatment. The 2020 EORTC/MSG criteria have integrated these newer diagnostic tools, reflecting advancements in nonculture-based methods that improve diagnostic accuracy and patient man- agement in fungal infections. These molecular diagnostics are pivotal in differentiating invasive infections from colo- nization, especially in immunocompromised patients, and have significantly refined the approach to invasive mold infection detection and management.