Diagnosing invasive fungal infections in the laboratory today: It's all good news?

Abstract

Despite the advances in medical science, invasive fungal infections (IFI) remain a diagnostic challenge. The increasing prevalence of IFI, driven by immunosuppressive therapies, advances in intensive care and emerging pathogens, underscores the need for early and accurate diagnosis. This review evaluates current laboratory methods for the diagnosis of IFI, highlighting their strengths and limitations.

Conventional techniques, including fungal culture, direct microscopy, and histopathology, remain the gold standard for the diagnosis of proven IFIs. These methods allow pathogen isolation, species identification and antifungal susceptibility testing. However, these techniques have limitations in terms of sensitivity and turnaround times. Although microscopy is a rapid technique, its sensitivity and species discrimination profile are limited.

Modern serological assays, such as β-d-glucan and galactomannan detection, have improved the diagnostic accuracy of probable IFI cases. Integration of these assays with clinical and radiological findings, enables earlier intervention, although this is accompanied by an increased risk of false positives and necessitates careful clinical correlation. Molecular diagnostics, particularly polymerase chain reaction (PCR), allow rapid, species-specific identification directly from clinical samples. The advent of MALDI-TOF mass spectrometry has further improved diagnostic efficiency, particularly for yeast identification, although challenges remain for filamentous fungi.

Innovative techniques, such as metagenomic sequencing, lateral-flow assays, and loop-mediated isothermal amplification, offer the potential for rapid and precise detection, even in resource-limited settings. The combination of conventional and innovative methods provides a comprehensive diagnostic framework. The continuous refinement of these tools, in conjunction with multidisciplinary collaboration, is imperative to improve the early diagnostic and targeted treatment of patients with IFI.

Resumen

A pesar de los avances en la ciencia médica, las infecciones fúngicas invasivas (IFI) siguen siendo un desafío diagnóstico. La creciente prevalencia de las IFI, impulsada por terapias inmunosupresoras, avances en cuidados intensivos y patógenos emergentes, subraya la necesidad de un diagnóstico temprano y preciso. Esta revisión evalúa los métodos de laboratorio actuales para el diagnóstico de las IFI, destacando sus fortalezas y limitaciones.

Las técnicas convencionales, incluyendo el cultivo micológico, la microscopía directa y la histopatología, siguen siendo el gold-estándar para el diagnóstico de IFI probada. Estos métodos permiten el aislamiento del patógeno, la identificación de especies y realizar pruebas de sensibilidad antifúngica. Sin embargo, estas técnicas tienen limitaciones de sensibilidad y tiempos de respuesta. Aunque la microscopía es una técnica rápida, tiene una sensibilidad limitada y no permite diferenciar las especies.

Las técnicas serológicas modernas, como la detección de β-D-glucano y galactomanano, han mejorado la precisión diagnóstica de los casos probables de IFI. La integración de estos ensayos con hallazgos clínicos y radiológicos permite una intervención más temprana, aunque esto conlleva un mayor riesgo de falsos positivos y requiere una cuidadosa correlación clínica. Los métodos moleculares, particularmente la reacción en cadena de la polimerasa (PCR), permiten una identificación rápida y específica de especies directamente a partir de muestras clínicas. La llegada de la espectrometría de masas MALDI-TOF ha mejorado aún más la eficiencia diagnóstica, especialmente para la identificación de levaduras, pero no es tan fiable para identificar hongos filamentosos.

Técnicas innovadoras como la secuenciación metagenómica, los ensayos de flujo lateral y la amplificación isotérmica mediada por bucle ofrecen el potencial para una detección rápida y precisa, incluso en países con recursos limitados. La combinación de métodos convencionales e innovadores proporciona un marco diagnóstico integral. El continuo perfeccionamiento de estas herramientas, junto con la colaboración multidisciplinaria, es imperativo para mejorar el diagnóstico precoz y el tratamiento dirigido en los pacientes con IFI.

Until the year 2020, around 140,000 fungal species were identified according to their different morphological (macro- and microscopic), physiological, and molecular characteristics. However, it is assumed that only about 700 have any human interaction, whether as pathogens or commensal species.⁷⁸

In recent decades, longer life expectancy and the advances in medicine have increased the susceptibility of humans to infections, including invasive fungal infections (IFI), especially due to the use of immunosuppressive therapies and advances in the management of critically ill patients. However, fungal infections are still underestimated by both the community and public health authorities.⁴⁷

Available estimates of fungal disease incidence and mortality are uncertain. Since 2013, the Leading International Fungal Education (LIFE) portal has facilitated the estimation of the burden of serious fungal infections on a country-by-country basis. According to these estimates, more than 300 million people suffer from systemic fungal infections worldwide every year. Of these, 6.5 million are affected by an immediately life-threatening fungal disease, and about 2.5 million dies from a fungal infection.^12,24 In addition, some pathogenic fungi benefit from climate change, gradually adapting to higher temperatures and becoming more frequent and possibly more virulent. Thus, an increase in IFIs is expected in the coming years.⁶⁸

IFIs can be caused by opportunistic fungi such as Candida, Cryptococcus, Aspergillus, Zygomycetes, Fusarium, Scedosporium, Lomentospora, as well as uncommon yeasts and molds, and those causing endemic mycoses including Blastomyces, Coccidioides, Emergomyces, Histoplasma, Paracoccidioides, Sporothrix and Talaromyces marneffei.

The majority of IFIs caused by opportunistic fungi are observed in immunocompromised patients (neutropenic or transplant recipients), critically ill patients, diabetics, as well as in those with severe viral infections, such as HIV, influenza, or COVID-19. Meanwhile, endemic mycoses can affect healthy people producing local infections or disseminated disease.¹²

In most cases, the successful outcome of patients with IFI is directly related to early diagnosis and prompt initiation of appropriate antifungal therapy. However, precise diagnosis of fungal infections is still hampered by the lack of specific clinical and imaging signs. Then, despite the advances achieved in recent decades and the incorporation of new diagnostic techniques, accurate and timely diagnosis remains a major challenge in clinical practice.²⁷

Current laboratory procedures for IFI diagnosis include direct microscopy, fungal culture and pathogen identification, histopathology, antigen detection, and PCR tests for fungal DNA detection.⁶³ It is important to note that the diagnostic classification of IFI (proven or probable), according to the criteria of published diagnostic guidelines for different host conditions, will depend on the technique used for diagnosis. Techniques that detect the fungal pathogen in sterile specimens by culture or histopathology will provide a diagnosis of proven IFI, whereas detection of antigens or fungal DNA by serological or molecular techniques, respectively, only provide a diagnosis of probable IFI.^11,26,32 This review presents the up-to-date situation of laboratory diagnosis of fungal infections, highlighting the advantages and limitations of the techniques and methodologies now available in the hospital setting.

1

Laboratory procedures for a proven IFI diagnosis

Despite the improvements in mycological diagnostic techniques reached in the last few years, only the mycological culture, the microscopy, or the histopathology of a sterile sample, obtained in a sterile manner, are still the gold standard methods to achieve a proven diagnosis of IFI. The main advantage of fungal culture is the possibility to isolate the pathogenic fungi, which will allow not only the subsequent species identification but also the in vitro antifungal susceptibility study. Conversely, the long response time and low sensitivity are the major drawbacks of this conventional methodology. If pathogenic fungi cannot be detected through culture or histopathological methods, a probable diagnosis of IFI can be established using serological, molecular, and other advanced techniques, along with an assessment of host factors and the clinical condition of the patient (Table 1).

Table 1.

Comparative summary of laboratory techniques used for the proven diagnosis of invasive fungal infections.

Technique	Strengths	Weaknesses	Diagnostic utility	References
Direct microscopy	• Rapid results (<5 min)• Easy to realize• Economic	• Limited sensitivity (50%)• Lack of specificity• Unable to identify genus or species• Expertise required	• Candidiasis• Hyphomycosis• Zygomycosis• Pheomycosis• Cryptococcosis	^26,65
Mycological culture	• Isolation of the fungal pathogen• Identification at species level• Detection of antifungal resistance• Long-term storage of strains• Quantitative information	• Time-consuming (>48 h)• Long-delayed targeted treatment• Prone to contaminations• False-negative results possible (50%)• Requires sterile/deep specimens to prevent false-positive results	• Candidiasis• Hyphomycosis• Zygomycosis• Pheomycosis• Cryptococcosis	^9,65
Blood culture	• Best diagnostic technique for candidemia and other yeast systemic infections• Detection of antifungal resistance	• Time-consuming (>48 h)• False-negative results possible (50%)• Lysis-centrifugation technique is recommended for endemic mycosis• Not useful for systemic filamentous fungal mycoses (with exceptions)	• Candidemia• Systemic mycosis by: ∘ Fusarium ∘ Scedosporium ∘ T. marneffei ∘ Lomentospora	^6,32,50,69
Histopathology	• Detection of fungal invasion in tissues• Detection of inflammatory host response• Gold standard for proven IFI	• Laborious and time-consuming• Similar histopathologic appearance of several fungus• No identification provide at genus or species level• Limited sensitivity	• Any mycoses caused by yeasts, filamentous, or dimorphic fungi	^25,37
Fungal identification from culture:
• Morphological identification	• Most common technique for identifying filamentous fungi	• Not useful for yeast• Time-consuming• May require the use of specific culture media• Requires trained personnel or reference laboratories	• Any mycoses caused by filamentous or dimorphic fungi	^34
• Chromogenic media	• Easy to use• Presumptive identification of most common Candida species• Cost-effective• Especially useful in mixed infections and surveillance cultures	• Time-consuming (48 h)• Rare or infrequent species cannot be identified	• Candidiasis	^56,61
• Biochemical identification	• Useful for yeasts common species• Cost-effective	• Laborious and time-consuming (>48 h)• High rate of misidentification for rare or new species• Displaced by mass spectrometry in clinical laboratories	• Common yeasts infections	^46
• MALDI-TOF	• Rapid results (<10 min)• Easy performance• Reduced cost per analysis• Accurate identification at species level of yeasts and some molds• Differentiation of closely related species• Possible direct identification from blood culture	• Requires colony growth on solid culture medium• Prior extraction step is required• Databases precise continuous updates to cover rarest and emerging species• Identifying most molds remains challenging: ∘ Limited number of mold spectra ∘ Spectra contamination by culture medium ∘ Longer protein extraction process ∘ Trained personal required	• Yeasts infections• Aspergillosis• Fusariosis• Scedosporiosis	^6,59
• Molecular identification by automated platforms	• Rapid results (<90 min)• Easy performance• High sensitivity and specificity• Accurate identification of common Candida and Cryptococcus species	• High price per test• Requires specialized equipment• Few species included on fungal panels• Only for blood culture isolates	• Fungemia	^15,44
• DNA sequencing for fungal isolates	• Gold standard for fungi molecular identification• Highest specificity• Cost-effective• Massive public databases	• Laborious and time-consuming (>24 h)• High technical expertise• Requires PCR amplification step• Data interpretation may be needed• Not widely commercialized	• Any mycoses caused by yeasts, filamentous, or dimorphic fungi	^76

1.1

Direct microscopic examination

In certain clinical samples, direct microscopy offers a significant diagnostic advantage that surpasses that of culture methods alone. Additionally, one of the key benefits of microscopy is the prompt availability of results, typically less than 1/2 h following the specimen's arrival in the laboratory. Then, quick processing and reporting of microscopy results are critical, as an IFI delayed diagnosis worsens the patient's prognosis due to the delay in starting appropriate antifungal treatment.

Direct visualization of fungi in sterile fluids or tissues facilitates the confirmation of IFI, particularly when fungal growth observed in culture may be attributed to contamination.²⁶ Centrifugation of liquid samples enhances the concentration of fungal elements and microscopic visualization using staining techniques, including Gram or Giemsa stains, India ink, or fluorescent brighteners (such as calcofluor white or blankophor), increasing the probability of their detection.

In addition, microscopy can distinguish whether an infection is caused by yeasts (Candida, Cryptococcus, etc.), septate molds (Aspergillus, Fusarium, etc.), or by non-septate molds as the organisms of the Zygomycetes class, which affects the choice of antifungal treatment.

Direct microscopy is particularly crucial when non-septate fungi are involved, as these fungi are often inadequately recovered by culture due to the damage sustained during refrigeration or tissue homogenization.⁶⁶ However, the global sensitivity of direct microscopy is not too high (around 50%), so a negative result does not exclude the possibility of IFI, but a positive direct examination in the clinical context of possible IFI substantially supports the diagnosis.

Microscopic observation of narrow, septate hyphae, branched at an acute angle suggests the presence of a hyphomycete, such Aspergillus, Fusarium, or Scedosporium, amongst others. On the other hand, the observation of wide, non-septate hyphae with an angle greater than 90° is suggestive of Zygomycetes. However, it should be noted that microscopic morphology in clinical samples does not allow identification of the genus or species of the fungus observed, as the structures visualized are indistinguishable from each other. Definitive identification of the species involved can only be achieved by mycological culture or fungal DNA detection.

Therefore, when IFI is clinically suspected, direct microscopic examination is recommended on all sterile samples (collected in a sterile manner) and also on difficult-to-obtain samples such as bronchoalveolar lavage (BAL), biopsies, and drainages. Then, biological fluids like cerebrospinal (CSF), pleural, synovial, and also BAL should be examined by microscopical visualization with staining techniques for fungal detection, preferably with optical brighteners; but, before that, the amount of sample sent to the laboratory should be checked and, if this is scarce, the sample should be reserved for culture, as the diagnostic value of culture is superior to that of microscopy.

Finally, it is important to note that microscopic visualization of fungal structures in clinical specimens requires trained mycologists who can distinguish these structures from human cells, artifacts, or other entities.

1.2

Mycological culture

Despite all the new diagnostic techniques incorporated in the laboratory in recent decades, the mycological culture remains the gold standard procedure for the IFI diagnosis, as it is the only technique that allows the isolation, complete identification, and preservation of the causal pathogen, as well as the subsequent study of its antifungal susceptibility.

To isolate the pathogenic fungus causing IFI it is necessary to collect representative biopsies of the involved tissue or sterile fluids such as blood, urine or CSF and inoculate them on appropriate culture media. Tissues should be sliced with scissors or scalpel and inoculated in small pieces into the culture medium to facilitate growth. The use of mechanical homogenizers is not recommended as it may destroy non-septate fungi. If sufficient sample volume is available, biological fluids (CSF, pleural, peritoneal, etc.) should be concentrated by centrifugation (1500–2500 g for 10 min) or filtration (0.2 μm pore size) prior to plating both the sample and the filter.

Although most fungi grow properly on culture media commonly used in clinical bacteriology (blood agar or chocolate agar), inoculating the samples on specific mycological media, like Sabouraud dextrose agar supplemented with chloramphenicol (SDCA), brain heart infusion agar (BHIA) or inhibitory mold agar (IMA) is recommended to improve the fungal growth; IMA is suggested for the proper isolation of filamentous fungi.⁹

After seeding, plates should be incubated at 30 °C, as this is the optimum growth temperature for most pathogenic fungi. If the incubator is not humidified, it is advisable to place a container of water near the culture plates. For valuable or difficult-to-obtain samples, fungal cultures should be incubated for 3–4 weeks before discarding. Incubation at 30 °C and 37 °C separately should be reserved when infection by Zygomycetes or dimorphic fungi is suspected.²⁰

Although culture remains the gold standard methodology, it is associated with low sensitivity (50%). A definitive diagnosis can be obtained if the microorganism grows in cultures from tissues obtained in sterile form or sterile fluids. However, when a filamentous fungus from a non-sterile fluid grows, such as BAL, the possibility of commensal organisms or environmental contamination must be considered.⁶⁶

Blood culture remains the best technique for the diagnosis of fungemia, despite its overall low sensitivity (50–95%). The same automated continuous monitoring systems used for bacteria are recommended for the recovery of yeasts from blood, but if a systemic infection by a dimorphic fungus is suspected, the lysis-centrifugation technique seems to offer greater advantages. However, for the recovery of most filamentous fungi, neither of these two blood culture systems has proven effective.⁴⁵

Currently, two blood culture automated systems are commercially available: BACT/ALERT® 3D (bioMérieux, France) and BD BACTEC™ FX (Becton Dickinson, USA). These systems detect microbial growth by measuring CO₂ production using colorimetry and fluorescence, respectively. The overall sensitivity of blood cultures for detecting yeasts ranges from 50% to 95%, while for molds the sensitivity is even lower, typically between 1% and 5%.⁶ For yeasts, blood cultures require 1–3 days for growth and an additional 1–2 days for organism identification and susceptibility testing, which frequently results in significant delays in the initiation of targeted treatment. For candidemia, the turnaround time can vary according to the species, typically ranging from 14 to 72 h, as some species, like Candida glabrata, grow more slowly than others. However, for deep-seated candidiasis without candidemia, the sensitivity of blood culture is significantly lower, and the detection time is considerably longer.⁷⁰

Blood cultures often have a low diagnostic yield in invasive mycoses caused by filamentous fungi, because these infections usually do not have a detectable hematogenous phase of dissemination. Therefore, growth of filamentous fungi in blood culture should be considered as contamination, with the notable exception of mycoses caused by Aspergillus terreus, Fusarium species, Pseudoallescheria/Scedosporium species, Lomentosporaprolificans or T. marneffei. Blood cultures are positive in 40%-50% of invasive cases of fusariosis and 70% of systemic infections by Pseudoallescheria/Scedosporium or L. prolificans. Then, the isolation of these species in blood typically indicates an invasive infection and is thus diagnostic of proven IFI.^33,51

The interpretation of culture results should always be related to the clinical situation of the patient. The isolation of a dimorphic fungus in any clinical specimen is of diagnostic value, although these infections are uncommon in non-endemic countries. Isolation of a fungus in tissue biopsies or sterile fluids (such as blood or CSF) is significant; however, it may also be due to accidental contamination, so it is recommended to confirm tissue invasion by histopathological studies to asses the existence of a host response to fungal aggression, although the inflammatory response may be diminished or abolished in immunocompromised patients.

Finally, it is important to remember that the value of fungal isolation in culture, even though it may be from a common contaminant, becomes significantly greater if the patient meets the clinical conditions necessary to be considered at high risk for IFI.

1.3

Methodologies for identification

Isolating the fungal pathogen from culture helps assess species identification and resistance to antifungals. However, simply growing pathogenic fungi in culture media is not enough to identify the specific fungal species causing the infection. Additional methods are needed for specific identification, which can lead to a more effective treatment.

1.3.1

Morphological identification of molds

Unlike yeasts, which cannot be identified by the morphological characteristics of their colonies on culture media, in clinical practice molds are identified by their macro- and microscopic properties. All filamentous fungi grown on a culture medium from a representative clinical sample in a patient at risk of IFI should be identified to species (or species complex) level by referral to a specialist laboratory if necessary.

The typical identification of molds relies on examining the macroscopic characteristics of the fungal colony, such as color, texture, and topography, as well as its distinctive microscopic features. These include reproductive structures like conidiogenous cells, vesicles, conidiophores, and sporangia, as well as other structures such as cleistothecia, Hülle cells, and chlamydospores. To facilitate the identification of the species involved, it is advisable to use the dichotomous keys and descriptions published in reference books.³⁵

The genus identification in Zygomycetes can be carried out from primary cultures of clinical samples on Sabouraud dextrose agar (SDA). However, species identification requires the use of more specific culture media, such as potato glucose agar (PDA) or malt extract agar (MEA), incubated in the dark at 25 °C and 35 °C, as some species are thermophilic.³⁰

For the isolation of hyphomycetes in clinical practice, it is common to use nutrient-rich culture media such as SDCA. However, the use of such media is not recommended for the identification of most opportunistic fungi as it stimulates mycelial production to the detriment of the sporulating structures, essential for the recognition of these fungi. To encourage sporulation, it is useful to use media like PDA, cornmeal agar (CMA), or oatmeal agar (OA) which provide carbon sources other than glucose. Nutrient-poor media such as potato-carrot agar (PCA) are mainly recommended to stimulate the sporulation of dematiaceous hyphomycetes.³⁰

1.3.2

Presumptive identification of yeasts by chromogenic culture media

Chromogenic media have been widely used in clinical microbiology to detect and identify bacteria, and so for Candida identification since 1994. This type of culture media allows the growth of specific pathogens with high specificity using chromogenic enzyme substrates. For mycological purposes, they are suitable for nonsterile samples, stimulating the growth of specific Candida species while inhibiting the growth of other microorganisms, such as bacteria.⁵⁷

All chromogenic media available for yeast detection include a substrate for the β-hexosaminidase enzyme to differentiate Candida albicans, the most frequent and clinically important Candida species. Combining chromogenic substrates, colonies from different species display distinct colors and morphology, enhancing the presumptive identification of C. albicans, Candida tropicalis, C. glabrata, Candida parapsilosis, Candida. krusei, and, recently, Candida auris.¹³ Examples of the chromogenic media commercially available include CHROMagar® Candida (CHROMagar, France), CHROMagar® Candida Plus (CHROMagar), CandiSelect™ 4 (CS4, Bio-Rad, France), HiCrome®Candida (HiMedia, India), chromID™ Candida Agar (CCA, bioMérieux, France), and Brilliance™ Candida Agar (Oxoid, UK), among others.

Chromogenic media have the unique ability to easily differentiate the presence of multiple species in a clinical sample. This makes them particularly useful for identifying mixed infections and for conducting epidemiological surveillance cultures where more than one type of yeast may be present. In addition, these media can be supplemented with antifungals, making it easy to isolate and identify fluconazole-resistant species such as C. auris, or emerging resistant strains of C. parapsilosis.⁶²

1.3.3

Biochemical phenotypic identification

Biochemical systems have been used for decades to identify the most common yeasts and bacteria in medical practice. Before the advent of mass spectrometry in microbiology laboratories, biochemical methods were the most widely used for yeast identification.

Using multiple microwell galleries, these systems assess the ability of yeasts to assimilate different nutrients, including sugars, organic acids, and enzyme substrates. To identify yeasts, various methods such as turbidimetric or colorimetric can be used. Each well produces a numerical result and, when combined, these numbers create a code that can be compared to a library of codes to determine the species.

For these methods, it is necessary to use a pure yeast colony, inoculate each gallery well with the yeast suspension, and then incubate the gallery at 37 °C for 24–48 h. After incubation, results can be read. Some commercial systems like API® 20C AUX (bioMérieux) or Auxacolor 2 (Bio-Rad) require manual reading, while in others, such as VITEK® 2 YST ID card (bioMérieux), the wells of the gallery are read automatically.⁴⁷

The accuracy of these systems for yeast identification is higher for commonly detected species than for rare ones, but the global misidentification rates remain high.³⁶ Furthermore, these techniques provide inaccurate identification of the emergent pathogen C. auris, mistaking this species with other phylogenetically related, such as Candida haemulonii, Candida famata, Candida sake, or Rhodotorula glutinis. The VITEK® 2 system database has recently been updated to include C. auris. However, the updated version shows limited capability to distinguish between C. auris and closely related species.³

1.3.4

Identification by mass spectrometry

In the last decade, mass spectrometry-based methods have become increasingly common in microbiology labs due to their capability to rapidly, easily, accurately, and cost-effectively identify bacteria and yeasts.

At present, the most popular method of mass spectrometry is Matrix-assisted laser desorption/ionization (MALDI-TOF MS). This system involves identifying unique characteristics of extracted proteins, particularly ribosomal and membrane proteins. The protein profile obtained for each sample is compared with comprehensive profile databases, allowing the identification of microorganisms up to the species level within a few minutes.^6,71 The accuracy of identification relies on previously created databases (libraries) that include thousands of reference spectra obtained from strains strictly identified by DNA sequencing.

MALDI-TOF MS enables direct species identification from a small amount of yeast colony. Although the colony can be directly transferred to the target plate and the matrix is added, in some cases a pre-extraction step is necessary. For this purpose, two extraction methods are available: (i) On-plate (or rapid) protein extraction using formic acid (the most common), and (ii) Off-plate (or long/full) extraction using ethanol/formic acid.⁶⁰

Unlike phenotypic techniques, MALDI-TOF MS methodology provides accurate and quick identification of yeasts. Furthermore, it significantly improves yeast identification by differentiating species within Candida complexes, including the C. parapsilosis complex, C. glabrata complex, and C. haemulonii complex.⁶⁰

The performance of certain spectrometry-based methods commercially available has been assessed. In a comparative study of the VITEK MS (bioMérieux) and MALDI Biotyper (Bruker, Germany) with 157 isolates, including non-C.albicans Candida species and rare yeast species, both systems showed a high sensitivity in yeast identification (96.8% and 98.7%, respectively).⁷¹ The performance of the Autof MS1000 (Autobio Diagnostics, China) and Vitek MS systems was compared using 1228 yeast isolates from 14 different species. The identification accuracy of all species complexes was 98.9–100% with Autof MS 1000 and 79.1–96.3% with VITEK MS. Both systems showed good performance in identifying C. auris.⁷⁹

Because over the past ten years MALDI-TOF MS has accumulated significant experience, this technique is now universally employed as the primary identification method for yeasts in clinical laboratories. However, identifying pathogenic filamentous fungi with MALDI-TOF MS remains challenging, partly due to the current limited number of mold spectra available in the libraries of these systems compared to yeasts. Furthermore, molds often grow within the solid media, which complicates the harvesting process and results in agar contamination of the spectra, hindering MALDI-TOF MS identification. In this way, a novel culture medium, ID Fungi Plates (Conidia, France) was developed to allow easier and faster harvesting of the isolates, and improve the identification of molds by MALDI-TOF MS.⁶¹

MALDI-TOF has been used for more than 10 years to identify species directly from positive blood culture bottles and, thus, significantly shortens the identification time of the causative agent compared to conventional techniques. Given the presence of blood cells in the sample, protocols must start with lysis or filtration steps to remove human proteins. In-house protocols utilizing the lytic agent saponin have been rapidly and successfully implemented for bacterial identification; however, only a limited number of assays have been developed for yeast-positive blood cultures. For this purpose, three commercial assays are available: the VITEK MS blood culture kit (bioMérieux), the MBT-Sepsityper® kit (Bruker), and the rapid BACpro® II kit (Nittobo Medical Co., Japan). The MBT Sepsityper® assay facilitates bacteria and yeasts identification in under 30 min. This kit has been further optimized with the ‘Rapid Sepsityper® protocol’, which reduces the number of centrifugation steps, enabling the identification of bacteria and yeasts within 10 min. Concurrently, Bruker developed a specialized MBT-Sepsityper spectral analysis module to enhance identification performance.⁶⁰

However, an analysis of this identification strategy, involving the Rapid Sepsityper and MBT-Sepsityper module for positive blood cultures, showed that the standard Sepsityper protocol, which utilizes ethanol and formic acid extraction, was more effective for yeast identification than the Rapid Sepsityper. Notably, the implementation of the specific MBT-Sepsityper module increased yeast identification rates by 38%. Moreover, regardless of the protocol applied, the addition of formic acid consistently enhanced identification rates.⁴⁴

In conclusion, although the use of MALDI-TOF directly from positive blood bottles may help implement a rapid detection strategy for fungemia in the routine workflow of clinical microbiology laboratories, this methodology requires skilled laboratory personnel and involves several manual steps that may limit the ability to process positive blood cultures uninterruptedly on a 24/7 basis.

1.3.5

Molecular methodologies

There are various PCR-based methodologies that can be used to identify pathogenic fungi once a fungal isolate has been recovered, usually from an automated blood culture system.

FilmArray® (bioMérieux) is a fully automated platform that integrates sample preparation, multiplex PCR amplification, and detection/identification of pathogens in approximately 1 h. Recently developed for this platform, the Biofire® Blood Culture Identification 2 (BCID2) Panel can detect up to 33 pathogens causing sepsis, including 26 bacteria and 7 yeast species (C. albicans, C. glabrata, C. krusei, C. parapsilosis, C. tropicalis, C. auris and Cryptococcus gattii/Cryptococcus neoformans), and 10 antimicrobial resistance genes. The panel's identification of microorganisms showed higher sensitivity and specificity (>98% and >99%, respectively) compared to cultures.⁴⁵

The ePlex® system (GenMark Diagnostics, USA) is a fully random-access multiplex PCR platform developed for syndromic diagnosis. It includes all the essential steps for identifying pathogens isolated in positive sterile sample cultures. Based on the Gram stain result of a positive blood culture, three panels have been developed to detect 56 pathogens causing sepsis and 10 antimicrobial resistance genes in 90 min. The Gram-negative Panel detects 21 bacterial genera or species and 6 resistance genes, while the Gram-positive Panel identifies 20 genera or species and 4 resistance genes. The Fungal Pathogen Panel detects 15 fungal genera or species: C. albicans, C. auris, Candida dubliniensis, C. famata, C. glabrata, Candida guilliermondii, Candida kefyr, Candida lusitaniae, C. parapsilosis, C. tropicalis, C. krusei, C. neoformans sensu lato, C. gattii sensu lato, Fusarium, and Rhodotorula. For fungal pathogens, the ePlex® system have shown a global sensitivity ranging from 99.8% to 100% in blood cultures, with 100% specificity.¹⁵

The recently commercialized Molecular Mouse™ system (Alifax, Italy) is the world's first handheld platform for the Real Time PCR. Ready-to-use lab-on-chip cartridges come with all lyophilized reagents in each micro-well, allowing for up to 6 simultaneous multiplex reactions. Starting from a positive blood culture, the Sepsis Panel's five cartridges can identify 44 major clinical microorganisms and 20 antibiotic resistance genes in about 1 h. The cartridges incorporate 15 Gram-negative genera or species, 20 Gram-positive genera or species, and 9 Candida species, including C. albicans, C. auris C. dubliniensis, C. glabrata, C. guilliermondii, C. krusei, C. lusitaniae, C. parapsilosis, and C. tropicalis. Currently, there is limited data on the utility of this new platform in a clinical setting, and none include bloodstream infections caused by Candida. For bacteria, the new instrument showed consistent agreement with routine protocols in cases of monomicrobial blood cultures but did not fully align with results for polymicrobial samples.⁴⁶

The identification of culture-grown filamentous fungi by their morphological, macro- and microscopic characteristics can be time-consuming and labor-intensive, heavily relying on the expertise of clinical mycologists in individual laboratories. Even the most skilled experts may not accurately identify all clinically relevant molds using phenotypic methods alone, and must send the strains to a reference laboratory for definite identification. Many clinical mycology laboratories now rely on DNA sequencing as the gold standard for identifying molds. This method improves both the consistency and accuracy of fungal pathogen identification. However, it has limitations. Accurately distinguishing fungi at the species level remains challenging and often requires data from multiple gene targets. Additionally, the cost of testing, limited commercial availability, and issues with comparison databases hinder the widespread use of sequencing for precise identification.⁷⁷

1.4

Mycological histopathology

Histopathological identification of fungal elements, together with a typical tissue reaction, remains the gold standard for a proven diagnosis of IFI. However, histopathological examination alone cannot identify the causative species due to the morphological similarities between the tissue forms of many fungal genera.³⁸ In many cases of severe invasive mycoses, histopathology is necessary to determine the presence of the causative agent in the tissue, assess the extent of the infection, and evaluate the host's inflammatory response. Additionally, surgical resection of the affected tissue is usually required for the adequate treatment of IFI.²⁵

However, as mentioned above, quickness is critical to achieving a prompt diagnosis of IFI, and the practice of evaluating hematoxylin–eosin stains of tissues before deciding whether to use specific fungal stains often leads to fatal delays for patients. Then, all biopsied tissues from immunocompromised patients with suspected infection should be stained with fungal stains, such as periodic acid-Schiff (PAS) or Gomori's methenamine silver (GMS) stains, in parallel with standard stains. The PAS staining allows better visualization of the morphology of the tissue adjacent to the fungi (that are colored magenta), while with GMS stains fungal cell walls are brownish-black making them easily visible; however, interpretation can be challenging because this staining technique also stains many structures besides fungal cell walls.²⁵ A drawback of using the GMS stain is the masking of the innate color of fungal elements, preventing determining whether it is a hyaline or dematiaceous fungus (crucial for diagnosing a pigmented fungus mycosis).

Histologically, the only way to achieve an accurate in situ etiological diagnosis in formalin-fixed tissue sections is by means of immunohistochemistry combined with in situ DNA detection techniques. Unfortunately, many specific reagents for immunohistochemistry are not commercially available, as most are based on polyclonal antibodies that require heterologous adsorption with cross-reacting antigens to ensure specificity. However, monoclonal antibodies for immunohistochemistry are accessible for diagnosing the most significant causes of IFI, including Candida, Aspergillus, Fusarium, Scedosporium, Zygomycetes, and those fungi producing endemic mycoses.³⁸

Fluorescence in situ hybridization (FISH) is effective for the specific identification of several IFI-causing species. By this method, fluorescently labeled oligonucleotide probes are targeted to specific complementary nucleic acid sequences within fungi. FISH techniques typically use probes that target ribosomal RNA (rRNA). However, as rRNA is rapidly degraded when irreversible damage has occurred, it mainly identifies viable cells.

In the last years, various types of probes have been successfully employed for the detection of fungi in fixed tissue sections: oligonucleotide DNA, peptide nucleic acids (PNA), and LNA (a combination of DNA and locked nucleic acid-modified nucleotides). Of them, PNA and LNA nucleotides are preferred as they strongly hybridize with their complementary RNA and DNA nucleotides and establish temperature-stable hybrids.⁴⁹

During the last decade, numerous probes targeting the most significant causes of invasive mycoses have been evaluated, and their sequences published.³⁷ However, it is important to note that since fungi identified by molecular probes depend on specifically designed probe sequences, only the targeted organisms will be detected. Therefore, universal (pan-fungal) probes should be used initially for screening the tissue sections whether the cause of IFI is unknown.

2

Laboratory procedures for a probable IFI diagnosis

2.1

Serological techniques

As discussed above, prompt diagnosis of IFI is crucial to start an antifungal therapy; however, the current methods for proven diagnosis are often delayed or insufficiently sensitive. Currently, serological techniques have become essential in diagnosing fungal infections, particularly those caused by Candida and Aspergillus species, because they can detect specific antigens or antibodies in the host's blood or sterile fluids, such as β-(1,3)-d-glucan, mannan antigen and anti-mannan antibody, anti-germ tube antibodies, galactomannan antigen, and glucuroxylomannan antigen. These methods allow for non-invasive probable IFI diagnosis, but still have their limitations (Table 2).

Table 2.

Comparative summary of commercial serological and PCR-based techniques for the diagnosis of invasive mycosis.

Technique	Strengths	Weaknesses	Diagnostic utility	References
Beta-d-glucan	• Pan-fungal• Quantitative• Sensitivity 85–96%• Specificity 80–92%	• False-positives due to several clinical conditions and microorganisms• Not useful for Cryptococcus and Zygomycetes infections	• Mycological criterion for probable IFI• Screening and monitoring tool for Candida and Pneumocystis infections	^18,33,71,79
Galactomannan	• Reliable for Aspergillus and other mold infections• Useful in BAL fluid and serum• Quantitative• Sensitivity 80–90%• Specificity 85–95%	• False-positives due to host conditions, beta-lactam antibiotics, chemotherapy, and others	• Mycological criterion for probable IA• Also positive in IFI by Scedosporium, Fusarium, Geotrichum, Histoplasma and other molds• Suitable for early detection and monitoring response to treatment	^2,17,26
Mannan & Antimannan	• Combined use (Ag+Ab) can be positive prior blood culture• Sensitivity 50–80%• Specificity 70–80%	• Limited to Candida infections• Effective when combined with PCR or additional serological biomarkers	• Mycological criterion for probable invasive candidiasis• Valuable for diagnosing and monitoring	^21,33,36
Anti-germ tube antibodies	• Discriminate between colonization and infection• High negative predictive value	• Low sensitivity and specificity	• Combined with BDG detection improves the diagnosis of invasive candidiasis.	^53,55,72
Glucuroxylomannan	• High sensibility, specificity, and predictive values• Fast performance• Low cost	• Lower accuracy in localized pulmonary or cutaneous infections	• Very useful in disseminated and meningeal cryptococcosis• Useful in combination with BG for diagnosis of invasive Trichosporon and Magnusiomyces infections• Valuable for diagnosing and monitoring	^8,42,58
Lateral flow assays	• Rapid results (<20 min)• High specificity (90–95%)• Easy performance• Low cost	• Lower sensitivity when used independently (70–85%)• Weakly positive results difficult to interpret without a digital readout	• Point-of-care tests• Available for diagnosis of Aspergillus,Cryptococcus, and Histoplasma infections• Useful in resource-limited settings	^1,58,74,75
PCR:Aspergillus spp.Candida spp.P. jiroveciiZygomycetes	• Highly sensitive• Sensitivity 75–95%• Specificity 80–90%	• Risk of contamination• False-negative results• Inability to distinguish colonization from infection in non-sterile samples	• Diagnosis of invasive aspergillosis, invasive candidiasis, mucormycosis, PCP, particularly in high-risk groups.• Used in combination with other biomarkers for improved accuracy	^23,36,49,64

2.1.1

β-(1,3)-d-Glucan assay

The β-(1,3)-d-glucan (BDG) is a carbohydrate component present in the cell walls of most pathogenic fungi, including Candida, Aspergillus, Fusarium, and Pneumocystis jirovecii. However, BDG is not present in other clinically significant fungi, such as Cryptococcus, Blastomyces (yeast form) and species of the Zygomycetes group.

The BDG assay is a pan-fungal diagnostic test, meaning that a positive result does not specify the genus of the fungi responsible for the infection. Therefore, to obtain an etiological diagnosis, the BDG result should be interpreted in conjunction with other diagnostic methods, including mycological culture, histological examination of biopsy samples, and radiological imaging. However, BDG false positives can occur due to intestinal translocation, particularly in patients who have undergone hemodialysis, received immunoglobulins, albumin, or β-lactam antibiotic treatment, as well as in individuals with certain medical conditions, or those who have had contact with gauze or surgical sponges.²⁸

Assays developed to measure BDG typically utilize serum samples and rely on the activation of a clotting cascade present in Limulus amebocyte lysate (LAL) by the serine protease zymogen Factor G. The sensitivity of the technique increases when performed serially, two times a week, in patients at risk of IFI and a repeatedly negative result rules out invasive mycosis by Candida, Aspergillus or P. jirovecii. BDG detection is useful in order to start an antifungal treatment as soon as possible and can be used as a biomarker to monitor treatment response, as decreasing values correlate with clinical improvement, while increasing BDG concentration is associated with treatment failure in candidiasis and invasive aspergillosis.

The Fungitell® kinetic assay (Associates of Cape Cod Inc, USA) was the first and the only FDA-cleared and CE marked in vitro diagnostic screening test for IFI (including Candida, Aspergillus and Pneumocystis) which detects and quantifies BDG in serum and CSF, within 3 h, using a colorimetric method. BDG values ≥80 pg/mL are interpreted as a positive result, although a positive result does not imply the presence of disease and should be used in conjunction with other diagnostic tools to establish a diagnosis. BDG values ranging from 60 to 79 pg/mL are considered inconclusive, and additional serum samples are recommended to confirm or rule out a diagnosis. The sensitivity and specificity of Fungitell® for diagnosing IFI have been reported to be 79.1% and 87.7%, respectively.³⁴ Conversely, the design of the technique requires batch processing to optimize reagent usage, precluding the individual processing of clinical samples and hindering the prompt diagnostics and treatment of patients.

The Beta-Glucan Test® (FUJIFILM Wako Chemicals Europe GmbH, Germany) is another commercial in vitro diagnostic assay designed for the quantitative determination of BDG in serum or plasma, also based on the activation of a clotting cascade in LAL. Unlike Fungitell®, this assay employs a kinetic turbidimetric method in a single test format, with continuous sample loading, and provides results in 90 min. These technical features facilitate its use in real-life management of patients with suspected IFI. In a comparative evaluation of the two commercially available BDG tests, the diagnostic sensitivity and specificity of Beta-Glucan Test®, using a cut-off value of 7 pg/mL, was 80% and 97.3% for invasive aspergillosis, 98.7% and 97.3% for invasive candidiasis, and 94.1% and 97.3% for P. jirovecii pneumonia, respectively.¹⁸

In high-risk surgical patients with intra-abdominal candidiasis, where blood cultures are often negative, BDG could be an indispensable biomarker. BDG concentrations facilitate early detection of intra-abdominal candidiasis, up to 10 days earlier than traditional culture-based methods, with a sensitivity of 65% and a specificity of 78%, allowing for a more specific and rapid initiation of antifungal treatment. However, clinicians should note that BDG levels may show non-specific elevations in such patients undergoing abdominal surgery, sepsis, or advanced liver cirrhosis. This underlines the need for a careful interpretation of BDG levels, ensuring that decisions regarding antifungal therapy are made with a comprehensive understanding of the patient's overall clinical context.^67,72

Additionally, BDG plays an essential role in the diagnosis of Pneumocystis pneumonia, with sensitivity and specificity levels reaching 85–95%, respectively, depending on the cut-off value used. Consequently, the lack of BDG detection can be used to exclude the diagnosis of P. jirovecii infection. The combination of BDG with lactate dehydrogenase (LDH) enhances the diagnostic accuracy of the test.⁶⁹ Moreover, BDG detection directly in BAL samples for diagnosis P. jirovecii pneumonia seems to be useful using a higher cut-off (128 pg/mL).⁸⁰

For its diagnostic value, detection of BDG is a mycological criterion for probable invasive mycosis according to the EORTC/MSG and ECIL-3 definitions.^5,10 It is also recommended by ESCMID/ECMM, IDSA and SEIMC for the diagnosis of candidemia and invasive candidiasis.^7,21,53

2.1.2

Galactomannan antigen

Galactomannan (GM) is a polysaccharide antigen found in the outer cell wall layer of Aspergillus and several other fungi. It is released during tissue invasion and can be detected in serum, BAL, and other biological fluids of patients with invasive aspergillosis (IA). Usually, GM detection is performed by enzyme immunoassays (EIA), but in recent years simpler and faster techniques, such as lateral flow assay (LFA) and chemiluminescence, have become available.

The Platelia™ Aspergillus EIA (Bio-Rad) is a quantitative and well standardized assay, the amount of specimen required for the analysis is small (300 μl), can be performed in 4 h, and it is considered the gold-standard methodology for the detection of GM in serum and BAL samples. However, the design of the technique prevents the processing of individual clinical samples separately, requiring batch processing to optimize reagents. This limitation hinders rapid diagnosis and real-time patient management.

For years, different diagnostic cut-off points have been proposed for probable IA diagnosis, depending on the clinical sample: serum (index 0.5), urine and BAL (index 1), CSF, pericardial or pleural fluid (index 0.5). However, the latest update of the EORCT/MSG Consensus Definitions proposes any of the following cut-off indexes for probable IA diagnosis: (i) single serum or plasma ≥1, (ii) BAL fluid ≥1, (iii) single serum or plasma ≥0.7, and BAL fluid ≥0.8.²⁶ False-positive results can be observed by intestinal translocation of GM in patients with chemotherapy, as well as in patients with graft-versus-host disease, or receiving beta-lactam antibiotic therapy, immunoglobulins or immunosuppressants. However, because other fungi also release GM, its detection in infections by Lomentospora/Scedosporium, Fusarium, Geotrichum, Histoplasma, Paecilomyces, Penicillium and Rhodotorula can facilitate the diagnosis of these severe invasive mycoses when used in combination with other diagnostic techniques.

In clinical settings, GM testing from serum and BAL has a sensitivity of 80–90%, with the specificity being particularly high when used in conjunction with PCR methods, thereby significantly improving the diagnostic accuracy for IA.^17,55 For screening, the reliability of GM detection is primarily influenced by the prevalence of IA and the impact of antifungal prophylaxis; thus, GM screening is unlikely to be beneficial and cost-effective when the probability of IA is low prior to testing, as in hematological patients receiving antifungal prophylaxis with mold-active agents. Therefore, GM detection should be reserved for IA diagnosis in high-risk populations.⁶⁶

Despite its utility, GM testing faces limitations, especially in early diagnosis, as the highest concentrations of GM are often only released during the terminal phases of invasive disease after angioinvasion has occurred. Furthermore, the sensitivity of the test varies between species, showing higher sensitivity in patients with non-Aspergillusfumigatus aspergillosis than in those with AI caused by A. fumigatus.⁶

Studies have shown that while serum GM can predict outcomes and assess response to antifungal therapy, testing BAL fluid and apply a higher cut-off index value may provide greater sensitivity and specificity for early pulmonary aspergillosis diagnosis. Generally, the optimal sensitivity and specificity of GM tests are achieved when combined with other diagnostic tests.^37,41

Recent studies have underscored the utility of GM test for detecting IA using proximal airway samples, such as induced sputum and tracheal aspirate. Findings revealed that GM detection in proximal airway samples not only offers a sensitivity of 93.1% and a specificity of 78.7%-93%, but also, in many cases, surpasses the diagnostic performance of serum and BAL.^19,64 This strategy can be especially beneficial in settings where bronchoscopy may be precluded due to patient conditions (COVID-19 infection) or resource limitations.

For its diagnostic value, GM detection in serum or BAL samples by EIA is a mycological criterion for probable invasive mycosis according to the EORTC/MSG and ECIL-3 definitions.^5,10

A major drawback of the EIA methodology in detecting GM is the relatively long turnaround times, although the time needed to perform the GM assay is only 3 h. Due to the practice of batching samples to enhance cost-effectiveness, many clinical microbiology laboratories typically conduct the assay only two or three times per week. Furthermore, some laboratories may not perform the assay on-site and must send samples to a reference laboratory, which further extends the time to obtain results. This diagnostic delay reduces the potential benefits of enhanced survival associated to the early diagnosis and treatment of IA.

To overcome the long turnaround time, several additional assays are available for the detection of Aspergillus antigens. These new assays significantly reduce the diagnosis time (<60 min), allowing testing individual samples. The Ag VIRCLIA® Monotest (Vircell SL, Spain) is an indirect chemiluminescent immunoassay (CLIA) in monotest format, recently developed for the qualitative detection of Aspergillus GM antigen in human serum, plasma and LBA samples. In a retrospective multicenter evaluation of EIA Platelia™ Aspergillus vs. Ag VIRCLIA® in 141 BAL fluid samples from hematological patients, the sensitivity and specificity of the two tests were quite comparable, and the overall qualitative agreement between EIA and CLIA results was 81–89%. The correlation of CLIA and EIA values was strong at 0.72 (95% confidence interval, 0.63–0.80), and CLIA has similar performance, compared to the gold-standard EIA, with the benefits of faster turnaround because batching is not required.¹⁶ In another prospective and comparative study of both techniques in 320 sera and 215 lower respiratory tract samples from patients with different clinical conditions, the overall sensitivity and specificity of CLIA assay for the diagnosis of proven/probable IA were 100% and 65%, respectively, and for Platelia EIA, 91.7% and 89.4%, respectively. The correlation between index values by both assays was strong for serum/BAL, and moderate for bronchial and tracheal aspirates.²

2.1.3

Mannan antigen and anti-mannan antibody

Mannan is a component of the Candida cell wall used as a diagnostic biomarker of invasive candidiasis for many years. Unfortunately, this antigen is rapidly metabolized and cleared from circulation and serial determinations are necessary; anti-mannan antibodies are detected many times after mannan disappearance, though. Thus, the combined and serial detection of mannan and anti-mannan antibodies is recommended for the diagnosis of candidemia in adults and neonates, as it can be positive prior to blood cultures.⁶⁷

Therefore, detection of mannan antigen in serum can be incorporated into the diagnostic tools for invasive candidiasis, particularly through assays that detect its presence in conjunction with anti-mannan antibodies. These mannan assays have a sensitivity range of 50–80%, with a notable improvement in sensitivity when combined with anti-mannan antibodies, resulting in a pooled sensitivity of about 70% and a specificity of 89.4%.³⁴

Commercially available assays like Platelia™ Candida Ag Plus EIA (Bio-Rad) and the CandTec latex agglutination test (Ramco Laboratories, USA) are tailored to detect mannan or mannoproteins. These tests are sensitive to the presence of Candida species and other commensal yeasts, but heavy colonization might lead to false positives. However, combining mannan antigen detection with anti-mannan antibody tests, like the Platelia™ Candida Ab Plus (Bio-Rad), generally yields the best diagnostic performance, particularly in immunocompromised individuals.³⁷

Interestingly, mannan and anti-mannan detection may be useful as an adjunct to blood culture for the early diagnosis of invasive candidiasis in neonates and young children. However, it is less useful in the diagnosis of invasive candidiasis caused by C. parapsilosis and other non-C.albicansCandida species.

For its diagnostic value, combined mannan and anti-mannan detection in serum is a mycological criterion of probable invasive candidiasis according to EORTC/MSG definitions,¹⁰ and it is recommended by ESCMID/ECMM, IDSA and SEIMC for the diagnosis of candidemia and invasive candidiasis.^7,21,53

2.1.4

Anti-germ tube antibodies

For decades, antigens expressed in the mycelial phase of Candida have been particularly studied in the hope that they could be specific for the invasive candidiasis diagnosis, as they are not expressed during the colonization state. An indirect immunofluorescence technique based on the detection of specific IgG antibodies against antigens located on the C. albicans germ-tubes (CAGTA) was developed years ago for the diagnosis and therapeutic monitoring of invasive candidiasis. This technique, known as Invasive Candidiasis CAGTA IFA IgG® (Vircell SL, Spain), enables the detection of specific CAGTA in serum or plasma within two hours using immunofluorescence. It may discriminate between the superficial colonization and infection with a high negative predictive value.

The sensitivity of this technique reaches 100% when the infection is a deep-seated candidiasis due to C. albicans, but for invasive candidiasis by other Candida species the sensitivity is lower.⁵⁶ The low sensitivity in these circumstances can be compensated if the technique is used in combination with other serologic tests, such the BDG assay.

To overcome the difficulties of immunofluorescence techniques (specific microscopy, trained personnel, long turnaround time, etc.) a new automated version of this technique, based on indirect chemiluminescent immunoassay (CLIA) in a single-test format, has recently been commercialized. This new product (VirClia® IgG MONOTEST, Vircell SL) includes ready-to-use reagents that provides rapid (45 min) and objective results.

In the context of invasive candidiasis, the CAGTA IFA IgG assay has shown sensitivity values between 51.7% and 69.2%, along with specificity values ranging from 75% to 80.3%. In comparison, the VirCLia® assay exhibited sensitivity and specificity values of 76.9% and 75.8%, respectively.^54,56 In a meta-analysis conducted with data from seven distinct studies utilizing the CAGTA IFA IgG assay, the results revealed a pooled sensitivity of 66% and a specificity of 76%.⁷³ These findings suggest that this assay possesses moderate accuracy for the diagnosis of invasive candidiasis.

2.1.5

Glucuroxylomannan

Glucuronoxylomannan (GXM) constitutes the main capsular polysaccharide of C. neoformans, and its detection in CSF or serum is the paradigm of a fast, sensitive and specific diagnostic technique with high diagnostic and prognostic values. For years, it has been widely accepted as a screening tool for the diagnosis of cryptococcal meningitis, due to its high sensitivity (93–100%). GXM detection is more accurate in disseminated and meningeal cryptococcosis, and lower in localized pulmonary or cutaneous infections. Any of the commercially available EIA (Premier® EIA Assay, Meridian Diagnostic, USA) or latex agglutination (Crypto-LA® test, Wampole Labs, USA) methods can be used to perform the test, using serum and CSF. Serial monitoring of GXM titer in CSF may correlate with clinical response to treatment.⁸

In the presence of the rheumatoid factor, Capnocytophaga canimorsus or Stomatococcus mucilaginosus bacteremias, and during invasive Trichosporon and Magnusiomyces capitatus infections, false positive results have been reported.⁸ Thus, some authors consider that the combined detection of GXM and BG could very useful for the early diagnosis of invasive trichosporonosis or M. capitatus infections.⁴³

2.1.6

Point-of-care tests in invasive fungal infections

Point-of-care tests (POCT) are diagnostic tools defined by the College of American Pathologists as tests conducted in close proximity to the patient, thereby directly impacting clinical decision-making and patient care. These tests range from simple, minimal-equipment assessments, such as home pregnancy kits, COVID-19 at-home tests, and blood glucose monitors, to more sophisticated ‘lab-on-a-chip’ (LOC) technologies. LOC tests operate on untreated samples like blood or urine right at the patient's bedside, combining convenience with advanced diagnostic techniques.

Their capacity to provide immediate results makes POCT valuable in both well-equipped clinics and resource-limited settings. These diagnostic tests can be classified into two categories: those that are simple to perform and require minimal equipment, making them suitable for global health application, and those that are more complex but offer rapid results and portability. The World Health Organization (WHO) emphasises that POCT should be affordable, sensitive, specific, robust, user-friendly, and portable, enhancing their utility in diverse environments.³⁷

2.1.6.1

Lateral-flow assays

Due to its easy performance, reliability, and rapid results (<45 min), lateral-flow assays (LFA) are the most commonly used POCT for the diagnosis of probable IFI. To date, these devices are able to detect by immunochromatography antigens from Aspergillus, Histoplasma, and Cryptococcus in serum, respiratory or urine samples.

AspLFD® and IMMY sōna Aspergillus galactomannan LFA (IMMY Diagnostics, USA) are the currently LFA commercially available tests for the diagnosis of probable IA. Both techniques exhibit high sensitivity and specificity in serum (97% and 98%, respectively),⁷⁵ and BAL from hematology patients (91% and 92%, respectively).⁴⁸ The implementation of digital readout significantly enhanced the diagnostic performance of both tests, particularly for weakly positive results. A recent comparative study of the IMMY sōna Aspergillus GM LFA and the GM Ag VIRCLIA® Monotest displayed comparable diagnostic performance, with a higher sensitivity for the GM-LFA if two consecutive positive samples were required for positivity.⁴⁰ Given its clinical utility, ease of use and low cost, in its latest Report of In Vitro Diagnostic (2021) the WHO recommends the use of LFA in BAL samples to facilitate the diagnosis of IA in immunocompromised patients.⁷⁶

LFA are also available for the detection and diagnosis of other fungal infections, including those caused by Cryptococcus and Histoplasma. For cryptococcosis, a new, rapid (<10 min), commercial method using a lateral flow device (CrAg® LFA, IMMY Diagnostics) has become available for detecting GXM. This method can be perform with serum, CSF or urine samples, and improve the early diagnosis of cryptococcal meningitis. This assay has high sensitivity and specificity in CSF, and it is also effective in detecting GXM in blood prior to the onset of meningitis symptoms. The WHO and numerous national HIV guidelines recommend blood CrAg® LFA screening in HIV-infected persons before the development of a fulminant meningitis, and preemptive treatment for those with a CrAg®-positive result.⁵⁹

The Histoplasma Urine Antigen LFA developed by MiraVista Diagnostics (Indianapolis, USA) represents a novel, CE-marked, rapid, and non-invasive method for detecting Histoplasma capsulatum antigens in urine samples within 30 min. It is designed for use in low-complexity clinical settings and requires no additional equipment. This qualitative assay exhibits a clinical sensitivity of 93.18% and a specificity of 96.97% when compared to culture-proven cases. The diagnostic accuracy of the MVista LFA is notably higher in patients with confirmed histoplasmosis (93.2%) when compared to those with probable disease (78.6%).¹ However, the MVista LFA shows a significant cross-reactivity of 30–85% with other endemic fungal infections, including blastomycosis, paracoccidioidomycosis, and coccidioidomycosis. Despite this limitation, this innovative methodology offers developing countries an easy and effective approach to diagnosing histoplasmosis.⁴⁷

2.2

PCR-based techniques

Various polymerase chain reaction (PCR) techniques, including conventional PCR, nested PCR, real-time PCR (RT-PCR), PCR-ELISA, multiplex PCR, and direct DNA sequencing, have significantly enhanced fungal diagnostics by offering rapid, sensitive, and species-specific identification. Although significant advances have been made in the development of commercial PCR techniques for the detection of fungal DNA, including Aspergillus, Candida, Pneumocystis, and zygomycetes detection, challenges as sensitivity, reproducibility, and the generation of false-negative results still persist (Table 2).³⁷

2.2.1

PCR for invasive aspergillosis

PCR techniques have revolutionized IA diagnostics, particularly when applied to BAL fluid, achieving sensitivities of 75–95% and specificities of 80–100%. Particularly notable are the commercialized tests MycAssay Aspergillus® (Myconostica, UK), AsperGenius® (PathoNostics BV, The Netherlands) and MycoGenie® (Ademtech, France). AsperGenius® and MycoGenie® not only detect Aspergillus DNA, but also resistance markers as TR34/L98H and TR46/Y121F/T289A, frequently associated with the environmental azole resistance. The nested PCR techniques, involving two rounds of amplification using two sets of primers, provide higher specificity and can detect extremely low levels of fungal DNA. Nonetheless, they are extremely susceptible to contamination and may not be suitable for samples with potential environmental contamination, such as specimens from endoscopic sinus surgery.³⁷

Combining PCR with other diagnostics methods, like GM detection, enhances overall accuracy.³¹ Other samples can be analyzed, such as fresh tissue or formalin-fixed, paraffin embedded tissue. However, cross reactivities have been reported with other molds, including species of Penicillium and Fusarium, and Rhizopus oryzae.⁴¹

2.2.2

PCR for invasive candidiasis

Candida DNA detection in blood samples by PCR-based methods has shown a higher sensitivity than other diagnostic techniques. The T2Candida® Panel (T2 Biosystems, USA) is the first and only FDA- and EMA-cleared diagnostic test for the detection of sepsis-causing fungal pathogens, directly in whole blood (4 mL). This panel gives positive or negative results for C. albicans/C. tropicalis, C. glabrata/C. krusei, and C. parapsilosis within 3–5 h, without requiring a positive blood culture for identification. T2Candida® sensitivity and specificity for candidemia were 91% and 98%, respectively, substantially superior to the values obtained with traditional culture methods (50–60% of sensitivity). Furthermore, this panel may be useful in monitoring the clearance of candidemia in patients undergoing antifungal therapy, timing for de-escalation and the optimal duration of treatment.⁵⁰

As already stated above, C. auris is an emerging pathogen known for its multi-drug resistance. Consequently, infections caused by this species necessitate rapid and reliable detection methods to both facilitate effective medical treatment and control hospital outbreaks. Traditional identification techniques are often susceptible to errors, which can result in misidentifications. In contrast, PCR-based assays offer dependable results with minimal turnaround times. Some of them are already commercially available, including AurisID (IMMY Diagnostics) and Fungiplex C. auris RUO Real-Time PCR (Bruker). They enable the detection of C. auris DNA in blood, other clinical samples, and hospital environments.⁶⁵

Additionally, the utility of multiplex PCR, which enables the simultaneous amplification of multiple targets, has been confirmed to be highly effective in diagnosing conditions like candidemia or deep-stated candidiasis, with excellent sensitivity and specificity.^29,47

2.2.3

PCR for Pneumocystis pneumonia

As P. jirovecii is a non-culturable fungus, the microscopic examination of lower respiratory tract or lung tissue samples, using different stains, including direct immunofluorescence, has been the traditional method for diagnosing Pneumocystis pneumonia (PCP). To overcome these diagnostic limitations, PCR methodology has been validated and is commercially available for diagnosing PCP from induced sputum, BAL, oropharyngeal lavage, serum, and blood. Among these, BAL is the sample of choice, with higher sensitivity and specificity rates. Although PCR offers greater sensitivity than microscopy techniques, its clinical utility may be diminished by false positives resulting from the detection of airway colonization or contamination during molecular processing.⁵⁸

For many years, in-house PCR techniques have been employed for the diagnosis of PCP in clinical laboratories, despite their lack of standardization and the variability in results. However, several commercially available kits have emerged, including PneumoGenius® (PathoNostics BV, The Netherlands), PneumID® (IMMY Diagnostics), FTD® P. jirovecii (Siemens Healthineers, Germany), LightMix® Modular P. jiroveci (Roche, Switzerland), MycoGENIE® P. jirovecii (Ademtech, France), and RealCycler PJIR® Kit (Progenie Molecular SLU, Spain).

PCR assays for PCP diagnosis, when used in conjunction with additional biomarkers such as lactate dehydrogenase (LDH) or BDG, have shown sensitivity rates between 85% and 97%, with specificity reaching up to 94%. PCR continues to be regarded as the gold standard for the diagnosis of PCP in various clinical contexts, especially among immunocompromised patients.^69,74

2.2.4

PCR for mucormycosis

Diagnosing mucormycosis is challenging and traditionally relies on recognizing host risk factors, clinical manifestations, mycological cultures, and both direct and histopathological visualization of hyphae in tissues in clinical samples. Histopathology demands significant expertise and is often hindered by hyphal fragmentation, which reduces diagnostic sensitivity. Furthermore, this infection is difficult to detect using conventional culture techniques due to the damage sustained during tissue homogenization prior to plating.

Contrasting with the conventional methods, molecular techniques have shown promise in diagnosing mucormycosis. Multiplex PCR assays targeting specific Zygomycetes species can identify fungal DNA in body fluids such as blood, BAL, and tissue samples. These techniques have demonstrated high sensitivity and specificity, offering an early diagnostic tool for this invasive infection. However, its utility is limited by a lack of standardization and variation in diagnostic targets across different species.⁴²

Currently, there are at least three commercial qPCR developed for mucormycosis diagnosis: MucorGenius® (PathoNostics BV), MycoGenie® Aspergillus-Mucorales species (Ademtech), and Fungiplex® (Bruker). These tests have shown overall sensitivities ranging from 75 to 90%, and specificities exceeding 90%.²³

Other panfungal PCR assays targeting the ITS1 region of the rDNA gene, followed by DNA sequencing, have shown sensitivities of up 97% in fresh tissue samples, and approximately 68% in formalin-fixed, paraffin-embedded specimens. These assays can detect any fungal DNA sequence, including uncultured or rare fungi. However, they require amplicon sequencing for species identification, which extends the turnaround time. There is also a risk of false positives due to contamination or colonization of non-sterile sites such as the airways or sinuses.²³

Overall, PCR techniques continue to be invaluable in fungal diagnostics, providing detailed insights into fungal presence in clinical samples. Their integration into routine diagnostic workflows, combined with improvements in standardization and contamination control, promises to enhance the reliability and effectiveness of fungal disease management.

3

Emerging and future techniques

New diagnostic methods, such as metagenomic sequencing and MALDI-TOF (mass spectrometry), are technologies that promise to enhance the identification of fungal pathogens and provide rapid results. Metagenomics sequencing, in particular, offers the ability to detect mixed infections and rare species.²²

The loop-mediated isothermal amplification (LAMP) is a molecular diagnostic method that is particularly advantageous in resource-limited settings, as it allows the rapid and efficient identification of fungal pathogens.³⁹ The technique operates under isothermal conditions and utilizes a set of four to six primers to amplify, with high specificity and efficiency, DNA of common and rare fungal species.⁴ This enables the detection within an hour without the necessity for sophisticated laboratory equipment, rendering LAMP especially valuable in situations where traditional diagnostic tools, like culture and PCR, are impractical due to cost, complexity, and time constraints.⁵²

Omics techniques, such as metabolomics, which involve measuring volatile organic compounds (VOCs) as biomarkers for diagnosing IFI, enable non-invasive breath tests to enhance diagnostic accuracy and promptness in pulmonary infections. These markers exploit changes in VOCs produced by both the host and the pathogen during infection and have proven effective in diagnosing infections caused by pathogens such as Aspergillus, Candida, and Coccidioides, achieving sensitivities up to 94% and specificities 93%.¹⁴ However, these techniques are still in early stages of development and have not yet been widely implemented in clinical practice.

In conclusion, serological and PCR-based techniques have markedly advanced fungal infections detection, offering earlier diagnosis and more precise species identification compared to traditional culture methods. However, both methodologies have their limitations, including false positives and lack of standardization. Future developments in molecular diagnostics, such as next-generation sequencing, hold the potential to further improve diagnostic accuracy and reduce time to diagnosis. A combined diagnostic approach, using both serological and PCR techniques, is likely the most effective strategy for timely diagnosis and treatment.

Funding

The publication of this article has been funded by Pfizer. Pfizer has neither taken part, nor intervened in the content of this article.

Conflict of interest

The authors declare that they have no conflict of interest.