Diagnostic methods for detection and characterisation of dimorphic fungi causing invasive disease in Africa: development and evaluation
Maphanga, Tsidiso Gugu
MetadataShow full item record
Emergomyces, Histoplasma, Blastomyces and Sporothrix species complex are thermally dimorphic fungi that have emerged as the causative agents of invasive fungal infections among immunocompetent and immunocompromised patients for the past 12 years. Accurate diagnoses of the mycoses caused by these fungi are challenging, not only because as there is a lack of clinical awareness globally, but also because these thermally dimorphic fungi are difficult to identify in the laboratory by conventional methods (i.e. culture, histology). In addition, cultured isolates require specialised biosafety facilities, which pose a tremendous challenge to resource-limited, diagnostic laboratories. As a result, estimates of the prevalence and burden of disease are poor. In addition, the lack of standardised antifungal susceptibility testing methods for these pathogens makes it difficult to determine the minimum inhibitory concentration (MIC), which may jeopardise patient treatment. These challenges indicate the need for alternative laboratory methods for identification. Furthermore, molecular methods are required for the correct identification of these pathogens in clinical settings, so that efficient antifungal agents may be used appropriately. We conducted several studies to address these challenges. In the first study, 50 Emergomyces africanus isolates from AIDS patients presenting with disseminated emergomycosis, from both public and private laboratories in South Africa during a nine-year period (2008 to 2016), were described and analysed. The Clinical Laboratory Standards Institute guideline (M27-A4 for yeast phase and M38-A2 for mould phase) was used to determine the MICs for the yeast and mould phase. Amphotericin B, itraconazole, voriconazole and posaconazole were more potent in vitro, while fluconazole, echinocandins and flucytosine were less potent agents. In the second study, we tested and analysed 212 urine samples from patients with suspected invasive fungal disease submitted to the mycology reference laboratory from August 2014 to December 2018 by public and private sector laboratories in South Africa. Corresponding fungal culture and histopathology results from several electronic laboratory information systems were obtained. Thirty-seven cultured fungal isolates were sent in parallel with the urine specimen for species-level identification (confirmed by phenotypic and molecular identification methods). The IMMY monoclonal Histoplasma capsulatum galactomannan enzyme immunoassay (EIA) had a sensitivity and specificity of respectively 88% and 72% when compared to culture-confirmed histoplasmosis. In contrast, the EIA showed a sensitivity and specificity of 83% and 93% when compared to histologically confirmed histoplasmosis/ emergomycosis. In the third study, we re-evaluated 20 cases of blastomycosis diagnosed among South African patients between 1967 and 2014. All isolates, initially identified as Blastomyces dermatitidis by histopathological and/or culture-based methods, were characterised phenotypically, genotypically and by whole genome sequencing. Morphological characteristics and phylogenetic analyses of multilocus sequence typing and whole genome sequences revealed two groups, both of which were closely related to but distinct from B. dermatitidis, and other related species. One group corresponded to the recently described Blastomyces percursus and the other group was described as Blastomyces emzantsi sp. nov. Whole genome sequencing confirmed distinct species identities and the absence of a full ortholog of the BAD-1 gene. Clinically, extrapulmonary disease was more common than lung involvement. In the fourth study, 156 dimorphic isolates from immunocompromised patients presenting with systemic endemic mycoses, from both public and private laboratories in South Africa during an 11-year period (2008 to 2019), were sent to the mycology reference laboratory for species- level identification. All isolates were initially identified by deoxyribonucleic acid (DNA) sequencing of the internal transcribed spacer (ITS) gene and those which failed sequencing were identified by sequencing of the large subunit (LSU) gene. The identity of the Sporothrix isolates was confirmed by sequencing the calmodulin gene. Only 63% of the dimorphic pathogens were correctly identified to genus level by culture-based methods at the diagnostic laboratories when compared to DNA sequencing methods. At the reference laboratory, culturebased methods correctly identified 95% of the dimorphic isolates. ITS sequencing correctly identified Emergomyces africanus, Histoplasma capsulatum, Blastomyces percursus, Emergomyces pasteurianus and Talaromyces marneffei. Calmodulin sequencing confirmed Sporothrix species and these clinical isolates belonged to Sporothrix schenckii sensu stricto. There was an increase in the number of cases of emergomycosis, sporotrichosis, histoplasmosis, blastomycosis and talaromycosis above what was previously reported. Skin lesions and pulmonary involvement were common. In conclusion, the in vitro susceptibility data for E. africanus supported the management of disseminated emergomycosis with amphotericin B, followed by itraconazole, voriconazole, or posaconazole. Fluconazole was a relatively less potent agent. The Histoplasma EIA provided a rapid alternative for diagnosis of histoplasmosis in an endemic setting. However, it should be noted that it may cross-react with patients infected with Emergomyces africanus. We observed that South African blastomycosis cases were caused by species that are distinct from B. dermatitidis, B. gilchristii and B. parvus, which might account for clinical differences between sub-Saharan and North American blastomycosis. Lastly, DNA sequencing was a valuable tool for accurate identification of dimorphic fungi to species level when compared to culture in the current setting, which could facilitate early diagnosis and initiation of proper treatment. However, in settings where DNA sequencing is not available, training of laboratory staff is necessary for the improvement of diagnosis of these pathogens in South Africa.