Superiority of a Novel Mp1p Antigen Detection Enzyme Immunoassay Compared to Standard BACTEC Blood Culture in the Diagnosis of Talaromycosis.

BACKGROUND: Talaromycosis is an invasive mycosis endemic in Southeast Asia and causes substantial morbidity and mortality in individuals with advanced human immunodeficiency virus (HIV) disease. Current diagnosis relies on isolating Talaromyces marneffei in cultures, which takes up to 14 days and is detectable only during late-stage infection, leading to high mortality. METHODS: In this retrospective case-control study, we assessed the accuracy of a novel Mp1p antigen-detecting enzyme immunoassay (EIA) in stored plasma samples of 372 patients who had culture-proven talaromycosis from blood or sterile body fluids (reference standard) and 517 individuals without talaromycosis (338 healthy volunteers; 179 with other infections). All participants were recruited between 2011 and 2017 in Vietnam. RESULTS: Of cases and controls, 66.1% and 75.4%, respectively, were male; the median age was 33 and 37, respectively. All cases were HIV infected; median CD4 count was 10 cells/µL. At an optical density cutoff of 0.5, the specificity was 98.1% (95% CI, 96.3%-99.0%); the sensitivity was superior to blood culture (86.3% [95% CI, 82.3%-89.5%] vs 72.8% [95% CI, 68.0%-77.2%]) (P < .001, McNemar test). The time to diagnosis was 6 hours vs 6.6 ± 3.0 days for blood culture. Paired plasma and urine testing in the same patients (n = 269) significantly increased sensitivity compared to testing plasma alone or testing urine alone (P < .001 and P = .02, respectively, McNemar test). CONCLUSIONS: The Mp1p EIA is highly specific and is superior in sensitivity and time to diagnosis compared to blood culture for the diagnosis of talaromycosis. Paired plasma and urine testing further increases sensitivity, introducing a new tool for rapid diagnosis, enabling early treatment and potentially reducing mortality.

Comparative genomics of Cryptococcus neoformans var. grubii associated with meningitis in HIV infected and uninfected patients in Vietnam.

The vast burden of cryptococcal meningitis occurs in immunosuppressed patients, driven by HIV, and is caused by Cryptococcus neoformans var. grubii. We previously reported cryptococcal meningitis in Vietnam arising atypically in HIV uninfected, apparently immunocompetent patients, caused by a single amplified fragment length polymorphism (AFLP) cluster of C. neoformans var. grubii (VNIγ). This variant was less common in HIV infected individuals; it remains unclear why this lineage is associated with apparently immunocompetent patients. To study this host tropism we aimed to further our understanding of clinical phenotype and genomic variation within Vietnamese C. neoformans var. grubii. After performing MLST on C. neoformans clinical isolates we identified 14 sequence types (STs); ST5 correlated with the VNIγ cluster. We next compared clinical phenotype by lineage and found HIV infected patients with cryptococcal meningitis caused by ST5 organisms were significantly more likely to have lymphadenopathy (11% vs. 4%, p = 0.05 Fisher's exact test) and higher blood lymphocyte count (median 0.76 versus 0.55 X109 cells/L, p = 0.001, Kruskal-Wallis test). Furthermore, survivors of ST5 infections had evidence of worse disability outcomes at 70 days (72.7% (40/55) in ST5 infections versus 57.1% (52/91) non-ST5 infections (OR 2.11, 95%CI 1.01 to 4.41), p = 0.046). To further investigate the relationship between strain and disease phenotype we performed genome sequencing on eight Vietnamese C. neoformans var. grubii. Eight genome assemblies exhibited >99% nucleotide sequence identity and we identified 165 kbp of lineage specific to Vietnamese isolates. ST5 genomes harbored several strain specific regions, incorporating 19 annotated coding sequences and eight hypothetical proteins. These regions included a phenolic acid decarboxylase, a DEAD-box ATP-dependent RNA helicase 26, oxoprolinases, a taurine catabolism dioxygenase, a zinc finger protein, membrane transport proteins and various drug transporters. Our work outlines the complexity of genomic pathogenicity in cryptococcal infections and identifies a number of gene candidates that may aid the disaggregation of the pathways associated with the pathogenesis of Cryptococcus neoformans var. grubii.